Megatest

#  Copyright 2006-2017, Matthew Welland.
# 
# This file is part of Megatest.
# 
#     Megatest is free software: you can redistribute it and/or modify
#     it under the terms of the GNU General Public License as published by
#     the Free Software Foundation, either version 3 of the License, or
#     (at your option) any later version.
# 
#     Megatest is distributed in the hope that it will be useful,
#     but WITHOUT ANY WARRANTY; without even the implied warranty of
#     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#     GNU General Public License for more details.
# 
#     You should have received a copy of the GNU General Public License
#     along with Megatest.  If not, see <http://www.gnu.org/licenses/>.

NOTE: This file gets copied occasionally into the wiki as "Roadmap DONE".
      Do not make changes in the wiki, they will be lost!

DONE
====

WW14
. Streamline compilation - DONE, all non-official egg modules are now bundled.

WW15
. syscheck; touch file in home, tmp, runs, links and start xterm [DONE]

WW16
. archiving improvements/extentions [DONE]
.. -get-data, -put-data [DONE]
.. use MT_ vars if defined and no switch present [DONE]
.. fix archive "first run" bug [DONE]
.. areas path1 path2 ... -> search path for archives [NOT NEEDED - use -start-dir]
.. -propagate -> move archive data forward when it is found in older bundles [NOT NEEDED - simply repost the data]

# make install CSCOPTS='-accumulate-profile -profile-name $(PWD)/profile-ww$(shell date +%V.%u)'
# rm <files>.o ; make install CSCOPTS='-profile' ; ... ;  chicken-profile | less
SHELL=/bin/bash
PREFIX=$(PWD)
CSCOPTS=
INSTALL=install
SRCFILES = common.scm items.scm launch.scm \
SRCFILES = common.scm items.scm launch.scm ods.scm runconfig.scm	\
   ods.scm runconfig.scm server.scm configf.scm \
   db.scm keys.scm margs.scm megatest-version.scm \
   process.scm runs.scm tasks.scm tests.scm genexample.scm \
   http-transport.scm filedb.scm tdb.scm \
           server.scm configf.scm db.scm keys.scm margs.scm		\
           process.scm runs.scm tasks.scm tests.scm genexample.scm	\
           http-transport.scm filedb.scm tdb.scm client.scm mt.scm	\
   client.scm mt.scm \
   ezsteps.scm lock-queue.scm sdb.scm \
   rmt.scm api.scm subrun.scm \
   portlogger.scm archive.scm env.scm diff-report.scm cgisetup/models/pgdb.scm 
           ezsteps.scm lock-queue.scm sdb.scm rmt.scm api.scm		\
           subrun.scm portlogger.scm archive.scm env.scm		\
           diff-report.scm cgisetup/models/pgdb.scm

# module source files
MSRCFILES = 
MSRCFILES = ftail.scm rmtmod.scm commonmod.scm


# ftail.scm rmtmod.scm commonmod.scm removed
# MSRCFILES = ducttape-lib.scm pkts.scm stml2.scm cookie.scm mutils.scm	\
#             mtargs.scm commonmod.scm dbmod.scm adjutant.scm ulex.scm	\
# Eggs to install (straightforward ones)
EGGS=matchable readline apropos base64 regex-literals format regex-case test coops trace csv \
dot-locking posix-utils posix-extras directory-utils hostinfo tcp-server rpc csv-xml fmt \
json md5 awful http-client spiffy uri-common intarweb spiffy-request-vars \
spiffy-directory-listing ssax sxml-serializer sxml-modifications iup canvas-draw sqlite3
#             rmtmod.scm apimod.scm

GUISRCF = dashboard-context-menu.scm dashboard-tests.scm		\
GUISRCF  = dashboard-context-menu.scm dashboard-tests.scm dashboard-guimonitor.scm gutils.scm dcommon.scm tree.scm vg.scm
          dashboard-guimonitor.scm gutils.scm dcommon.scm tree.scm	\
          vg.scm

OFILES   = $(SRCFILES:%.scm=%.o)
GOFILES  = $(GUISRCF:%.scm=%.o)

MOFILES = $(addprefix mofiles/,$(MSRCFILES:%.scm=%.o))
# compiled import files
MOIMPFILES = $(MSRCFILES:%.scm=%.import.o)

%.import.o : %.import.scm
	csc $(CSCOPTS) -unit $*.import -c $*.import.scm -o $*.import.o

# I'm not sure the cp is a good idea, changed a lot of things and it may not have been necessary...
# mofiles/%.o  %.import.scm : %.scm megatest-fossil-hash.scm
#	@[ -e mofiles ] || mkdir -p mofiles
#	csc $(CSCOPTS) -I mofiles -I $* -J -c $< -o $*.o
#	cp $*.o mofiles/$*.o
#	@touch $*.import.scm # ensure it is touched after the .o is made

mofiles/%.o : %.scm
	mkdir -p mofiles
	csc $(CSCOPTS) -J -c $< -o mofiles/$*.o

ADTLSCR=mt_laststep mt_runstep mt_ezstep
HELPERS=$(addprefix $(PREFIX)/bin/,$(ADTLSCR))
DEPLOYHELPERS=$(addprefix deploytarg/,$(ADTLSCR))
MTESTHASH=$(shell fossil info|grep checkout:| awk '{print $$2}')

ifeq ($(MTESTHASH),)
$(error MTESTHASH is broken!)
endif

CSIPATH=$(shell which csi)
CKPATH=$(shell dirname $(shell dirname $(CSIPATH)))
# CSIPATH=$(shell which csi)
# CKPATH=$(shell dirname $(shell dirname $(CSIPATH)))
# ARCHSTR=$(shell uname -m)_$(shell uname -r)
# BASH_MACHTYPE=$(shell bash -c "echo \$$MACHTYPE")
# ARCHSTR=$(BASH_MACHTYPE)_$(shell lsb_release -sr)
ARCHSTR=$(shell if [[ -e /usr/bin/sw_vers ]]; then /usr/bin/sw_vers -productVersion; else lsb_release -sr; fi)
# ARCHSTR=$(shell bash -c "echo \$$MACHTYPE")

PNGFILES = $(shell cd docs/manual;ls *png)

#all : $(PREFIX)/bin/.$(ARCHSTR) mtest dboard mtut ndboard
# all : $(PREFIX)/bin/.$(ARCHSTR) mtest dboard mtut ndboard
all : $(PREFIX)/bin/.$(ARCHSTR) mtest dboard mtut tcmt

mtest: $(OFILES) readline-fix.scm megatest.o $(MOFILES) mofiles/ftail.o mofiles/rmtmod.o mofiles/commonmod.o
	csc $(CSCOPTS) $(OFILES) $(MOFILES) megatest.o -o mtest
mtest: $(OFILES) readline-fix.scm megatest.o $(MOFILES) $(MOIMPFILES)  megatest-version.scm
	csc $(CSCOPTS) $(OFILES) $(MOFILES) $(MOIMPFILES) megatest.o -o mtest

showmtesthash:
	@echo $(MTESTHASH)

dboard : $(OFILES) $(GOFILES) dashboard.scm $(MOFILES)
	csc $(CSCOPTS) $(OFILES) dashboard.scm $(GOFILES) $(MOFILES) -o dboard
dboard : $(OFILES) $(GOFILES) dashboard.scm $(MOFILES) $(MOIMPFILES) megatest-version.scm megatest-fossil-hash.scm
	csc $(CSCOPTS) $(OFILES) dashboard.scm $(GOFILES) $(MOFILES) $(MOIMPFILES) -o dboard

ndboard : newdashboard.scm $(OFILES) $(GOFILES)
	csc $(CSCOPTS) $(OFILES) $(GOFILES) newdashboard.scm -o ndboard

mtut: $(OFILES) $(MOFILES) megatest-fossil-hash.scm mtut.scm
mtut: $(OFILES) $(MOFILES) megatest-fossil-hash.scm mtut.scm megatest-version.scm
	csc $(CSCOPTS) $(OFILES) $(MOFILES) mtut.scm -o mtut

# include makefile.inc

TCMTOBJS = \
	api.o \
	archive.o \
	cgisetup/models/pgdb.o \
	client.o \
	common.o \
	configf.o \
	db.o \
	env.o \
	http-transport.o \
	items.o \
	keys.o \
	launch.o \
	lock-queue.o \
	margs.o \
	mt.o \
	megatest-version.o \
	ods.o \
	portlogger.o \
	process.o \
	rmt.o \
        mofiles/rmtmod.o \
        mofiles/commonmod.o \
        rpc-transport.o \
	runconfig.o \
	runs.o \
	server.o \
	tasks.o \
	tdb.o \
	tests.o \
	subrun.o \
        ezsteps.o

#        mofiles/rmtmod.o \
#        mofiles/commonmod.o \

tcmt : $(TCMTOBJS) tcmt.scm
	csc $(CSCOPTS) $(TCMTOBJS) tcmt.scm -o tcmt
tcmt : $(TCMTOBJS) tcmt.scm megatest-version.scm
	csc $(CSCOPTS) $(TCMTOBJS) $(MOFILES) $(MOIMPFILES) tcmt.scm -o tcmt

# install documentation to $(PREFIX)/docs
# DOES NOT REBUILD DOCS
#
$(PREFIX)/share/docs/megatest_manual.html : docs/manual/megatest_manual.html
	mkdir -p $(PREFIX)/share/docs
	$(INSTALL) docs/manual/megatest_manual.html $(PREFIX)/share/docs/megatest_manual.html
	for png in $(PNGFILES);do $(INSTALL) docs/manual/$$png $(PREFIX)/share/docs/$$png;done

# add a fake dependency so this doens't copy everytime
$(PREFIX)/share/js/jquery-3.1.0.slim.min.js : # .fslckout
	mkdir -p $(PREFIX)/share/js
	fossil wiki export java-script-lib > $(PREFIX)/share/js/jquery-3.1.0.slim.min.js

$(PREFIX)/share/db/mt-pg.sql : mt-pg.sql
	mkdir -p $(PREFIX)/share/db
	$(INSTALL) mt-pg.sql $(PREFIX)/share/db/mt-pg.sql

#multi-dboard : multi-dboard.scm $(OFILES) $(GOFILES)
#	csc $(CSCOPTS) $(OFILES) $(GOFILES) multi-dboard.scm -o multi-dboard
# Special dependencies for the includes
$(MOFILE) $(MOIMPFILES) : megatest-fossil-hash.scm

# common.o : mofiles/commonmod.o megatest-fossil-hash.scm
# 

# $(PREFIX)/bin/revtagfsl : utils/revtagfsl.scm
#	csc utils/revtagfsl.scm -o $(PREFIX)/bin/revtagfsl
# commonmod.o dashboard.o megatest.o tcmt.o apimod.o : megatest-fossil-hash.scm

# Special dependencies for the includes
tests.o db.o launch.o runs.o dashboard-tests.o dashboard-context-menu.o dashboard-guimonitor.o dashboard-main.o monitor.o dashboard.o  \
archive.o megatest.o : db_records.scm
tests.o db.o launch.o runs.o dashboard-tests.o				\
dashboard-context-menu.o dashboard-guimonitor.o dashboard-main.o	\
monitor.o dashboard.o archive.o megatest.o : db_records.scm megatest-fossil-hash.scm

tests.o runs.o dashboard.o dashboard-tests.o dashboard-context-menu.o dashboard-main.o  : run_records.scm

db.o ezsteps.o keys.o launch.o megatest.o monitor.o runs-for-ref.o runs.o tests.o : key_records.scm

tests.o tasks.o dashboard-tasks.o : task_records.scm

runs.o : test_records.scm

megatest.o : megatest-fossil-hash.scm
rmt.scm client.scm common.scm configf.scm dashboard-guimonitor.scm dashboard-tests.scm dashboard.scm db.scm dcommon.scm ezsteps.scm fs-transport.scm http-transport.scm index-tree.scm items.scm keys.scm launch.scm megatest.scm monitor.scm mt.scm newdashboard.scm runconfig.scm runs.scm server.scm tdb.scm tests.scm tree.scm : common_records.scm
megatest.o : megatest-fossil-hash.scm megatest-version.scm

rmt.scm client.scm common.scm configf.scm dashboard-guimonitor.scm dashboard-tests.scm dashboard.scm db.scm dcommon.scm ezsteps.scm fs-transport.scm http-transport.scm index-tree.scm items.scm keys.scm launch.scm megatest.scm monitor.scm mt.scm newdashboard.scm runconfig.scm runs.scm server.scm tdb.scm tests.scm tree.scm : common_records.scm megatest-version.scm

common_records.scm : altdb.scm

# mofiles/stml2.o : mofiles/cookie.o
# configf.o : mofiles/commonmod.o

vg.o dashboard.o : vg_records.scm
vg.o dashboard.o : vg_records.scm megatest-version.scm

dcommon.o : run_records.scm

mofiles/stml2.o : mofiles/cookie.o

# # special include based modules
# mofiles/pkts.o      : pkts/pkts.scm
# mofiles/stml2.o     : cookie.o
# # mofiles/mtargs.o    : mtargs/mtargs.scm
# # mofiles/mtconfigf.o : mtconfigf/mtconfigf.scm
# mofiles/ulex.o      : ulex/ulex.scm
# mofiles/mutils.o    : mutils/mutils.scm
# mofiles/cookie.o    : stml2/cookie.scm
# mofiles/stml2.o     : stml2/stml2.scm

# Temporary while transitioning to new routine
# runs.o : run-tests-queue-classic.scm  run-tests-queue-new.scm

# for the modularized stuff
mofiles/rmtmod.o : mofiles/commonmod.o

megatest-fossil-hash.scm : $(SRCFILES) megatest.scm *_records.scm

$(PREFIX)/bin/.$(ARCHSTR)/tcmt : tcmt
	$(INSTALL) tcmt $(PREFIX)/bin/.$(ARCHSTR)/tcmt

$(PREFIX)/bin/tcmt : $(PREFIX)/bin/.$(ARCHSTR)/tcmt utils/mk_wrapper
	utils/mk_wrapper $(PREFIX) tcmt $(PREFIX)/bin/tcmt
	chmod a+x $(PREFIX)/bin/tcmt

# $(PREFIX)/bin/.$(ARCHSTR)/mdboard : multi-dboard
#	$(INSTALL) multi-dboard $(PREFIX)/bin/.$(ARCHSTR)/mdboard

# $(PREFIX)/bin/mdboard : $(PREFIX)/bin/.$(ARCHSTR)/mdboard  utils/mk_wrapper
# 	utils/mk_wrapper $(PREFIX) mdboard $(PREFIX)/bin/mdboard
# 	chmod a+x $(PREFIX)/bin/mdboard

# $(HELPERS) : utils/%
# 	$(INSTALL) $< $@
# 	chmod a+x $@

$(PREFIX)/bin/mt_laststep : utils/mt_laststep
	$(INSTALL) $< $@
	chmod a+x $@

$(PREFIX)/bin/mt_runstep : utils/mt_runstep
	$(INSTALL) $< $@
	chmod a+x $@

	$(INSTALL) $< $@
	chmod a+x $@

$(PREFIX)/bin/mtrunner : utils/mtrunner
	$(INSTALL) $< $@
	chmod a+x $@

# $(PREFIX)/bin/refdb : refdb
# 	$(INSTALL) $< $@
# 	chmod a+x $@

deploytarg/nbfake : utils/nbfake
	$(INSTALL) $< $@
	chmod a+x $@

deploytarg/viewscreen : utils/viewscreen
	$(INSTALL) $< $@
	chmod a+x $@

	mkdir -p ext-tests
	cd ext-tests;fossil open --nested $(MTQA_FOSSIL)

$(MTQA_FOSSIL) :
	fossil clone https://www.kiatoa.com/fossils/megatest_qa $(MTQA_FOSSIL)

clean : 
	rm -f $(OFILES) $(GOFILES) $(MOFILES) $(TCMTOBJS) $(PREFIX)/megatest $(PREFIX)/dashboard mtest mtutil mtut tcmt ftail.import.scm readline-fix.scm serialize-env dboard dboard.o megatest.o dashboard.o megatest-fossil-hash.* altdb.scm mofiles/*.o vg.o
	rm -f $(OFILES) $(GOFILES) $(MOFILES) $(TCMTOBJS)		\
            $(PREFIX)/megatest $(PREFIX)/dashboard mtest mtutil mtut	\
            tcmt readline-fix.scm serialize-env dboard *.o		\
            megatest-fossil-hash.* altdb.scm mofiles/*.o                \
            mofiles/*.o vg.o cookie.o dashboard-main.o	                \
            ducttape-lib.o ftail.o mutils.o pkts.o rmtmod.o stml2.o	\
            tcmt.o *.import.scm *.import.o
	rm -f $(OFILES) $(GOFILES) $(MOFILES) $(TCMTOBJS)		\
            $(PREFIX)/megatest $(PREFIX)/dashboard mtest mtutil mtut	\
            tcmt ftail.import.scm readline-fix.scm serialize-env	\
            dboard dboard.o megatest.o dashboard.o			\
            megatest-fossil-hash.* altdb.scm mofiles/*.o vg.o
	rm -rf share

#======================================================================
# Make the records files
#======================================================================

# vg_records.scm : records.sh
#	./records.sh

#======================================================================
# Deploy section (not complete yet)
#======================================================================

$(DEPLOYHELPERS) : utils/mt_*
	$(INSTALL) $< $@
	chmod a+X $@

deploytarg/apropos.so : Makefile
	chicken-install -p deploytarg -deploy -keep-installed $(EGGS)

#	for i in apropos base64 canvas-draw csv-xml directory-utils dot-locking extras fmt format hostinfo http-client intarweb json md5 message-digest posix posix-extras readline regex regex-case s11n spiffy spiffy-request-vars sqlite3 srfi-1 srfi-18 srfi-69 tcp test uri-common check-errors synch matchable sql-null tcp-server rpc blob-utils string-utils variable-item defstruct uri-generic sendfile opensll openssl lookup-table list-utils stack; do \
#	chicken-install -prefix deploytarg -deploy $$i;done

# deploytarg/libsqlite3.so : 
# 	CSC_OPTIONS="-Ideploytarg -Ldeploytarg" $CHICKEN_INSTALL -prefix deploytarg -deploy sqlite3

deploy : deploytarg/mtest deploytarg/dboard $(DEPLOYHELPERS) deploytarg/nbfake deploytarg/remrun deploytarg/viewsceen deploytarg/nbfind deploytarg/apropos.so

# deploytarg/libiupcd.so : $(CKPATH)/lib/libiupcd.so
# 	for i in iup im cd av call sqlite; do \
# 	  cp $(CKPATH)/lib/lib$$i* deploytarg/ ; \
# 	done
# 	cp $(CKPATH)/include/*.h deploytarg

# puts deployed megatest in directory "megatest"
deploytarg/mtest : $(OFILES) megatest.o deploytarg/apropos.so
	csc -deploy $(CSCOPTS) $(OFILES) megatest.scm -o deploytarg
	mv deploytarg/deploytarg deploytarg/mtest

deploytarg/dboard :  $(OFILES) $(GOFILES) dashboard.scm deploytarg/apropos.so
	csc -deploy $(OFILES) $(GOFILES) dashboard.scm -o deploytarg
	mv deploytarg/deploytarg deploytarg/dboard

# DATASHAREO=configf.o common.o process.o tree.o dcommon.o margs.o launch.o gutils.o db.o synchash.o server.o \
#            megatest-version.o tdb.o ods.o mt.o keys.o
datashare-testing/sd : datashare.scm $(OFILES)
	csc $(CSCOPTS) datashare.scm $(OFILES) -o datashare-testing/sd

datashare-testing/sdat: sharedat.scm $(OFILES)
	csc $(CSCOPTS) sharedat.scm $(OFILES) -o datashare-testing/sdat

sd : datashare-testing/sd
	mkdir -p /tmp/$(USER)/datashare/disk1 /tmp/$(USER)/basepath

xterm : sd
	(export BASEPATH=/tmp/$(USER)/basepath ; export PATH="$(PWD)/datashare-testing:$(PATH)" ; xterm &)

datashare-testing/spublish : spublish.scm $(OFILES)
	csc $(CSCOPTS) spublish.scm  megatest-version.o margs.o process.o common.o  -o datashare-testing/spublish
datashare-testing/spublish : spublish.scm $(OFILES) megatest-version.scm
	csc $(CSCOPTS) spublish.scm margs.o process.o common.o  -o datashare-testing/spublish

datashare-testing/sretrieve : sretrieve.scm $(OFILES)
	csc $(CSCOPTS) sretrieve.scm  megatest-version.o margs.o process.o common.o  -o datashare-testing/sretrieve
datashare-testing/sretrieve : sretrieve.scm $(OFILES) megatest-version.scm
	csc $(CSCOPTS) sretrieve.scm margs.o process.o common.o  -o datashare-testing/sretrieve


datashare-testing/sauthorize : sauthorize.scm $(OFILES)
	 csc $(CSCOPTS) sauthorize.scm  megatest-version.o margs.o process.o common.o  -o datashare-testing/sauthorize
datashare-testing/sauthorize : sauthorize.scm $(OFILES) megatest-version.scm
	 csc $(CSCOPTS) sauthorize.scm margs.o process.o common.o  -o datashare-testing/sauthorize

sauth-init:
	mkdir -p  datashare-testing
	rm  datashare-testing/sauthorize
	rm  datashare-testing/sretrieve 
	rm  datashare-testing/spublish

sauth : sauth-init datashare-testing/sauthorize datashare-testing/sretrieve datashare-testing/spublish 


# base64 dot-locking \
#             csv-xml z3

#  "(define (toplevel-command . a) #f)"
# if egrep 'version.*3.0' $(shell dirname $(shell dirname $(shell which csi)))/lib/chicken/7/readline.setup-info;then \

readline-fix.scm :
	if [[ $(shell chicken-status | grep readline | awk '{print $4}' | cut -d. -f1) -gt 3 ]];then \
	   echo "(define *use-new-readline* #f)" > readline-fix.scm; \
	else \
	   echo "(define *use-new-readline* #t)" > readline-fix.scm;\
	fi

altdb.scm :
	echo ";; optional alternate db setup" > altdb.scm
	echo "(define *available-db* (make-hash-table))" >> altdb.scm
	if  csi -ne '(use mysql-client)';then \
           echo "(use mysql-client)(hash-table-set! *available-db* 'mysql #t)" >> altdb.scm; \
	fi
	if csi -ne '(use postgresql)';then \
	   echo "(use postgresql)(hash-table-set! *available-db* 'postgresql #t)" >> altdb.scm;\
	fi

portlogger-example : portlogger-example.scm api.o archive.o client.o common.o configf.o dashboard-tests.o dashboard-context-menu.o db.o dcommon.o ezsteps.o filedb.o genexample.o gutils.o http-transport.o items.o keys.o launch.o lock-queue.o margs.o megatest-version.o mt.o ods.o portlogger.o process.o rmt.o rpc-transport.o runconfig.o runs.o sdb.o server.o synchash.o tasks.o tdb.o tests.o tree.o
	csc $(CSCOPTS) portlogger-example.scm api.o archive.o client.o common.o configf.o dashboard-tests.o dashboard-context-menu.o db.o dcommon.o ezsteps.o filedb.o genexample.o gutils.o http-transport.o items.o keys.o launch.o lock-queue.o margs.o megatest-version.o mt.o ods.o portlogger.o process.o rmt.o rpc-transport.o runconfig.o runs.o sdb.o server.o synchash.o tasks.o tdb.o tests.o tree.o
portlogger-example : portlogger-example.scm api.o archive.o client.o common.o configf.o dashboard-tests.o dashboard-context-menu.o db.o dcommon.o ezsteps.o filedb.o genexample.o gutils.o http-transport.o items.o keys.o launch.o lock-queue.o margs.o mt.o ods.o portlogger.o process.o rmt.o runconfig.o runs.o sdb.o server.o synchash.o tasks.o tdb.o tests.o tree.o
	csc $(CSCOPTS) portlogger-example.scm api.o archive.o client.o common.o configf.o dashboard-tests.o dashboard-context-menu.o db.o dcommon.o ezsteps.o filedb.o genexample.o gutils.o http-transport.o items.o keys.o launch.o lock-queue.o margs.o mt.o ods.o portlogger.o process.o rmt.o runconfig.o runs.o sdb.o server.o synchash.o tasks.o tdb.o tests.o tree.o

# create a pdf dot graphviz diagram from notations in rmt.scm
rmt.pdf : rmt.scm
	grep ';;DOT' rmt.scm | sed -e 's/.*;;DOT //' > rmt.dot;dot -Tpdf rmt.dot -o rmt.pdf

buildmanual:
	cd docs/manual && make

#  Copyright 2006-2017, Matthew Welland.
#  Copyright 2006-2020, Matthew Welland.
# 
# This file is part of Megatest.
# 
#     Megatest is free software: you can redistribute it and/or modify
#     it under the terms of the GNU General Public License as published by
#     the Free Software Foundation, either version 3 of the License, or
#     (at your option) any later version.
# 
#     Megatest is distributed in the hope that it will be useful,
#     but WITHOUT ANY WARRANTY; without even the implied warranty of
#     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#     GNU General Public License for more details.
# 
#     You should have received a copy of the GNU General Public License
#     along with Megatest.  If not, see <http://www.gnu.org/licenses/>.

TODO
====

WW38
. Add test_rundat to no-sync ==> correction, put in <testdir>/.meta/test-run.dat
. Add STATE/STATUS transitions to .meta/test-run.dat or similar
. Swizzle update-test-rundat to operate on no-sync
. Swizzle update-run-duration, -uname-host and cpuload-diskfree to no-sync
. On state/status change update tests table with duration
. Dashboard should resist running from non-homehost

WW15
. fill newview matrix with data, filter pipeline gui elements
. improve [script], especially indent handling

WW16
. split db into megatest.db (runs etc.) db/<something>.db
. release basic newview implementation

WW18
. release split db implementation
. mtutil calls from dashboard (for remote control)
. logs browser (esp. for surfacing mtutil related activities)

WW19
. break command line into sections; all, run control, queries, utilities etc.
. pull in ftfplan (not integrated, just code pulled in)

WW20
. ./configure => ubuntu, sles11, sles12, rh7
. Add output flushing in teamcity support
. Switch to using simple runs query everywhere
. Add end_time to runs and add a rollup call that sets state, status and end_time

Future
. Switch to scsh-process pipeline management for job execution/control
. Use call-with-environment-variables more.

Migration to inmem db plus per run db
-------------------------------------

. Re-work the dbstruct data structure?
.. Move main.db to global?
.. [ run-id.db inmemdb last-mod last-read last-sync inuse ]
. Re-work all queries to use run-id to dereference server
. Open main.db directly in calls to -runtests etc. No need to talk remote?
. remove common:faux-lock

;;======================================================================
;; Copyright 2017, Matthew Welland.
;; 
;; This file is part of Megatest.
;; 
;;     Megatest is free software: you can redistribute it and/or modify
;;     it under the terms of the GNU General Public License as published by
;;     the Free Software Foundation, either version 3 of the License, or
;;     (at your option) any later version.
;; 
;;     Megatest is distributed in the hope that it will be useful,
;;     but WITHOUT ANY WARRANTY; without even the implied warranty of
;;     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
;;     GNU General Public License for more details.
;; 
;;     You should have received a copy of the GNU General Public License
;;     along with Megatest.  If not, see <http://www.gnu.org/licenses/>.

;;======================================================================

(declare (unit adjutant))

(module adjutant *

(import scheme chicken data-structures extras files)
(import (prefix sqlite3 sqlite3:) posix typed-records srfi-18 srfi-69
	md5 message-digest
	regex srfi-1)

(define (adjutant-run)
  (print "Running the adjutant!"))

)

    get-run-times
    get-targets
    get-target
    ;; register-run
    get-tests-tags
    get-test-times
    get-tests-for-run
    get-tests-for-run-state-status
    get-test-id
    get-tests-for-runs-mindata
    get-tests-for-run-mindata
    get-run-name-from-id
    get-runs
    simple-get-runs
    get-num-runs

    read-test-data*
    login
    tasks-get-last
    testmeta-get-record
    have-incompletes?
    ;; synchash-get
    get-changed-record-ids
		get-run-record-ids 
    get-run-record-ids 
    get-not-completed-cnt))

(define api:write-queries
  '(
    get-keys-write ;; dummy "write" query to force server start

    ;; SERVERS
    start-server
    kill-server

    ;; TESTS
    test-set-state-status-by-id
    delete-test-records
    delete-old-deleted-test-records
    test-set-state-status
    test-set-top-process-pid
    set-state-status-and-roll-up-items
     
    
    update-pass-fail-counts
    top-test-set-per-pf-counts ;; (db:top-test-set-per-pf-counts (db:get-db *db* 5) 5 "runfirst")

    ;; RUNS
    register-run
    set-tests-state-status
    delete-run

    csv->test-data

    ;; MISC
    sync-inmem->db
    drop-all-triggers
    create-all-triggers
    update-tesdata-on-repilcate-db 

    ;; TESTMETA
    testmeta-add-record
    testmeta-update-field

    ;; TASKS
    tasks-add
    tasks-set-state-given-param-key
    ))

;; These are called by the server on recipt of /api calls
;;    - keep it simple, only return the actual result of the call, i.e. no meta info here
;;
;;    - returns #( flag result )
;;
(define (api:execute-requests dbstruct dat)
  (handle-exceptions
   exn
   (let ((call-chain (get-call-chain)))
     (debug:print 0 *default-log-port* "WARNING: api:execute-requests received an exception from peer, dat=" dat)
     (debug:print 0 *default-log-port* "WARNING: api:execute-requests received an exception from peer, dat=" dat ", exn=" exn)
     (print-call-chain (current-error-port))
     (debug:print 0 *default-log-port* " message: "  ((condition-property-accessor 'exn 'message) exn))       
     (vector #f (vector exn call-chain dat))) ;; return some stuff for debug if an exception happens
   (cond
    ((not (vector? dat))                    ;; it is an error to not receive a vector
     (vector #f (vector #f "remote must be called with a vector")))
    ((> *api-process-request-count* 20) ;; 20)

                   ;; MISC
                   ((sync-inmem->db)               (let ((run-id (car params)))
                                                     (db:sync-touched dbstruct run-id force-sync: #t)))
                   ((mark-incomplete)              (apply db:find-and-mark-incomplete dbstruct params))
                   ((create-all-triggers)          (db:create-all-triggers dbstruct))
                   ((drop-all-triggers)            (db:drop-all-triggers dbstruct)) 

                   ;; TESTMETA
                   ((testmeta-add-record)       (apply db:testmeta-add-record dbstruct params))
                   ((testmeta-update-field)     (apply db:testmeta-update-field dbstruct params))
                   ((get-tests-tags)            (db:get-tests-tags dbstruct))

                   ;; TASKS
                   ((tasks-add)                 (apply tasks:add dbstruct params))   
                   ((tasks-set-state-given-param-key) (apply tasks:set-state-given-param-key dbstruct params))
                   ((tasks-get-last)            (apply tasks:get-last dbstruct params))

		   ;; NO SYNC DB
		   ((no-sync-set)               (apply db:no-sync-set         *no-sync-db* params))
		   ((no-sync-get/default)       (apply db:no-sync-get/default *no-sync-db* params))
		   ((no-sync-del!)              (apply db:no-sync-del!        *no-sync-db* params))
		   ((no-sync-get-lock)          (apply db:no-sync-get-lock    *no-sync-db* params))
		 
		   
                   ;; ARCHIVES
                   ;; ((archive-get-allocations)   
                   ((archive-register-disk)     (apply db:archive-register-disk dbstruct params))
                   ((archive-register-block-name)(apply db:archive-register-block-name dbstruct params))
                   ;; ((archive-allocate-testsuite/area-to-block)(apply db:archive-allocate-testsuite/area-to-block dbstruct block-id testsuite-name areakey))

                   ;;======================================================================

                   ((get-run-info)                 (apply db:get-run-info dbstruct params))
                   ((get-run-status)               (apply db:get-run-status dbstruct params))
                   ((get-run-state)                (apply db:get-run-state dbstruct params))
                   ((set-run-status)               (apply db:set-run-status dbstruct params))
                   ((set-run-state-status)  			 (apply db:set-run-state-status dbstruct params))
                   ((update-tesdata-on-repilcate-db) (apply db:update-tesdata-on-repilcate-db  dbstruct params)) 
                   ((get-tests-for-run)            (apply db:get-tests-for-run dbstruct params))
                   ((get-tests-for-run-state-status) (apply db:get-tests-for-run-state-status dbstruct params))
                   ((get-test-id)                  (apply db:get-test-id dbstruct params))
                   ((get-tests-for-run-mindata)    (apply db:get-tests-for-run-mindata dbstruct params))
                   ;; ((get-tests-for-runs-mindata)   (apply db:get-tests-for-runs-mindata dbstruct params))
                   ((get-runs)                     (apply db:get-runs dbstruct params))
                   ((simple-get-runs)              (apply db:simple-get-runs dbstruct params))
                   ((get-num-runs)                 (apply db:get-num-runs dbstruct params))
                   ((get-runs-cnt-by-patt)         (apply db:get-runs-cnt-by-patt dbstruct params))

;;======================================================================
;; Copyright 2017, Matthew Welland.
;; 
;; This file is part of Megatest.
;; 
;;     Megatest is free software: you can redistribute it and/or modify
;;     it under the terms of the GNU General Public License as published by
;;     the Free Software Foundation, either version 3 of the License, or
;;     (at your option) any later version.
;; 
;;     Megatest is distributed in the hope that it will be useful,
;;     but WITHOUT ANY WARRANTY; without even the implied warranty of
;;     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
;;     GNU General Public License for more details.
;; 
;;     You should have received a copy of the GNU General Public License
;;     along with Megatest.  If not, see <http://www.gnu.org/licenses/>.

;;======================================================================

(declare (unit apimod))
(declare (uses commonmod))
(declare (uses ulex))

(module apimod
	*
	
(import scheme chicken data-structures extras)
(import (prefix sqlite3 sqlite3:) posix typed-records srfi-18)
(import commonmod)
(import (prefix ulex ulex:))


(define (api:execute-requests params)
  #f)

)

	(hash-table-ref blockid-cache key)
	(let* ((pscript     (configf:lookup *configdat* "archive" "pathscript"))
	       (pscript-cmd (conc pscript " " testsuite-name " " target " " run-name " " test-name))
	       (apath       (if pscript
				(handle-exceptions
				 exn
				 (begin
				   (debug:print 0 *default-log-port* "ERROR: script \"" pscript-cmd "\" failed to run properly.")
				   (debug:print 0 *default-log-port* "ERROR: script \"" pscript-cmd "\" failed to run properly. exn=" exn)
				   (exit 1))
				 (with-input-from-pipe
				  pscript-cmd
				  read-line))
				#f)) ;; this is the user-calculated archive path
	       (adisks    (archive:get-archive-disks))
	       (best-disk (common:get-disk-with-most-free-space adisks dneeded)))

		(rmt:test-set-archive-block-id run-id test-id archive-id)
		(if (member (symbol->string archive-command) '("save-remove"))
                    (begin 
                     (debug:print-info 0 *default-log-port* "remove testdat")
		    (runs:remove-test-directory test-dat 'archive-remove)))))
	    (hash-table-ref test-groups test-base)))))
       (hash-table-keys disk-groups))
      
    #t))

(define (archive:megatest-db target-patt run-patt)
 (let* ((blockid-cache  (make-hash-table))
        (tsname         (common:get-testsuite-name))
        (min-space      (string->number (or (configf:lookup *configdat* "archive" "minspace") "1000")))
        (bup-exe        (or (configf:lookup *configdat* "archive" "bup") "bup"))

#  Copyright 2006-2017, Matthew Welland.
# 
# This file is part of Megatest.
# 
#     Megatest is free software: you can redistribute it and/or modify
#     it under the terms of the GNU General Public License as published by
#     the Free Software Foundation, either version 3 of the License, or
#     (at your option) any later version.
# 
#     Megatest is distributed in the hope that it will be useful,
#     but WITHOUT ANY WARRANTY; without even the implied warranty of
#     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#     GNU General Public License for more details.
# 
#     You should have received a copy of the GNU General Public License
#     along with Megatest.  If not, see <http://www.gnu.org/licenses/>.


#======================================================================
# Chicken build
#======================================================================

# CHICKEN_BIN_DIR=$(shell dirname $(shell which csi))
# if have csi on path use that, else use default
# CSIPATH=$(shell which csi)
# CKPATH=$(shell dirname $(shell dirname $(CSIPATH)))
sCHICKEN_PREFIX=$(or $(CKPATH),$(PREFIX)/bin/.$(ARCHSTR))

whatever :
	@echo "CHICKEN_PREFIX=$(CHICKEN_PREFIX)"

tgz-$(USER)/postgresql-9.6.4.tar.gz :
	mkdir -p tgz-$(USER)
	wget -c https://ftp.postgresql.org/pub/source/v9.6.4/postgresql-9.6.4.tar.gz
	mv postgresql-9.6.4.tar.gz tgz-$(USER)/

tgz-$(USER)/sqlite-autoconf-3090200.tar.gz :
	mkdir -p tgz-$(USER)
	curl http://www.sqlite.org/2015/sqlite-autoconf-3090200.tar.gz > tgz-$(USER)/sqlite-autoconf-3090200.tar.gz 

tgz-$(USER)/nanomsg-1.0.0.tar.gz :
	wget --no-check-certificate https://github.com/nanomsg/nanomsg/archive/1.0.0.tar.gz
	mv 1.0.0.tar.gz tgz-$(USER)/nanomsg-1.0.0.tar.gz	

tgz-$(USER)/chicken-4.13.0.tar.gz :
	mkdir -p tgz-$(USER)
	curl https://code.call-cc.org/releases/4.13.0/chicken-4.13.0.tar.gz > tgz-$(USER)/chicken-4.13.0.tar.gz

tgz-$(USER)/ffcall.tar.gz :
	wget -c -O tgz-$(USER)/ffcall.tar.gz 'http://www.kiatoa.com/fossils/ffcall/tarball?name=ffcall&uuid=trunk'

$(CHICKEN_PREFIX)/bin/pg_config : tgz-$(USER)/postgresql-9.6.4.tar.gz
	mkdir -p build-$(USER)/
	tar xfz tgz-$(USER)/postgresql-9.6.4.tar.gz -C build-$(USER)
	cd build-$(USER)/postgresql-9.6.4; ./configure --prefix=$(CHICKEN_PREFIX) --with-openssl; make; make install

build-$(USER)/sqlite-autoconf-3090200/configure : tgz-$(USER)/sqlite-autoconf-3090200.tar.gz
	mkdir -p build-$(USER);
	cd build-$(USER); tar xf ../tgz-$(USER)/sqlite-autoconf-3090200.tar.gz

$(CHICKEN_PREFIX)/lib/libnanomsg.so : tgz-$(USER)/nanomsg-1.0.0.tar.gz
	cd tgz-$(USER); tar -xzvf nanomsg-1.0.0.tar.gz
	cd tgz-$(USER)/nanomsg-1.0.0; mkdir build-$(USER); cd build-$(USER);
	cd tgz-$(USER)/nanomsg-1.0.0/build-$(USER); cmake ../ -DCMAKE_INSTALL_PREFIX=$(CHICKEN_PREFIX)
	cd tgz-$(USER)/nanomsg-1.0.0/build-$(USER); make; make install

$(CHICKEN_PREFIX)/chicken-4.13.0/LICENSE : tgz-$(USER)/chicken-4.13.0.tar.gz
	mkdir -p build-$(USER)/eggs-installed
	cd build-$(USER);tar xf ../tgz-$(USER)/chicken-4.13.0.tar.gz

tgz-$(USER)/opensrc.fossil :
	cd tgz-$(USER); fossil clone http://www.kiatoa.com/fossils/opensrc opensrc.fossil
	mkdir tgz-$(USER)/opensrc
	cd tgz-$(USER)/opensrc; fossil open --nested ../opensrc.fossil; fossil up; fossil uv sync

$(CHICKEN_PREFIX)/lib/libiupweb.so : tgz-$(USER)/opensrc.fossil
	cd tgz-$(USER)/opensrc; fossil unversioned cat libs/cd/cd-5.10_Linux26g4_64_lib.tar.gz > ../cd.tgz
	cd tgz-$(USER)/opensrc; fossil unversioned cat libs/im/im-3.11_Linux26g4_64_lib.tar.gz > ../im.tgz
	cd tgz-$(USER)/opensrc; fossil unversioned cat libs/iup/iup-3.19.1_Linux26g4_64_lib.tar.gz > ../iup.tgz
	cd tgz-$(USER); tar -xzf cd.tgz;
	cd tgz-$(USER); tar -xzf im.tgz;
	cd tgz-$(USER); tar -xzf iup.tgz;
	cp tgz-$(USER)/include/* $(CHICKEN_PREFIX)/include/
	cp tgz-$(USER)/*.so $(CHICKEN_PREFIX)/lib/
	cp tgz-$(USER)/*.a  $(CHICKEN_PREFIX)/lib/
	cp tgz-$(USER)/ftgl/lib/*/* $(CHICKEN_PREFIX)/lib/

EGGS=srfi-69 srfi-42 sqlite3 iup canvas-draw typed-records md5 regex-case base64 \
format dot-locking csv-xml z3 udp hostinfo directory-utils stack dbi crypt sha1 \
posix-extras pathname-expand csv call-with-environment-variables s11n spiffy \
uri-common intarweb http-client spiffy-request-vars spiffy-directory-listing \
ansi-escape-sequences test slice rfc3339 uuid-lib filepath srfi-19 sparse-vectors \
sql-de-lite fmt readline apropos json simple-exceptions rpc trace logpro refdb postgresql nanomsg
EGGSTARG=$(addsuffix .done,$(addprefix build-$(USER)/eggs-installed/,$(EGGS)))
EGGSTARG2=$(addsuffix .done, $(EGGS))

$(CHICKEN_PREFIX)/lib/libcallback.a : tgz-$(USER)/ffcall.tar.gz
	cd tgz-$(USER); tar -xzvf ffcall.tar.gz 
	cd tgz-$(USER)/ffcall; ./configure --prefix=$(CHICKEN_PREFIX) --enable-shared
	cd tgz-$(USER)/ffcall; make CC="gcc -fPIC"; make install	

$(CHICKEN_PREFIX)/bin/sqlite3 : build-$(USER)/sqlite-autoconf-3090200/configure 
	cd build-$(USER)/sqlite-autoconf-3090200; ./configure --prefix=$(CHICKEN_PREFIX); make; make install

$(CHICKEN_PREFIX)/bin/csi : $(CHICKEN_PREFIX)/bin/sqlite3 $(CHICKEN_PREFIX)/lib/libiupweb.so $(CHICKEN_PREFIX)/chicken-4.13.0/LICENSE
	cd build-$(USER)/chicken-4.13.0;make PLATFORM=linux PREFIX=$(CHICKEN_PREFIX) 
	cd build-$(USER)/chicken-4.13.0;make PLATFORM=linux PREFIX=$(CHICKEN_PREFIX) install

ALL_CKBIN=chicken chicken-bind chicken-bug chicken-dump			\
chicken-install chicken-profile chicken-sqlite3 chicken-status		\
chicken-uninstall csc csi feathers nanocat sqlite3 vacuumdb logpro	\
refdb

CKBIN_WRAPPERS=$(addprefix $(PREFIX)/bin/,$(ALL_CKBIN))

$(PREFIX)/bin/% : $(CHICKEN_PREFIX)/bin/% $(CHICKEN_PREFIX)/bin/csi $(EGGSTARG2)
	utils/mk_wrapper_tool $(PREFIX) $* $(PREFIX)/bin/$*
	chmod a+x $(PREFIX)/bin/$*

$(PREFIX)/bin :
	mkdir -p $(PREFIX)/bin $(CHICKEN_PREFIX)/bin

chicken : $(PREFIX)/bin $(CHICKEN_PREFIX)/bin/csi  binwrappers
	@echo "Fake target to build prefix chicken"

binwrappers : $(CKBIN_WRAPPERS)

postgresql.done : $(CHICKEN_PREFIX)/bin/pg_config
	CSC_OPTIONS="-I$(CHICKEN_PREFIX)/include -L$(CHICKEN_PREFIX)/lib -L$(CHICKEN_PREFIX)/lib64" $(CHICKEN_PREFIX)/bin/chicken-install postgresql > postgresql.done

nanomsg.done : $(CHICKEN_PREFIX)/lib/libnanomsg.so
	CSC_OPTIONS="-I$(CHICKEN_PREFIX)/include -L$(CHICKEN_PREFIX)/lib -L$(CHICKEN_PREFIX)/lib64" $(CHICKEN_PREFIX)/bin/chicken-install nanomsg > nanomsg.done

iup.done : $(CHICKEN_PREFIX)/lib/libcallback.a
	CSC_OPTIONS="-I$(CHICKEN_PREFIX)/include -L$(CHICKEN_PREFIX)/lib" $(CHICKEN_PREFIX)/bin/chicken-install -D no-library-checks -feature disable-iup-web -feature disable-iup-pplot -feature disable-iup-matrixex iup > iup.done

canvas-draw.done :
	CSC_OPTIONS="-I$(CHICKEN_PREFIX)/include -L$(CHICKEN_PREFIX)/lib" $(CHICKEN_PREFIX)/bin/chicken-install -D no-library-checks canvas-draw > canvas-draw.done

sqlite3.done :
	CSC_OPTIONS="-I$(CHICKEN_PREFIX)/include -L$(CHICKEN_PREFIX)/lib" $(CHICKEN_PREFIX)/bin/chicken-install sqlite3 > sqlite3.done

sql-de-lite.done :
	CSC_OPTIONS="-I$(CHICKEN_PREFIX)/include -L$(CHICKEN_PREFIX)/lib" $(CHICKEN_PREFIX)/bin/chicken-install sql-de-lite > sql-de-lite.done

dbi.done : postgresql.done sqlite3.done sql-de-lite.done
	CSC_OPTIONS="-I$(CHICKEN_PREFIX)/include -L$(CHICKEN_PREFIX)/lib" $(CHICKEN_PREFIX)/bin/chicken-install dbi > dbi.done

%.done :
	$(CHICKEN_PREFIX)/bin/chicken-install $* > $*.done

build-$(USER)/eggs-installed/%.done : $(CHICKEN_PREFIX)/bin/csi $(EGGS)
	$(CHICKEN_PREFIX)/bin/chicken-install $* > build-$(USER)/eggs-installed/$*.done

build-clean :
	rm -rf build-$(USER) bin

     matchable regex posix (srfi 18) extras ;; tcp 
     (prefix nanomsg nmsg:)
     (prefix sqlite3 sqlite3:)
     pkts (prefix dbi dbi:)
     )

(declare (unit common))
(declare (uses commonmod))
(import commonmod)
;; (declare (uses commonmod))
;; (import commonmod)

(include "common_records.scm")


;; (require-library margs)
;; (include "margs.scm")

;; execute thunk, return value.  If exception thrown, trap exception, return #f, and emit nonfatal condition note to *default-log-port* .
;; arguments - thunk, message
(define (common:fail-safe thunk warning-message-on-exception)
  (handle-exceptions
   exn
   (begin
     (debug:print-info 0 *default-log-port* "notable but nonfatal condition - "warning-message-on-exception)
     (debug:print-info 0 *default-log-port* "notable but nonfatal condition - "warning-message-on-exception", exn=" exn)
     (debug:print-info 0 *default-log-port*
                       (string-substitute "\n?Error:" "nonfatal condition:"
                                          (with-output-to-string
                                            (lambda ()
                                              (print-error-message exn) ))))
     (debug:print-info 0 *default-log-port* "    -- continuing after nonfatal condition...")
     #f)
   (thunk)))

(define getenv get-environment-variable)
(define (safe-setenv key val)
  (if (or (substring-index "!" key)
  (if (or (substring-index "!" key) (substring-index ":" key)) ;; variables containing : are for internal use and cannot be environment variables.
	  (substring-index ":" key)  ;; variables containing : are for internal use and cannot be environment variables.
	  (substring-index "." key)) ;; periods are not allowed in environment variables
      (debug:print-error 4 *default-log-port* "skip setting internal use only variables containing \":\" or starting with \"!\"")
      (if (and (string? val)
	       (string? key))
	  (handle-exceptions
	      exn
	      (debug:print-error 0 *default-log-port* "bad value for setenv, key=" key ", value=" val)
	      (debug:print-error 0 *default-log-port* "bad value for setenv, key=" key ", value=" val ", exn=" exn)
	    (setenv key val))
	  (debug:print-error 0 *default-log-port* "bad value for setenv, key=" key ", value=" val))))

(define home (getenv "HOME"))
(define user (getenv "USER"))

            (caddr argv))
           (else (car argv))))
         (fullpath (realpath this-script)))
    fullpath))
(define *common:this-exe-fullpath* (common:get-this-exe-fullpath))
(define *common:this-exe-dir*      (pathname-directory *common:this-exe-fullpath*))
(define *common:this-exe-name*     (pathname-strip-directory *common:this-exe-fullpath*))

;; when called from a wrapper I need sometimes to find the calling
;; wrapper, this is for dashboard to find the correct megatest.
;;
(define (common:find-local-megatest #!optional (progname "megatest"))
  (let ((res (filter file-exists?
		     (map (lambda (updir)
			    (let* ((lm  (car (argv)))
				   (dir (pathname-directory lm))
				   (exe (pathname-strip-directory lm)))
			      (conc (if dir (conc dir "/") "")
				    (case (string->symbol exe)
				      ((dboard)    (conc updir progname))
				      ((mtest)     (conc updir progname))
				      ((dashboard) progname)
				      (else exe)))))
			  '("../../" "../")))))
    (if (null? res)
	(begin
	  (debug:print 0 *current-log-port* "Failed to find this executable! Using what can be found on the path")
	  progname)
	(car res))))

(define *common:logpro-exit-code->status-sym-alist*
  '( ( 0 . pass )
     ( 1 . fail )
     ( 2 . warn )
     ( 3 . check )
     ( 4 . waived )

	(max-allowed (string->number (or (configf:lookup *configdat* "setup" "max-logfiles") "300")))) ;; name -> age
    (if (not (directory-exists? "logs"))(create-directory "logs"))
    (directory-fold 
     (lambda (file rem)
       (handle-exceptions
	exn
	(begin
	  (debug:print-info 2 *default-log-port* "unable to rotate log " file ", probably handled by another process, this is safe to ignore.")
	  (debug:print-info 2 *default-log-port* "unable to rotate log " file ", probably handled by another process, this is safe to ignore. exn=" exn)
	  (debug:print 2 *default-log-port* " message: " ((condition-property-accessor 'exn 'message) exn))
	  ;; (print-call-chain (current-error-port)) ;; 
	  )
	(let* ((fullname  (conc "logs/" file))
	       (mod-time  (file-modification-time fullname))
	       (file-age  (- (current-seconds) mod-time))
	       (file-old  (> file-age (* 48 60 60)))

	      (if (and (> file-age (* (string->number (or (configf:lookup *configdat* "setup" "log-expire-days") "30")) 24 3600))
		       (file-exists? fullname)) ;; just in case it was gzipped - will get it next time
		  (handle-exceptions
		   exn
		   #f
		   (if (directory? fullname)
		       (begin
			 (debug:print-error 0 *default-log-port* fullname " in logs directory is a directory! Cannot rotate it, it is best to not put subdirectories in the logs dir.")
			 (debug:print-info 0 *default-log-port* fullname " in logs directory is a directory! Cannot rotate it, it is best to not put subdirectories in the logs dir.")
			 (inc-stat "directories"))
		       (begin
			 (delete-file* fullname)
			 (inc-stat "deleted")))
		   (hash-table-delete! all-files file)))))))
     '()
     "logs")

				   (lambda (a b)
				     (< (hash-table-ref all-files a)(hash-table-ref all-files b))))
			     (- num-logs max-allowed))))
	    (for-each
	     (lambda (file)
	       (let* ((fullname (conc "logs/" file)))
		 (if (directory? fullname)
		     (debug:print-error 0 *default-log-port* fullname " in logs directory is a directory! Cannot rotate it, it is best to not put subdirectories in the logs dir.")
		     (debug:print-info 0 *default-log-port* fullname " in logs directory is a directory! Cannot rotate it, it is best to not put subdirectories in the logs dir.")
		     (handle-exceptions
		      exn
		      (debug:print-error 0 *default-log-port* "failed to remove " fullname)
		      (debug:print-info 0 *default-log-port* "failed to remove " fullname ", exn=" exn)
		      (delete-file* fullname)))))
	     files)
	    (debug:print-info 0 *default-log-port* "Deleted " (length files) " files from logs, keeping " max-allowed " files."))))))
  
;; Force a megatest cleanup-db if version is changed and skip-version-check not specified
;; Do NOT check if not on homehost!
;;

             ((get-environment-variable "MT_SKIP_DB_MIGRATE") #t)
             ((and (common:file-exists? mtconf) (common:file-exists? dbfile) (not read-only)
                   (eq? (current-user-id)(file-owner mtconf))) ;; safe to run -cleanup-db
              (debug:print 0 *default-log-port* "   I see you are the owner of megatest.config, attempting to cleanup and reset to new version")
              (handle-exceptions
               exn
               (begin
                 (debug:print 0 *default-log-port* "Failed to switch versions.")
                 (debug:print 0 *default-log-port* "Failed to switch versions. exn=" exn)
                 (debug:print 0 *default-log-port* " message: " ((condition-property-accessor 'exn 'message) exn))
                 (print-call-chain (current-error-port))
                 (exit 1))
               (common:cleanup-db dbstruct)))
             ((not (common:file-exists? mtconf))
              (debug:print 0 *default-log-port* "   megatest.config does not exist in this area.  Cannot proceed with megatest version migration.")
              (exit 1))

(define (common:read-encoded-string instr)
  (handle-exceptions
   exn
   (handle-exceptions
    exn
    (begin
      (debug:print-error 0 *default-log-port* "received bad encoded string \"" instr "\", message: " ((condition-property-accessor 'exn 'message) exn))
      (debug:print-error 0 *default-log-port* "received bad encoded string \"" instr "\", message: " ((condition-property-accessor 'exn 'message) exn) ", exn=" exn)
      (print-call-chain (current-error-port))
      #f)
    (read (open-input-string (base64:base64-decode instr))))
   (read (open-input-string (z3:decode-buffer (base64:base64-decode instr))))))

;; dot-locking egg seems not to work, using this for now
;; if lock is older than expire-time then remove it and try again

  (let ((res (assoc key lst)))
    (if res (cadr res)(if (null? default) #f (car default)))))

(define (common:get-testsuite-name)
  (or (configf:lookup *configdat* "setup" "area-name") ;; megatest is a flexible tool, testsuite is too limiting a description.
      (configf:lookup *configdat* "setup" "testsuite" )
      (getenv "MT_TESTSUITE_NAME")
      (if (string? *toppath* )
          (pathname-file *toppath*)
          #f))) ;; (pathname-file (current-directory)))))
      (pathname-file (or (if (string? *toppath* )
			     (pathname-file *toppath*)
			     #f)
			 (common:get-topath #f)))
      "please-set-setup-area-name")) ;; (pathname-file (current-directory)))))

;; safe getting of toppath
(define (common:get-toppath areapath)
  (or *toppath*
      (if areapath
	  (begin
	    (set! *toppath* areapath)

(define (common:get-db-tmp-area . junk)
  (if *db-cache-path*
      *db-cache-path*
      (if *toppath* ;; common:get-create-writeable-dir
	  (handle-exceptions
	      exn
	      (begin
		(debug:print-error 0 *default-log-port* "Couldn't create path to " *db-cache-path*)
		(debug:print-error 0 *default-log-port* "Couldn't create path to " *db-cache-path* ", exn=" exn)
		(exit 1))
	      (let* ((tsname (common:get-testsuite-name))
	    (let ((dbpath (common:get-create-writeable-dir
			   (list (conc "/tmp/" (current-user-name)
				       "/megatest_localdb/"
				       (common:get-testsuite-name) "/"
				       (string-translate *toppath* "/" ".")))))) ;;  #t))))
	      (set! *db-cache-path* dbpath)
	      dbpath))
		     (dbpath (common:get-create-writeable-dir
			      (list (conc "/tmp/" (current-user-name)
					  "/megatest_localdb/"
					  tsname "/"
					  (string-translate *toppath* "/" "."))
				    (conc "/tmp/" (current-process-id) ;; just in case we have an issue with the dir by own user name
					  "/megatest_localdb/"
					  tsname
					  (string-translate *toppath* "/" "."))
				    ))))
		(set! *db-cache-path* dbpath)
		dbpath))
	  #f)))

(define (common:get-area-path-signature)
  (message-digest-string (md5-primitive) *toppath*))

(define (common:get-signature str)
  (message-digest-string (md5-primitive) str))

		 (tal (cdr dirs)))
	(let ((res (or (and (directory? hed)
			    (file-write-access? hed)
			    hed)
		       (handle-exceptions
			   exn
			   (begin
			     (debug:print-info 0 *default-log-port* "could not create " hed ", this might cause problems down the road.")
			     (debug:print-info 0 *default-log-port* "could not create " hed
					       ", this might cause problems down the road. exn=" exn)
			     #f)
			(create-directory hed #t)))))
	  (if (and (string? res)
		   (directory? res))
	      res
	      (if (null? tal)
		  #f

      rtestpatt)
     (else 
      (debug:print-info 0 *default-log-port* "using testpatt " args-testpatt " rtestpatt:" rtestpatt)
      args-testpatt))))



;; 
(define (common:false-on-exception thunk #!key (message #f))
  (handle-exceptions exn
(define (common:false-on-exception thunk #!key (message #f)(tries 1))
  (handle-exceptions
      exn
                     (begin
                       (if message
                           (debug:print-info 0 *default-log-port* message))
                       #f) (thunk) ))
          (debug:print-info 0 *default-log-port* message " exn=" exn))
      (if (> tries 1)
	  (begin
	    (thread-sleep! 1)
	    (common:false-on-exception thunk message: message tries: (- tries 1)))
	  #f))
    (thunk)))

(define (common:file-exists? path-string #!key (silent #f))
  ;; this avoids stack dumps in the case where 
(define (common:file-exists? path-string #!key (silent #f)(tries 1))
  ;; this avoids stack dumps in the case where NFS is slow or flakey

  ;;;; TODO: catch permission denied exceptions and emit appropriate warnings, eg:  system error while trying to access file: "/nfs/pdx/disks/icf_env_disk001/bjbarcla/gwa/issues/mtdev/randy-slow/reproduce/q...
  (common:false-on-exception (lambda () (file-exists? path-string))
  (common:false-on-exception
   (lambda ()(file-exists? path-string))
                             message: (if (not silent)
                                          (conc "Unable to access path: " path-string)
                                          #f)
   tries: tries
                             ))

(define (common:directory-exists? path-string)
  ;;;; TODO: catch permission denied exceptions and emit appropriate warnings, eg:  system error while trying to access file: "/nfs/pdx/disks/icf_env_disk001/bjbarcla/gwa/issues/mtdev/randy-slow/reproduce/q...
  (common:false-on-exception (lambda () (directory-exists? path-string))
                             message: (conc "Unable to access path: " path-string)
                             ))

;; does the directory exist and do we have write access?
;;
;;    returns the directory or #f
;;
(define (common:directory-writable? path-string)
  (handle-exceptions
   exn
   #f
      exn
    (begin
      (debug:print 0 *default-log-port* "Failed to identify access to " path-string ", exn=" exn)
      #f)
   (if (and (directory-exists? path-string)
            (file-write-access? path-string))
       path-string
       #f)))

(define (common:get-linktree)
  (or (getenv "MT_LINKTREE")

	   ;; first look in config, then look in file .homehost, create it if not found
	   (homehost (or (configf:lookup *configdat* "server" "homehost" )
			 (handle-exceptions
			     exn
			     (if (> trynum 0)
				 (let ((delay-time (* (- 5 trynum) 5)))
				   (mutex-unlock! *homehost-mutex*)
				   (debug:print 0 *default-log-port* "ERROR: ["(common:human-time)"] Failed to read .homehost file, delaying " delay-time " seconds and trying again, message: "  ((condition-property-accessor 'exn 'message) exn))
				   (debug:print 0 *default-log-port* "ERROR: ["(common:human-time)"] Failed to read .homehost file, delaying "
						delay-time " seconds and trying again, message: "  ((condition-property-accessor 'exn 'message) exn)
						", exn=" exn)
				   (thread-sleep! delay-time)
				   (common:get-homehost trynum: (- trynum 1)))
				 (begin
				   (mutex-unlock! *homehost-mutex*)
				   (debug:print 0 *default-log-port* "ERROR: ["(common:human-time)"] Failed to read .homehost file after trying five times. Giving up and exiting, message: "  ((condition-property-accessor 'exn 'message) exn))
				   (debug:print 0 *default-log-port* "ERROR: ["(common:human-time)
						"] Failed to read .homehost file after trying five times. Giving up and exiting, message: "
						((condition-property-accessor 'exn 'message) exn))
				   (exit 1)))
			   (let ((hhf (conc *toppath* "/.homehost")))
			     (if (common:file-exists? hhf)
				 (with-input-from-file hhf read-line)
				 (if (file-write-access? *toppath*)
				     (begin
				       (with-output-to-file hhf

;;======================================================================

;; lazy-safe get file mod time. on any error (file not existing etc.) return 0
;;
(define (common:lazy-modification-time fpath)
  (handle-exceptions
      exn
    (begin
      (debug:print 0 *default-log-port* "Failed to get modifcation time for " fpath ", treating it as zero. exn=" exn)
      0
    (file-modification-time fpath)))
      0)
    (if (file-exists? fpath)
	(file-modification-time fpath)
	0)))

;; find timestamp of newest file associated with a sqlite db file
(define (common:lazy-sqlite-db-modification-time fpath)
  (let* ((glob-list (handle-exceptions
			exn
		      (begin
			(debug:print 0 *default-log-port* "Failed to glob " fpath "*, exn=" exn)
			`(,(conc "/no/such/file, message: " ((condition-property-accessor 'exn 'message) exn)))
			`(,(conc "/no/such/file, message: " ((condition-property-accessor 'exn 'message) exn))))
		      (glob (conc fpath "*"))))
         (file-list (if (eq? 0 (length glob-list))
			'("/no/such/file")
			glob-list)))
  (apply max
   (map
    common:lazy-modification-time 
    file-list))))
	 (map
	  common:lazy-modification-time 
	  file-list))))

;; return a nice clean pathname made absolute
(define (common:nice-path dir)
  (let ((match (string-match "^(~[^\\/]*)(\\/.*|)$" dir)))
    (if match ;; using ~ for home?
	(common:nice-path (conc (common:read-link-f (cadr match)) "/" (caddr match)))
	(normalize-pathname (if (absolute-pathname? dir)
				dir
				(conc (current-directory) "/" dir))))))

;; make "nice-path" available in config files and the repl
(define nice-path common:nice-path)

(define (common:read-link-f path)
  (handle-exceptions
      exn
      (begin
	(debug:print-error 0 *default-log-port* "command \"/bin/readlink -f " path "\" failed.")
	(debug:print-error 0 *default-log-port* "command \"/bin/readlink -f " path "\" failed. exn=" exn)
	path) ;; just give up
    (with-input-from-pipe
	(conc "/bin/readlink -f " path)
      (lambda ()
	(read-line)))))

;; returns *effective load* (not normalized)
;;
(define (common:get-intercept onemin fivemin)
  (if (< onemin fivemin) ;; load is decreasing, just use the onemin load
      onemin
      (let* ((load-change (- onemin fivemin))
	     (tchange     (- 300 60)))
	(max (+ onemin (* 60 (/ load-change tchange))) 0))))

;; calculate a delay number based on a droop curve
;;    inputs are:
;;      - load-in, load as from uptime, NOT normalized
;;      - numcpus, number of cpus, ideally use the real cpus, not threads
;;
(define (common:get-delay load-in numcpus)
  (let* ((ratio (/ load-in numcpus))
         (new-option (configf:lookup *configdat* "load" "new-load-method"))
	 (paramstr   (or (configf:lookup *configdat* "load" "exp-params")
			 "15 12 1281453987.9543 0.75")) ;; 5 4 10 1"))
	 (paramlst   (map string->number (string-split paramstr))))
    (if new-option
	(begin
          (cond ((and (>= ratio 0) (< ratio .5))
		 0)
		((and (>= ratio 0.5) (<= ratio .9))
		 (* ratio (/ 5 .9)))
		((and (> ratio .9) (<= ratio 1.1))
		 (+ 5 (* (- ratio .9) (/ 55 .2)))) 
		((> ratio 1.1)
		 60)))
	(match paramlst
	  ((r1 r2 s1 s2)
	   (debug:print 3 *default-log-port* "Using params r1=" r1 " r2=" r2 " s1=" s1 " s2=" s2)
	   (min (max (/ (expt r1 (* r2 s2 ratio)) s1) 0) 30))
	  (else
	   (debug:print 0 *default-log-port* "BAD exp-params, should be \"r1 r2 s1 s2\" but got " paramstr)
	   30)))))

(define (common:print-delay-table)
  (let loop ((x 0))
    (print x "," (common:get-delay x 1))
    (if (< x 2)
	(loop (+ x 0.1)))))

(define (get-cpu-load #!key (remote-host #f))
  (car (common:get-cpu-load remote-host)))
;;   (let* ((load-res (process:cmd-run->list "uptime"))
;; 	 (load-rx  (regexp "load average:\\s+(\\d+)"))
;; 	 (cpu-load #f))
;;     (for-each (lambda (l)
;; 		(let ((match (string-search load-rx l)))
;; 		  (if match
;; 		      (let ((newval (string->number (cadr match))))
;; 			(if (number? newval)
;; 			    (set! cpu-load newval))))))
;; 	      (car load-res))
;;     cpu-load))

;; get values from cached info from dropping file in logs dir
;;  e.g. key is host and dtype is normalized-load
;;
(define (common:get-cached-info key dtype #!key (age 10))
  (if *toppath*
      (let* ((fullpath (conc *toppath* "/.sysdata/" key "-" dtype ".log")))
      (let* ((fullpath (conc *toppath* "/.sysdata/" key "-" dtype ".log"))
	     (delfile  (lambda ()
			 (debug:print-info 1 *default-log-port* " removing bad file " fullpath ", exn=" exn)
			 (delete-file* fullpath)
			 #f)))
	(if (and (file-exists? fullpath)
		 (file-read-access? fullpath))
	    (handle-exceptions
	     exn
	     #f
		exn
	      (begin
		(debug:print 0 *default-log-port* "failed to get cached info from " fullpath ", exn=" exn)
		#f)
	     (debug:print 2 *default-log-port* "reading file " fullpath)
	     (let ((real-age (- (current-seconds)(file-change-time fullpath)))) 
	     (let ((real-age (- (current-seconds)
				(handle-exceptions
				    exn
				  (begin
				    (debug:print 1 *default-log-port* "Failed to read mod time on file "
						 fullpath ", using 0, exn=" exn)
				    0)
				  (file-change-time fullpath)))))
	       (if (< real-age age)
		   (handle-exceptions
		       exn
		     (delfile)
		     (let* ((res (with-input-from-file fullpath read)))
		       (if (eof-object? res)
		     (begin
			   (begin
		       (debug:print-info 1 *default-log-port* " removing bad file " fullpath)
		       (delete-file* fullpath)
		       #f)
		     (with-input-from-file fullpath read))
			     (delfile)
			     #f)
			   res)))
		   (begin
		     (debug:print-info 2 *default-log-port* "file " fullpath " is too old (" real-age" seconds) to trust, skipping reading it")
		     (debug:print-info 2 *default-log-port* "file " fullpath
				       " is too old (" real-age" seconds) to trust, skipping reading it")
		     #f))))
	    (begin
	      (debug:print 2 *default-log-port* "not reading file " fullpath)
	      #f)))
      #f))
 
(define (common:write-cached-info key dtype dat)
  (if *toppath*
      (let* ((fulldir  (conc *toppath* "/.sysdata"))
	     (fullpath (conc fulldir "/" key "-" dtype ".log")))
	(if (not (file-exists? fulldir))(create-directory fulldir #t))
	(handle-exceptions
	 exn
	 #f
	 (with-output-to-file fullpath (lambda ()(pp dat)))))
	    exn
	  (begin
	    (debug:print 0 *default-log-path* "failed to write file " fullpath ", exn=" exn)
	    #f)
	  (with-output-to-file fullpath (lambda ()(pp dat)))))
      #f))
  

(define (common:raw-get-remote-host-load remote-host)
  (handle-exceptions
      exn
    (begin
      (debug:print 0 *default-log-port* "failed to ssh to " remote-host " and get loadavg. exn=" exn)
      #f) ;; more specific handling of errors needed
    (with-input-from-pipe 
     (conc "ssh " remote-host " cat /proc/loadavg")
     (lambda ()(list (read)(read)(read))))))

;; get cpu load by reading from /proc/loadavg, return all three values
;;
(define (common:get-cpu-load remote-host)
  (handle-exceptions
   exn
   '(99 99 99)
   (let* ((actual-hostname (or remote-host (get-host-name) "localhost")))
     (or (common:get-cached-info actual-hostname "cpu-load")
	 (let ((result (if remote-host
			   (map (lambda (res)
				  (if (eof-object? res) 9e99 res))
			        (with-input-from-pipe 
				 (conc "ssh " remote-host " cat /proc/loadavg")
				 (lambda ()(list (read)(read)(read)))))
			   (with-input-from-file "/proc/loadavg" 
			     (lambda ()(list (read)(read)(read)))))))
	   (common:write-cached-info actual-hostname "cpu-load" result)
	                     result)))))
      exn
    (begin
      (debug:print 0 *default-log-port* "failed to ssh or read loadavg from host " remote-host ", exn=" exn)
      '(-99 -99 -99))
    (let* ((actual-hostname (or remote-host (get-host-name) "localhost")))
      (or (common:get-cached-info actual-hostname "cpu-load")
	  (let ((result (if remote-host
			    (map (lambda (res)
				   (if (eof-object? res) 9e99 res))
			         (with-input-from-pipe 
				  (conc "ssh " remote-host " cat /proc/loadavg")
				  (lambda ()(list (read)(read)(read)))))
			    (with-input-from-file "/proc/loadavg" 
			      (lambda ()(list (read)(read)(read)))))))
	    (match
		result
	      ((l1 l2 l3)
	       (if (and (number? l1)
		      (number? l2)
		      (number? l3))
		   (begin
		     (common:write-cached-info actual-hostname "cpu-load" result)
		     result)
		   '(-1 -1 -1))) ;; -1 is bad result
	      (else '(-2 -2 -2))))))))

;; get normalized cpu load by reading from /proc/loadavg and /proc/cpuinfo return all three values and the number of real cpus and the number of threads
;; returns alist '((adj-cpu-load . normalized-proc-load) ... etc.
;;  keys: adj-proc-load, adj-core-load, 1m-load, 5m-load, 15m-load
;;
(define (common:get-normalized-cpu-load remote-host)
  (let ((res (common:get-normalized-cpu-load-raw remote-host))

	      (if new-best
		  (begin ;; found a host, return it
		    (debug:print 0 *default-log-port* "INFO: Found host: " new-best " load: " load " last-used: " delta " seconds ago, with job-rate: " job-rate)
		    (host-last-used-set! rec curr-time)
		    new-best)
		  (if (null? tal) #f (loop (car tal)(cdr tal) best-host)))))))))

(define (common:wait-for-homehost-load maxload msg)
(define (common:wait-for-homehost-load maxnormload msg)
  (let* ((hh-dat (if (common:on-homehost?) ;; if we are on the homehost then pass in #f so the calls are local.
                     #f
                     (common:get-homehost)))
         (hh     (if hh-dat (car hh-dat) #f)))
    (common:wait-for-normalized-load maxload msg hh)))
    (common:wait-for-normalized-load maxnormload msg hh)))

(define *numcpus-cache* (make-hash-table))
(define (common:get-num-cpus remote-host)
  (let* ((actual-host (or remote-host (get-host-name))))
    ;; hosts had better not be changing the number of cpus too often!
    (or (hash-table-ref/default *numcpus-cache* actual-host #f)
    (or (common:get-cached-info actual-host "num-cpus" age: (+ 2592000 (random 3600))) ;; hosts had better not be changing the number of cpus too often!
	(let* ((proc   (lambda ()
			 (let loop ((numcpu 0)
				    (inl    (read-line)))
			   (if (eof-object? inl)
			       (if (> numcpu 0)
				   numcpu
				   #f) ;; if zero return #f so caller knows that things are not working
			       (loop (if (string-match "^processor\\s+:\\s+\\d+$" inl)
					 (+ numcpu 1)
					 numcpu)
				     (read-line))))))
	       (result (if remote-host
			   (with-input-from-pipe 
			       (conc "ssh " remote-host " cat /proc/cpuinfo")
			     proc)
			   (with-input-from-file "/proc/cpuinfo" proc))))
	  (if (and (number? result)
		   (> result 0))
	      (common:write-cached-info actual-host "num-cpus" result))
	  result))))
	(let* ((numcpus (or (common:get-cached-info actual-host "num-cpus" age: (+ 2592000 (random 3600)))
			    (let* ((proc   (lambda ()
					     (let loop ((numcpu 0)
							(inl    (read-line)))
					       (if (eof-object? inl)
						   (if (> numcpu 0)
						       numcpu
						       #f) ;; if zero return #f so caller knows that things are not working
						   (loop (if (string-match "^processor\\s+:\\s+\\d+$" inl)
							     (+ numcpu 1)
							     numcpu)
							 (read-line))))))
				   (result (if remote-host
					       (with-input-from-pipe 
						(conc "ssh " remote-host " cat /proc/cpuinfo")
						proc)
					       (with-input-from-file "/proc/cpuinfo" proc))))
			      (if (and (number? result)
				       (> result 0))
				  (common:write-cached-info actual-host "num-cpus" result))
			      result))))
	  (hash-table-set! *numcpus-cache* actual-host numcpus)
	  numcpus))))

;; wait for normalized cpu load to drop below maxload
;;
(define (common:wait-for-normalized-load maxload msg remote-host #!optional (rem-tries 5))
(define (common:wait-for-normalized-load maxnormload msg remote-host #!optional (rem-tries 5))
  (let ((num-cpus (common:get-num-cpus remote-host)))
    (if num-cpus
	(common:wait-for-cpuload maxload num-cpus 15 msg: msg remote-host: remote-host)
	(common:wait-for-cpuload maxnormload num-cpus 15 msg: msg remote-host: remote-host)
	(begin
	  (thread-sleep! (random 60)) ;; we failed to get num cpus. wait a bit and try again
	  (if (> rem-tries 0)
	      (common:wait-for-normalized-load maxload msg remote-host (- rem-tries 1))
	      (common:wait-for-normalized-load maxnormload msg remote-host (- rem-tries 1))
	      #f)))))

;; DO NOT CALL THIS DIRECTLY. It is called from common:wait-for-normalized-load
;;   count     - count down to zero, at some point we'd give up if the load never drops
;;   num-tries - count down to zero number tries to get numcpus
;;
(define (common:wait-for-cpuload maxload-in numcpus-in waitdelay #!key (count 1000) (msg #f)(remote-host #f)(force-maxload #f))
  (let* ((loadavg (common:get-cpu-load remote-host))
	 (numcpus (if (<= 1 numcpus-in) ;; not possible to have zero.  If we get 1, it's possible that we got the previous default, and we should check again
		      (common:get-num-cpus remote-host)
		      numcpus-in))
(define (common:wait-for-cpuload maxnormload numcpus-in
				 #!key (count 1000)
				 (msg #f)(remote-host #f)(num-tries 5))
  (let* ((loadavg     (common:get-cpu-load remote-host))
	  ;; not possible to have zero.  If we get 1, it's possible that we got the previous default, and we should check again
	 (numcpus     (if (<= 1 numcpus-in)
			  (common:get-num-cpus remote-host)
			  numcpus-in))
	 (maxload (if force-maxload
		      maxload-in
		      (if (number? maxload-in)
			  (max maxload-in 0.5)
			  0.5))) ;; so maxload must be greater than 0.5 for now BUG - FIXME?
	 (first   (car loadavg))
	 (next    (cadr loadavg))
	 (adjload (* maxload (max 1 numcpus))) ;; possible bug where numcpus (or could be maxload) is zero, crude fallback is to at least use 1
	 (loadjmp (- first next))
	 (first       (car loadavg))
	 (next        (cadr loadavg))
	 (adjmaxload  (* maxnormload (max 1 numcpus))) ;; possible bug
						   ;; where numcpus
						   ;; (or could be
						   ;; maxload) is
         (adjwait (min (+ 300 (random 10)) (abs (* (+ (random 10)(/ (- 1000 count) 10) waitdelay) (- first adjload) ))  )));; add some randomness to the time to break any alignment where netbatch dumps many jobs to machines simultaneously
    ;; let's let the user know once in a long while that load checking is happening but not constantly report it
    (if (> (random 100) 75) ;; about 25% of the time
	(debug:print-info 1 *default-log-port* "Checking cpuload on " (or remote-host "localhost") ", maxload: " maxload
			  ", load: " first ", adjload: " adjload ", loadjmp: " loadjmp))
						   ;; zero, crude
						   ;; fallback is to
						   ;; at least use 1
	 ;; effective load accounts for load jumps, this should elminate all the first-next-avg, adjwait, load-jump-limit
	 ;; etc.
	 (effective-load    (common:get-intercept first next))
	 (recommended-delay (common:get-delay effective-load numcpus))
	 (effective-host    (or remote-host "localhost"))
	 (normalized-effective-load (/ effective-load numcpus))
	 (will-wait                 (> normalized-effective-load maxnormload)))
    (if (> recommended-delay 0)
	(let* ((actual-delay (min recommended-delay 30)))
	  (if (common:low-noise-print 30 (conc (round actual-delay) "-safe-load"))
	      (debug:print-info 0 *default-log-port* "Load control, delaying "
				actual-delay " seconds to maintain safe load. current normalized effective load is "
				normalized-effective-load"."))
	  (thread-sleep! actual-delay)))
    
    (cond
     ;; bad data, try again to get the data
     ((not will-wait)
      (if (common:low-noise-print 30 (conc (round normalized-effective-load) "-load-acceptable-" effective-host))
	  (debug:print 0 *default-log-port* "Effective load on " effective-host " is acceptable at " effective-load " continuing.")))
     ((and (> first adjload)
	   (> count 0))
      (debug:print-info 0 *default-log-port* "server start delayed " adjwait " seconds due to load " first " exceeding max of " adjload " on server " (or remote-host (get-host-name)) " (normalized load-limit: " maxload ") " (if msg msg ""))
      (thread-sleep! adjwait)
      (common:wait-for-cpuload maxload numcpus waitdelay count: (- count 1) msg: msg remote-host: remote-host))
     ((and (> loadjmp numcpus)
     ((and (< first 0) ;; this indicates the loadavg data is bad - machine may not be reachable
	   (> num-tries 0))
      (debug:print 0 *default-log-port* "WARNING: received bad data from get-cpu-load "
		   first ", we'll sleep 10s and try " num-tries " more times.")
      (thread-sleep! 10)
      (common:wait-for-cpuload maxnormload numcpus-in
			       count: count remote-host: remote-host num-tries: (- num-tries 1)))
     ;; need to wait for load to drop
     ((and will-wait ;; (> first adjmaxload)
	   (> count 0))
      (debug:print-info 0 *default-log-port* "waiting " adjwait " seconds due to load jump " loadjmp " > numcpus " numcpus (if msg msg ""))
      (thread-sleep! adjwait)
      (common:wait-for-cpuload maxload numcpus waitdelay count: (- count 1) msg: msg remote-host: remote-host)))))

      (debug:print-info 0 *default-log-port*
			"Delaying 15" ;; adjwait
			" seconds due to normalized effective load " normalized-effective-load ;; first
			" exceeding max of " adjmaxload
			" on server " (or remote-host (get-host-name))
			" (normalized load-limit: " maxnormload ") " (if msg msg ""))
      (thread-sleep! 15) ;; adjwait)
      (common:wait-for-cpuload maxnormload numcpus count: (- count 1) msg: msg remote-host: remote-host)
      ;; put the message here to indicate came out of waiting
      (debug:print-info 1 *default-log-port*
			"On host: " effective-host
			", effective load: " effective-load
			", numcpus: " numcpus
			", normalized effective load: " normalized-effective-load
			))
     ;; overloaded and count expired (i.e. went to zero)
     (else
      (if (> num-tries 0) ;; should be "num-tries-left". 
	  (if (common:low-noise-print 30 (conc (round effective-load) "-load-acceptable-" effective-host))
	      (debug:print 0 *default-log-port* "Load on " effective-host " is acceptable at effective normalized load of "
			   effective-normalized-load "  continuing."))
	  (debug:print 0 *default-log-port* "Load on " effective-host ", "
		       first" could not be retrieved. Giving up and continuing."))))))

;; DO NOT CALL THIS DIRECTLY. It is called from common:wait-for-normalized-load
;;
;; (define (common:wait-for-cpuload maxload-in numcpus-in waitdelay #!key (count 1000) (msg #f)(remote-host #f)(force-maxload #f)(num-tries 5))
;;   (let* ((loadavg (common:get-cpu-load remote-host))
;; 	 (numcpus (if (<= 1 numcpus-in) ;; not possible to have zero.  If we get 1, it's possible that we got the previous default, and we should check again
;; 		      (common:get-num-cpus remote-host)
;; 		      numcpus-in))
;; 	 (maxload (if force-maxload
;; 		      maxload-in
;; 		      (if (number? maxload-in)
;; 			  (max maxload-in 0.5)
;; 			  0.5))) ;; so maxload must be greater than 0.5 for now BUG - FIXME?
;; 	 (first   (car loadavg))
;; 	 (next    (cadr loadavg))
;; 	 (adjmaxload (* maxload (max 1 numcpus))) ;; possible bug where
;; 					       ;; numcpus (or could be
;; 					       ;; maxload) is zero,
;; 					       ;; crude fallback is to
;; 					       ;; at least use 1
;; 	 (loadjmp (- first (if (> next (* numcpus 0.7)) ;; could do something with average of first and next?
;; 			       0
;; 			       next))) ;; we will force a conservative calculation any time next is large.
;; 	 (first-next-avg    (/ (+ first next) 2))
;; 	 ;; add some randomness to the time to break any alignment
;; 	 ;; where netbatch dumps many jobs to machines simultaneously
;;          (adjwait           (min (+ 300 (random 10)) (abs (* (+ (random 10)
;; 								(/ (- 1000 count) 10)
;; 								waitdelay)
;; 							     (- first adjmaxload) ))))
;; 	 (load-jump-limit   (configf:lookup-number *configdat* "setup" "load-jump-limit"))
;; 	 ;; effective load accounts for load jumps, this should elminate all the first-next-avg, adjwait, load-jump-limit
;; 	 ;; etc.
;; 	 (effective-load    (common:get-intercept first next))
;; 	 (effective-host    (or remote-host "localhost"))
;; 	 (normalized-effective-load (/ effective-load numcpus))
;; 	 (will-wait                 (> normalized-effective-load maxload)))
;; 	 
;;     ;; let's let the user know once in a long while that load checking
;;     ;; is happening but not constantly report it
;;     #;(if (common:low-noise-print 30 (conc "cpuload" (or remote-host "localhost"))) ;; (> (random 100) 75) ;; about 25% of the time
;; 	(debug:print-info 1 *default-log-port* "Checking cpuload on " (or remote-host "localhost") ", maxload: " maxload
;;     ", load: " first ", adjmaxload: " adjmaxload ", loadjmp: " loadjmp))
;; 
;;     (debug:print-info 1 *default-log-port*
;; 		      "On host: " effective-host
;; 		      ", effective load: " effective-load
;; 		      ", numcpus: " numcpus
;; 		      ", normalized effective load: " normalized-effective-load
;; 		      )
;;     
;;     (cond
;;      ;; bad data, try again to get the data
;;      ((and (< first 0) ;; this indicates the loadavg data is bad - machine may not be reachable
;; 	   (> num-tries 0))
;;       (debug:print 0 *default-log-port* "WARNING: received bad data from get-cpu-load " first ", we'll sleep 10s and try " num-tries " more times.")
;;       (thread-sleep! 10)
;;       (common:wait-for-cpuload maxload-in numcpus-in waitdelay
;; 			       count: count remote-host: remote-host force-maxload: force-maxload num-tries: (- num-tries 1)))
;;      ;; need to wait for load to drop
;;      ((and will-wait ;; (> first adjmaxload)
;; 	   (> count 0))
;;       (debug:print-info 0 *default-log-port*
;; 			"Delaying " 15 ;; adjwait
;; 			" seconds due to normalized effective load " normalized-effective-load ;; first
;; 			" exceeding max of " adjmaxload
;; 			" on server " (or remote-host (get-host-name))
;; 			" (normalized load-limit: " maxload ") " (if msg msg ""))
;;       (thread-sleep! 15) ;; adjwait)
;;       (common:wait-for-cpuload maxload numcpus waitdelay count: (- count 1) msg: msg remote-host: remote-host))
;;      ((and (> loadjmp (cond
;; 		       (load-jump-limit load-jump-limit)
;; 		       ((> numcpus 8)(/ numcpus 2))
;; 		       ((> numcpus 4)(/ numcpus 1.2))
;; 		       (else 0.5)))
;; 	   (> count 0))
;;       (debug:print-info 0 *default-log-port* "waiting " adjwait " seconds due to possible load jump " loadjmp ". "
;; 			(if msg msg ""))
;;       (thread-sleep! adjwait)
;;       (common:wait-for-cpuload maxload numcpus waitdelay count: (- count 1) msg: msg remote-host: remote-host))
;;      (else
;;       (if (> num-tries 0)
;; 	  (if (common:low-noise-print 30 (conc (round first) "-load-acceptable-" (or remote-host "localhost")))
;; 	      (debug:print 0 *default-log-port* "Load on " (or remote-host "localhost") " is acceptable at " first " continuing."))
;; 	  (debug:print 0 *default-log-port* "Load on " (or remote-host "localhost") ", "first" could not be retrieved. Giving up and continuing."))))))
;; 
(define (get-uname . params)
  (let* ((uname-res (process:cmd-run->list (conc "uname " (if (null? params) "-a" (car params)))))
	 (uname #f))
    (if (null? (car uname-res))
	"unknown"
	(caar uname-res))))

	  (let ((pktsdir (or (hash-table-ref/default *pkts-info* 'pkts-dir #f)
			     (let* ((pktsdirs (common:get-pkts-dirs mtconf use-lt))
				    (pktsdir   (car pktsdirs))) ;; assume it is there
			       (hash-table-set! *pkts-info* 'pkts-dir pktsdir)
			       pktsdir))))
            (handle-exceptions
             exn
             (debug:print-info 0 "failed to write out packet to " pktsdir) ;; don't care if this failed for now but MUST FIX - BUG!!
             (debug:print-info 0 "failed to write out packet to " pktsdir ", exn=" exn) ;; don't care if this failed for now but MUST FIX - BUG!!
             (if (not (file-exists? pktsdir))
                 (create-directory pktsdir #t))
             (with-output-to-file
                 (conc pktsdir "/" uuid ".pkt")
               (lambda ()
                 (print pkt)))))))))
	

(define (common:join-backgrounded-threads)
  ;; may need to trap and ignore exceptions -- dunno how atomic threads are...
  (for-each
   (lambda (thread-name)
     (let* ((thread (hash-table-ref/default *common:thread-punchlist* thread-name #f)))
       (if thread
           (handle-exceptions
           exn
           #t ;; just ignore it, it might have died in the meantime so joining it will throw an exception
           (thread-join! thread))
               exn
	     (begin
	       (debug:print 0 *default-log-port* "joining threads failed. exn=" exn)
               #t) ;; just ignore it, it might have died in the meantime so joining it will throw an exception
             (thread-join! thread))
           )))
   (hash-table-keys *common:thread-punchlist*)))

(define *common:telemetry-log-state* 'startup)
(define *common:telemetry-log-socket* #f)

(define (common:telemetry-log-open)

    ((_ (name arg ...) body ...)
     (define-syntax name (syntax-rules () ((name arg ...) (begin body ...)))))))

;; (define-syntax common:handle-exceptions
;;   (syntax-rules ()
;;     ((_ exn-in errstmt ...)(handle-exceptions exn-in errstmt ...))))

;; this works, why didn't I use it more?
(define-syntax common:debug-handle-exceptions
  (syntax-rules ()
    ((_ debug exn errstmt body ...)
     (if debug
	 (begin body ...)
	 (handle-exceptions exn errstmt body ...)))))

    (not (null? (lset-intersection! eq? *verbosity* n))))
   ((and (number? *verbosity*)
	 (list? n))
    (member *verbosity* n))))

(define (debug:setup)
  (let ((debugstr (or (args:get-arg "-debug")
		      (args:get-arg "-debug-noprop")
		      (getenv "MT_DEBUG_MODE"))))
    (set! *verbosity* (debug:calc-verbosity debugstr))
    (debug:check-verbosity *verbosity* debugstr)
    ;; if we were handed a bad verbosity rule then we will override it with 1 and continue
    (if (not *verbosity*)(set! *verbosity* 1))
    (if (and (not (args:get-arg "-debug-noprop"))
    (if (or (args:get-arg "-debug")
	    (not (getenv "MT_DEBUG_MODE")))
	     (or (args:get-arg "-debug")
		 (not (getenv "MT_DEBUG_MODE"))))
	(setenv "MT_DEBUG_MODE" (if (list? *verbosity*)
				    (string-intersperse (map conc *verbosity*) ",")
				    (conc *verbosity*))))))
  
(define (debug:print n e . params)
  (if (debug:debug-mode n)
      (with-output-to-port (or e (current-error-port))

;;======================================================================

(declare (unit commonmod))

(module commonmod
	*
	
(import scheme chicken data-structures extras)
(import (prefix sqlite3 sqlite3:) posix typed-records srfi-18)
(import scheme chicken data-structures extras files)
(import (prefix sqlite3 sqlite3:) posix typed-records srfi-18 srfi-69
	md5 message-digest
	regex srfi-1)

;;======================================================================
;; CONTENTS
;;
;;  config file utils
;;  misc conversion, data manipulation functions
;;  testsuite and area utilites
;;
;;======================================================================

(include "megatest-version.scm")
(include "megatest-fossil-hash.scm")

(define (get-full-version)
  (conc megatest-version "-" megatest-fossil-hash))

(define (version-signature)
  (conc megatest-version "-" (substring megatest-fossil-hash 0 4)))


;;======================================================================
;; config file utils
;;======================================================================

(define (lookup cfgdat section var)
  (if (hash-table? cfgdat)
      (let ((sectdat (hash-table-ref/default cfgdat section '())))
	(if (null? sectdat)
	    #f
	    (let ((match (assoc var sectdat)))
	      (if match ;; (and match (list? match)(> (length match) 1))
		  (cadr match)
		  #f))
	    ))
      #f))

;; returns var key1=val1; key2=val2 ... as alist
(define (get-key-list cfgdat section var)
  ;; convert string a=1; b=2; c=a silly thing; d=
  (let ((valstr (lookup cfgdat section var)))
    (if valstr
	(val->alist valstr)
	'()))) ;; should it return empty list or #f to indicate not set?


(define (get-section cfgdat section)
  (hash-table-ref/default cfgdat section '()))

;;======================================================================
;; misc conversion, data manipulation functions
;;======================================================================

;; if it looks like a number -> convert it to a number, else return it
;;
(define (lazy-convert inval)
  (let* ((as-num (if (string? inval)(string->number inval) #f)))
    (or as-num inval)))

;; to '((a . 1)(b . 2)(c . "a silly thing")(d . ""))
;;
(define (val->alist val #!key (convert #f))
  (let ((val-list (string-split-fields ";\\s*" val #:infix)))
    (if val-list
	(map (lambda (x)
	       (let ((f (string-split-fields "\\s*=\\s*" x #:infix)))
		 (case (length f)
		   ((0) `(,#f))  ;; null string case
		   ((1) `(,(string->symbol (car f))))
		   ((2) `(,(string->symbol (car f)) .
			  ,(let ((inval (cadr f)))
			     (if convert (lazy-convert inval) inval))))
		   (else f))))
	     (filter (lambda (x)
		       (not (string-match "^\\s*" x)))
		     val-list))
	'())))

;;======================================================================
;; testsuite and area utilites
;;======================================================================

(define (get-testsuite-name toppath configdat)
  (or (lookup configdat "setup" "area-name")
      (lookup configdat "setup" "testsuite")
      (get-environment-variable "MT_TESTSUITE_NAME")
      (if (string? toppath)
          (pathname-file toppath)
          #f)))

(define (get-area-path-signature toppath #!optional (short #f))
  (let ((res (message-digest-string (md5-primitive) toppath)))
    (if short
	(substring res 0 4)
	res)))

(define (get-area-name configdat toppath #!optional (short #f))
  ;; look up my area name in areas table (future)
  ;; generate auto name
  (conc (get-area-path-signature toppath short)
	"-"
	(get-testsuite-name toppath configdat)))

;; need generic find-record-with-var-nmatching-val
;;
(define (path->area-record cfgdat path)
  (let* ((areadat (get-cfg-areas cfgdat))
	 (all     (filter (lambda (x)
			    (let* ((keyvals (cdr x))
				   (pth     (alist-ref 'path keyvals)))
			      (equal? path pth)))
			  areadat)))
    (if (null? all)
	#f
	(car all)))) ;; return first match

;; given a config return an alist of alists
;;   area-name => data
;;
(define (get-cfg-areas cfgdat)
  (let ((adat (get-section cfgdat "areas")))
    (map (lambda (entry)
	   `(,(car entry) . 
	     ,(val->alist (cadr entry))))
	 adat)))
	 
;; (define (debug:print . params) #f)
;; (define (debug:print-info . params) #f)
;; 
;; (define (set-functions dbgp dbgpinfo)
;;   (set! debug:print dbgp)
;;   (set! debug:print-info dbgpinfo))

(define (config:eval-string-in-environment str)
  ;; (if (or (string-null? str)
  ;;	  (equal? "!" (substring str 0 1))) ;; null string or starts with ! are preserved but NOT set in the environment
      str
      (handle-exceptions
       exn
       (begin
	 (debug:print-error 0 *default-log-port* "problem evaluating \"" str "\" in the shell environment")
	 (debug:print-error 0 *default-log-port* "problem evaluating \"" str "\" in the shell environment, exn=" exn)
	 #f)
       (let ((cmdres (process:cmd-run->list (conc "echo " str))))
	 (if (null? cmdres) ""
	     (caar cmdres))))) ;; )

;;======================================================================
;; Make the regexp's needed globally available

				((runconfigs-get rget) (conc "(lambda (ht)(runconfigs-get ht \"" cmd "\"))"))
				;; ((rget)           (conc "(lambda (ht)(runconfigs-get ht \"" cmd "\"))"))
				(else "(lambda (ht)(print \"ERROR\") \"ERROR\")"))))
		;; (print "fullcmd=" fullcmd)
		(handle-exceptions
		 exn
		 (begin
		   (debug:print 0 *default-log-port* "WARNING: failed to process config input \"" l "\"")
		   (debug:print 0 *default-log-port* "WARNING: failed to process config input \"" l "\", exn=" exn)
		   (debug:print 0 *default-log-port* " message: " ((condition-property-accessor 'exn 'message) exn))
		   ;; (print "exn=" (condition->list exn))
		   (set! result (conc "#{( " cmdtype ") " cmd "}, full expansion: " fullcmd)))
		 (if (or allow-system
			 (not (member cmdtype '("system" "shell" "sh"))))
		     (with-input-from-string fullcmd
		       (lambda ()

		       ;;	   (not (eq? allow-processing 'return-string)))
		       ((#t #f)
			(configf:process-line inl ht allow-processing))
		       ((return-string)
			inl)
		       (else
			(configf:process-line inl ht allow-processing)))))
	    (if (and (string? res)
		     (not (equal? (hash-table-ref/default settings "trim-trailing-spaces" "no") "no")))
	    (if (and (string? res)  ;; must set to "no" to force NOT trimming trailing spaces
		     (not (equal? (hash-table-ref/default settings "trim-trailing-spaces" "yes") "no")))
		(string-substitute "\\s+$" "" res)
		res))))))

(define (configf:cfgdat->env-alist section cfgdat-ht allow-system)
  (filter
   (lambda (pair)
     (let* ((var (car pair))

                                          (full-conf     (if (and (absolute-pathname? include-file) (file-exists? include-file))
                                                             include-file
                                                             (common:nice-path 
                                                              (conc (if curr-conf-dir
                                                                        curr-conf-dir
                                                                        ".")
                                                                    "/" include-file)))))
				     (let ((all-matches (sort (handle-exceptions exn (list) (glob full-conf)) string<=?)))
				     (let ((all-matches (sort (handle-exceptions exn
								(begin
								 (debug:print '(2 9) *default-log-port* "glob of " full-conf " gave no match. , exn=" exn)
								 (list))
								(glob full-conf)) string<=?)))
				       (if (null? all-matches)
					   (begin
					     (debug:print '(2 9) #f "INFO: include file(s) matching " include-file " not found (called from " path ")")
					     (debug:print 2 *default-log-port* "        " full-conf))
					   (for-each
					    (lambda (fpath)
					      ;; (push-directory conf-dir)

       (hash-table-set! ht (car section)(cdr section)))
     adat)
    ht))

;; if 
(define (configf:read-alist fname)
  (handle-exceptions
   exn
   #f
   (configf:alist->config
    (with-input-from-file fname read))))
      exn
    (begin
      (debug:print 0 *default-log-port* "read of alist " fname " failed. exn=" exn)
      #f)
    (configf:alist->config
     (with-input-from-file fname read))))

(define (configf:write-alist cdat fname)
  (if (not (common:faux-lock fname))
      (debug:print 0 *default-log-port* "INFO: Could not get lock on " fname))
  (let* ((dat  (configf:config->alist cdat))
         (res
          (begin
            (with-output-to-file fname ;; first write out the file
              (lambda ()
                (pp dat)))
            
            (if (common:file-exists? fname)   ;; now verify it is readable
                (if (configf:read-alist fname)
                    #t ;; data is good.
                    (begin
                      (handle-exceptions
                       exn
                       #f
                       (debug:print 0 *default-log-port* "WARNING: content " dat " for cache " fname " is not readable. Deleting generated file.")
                       (delete-file fname))
			  exn
			(begin
			  (debug:print 0 *default-log-port* "deleting " fname " failed, exn=" exn)
			  #f)
			(debug:print 0 *default-log-port* "WARNING: content " dat " for cache " fname " is not readable. Deleting generated file.")
			(delete-file fname))
                      #f))
                #f))))
    (common:faux-unlock fname)
    res))
  
;; convert hierarchial list to ini format
;;

#!/bin/bash

#  Copyright 2006-2017, Matthew Welland.
# 
# This file is part of Megatest.
# 
#     Megatest is free software: you can redistribute it and/or modify
#     it under the terms of the GNU General Public License as published by
#     the Free Software Foundation, either version 3 of the License, or
#     (at your option) any later version.
# 
#     Megatest is distributed in the hope that it will be useful,
#     but WITHOUT ANY WARRANTY; without even the implied warranty of
#     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#     GNU General Public License for more details.
# 
#     You should have received a copy of the GNU General Public License
#     along with Megatest.  If not, see <http://www.gnu.org/licenses/>.

# Configure the build

if [[ "$1"x == "x" ]];then
    PREFIX=$PWD
else
    PREFIX=$1
fi


#======================================================================
# Configure stuff needed for eggs
#======================================================================

function configure_dependencies () {

    #======================================================================
    # libnanomsg
    #======================================================================

    if [[ ! $(ls /usr/lib/*/libnanomsg*) ]];then
	echo "libnanomsg build needed."
	echo "BUILD_NANOMSG=yes" >> makefile.inc
    fi

    #======================================================================
    # postgresql libraries
    #======================================================================

    if [[ ! $(ls /usr/lib/*/libpq.*) ]];then
	echo "Postgresql build needed."
	echo "BUILD_POSTGRES=yes" >> makefile.inc
    fi

    if [[ ! $(ls /usr/lib/*/libsqlite3.*) ]];then
    echo "Sqlite3 build needed."
	echo "BUILD_SQLITE3=yes" >> makefile.inc
    fi

}

#======================================================================
# Initialize makefile.inc
#======================================================================

echo "" > makefile.inc

#======================================================================
# Do we need Chicken?
#======================================================================

if [[ -e /usr/bin/sw_vers ]]; then
    ARCHSTR=$(/usr/bin/sw_vers -productVersion)
else
    ARCHSTR=$(lsb_release -sr)
fi

echo "CHICKEN_PREFIX=$PREFIX/.$ARCHSTR" >> makefile.inc
CHICKEN_PREFIX=$PREFIX/bin/.$ARCHSTR

if [[ ! $(type csi) ]];then
    echo "Chicken build needed."
    echo "BUILD_CHICKEN=yes" >> makefile.inc
    configure_dependencies
    echo "include chicken.makefile" >> makefile.inc
else
    echo "CSIPATH=$(which csi)" >> makefile.inc
    CSIPATH=$(which csi)
    echo "CKPATH=$(dirname $(dirname $CSIPATH))" >> makefile.inc
fi

# Make setup scripts
echo "#!/bin/bash" > setup.sh
echo "export PATH=$CHICKEN_PREFIX/bin:\$PATH" >> setup.sh
echo "export LD_LIBRARY_PATH=$CHICKEN_PREFIX/lib" >> setup.sh
echo 'exec "$@"' >> setup.sh
chmod a+x setup.sh

echo "setenv PATH $CHICKEN_PREFIX/bin:\$PATH" > setup.csh
echo "setenv LD_LIBRARY_PATH $CHICKEN_PREFIX/lib" >> setup.csh

echo "All done creating makefile.inc, feel free to edit it!"
echo "run \"setup.sh bash\" or source setup.csh to get PATH and LD_LIBRARY_PATH adjusted"

;;======================================================================
;; Copyright 2019, Matthew Welland.
;; 
;; This file is part of Megatest.
;; 
;;     Megatest is free software: you can redistribute it and/or modify
;;     it under the terms of the GNU General Public License as published by
;;     the Free Software Foundation, either version 3 of the License, or
;;     (at your option) any later version.
;; 
;;     Megatest is distributed in the hope that it will be useful,
;;     but WITHOUT ANY WARRANTY; without even the implied warranty of
;;     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
;;     GNU General Public License for more details.
;; 
;;     You should have received a copy of the GNU General Public License
;;     along with Megatest.  If not, see <http://www.gnu.org/licenses/>.

;;======================================================================

(declare (unit cookie))

(include "stml2/cookie.scm")

(declare (uses subrun))

(include "common_records.scm")
(include "db_records.scm")
(include "run_records.scm")

(define (dboard:launch-testpanel run-id test-id)
  (let* (;; (cfg-sh  (conc *common:this-exe-dir* "/cfg.sh"))
         ;; (cmd (conc
  (let* ((dboardexe (common:find-local-megatest "dashboard"))
         (cmd (conc dboardexe
         ;;       (if (common:file-exists? cfg-sh)
         ;;           (conc "source "cfg-sh" && ")
         ;;           "")
         ;;       *common:this-exe-fullpath*
         ;;       " -test " run-id "," test-id
         ;;       " &"))
         (cmd (conc *common:this-exe-dir*"/../dashboard "
                    "-test " run-id "," test-id
                    " -test " run-id "," test-id
                    " &")))
    (system cmd)))


(define (dashboard:run-menu-items  run-id test-id target runname test-name testpatt item-test-path test-info)
  (list
   (iup:menu-item

                     (runs:get-mt-env-alist run-id run-name target test-name item-path)
                     
                     (lambda ()
                       (if scheme-match
                           (begin
                             (handle-exceptions
                              exn
                              (print "error with custom menu scheme")
                              (print "error with custom menu scheme, exn=" exn)
                              (begin
                                ;;(BB> "gonna eval it!")
                                (eval (with-input-from-string (cadr scheme-match) read)))))
                           (common:run-a-command command-line with-vars: #t))))))))
             #f)))
     vars)))

			    "/"))
	       (item-path  (db:test-get-item-path testdat))
	       ;; this next block was added to fix a bug where variables were
               ;; needed. Revisit this.
	       (runconfig  (let ((runconfigf (conc  *toppath* "/runconfigs.config"))) ;; no rush but it would be good to convert this call to use runconfig:read
	 		     (if (common:file-exists? runconfigf)
	 			 (handle-exceptions
                                   exn
                                   #f  ;; do nothing, just keep on trucking ....
                                     exn
				   (begin
				     (debug:print 0 *default-log-port* "failed to set up environment for " runconfigf ", exn=" exn)
                                     #f)  ;; do nothing, just keep on trucking ....
                                   (setup-env-defaults runconfigf run-id (make-hash-table) keydat environ-patt: keystring))
	 			 (make-hash-table))))
	       (testconfig    (begin
				;; (runs:set-megatest-env-vars run-id inrunname: runname testname: test-name itempath: item-path)
				(runs:set-megatest-env-vars run-id inkeyvals: keydat inrunname: runname intarget: keystring testname: testname itempath: item-path) ;; these may be needed by the launching process
				(handle-exceptions
				 exn  ;; NOTE: I've no idea why this was written this way. Research, study and fix needed!
				 (tests:get-testconfig (db:test-get-testname testdat) (db:test-get-item-path testdat) test-registry #f allow-write-cache: #f)
				 (tests:get-testconfig (db:test-get-testname testdat) item-path test-registry #t allow-write-cache: #f))))
				    exn  ;; NOTE: I've no idea why this was written this way. Research, study and fix needed!
				  (begin
				    (debug:print 0 *default-log-port* "testconfig load using " item-path " failed, trying " (db:test-get-item-path testdat) ", exn=" exn)
				    (tests:get-testconfig (db:test-get-testname testdat) (db:test-get-item-path testdat) test-registry #f allow-write-cache: #f))
				  (tests:get-testconfig (db:test-get-testname testdat) item-path test-registry #t allow-write-cache: #f))))
	       (viewlog    (lambda (x)
			     (if (common:file-exists? logfile)
					;(system (conc "firefox " logfile "&"))
				 (dcommon:run-html-viewer logfile)
				 (message-window (conc "File " logfile " not found")))))
	       (view-a-log (lambda (lfile) 
			     (let ((lfilename (conc rundir "/" lfile)))

				    (need-update   (or (and (>= curr-mod-time db-mod-time)
							    (> (current-milliseconds)(+ last-update 250))) ;; every half seconds if db touched
						       (> (current-milliseconds)(+ last-update 10000))     ;; force update even 10 seconds
						       request-update))
				    (newtestdat (if need-update 
						    ;; NOTE: BUG HIDER, try to eliminate this exception handler
						    (handle-exceptions
						     exn 
						     (debug:print-info 0 *default-log-port* "test db access issue in examine test for run-id " run-id ", test-id " test-id ": " ((condition-property-accessor 'exn 'message) exn))
						     (rmt:get-test-info-by-id run-id test-id )))))
							exn
						      (begin
							(debug:print-info 0 *default-log-port* "test db access issue in examine test for run-id " run-id
									  ", test-id " test-id ": " ((condition-property-accessor 'exn 'message) exn) ", exn=" exn)
							#f)
						      (rmt:get-test-info-by-id run-id test-id)))))
			       ;; (print "INFO: need-update= " need-update " curr-mod-time = " curr-mod-time)
			       (cond
				((and need-update newtestdat)
				 (set! testdat newtestdat)
				 (set! teststeps    (augment-teststeps (tests:get-compressed-steps run-id test-id)))
				 (set! logfile      (conc (db:test-get-rundir testdat) "/" (db:test-get-final_logf testdat)))
				 (set! rundir       ;; (filedb:get-path *fdb* 

(declare (uses dashboard-guimonitor))
(declare (uses tree))
(declare (uses dcommon))
(declare (uses dashboard-context-menu))
(declare (uses vg))
(declare (uses subrun))
;; (declare (uses dashboard-main))
(declare (uses megatest-version))
(declare (uses mt))

(include "common_records.scm")
(include "db_records.scm")
(include "run_records.scm")
(include "task_records.scm")
(include "megatest-version.scm")
(include "megatest-fossil-hash.scm")
(include "vg_records.scm")

(define help (conc 
	      "Megatest Dashboard, documentation at http://www.kiatoa.com/fossils/megatest
  version " megatest-version "
  license GPL, Copyright (C) Matt Welland 2012-2017

			"-main"
			"-v"
			"-q"
			"-use-db-cache"
			"-skip-version-check"
			"-repl"
                        "-rh5.11" ;; fix to allow running on rh5.11

			"-:p"     ;; ignore the built in chicken profiling switch
			;; placeholder
			;; "-:p"
			)
		 args:arg-hash
		 0))

;; check for MT_* environment variables and exit if found
(if (not (args:get-arg "-test"))
    (begin

;;
(define (dboard:common-set-tabdat! commondat tabnum tabdat)
  (hash-table-set!
   (dboard:commondat-tabdats commondat)
   tabnum
   tabdat))

(define *updater-running* #f) ;; move this into one of the stucts

;; gets and calls updater list based on curr-tab-num
;;
(define (dboard:common-run-curr-updaters commondat #!key (tab-num #f))
  (if (dboard:common-get-tabdat commondat tab-num: tab-num) ;; only update if there is a tabdat
      (let* ((tnum     (or tab-num (dboard:commondat-curr-tab-num commondat)))
	     (updaters (hash-table-ref/default (dboard:commondat-updaters commondat)
					       tnum

	 (curr-updaters (hash-table-ref/default (dboard:commondat-updaters commondat) tnum '())))
    (hash-table-set! (dboard:commondat-updaters commondat)
		     tnum
		     (cons updater curr-updaters))))

;; data for each specific tab goes here
;;
(defstruct dboard:tabdat 
(defstruct dboard:tabdat
  ;; flags
  ((already-running #f)                  : boolean)     ;; the updater is already running. skip
  ;; runs
  ((allruns         '())                 : list)        ;; list of dboard:rundat records
  ((allruns-by-id    (make-hash-table))  : hash-table)  ;; hash of run-id -> dboard:rundat records
  ((done-runs       '())                 : list)        ;; list of runs already drawn
  ((not-done-runs   '())                 : list)        ;; list of runs not yet drawn
  (header            #f)                                ;; header for decoding the run records
  (keys              #f)                                ;; keys for this run (i.e. target components)

  ((tests-by-name (make-hash-table)) : hash-table) ;; hash of testfullname => testdat
  key-vals
  ((last-update   0)                 : number)    ;; last query to db got records from before last-update
  ((last-db-time  0)                 : number)    ;; last timestamp on megatest.db
  ((data-changed  #f)                : boolean)   
  ((run-data-offset  0)              : number)      ;; get only 100 items per call, set back to zero when received less than 100 items
  (db-path #f))

;; for the new runs view lets build up a few new record types and then consolidate later
;;
;; this is a two level deep pipeline for the incoming data:
;;   sql query data ==> filters ==> data for display
;;
(defstruct dboard:rdat
  ;; view related items
  (runnum    0) ;; which column we are processing, index into runsbynum, we sweep across all these runs then start over
  (leftcol   0) ;; number of the leftmost visible column
  (toprow    0) ;; topmost visible row
  (numcols  24) ;; number of columns visible
  (numrows  20) ;; number of rows visible
  
  ;; data from sql db
  (keys       (rmt:get-keys))         ;; to be removed when targets handling is refactored
  (runs       (make-sparse-vector))   ;; id => runrec
  (runsbynum  (make-vector 100 #f))   ;; vector num => runrec 
  (targ-runid (make-hash-table))      ;; area/target/runname => run-id  ;; not sure this will be needed
  (tests      (make-hash-table))      ;; test[/itempath] => list of test rec

  ;; run sql filters 
  (targ-sql-filt        "%")
  (runname-sql-filt     "%")
  (run-state-sql-filt   "%")
  (run-status-sql-filt  "%")

  ;; test sql filter
  (testname-sql-filt    "%")
  (itempath-sql-filt    "%")
  (test-state-sql-filt  "%")
  (test-status-sql-filt "%")

  ;; other sql related fields
  (last-updates (make-sparse-vector 0))  ;; run-id -> timestamp of the last update from sql db, set to zero on any field changes

  ;; filtered data
  (cols  (make-sparse-vector))   ;; columnnum => run-id
  (tests (make-hash-table))      ;; test[/itempath] => (vector columnnum => test rec)

  ;; various
  (prev-run-ids  '())            ;; push previously looked at runs on this
  (view-changed #f)

  ;; widgets
  (runs-tree #f)                 ;; 
  )

(define (dboard:rdat-push-run-id rdat run-id)
  (dboard:rdat-prev-run-ids-set! rdat (cons run-id (dboard:rdat-prev-run-ids rdat))))

(defstruct dboard:runrec
  id
  target  ;; a/b/c...
  tdef    ;; for future use
  )
     
(defstruct dboard:testrec
  id
  runid
  testname  ;; test[/itempath]
  state
  status
  start-time
  duration
  )

;; register dboard:rundat with BBpp
;; this is used by BBpp (Brandon's pretty printer) to convert dboard:rundat into a composition of lists that pp will handle
(hash-table-set! *BBpp_custom_expanders_list* RUNDAT:
                 (cons dboard:rundat?
                       (lambda (tabdat-item)
                         (filter

;;
;;    NOTE: Yes, this is used
;;
(define (dboard:get-tests-for-run-duplicate tabdat run-id run testnamepatt key-vals)
  (let* ((start-time   (current-seconds))
	 (access-mode  (dboard:tabdat-access-mode tabdat))
         (num-to-get   (string->number (or (configf:lookup *configdat* "setup" "num-tests-to-get")
                                           "200")))
                                           "50"))) ;; was 200, which is fine in a normal run area.
	 (states       (hash-table-keys (dboard:tabdat-state-ignore-hash tabdat)))
	 (statuses     (hash-table-keys (dboard:tabdat-status-ignore-hash tabdat)))
         (do-not-use-db-file-timestamps #t) ;; (configf:lookup *configdat* "setup" "do-not-use-db-file-timestamps")) ;; this still hosts runs-summary-tab
         (do-not-use-query-timestamps   #t) ;; (configf:lookup *configdat* "setup" "do-not-use-query-timestamps")) ;; this no longer troubles runs-summary-tab
	 (sort-info    (get-curr-sort))
	 (sort-by      (vector-ref sort-info 1))
	 (sort-order   (vector-ref sort-info 2))

    ;;
    ;; (debug:print 0 *default-log-port* "got-all: " got-all " multi-get: " multi-get " num-to-get: " num-to-get " (length tmptests): " (length tmptests) " db-modified: " db-modified " db-mod-time: " db-mod-time " db-path: " db-path)
    (if got-all
	(begin
	  (dboard:rundat-last-update-set!     run-dat (- start-time 2))
	  (dboard:rundat-run-data-offset-set! run-dat 0))
	(begin
	  ;;; (thread-sleep! 0.25) ;; give the rest of the gui some time to update. <-- this did NOT help
	  (dboard:rundat-run-data-offset-set! run-dat
					      (+ num-to-get (dboard:rundat-run-data-offset run-dat)))))

    (for-each 
     (lambda (tdat)
       (let ((test-id (db:test-get-id tdat))
	     (state   (db:test-get-state tdat)))

;; this calls dboard:get-tests-for-run-duplicate for each run
;;
;; create a virtual table of all the tests
;; keypatts: ( (KEY1 "abc%def")(KEY2 "%") )
;;
(define (dboard:update-rundat tabdat runnamepatt numruns testnamepatt keypatts)
  (dboard:tabdat-already-running-set! tabdat #t)
  (let* (;; (already-running  (dboard:tabdat-already-running tabdat))
  (let* ((access-mode      (dboard:tabdat-access-mode tabdat))
	 (access-mode      (dboard:tabdat-access-mode tabdat))
         (keys             (dboard:tabdat-keys tabdat)) ;; (db:dispatch-query access-mode rmt:get-keys db:get-keys)))
	 (last-runs-update (- (dboard:tabdat-last-runs-update tabdat) 2))
         (allruns          (rmt:get-runs runnamepatt numruns (dboard:tabdat-start-run-offset tabdat) keypatts))
         ;;(allruns-tree (rmt:get-runs-by-patt (dboard:tabdat-keys tabdat) "%" #f #f #f #f))
         (allruns-tree    (rmt:get-runs-by-patt keys "%" #f #f #f #f 0)) ;; last-runs-update));;'("id" "runname")
	 (header      (db:get-header allruns))
	 (runs        (db:get-rows   allruns)) ;; RA => Filtered as per runpatt selected

		   (elapsed-time (- (current-seconds) start-time)))
	      (if (null? all-test-ids)
		  (hash-table-delete! (dboard:tabdat-allruns-by-id tabdat) run-id)
		  (hash-table-set!    (dboard:tabdat-allruns-by-id tabdat) run-id run-struct))
	      (if (or (null? tal)
		      (> elapsed-time 2)) ;; stop loading data after 5 seconds, on the next call more data *should* be loaded since get-tests-for-run uses last update
		  (begin
		    (when (> elapsed-time 2)   
		    #;(when (> elapsed-time 2)   
                      (debug:print 0 *default-log-port* "NOTE: updates are taking a long time, " elapsed-time "s elapsed.")
                      (let* ((old-val (iup:attribute *tim* "TIME"))
                             (new-val (number->string (inexact->exact (floor (* 2  (string->number old-val)))))))
                        (if (< (string->number new-val) 5000)
                            (begin
                            ((debug:print 0 *default-log-port* "NOTE: increasing poll interval from "old-val" to "new-val)
                            (iup:attribute-set! *tim* "TIME" new-val))))
			      (debug:print 0 *default-log-port* "NOTE: increasing poll interval from "old-val" to "new-val)
			      (iup:attribute-set! *tim* "TIME" new-val)))))


                      )
		    (dboard:tabdat-allruns-set! tabdat new-res)
		    maxtests)
		  (if (> (dboard:rundat-run-data-offset run-struct) 0)
		      (loop run tal new-res newmaxtests) ;; not done getting data for this run
		      (loop (car tal)(cdr tal) new-res newmaxtests)))))))
    (dboard:tabdat-filters-changed-set! tabdat #f)
    (dboard:update-tree tabdat runs-hash header tb)))
    (dboard:update-tree tabdat runs-hash header tb)
    (dboard:tabdat-already-running-set! tabdat #f)))

(define *collapsed* (make-hash-table))

(define (toggle-hide lnum uidat) ; fulltestname)
  (let* ((btn (vector-ref (dboard:uidat-get-lftcol uidat) lnum))
	 (fulltestname (iup:attribute btn "TITLE"))
	 (parts        (string-split fulltestname "("))

 ;;	 (dboard:tabdat-logs-textbox-set! tabdat logs-tb)
 ;;	 logs-tb))

;; browse runs as a tree. Used in both "Runs" tab and
;; in the runs control panel.
;;
(define (dboard:runs-tree-browser commondat tabdat)
  (let* (
	 (txtbox (iup:textbox #:action (lambda (val a b)
					 (debug:catch-and-dump
					  (lambda ()
					    ;; for the Runs view we put the list of keyvals into tabdat target
					    ;; for the Run Controls we put then update the run-command
					    (if b (dboard:tabdat-target-set! tabdat (string-split b "/")))
					    (dashboard:update-run-command tabdat))
					  "command-testname-selector tb action"))
			      #:value (dboard:test-patt->lines
				       (dboard:tabdat-test-patts-use tabdat))
			      #:expand "HORIZONTAL"
			      ;; #:size "10x30"
			      ))
  (let* ((txtbox (iup:textbox
		  #:action (lambda (val a b)
			     (debug:catch-and-dump
			      (lambda ()
				;; for the Runs view we put the list
				;; of keyvals into tabdat target for
				;; the Run Controls we put then update
				;; the run-command
				(if b (dboard:tabdat-target-set! tabdat
								 (string-split b "/")))
				(dashboard:update-run-command tabdat))
			      "command-testname-selector tb action"))
		  #:value (dboard:test-patt->lines
			   (dboard:tabdat-test-patts-use tabdat))
		  #:expand "HORIZONTAL"
		  ;; #:size "10x30"
		  ))
	 (tb
          (iup:treebox
           #:value 0
           #:title "Runs"     ;;  was #:name -- iup 3.19 changed
			      ;;  this... "Changed: [DEPRECATED
           #:title "Runs" ;;  was #:name -- iup 3.19 changed this... "Changed: [DEPRECATED REMOVED] removed the old attribute NAMEid from IupTree to avoid conflict with the common attribute NAME. Use the TITLEid attribute."
			      ;;  REMOVED] removed the old attribute
			      ;;  NAMEid from IupTree to avoid
			      ;;  conflict with the common attribute
			      ;;  NAME. Use the TITLEid attribute."
           #:expand "YES"
           #:addexpanded "YES"
           #:size "10x"
           #:selection-cb
           (lambda (obj id state)
             (debug:catch-and-dump
              (lambda ()
                (let* ((run-path (tree:node->path obj id))
                       (run-id    (tree-path->run-id tabdat (cdr run-path))))
                  ;; (dboard:tabdat-view-changed-set! tabdat #t) ;; ?? done below when run-id is a number
                  (dboard:tabdat-target-set! tabdat (cdr run-path)) ;; (print "run-path: " run-path)
		  (iup:attribute-set! txtbox "VALUE" (string-intersperse (cdr run-path) "/"))
                  ;; (dboard:tabdat-view-changed-set! tabdat #t) ;; ??
                  ;; done below when run-id is a number
                  (dboard:tabdat-target-set! tabdat (cdr run-path)) ;; (print
								    ;; "run-path:
								    ;; "
								    ;; run-path)
		  (iup:attribute-set! txtbox "VALUE"
				      (string-intersperse (cdr run-path) "/"))
		  (dashboard:update-run-command tabdat)
                  (dboard:tabdat-layout-update-ok-set! tabdat #f)
                  (if (number? run-id)
                      (begin
                        ;; capture last two in tabdat.
                        (dboard:tabdat-prev-run-id-set!
                         tabdat
                         (dboard:tabdat-curr-run-id tabdat))
                        (dboard:tabdat-curr-run-id-set! tabdat run-id)
                        (dboard:tabdat-view-changed-set! tabdat #t))
                      (debug:print-error 5 *default-log-port* "tree-path->run-id returned non-number " run-id))))
              "treebox"))
           ;; (print "path: " (tree:node->path obj id) " run-id: " run-id)
           )))
    (dboard:tabdat-runs-tree-set! tabdat tb)
    (iup:detachbox
     (iup:vbox 
      txtbox
      tb
      ))))

;; browse runs as a tree. Used in both "Runs" tab and
;; in the runs control panel.
;;
;;  THIS IS THE NEW ONE
;;
(define (dboard:runs-tree-new-browser commondat rdat)
  (let* ((txtbox (iup:textbox
		  #:action (lambda (val a b)
			     (debug:catch-and-dump
			      (lambda ()
				;; for the Runs view we put the list
				;; of keyvals into tabdat target for
				;; the Run Controls we put then update
				;; the run-command
				(if b (dboard:rdat-targ-sql-filt-set! rdat
								 (string-split b "/")))
				#;(dashboard:update-run-command tabdat))
			      "command-testname-selector tb action"))
		  ;; #:value (dboard:test-patt->lines  ;; This seems like it was wrong, BUG in code where it was copied from?
	          ;;		   (dboard:tabdat-test-patts-use tabdat))
		  #:expand "HORIZONTAL"
		  ;; #:size "10x30"
		  ))
	 (tb
          (iup:treebox
           #:value 0
           #:title "Runs"     ;;  was #:name -- iup 3.19 changed
			      ;;  this... "Changed: [DEPRECATED
			      ;;  REMOVED] removed the old attribute
			      ;;  NAMEid from IupTree to avoid
			      ;;  conflict with the common attribute
			      ;;  NAME. Use the TITLEid attribute."
           #:expand "YES"
           #:addexpanded "YES"
           #:size "10x"
           #:selection-cb
           (lambda (obj id state)
             (debug:catch-and-dump
              (lambda ()
                (let* ((run-path (tree:node->path obj id))
                       (run-id    (new-tree-path->run-id rdat (cdr run-path))))
                  ;; (dboard:tabdat-view-changed-set! tabdat #t) ;; ??
                  ;; done below when run-id is a number
                  (dboard:rdat-targ-sql-filt-set! rdat (cdr run-path)) ;; (print
								    ;; "run-path:
								    ;; "
								    ;; run-path)
		  (iup:attribute-set! txtbox "VALUE"
				      (string-intersperse (cdr run-path) "/"))
		  #;(dashboard:update-run-command tabdat)
                  #;(dboard:tabdat-layout-update-ok-set! tabdat #f)
                  (if (number? run-id)
                      (begin
                        ;; capture last two in tabdat.
                        (dboard:rdat-push-run-id rdat run-id)
			(dboard:rdat-view-changed-set! rdat #t))
                      (debug:print-error 5 *default-log-port* "tree-path->run-id returned non-number " run-id))))
              "treebox"))
           ;; (print "path: " (tree:node->path obj id) " run-id: " run-id)
           )))
    (dboard:rdat-runs-tree-set! rdat tb)
    (iup:detachbox
     (iup:vbox 
      txtbox))))
      txtbox
      tb
      ))))

;;======================================================================
;; R U N   C O N T R O L S
;;======================================================================
;;
;; A gui for launching tests
;;

;;
;; display and manage a single run at a time

(define (tree-path->run-id tabdat path)
  (if (not (null? path))
      (hash-table-ref/default (dboard:tabdat-path-run-ids tabdat) path #f)
      #f))

(define (new-tree-path->run-id rdat path)
  (if (not (null? path))
      (hash-table-ref/default (dboard:rdat-path-run-ids tabdat) path #f)
      #f))

;; (define (dboard:get-tests-dat tabdat run-id last-update)
;;   (let* ((access-mode     (dboard:tabdat-access-mode tabdat))
;;          (tdat (if run-id (db:dispatch-query access-mode rmt:get-tests-for-run db:get-tests-for-run
;;                                              run-id 
;; 					     (hash-table-ref/default (dboard:tabdat-searchpatts tabdat) "test-name" "%/%")
;; 					     (hash-table-keys (dboard:tabdat-state-ignore-hash tabdat))  ;; '()

	 (result-child #f))
    (if (and (common:file-exists? source)
	     (file-read-access? source))
	(handle-exceptions
	 exn
	 (begin
	   (print-call-chain)
	   (debug:print 0 *default-log-port* " message: " ((condition-property-accessor 'exn 'message) exn))
	   (debug:print 0 *default-log-port* " message: " ((condition-property-accessor 'exn 'message) exn) ", exn=" exn)
	   (debug:print 0 *default-log-port* "ERROR: failed to load " source ", try loading in the repl: megatest -repl")
	   (set! success #f))
	 (load source))
	(begin
	  (debug:print 0 *default-log-port* "ERROR: cannot find file to load: \"" source "\" for user view " view-name)))
    ;; now run the user supplied definition for the tab view
    (if success
	(handle-exceptions
	 exn
	 (begin
	   (print-call-chain)
	   (debug:print 0 *default-log-port* " message: " ((condition-property-accessor 'exn 'message) exn))
	   (debug:print 0 *default-log-port* " message: " ((condition-property-accessor 'exn 'message) exn) ", exn=" exn)
	   (debug:print 0 *default-log-port* "ERROR: failed call procedure " viewgen
			", with; tab-num=" tab-num ", view-name=" view-name
			", and views-cfgdat and megatest configdat as parameters. To debug try loading in the repl: megatest -repl")
	   (set! success #f))
	 (print "Adding tab " view-name " with proc " viewgen)
	 ;; (iup:child-add! tabs
	 (set! result-child 
	       ((eval (string->symbol viewgen)) commondat tabs tab-num view-name views-cfgdat *configdat*))))
    ;; and finally set the updater
    (if success
	(dboard:commondat-add-updater commondat
				      (lambda ()
					(handle-exceptions
					 exn
					 (begin
					   (print-call-chain)
					   (debug:print 0 *default-log-port* " message: " ((condition-property-accessor 'exn 'message) exn))
					   (debug:print 0 *default-log-port* " message: " ((condition-property-accessor 'exn 'message) exn) ", exn=" exn)
					   (debug:print 0 *default-log-port* "ERROR: failed call procedure \"" updater
							"\", with; tabnum=" tab-num ", view-name=" view-name
							", and views-cfgdat and megatest configdat as parameters. To debug try loading in the repl: megatest -repl")
					   (set! success #f))
					 (debug:print 4 *default-log-port* "Running updater for tab " view-name " with proc " updater " and tab-num: " tab-num)
					 ((eval (string->symbol updater)) commondat tabs tab-num view-name views-cfgdat *configdat*)))
				      tab-num: tab-num))

            sort-lb))) 
	)

        ;; insert extra widget here
        (if extra-widget
            extra-widget
            (iup:hbox)) ;; empty widget

        

        
        )))

     (let* ((status-toggles (map (lambda (status)
				   (iup:toggle (conc status)
					       #:fontsize 8 ;; btn-fontsz ;; "10"
					       ;; #:expand "HORIZONTAL"
					       #:action   (lambda (obj val)

				      (debug:print 6 *default-log-port* "(dboard:tabdat-start-run-offset tabdat) " (dboard:tabdat-start-run-offset tabdat) " maxruns: " maxruns ", val: " val " oldmax: " oldmax)
				      (iup:attribute-set! obj "MAX" (* maxruns 10))))
		#:expand "HORIZONTAL"
		#:max (* 10 (max (hash-table-size (dboard:tabdat-allruns-by-id tabdat)) 10))
		#:min 0
		#:step 0.01))

;; make-simple-run                   procedure (target3772 id3773 runname3774 state3775 status3776 owner3777 event_time3778)
;; rmt:simple-get-runs               procedure (runpatt1001 count1002 offset1003 target1004)
;; simple-run-event_time             procedure (x3834)
;; simple-run-event_time-set!        procedure (x3830 val3831)
;; simple-run-id                     procedure (x3794)
;; simple-run-id-set!                procedure (x3790 val3791)
;; simple-run-owner                  procedure (x3826)
;; simple-run-owner-set!             procedure (x3822 val3823)
;; simple-run-runname                procedure (x3802)
;; simple-run-runname-set!           procedure (x3798 val3799)
;; simple-run-state                  procedure (x3810)
;; simple-run-state-set!             procedure (x3806 val3807)
;; simple-run-status                 procedure (x3818)
;; simple-run-status-set!            procedure (x3814 val3815)
;; simple-run-target                 procedure (x3786)
;; simple-run-target-set!            procedure (x3782 val3783)
;; simple-run?                       procedure (x3780)


;;======================================================================
;; Extracting the data to display for runs
;;
;; This needs to be re-entrant such that it does one column per call
;; on the zeroeth call update runs data
;; on each subsequent call update one run (configurable - could do two, three ... or update until tdelta exceeded
;; on last run reset to zeroeth
;;
;;   1. select with run filters; area, target, runname, runstate, runstatus, starttime, duration
;;       - put this information into two data structures:
;;         a. hash of area/target/runname => runstruct #< ordernun, id, area, target, runname, state,
;;                                                        status, starttime, duration, non-deleted testcount>
;;            ordernum reflects order as received from sql query
;;         b. sparsevec of id => runstruct
;;   2. for each run in runshash ordered by ordernum do:
;;         retrieve data since last update for that run
;;         if there is a deleted test - retrieve full data
;;         if there are non-deleted tests register this run in the columns sparsevec
;;         if this is the zeroeth column regenerate the rows sparsevec
;;         if this column is in the visible zone update visible cells
;;
;; Other factors:
;;   1. left index handling:
;;       - add test/itempaths to left index as discovered, re-order and
;;         update row -> test/itempath mapping on each read run
;;======================================================================

;; runs is <vec header runs>
;;   get ALL runs info
;;   update rdat-targ-run-id
;;   update rdat-runs
;;
(define (dashboard:update-runs-data rdat)
  (let* ((tb               (dboard:rdat-runs-tree rdat))
	 (targ-sql-filt    (dboard:rdat-targ-sql-filt    rdat))
	 (runname-sql-filt (dboard:rdat-runname-sql-filt rdat))
	 (state-sql-filt   (dboard:rdat-run-state-sql-filt   rdat))
	 (status-sql-filt  (dboard:rdat-run-status-sql-filt  rdat))
	 ;; Use (db:get-value-by-header (db:get-header runinfo)(db:get-rows runinfo))
	 (data             (rmt:simple-get-runs runname-sql-filt #f #f targ-sql-filt #f))
	 (numruns          (length data)))
    ;; store in the runsbynum vector
    (dboard:rdat-runsbynum-set! rdat (list->vector data))
    ;; update runs       id              => runrec
    ;; update targ-runid target/runname  => run-id
    (for-each
     (lambda (runrec)
       (let* ((run-id (simple-run-id runrec))
	      (full-targ-runname (conc (simple-run-target runrec) "/"
				       (simple-run-runname runrec))))
	 (debug:print 0 *default-log-port* "Update run  " run-id)
	 (sparse-vector-set! (dboard:rdat-runs rdat) run-id runrec)
	 (hash-table-set! (dboard:rdat-targ-runid rdat) full-targ-runname run-id)
	 ))
     data)
    numruns))

;; NOTE: runnum is NOT the run-id, it is a pointer into the runsbynum vector
;;
(define (dashboard:update-run-data runnum rdat)
  (let* ((curr-time            (current-seconds))
	 (runrec               (vector-ref (dboard:rdat-runsbynum rdat) runnum))
	 (run-id               (simple-run-id runrec))
	 (last-update          (sparse-vector-ref (dboard:rdat-last-updates rdat) run-id))
	 ;; filters
	 (testname-sql-filt    (dboard:rdat-testname-sql-filt    rdat))
	 ;; (itempath-sql-filt    (dboard:rdat-itempath-sql-filt    rdat))
	 (test-state-sql-filt  (dboard:rdat-test-state-sql-filt  rdat))  ;; not used yet
	 (test-status-sql-filt (dboard:rdat-test-status-sql-filt rdat))  ;; not used yet
	 (tests                (rmt:get-tests-for-run-state-status run-id
						      testname-sql-filt
						      last-update                ;; last-update
						      )))
    (sparse-vector-set! (dboard:rdat-last-updates rdat) run-id (- curr-time 1))
    (debug:print 0 *default-log-port* "Got " (length tests) " tests for run-id "
		 run-id " testname-sql-filt " testname-sql-filt " and last-update " last-update) 
    (length tests)))

(define (new-runs-updater commondat rdat)
  (let* ((runnum           (dboard:rdat-runnum          rdat))
	 (start-time       (current-milliseconds))
	 (tot-runs         #f))
    (if (eq? runnum 0)(dashboard:update-runs-data rdat))
    (set! tot-runs (vector-length (dboard:rdat-runsbynum rdat)))
    (let loop ((rn   runnum))
      (if (and (< (- (current-milliseconds) start-time) 250)
	       (< rn tot-runs))
	  (let* ((newrn (if (>= runnum (vector-length (dboard:rdat-runsbynum rdat)))
			    0 ;; start over
			    (+ rn 1)))) ;; (+ runnum 1)))
	    (dashboard:update-run-data rn rdat)
	    (dboard:rdat-runnum-set! rdat newrn)
	    (if (> newrn 0)
		(loop newrn)))))
    (if (>=  (dboard:rdat-runnum rdat) tot-runs)
	(dboard:rdat-runnum-set! rdat 0))
    ;; (dboard:rdat-runnum-set! rdat rn))) ;; not needed as it is set above
    ;; (dboard:rdat-last-update-set! rdat (- (current-seconds) 10))
    ;;    	 (tree:add-node tb "Runs" (string-split full-targ-runname "/"))
    '()))

(define (dboard:runs-new-matrix commondat rdat)
  (iup:matrix
   #:alignment1 "ALEFT"
   ;; #:expand "YES" ;; "HORIZONTAL"
   #:scrollbar "YES"
   #:numcol 10
   #:numlin 20
   #:numcol-visible 5 ;; (min 8)
   #:numlin-visible 1
   #:click-cb
   (lambda (obj row col status)
     (let* ((cell (conc row ":" col)))
       #f))
   ))
	 
(define (make-runs-view commondat rdat tab-num)
  ;; register an updater
  (dboard:commondat-add-updater
   commondat
   (lambda ()
     (new-runs-updater commondat rdat))
   tab-num: tab-num)

  (iup:vbox
   (iup:split
    #:orientation "VERTICAL" ;; "HORIZONTAL"
    #:value 100
    (dboard:runs-tree-new-browser commondat rdat)
    (dboard:runs-new-matrix commondat rdat)
    )))

(define (make-dashboard-buttons commondat) ;;  runs-sum-dat new-view-dat)
  (let* ((stats-dat       (dboard:tabdat-make-data))
	 (runs-dat        (dboard:tabdat-make-data))
	 (runs2-dat       (make-dboard:rdat)) ;; (dboard:tabdat-make-data))
	 (onerun-dat      (dboard:tabdat-make-data)) ;; name for run-summary structure 
	 (runcontrols-dat (dboard:tabdat-make-data))
	 (runtimes-dat    (dboard:tabdat-make-data))
	 (nruns           (dboard:tabdat-numruns runs-dat))
	 (ntests          (dboard:tabdat-num-tests runs-dat))
	 (keynames        (dboard:tabdat-dbkeys runs-dat))
	 (nkeys           (length keynames))

	 (btn-height      (dboard:tabdat-runs-btn-height runs-dat))
	 (btn-fontsz      (dboard:tabdat-runs-btn-fontsz runs-dat))
	 (cell-width      (dboard:tabdat-runs-cell-width runs-dat)))
    ;; controls (along bottom)
    ;; (set! controls (dboard:make-controls commondat runs-dat))
    
    ;; create the left most column for the run key names and the test names 
    (set! lftlst (list (iup:hbox
			(iup:label) ;; (iup:valuator)
			(apply iup:vbox 
			       (map (lambda (x)		
				      (let ((res (iup:hbox #:expand "HORIZONTAL"
							   (iup:label x #:size (conc 40 btn-height) #:fontsize btn-fontsz #:expand "NO") ;; "HORIZONTAL")
							   (iup:textbox #:size (conc 35 btn-height) #:fontsize btn-fontsz #:value "%" #:expand "NO" ;; "HORIZONTAL"
									#:action (lambda (obj unk val)
										   ;; each field (field name is "x" var) live updates
										   ;; the search filter as it is typed
										   (dboard:tabdat-target-set! runs-dat #f) ;; ensure the fields text boxes are used and not the info from the tree
										   (mark-for-update runs-dat)
										   (update-search commondat runs-dat x val))))))
					(set! i (+ i 1))
					res))
				    keynames)))))
    (set! lftlst
	  (list (iup:hbox
		 (iup:label) ;; (iup:valuator)
		 (apply iup:vbox 
			(map (lambda (x)		
			       (let ((res (iup:hbox
					   #:expand "HORIZONTAL"
					   (iup:label x
						      #:size (conc 40 btn-height)
						      #:fontsize btn-fontsz
						      #:expand "NO") ;; "HORIZONTAL")
					   (iup:textbox
					    #:size (conc 35 btn-height)
					    #:fontsize btn-fontsz
					    #:value "%"
					    #:expand "NO" ;; "HORIZONTAL"
					    #:action (lambda (obj unk val)
						       ;; each field
						       ;; (field name is "x" var) live updates
						       ;; the search filter as it is typed
						       (dboard:tabdat-target-set! runs-dat #f)
						       ;; ensure fields text boxes are used
						       ;; and not the info from the tree
						       (mark-for-update runs-dat)
						       (update-search commondat runs-dat x val))))))
				 (set! i (+ i 1))
				 res))
			     keynames)))))
    (let loop ((testnum  0)
	       (res      '()))
      (cond
       ((>= testnum ntests)
	;; now lftlst will be an hbox with the test keys and the test name labels
	(set! lftlst
	      (append
	       lftlst
	       (list
		(iup:hbox
	(set! lftlst (append lftlst (list (iup:hbox  #:expand "HORIZONTAL"
						     (iup:valuator #:valuechanged_cb (lambda (obj)
										       (let ((val (string->number (iup:attribute obj "VALUE")))
											     (oldmax  (string->number (iup:attribute obj "MAX")))
											     (newmax  (* 10 (length (dboard:tabdat-all-test-names runs-dat)))))
											 (dboard:commondat-please-update-set! commondat #t)
											 (dboard:tabdat-start-test-offset-set! runs-dat (inexact->exact (round (/ val 10))))
											 (debug:print 6 *default-log-port* "(dboard:tabdat-start-test-offset runs-dat) " (dboard:tabdat-start-test-offset runs-dat) " val: " val " newmax: " newmax " oldmax: " oldmax)
											 (if (< val 10)
											     (iup:attribute-set! obj "MAX" newmax))
											 ))
								   #:expand "VERTICAL" 
								   #:orientation "VERTICAL"
								   #:min 0
								   #:step 0.01)
						     (apply iup:vbox (reverse res)))))))
		 #:expand "HORIZONTAL"
		 (iup:valuator
		  #:valuechanged_cb
		  (lambda (obj)
		    (let ((val (string->number (iup:attribute obj "VALUE")))
			  (oldmax  (string->number (iup:attribute obj "MAX")))
			  (newmax  (* 10 (length (dboard:tabdat-all-test-names runs-dat)))))
		      (dboard:commondat-please-update-set! commondat #t)
		      (dboard:tabdat-start-test-offset-set! runs-dat
							    (inexact->exact (round (/ val 10))))
		      (debug:print 6 *default-log-port*
				   "(dboard:tabdat-start-test-offset runs-dat) "
				   (dboard:tabdat-start-test-offset runs-dat) " val: " val
				   " newmax: " newmax " oldmax: " oldmax)
		      (if (< val 10)
			  (iup:attribute-set! obj "MAX" newmax))
		      ))
		  #:expand "VERTICAL" 
		  #:orientation "VERTICAL"
		  #:min 0
		  #:step 0.01)
		 (apply iup:vbox (reverse res)))))))
       (else
	(let ((labl  (iup:button "" ;; the testname labels
				 #:flat "YES" 
				 #:alignment "ALEFT"
	(let ((labl  (iup:button
		      "" ;; the testname labels
		      #:flat "YES" 
		      #:alignment "ALEFT"
					; #:image img1
					; #:impress img2
				 #:size  (conc cell-width btn-height)
				 #:expand  "HORIZONTAL"
				 #:fontsize btn-fontsz
				 #:action (lambda (obj)
					    (mark-for-update runs-dat)
					    (toggle-hide testnum (dboard:commondat-uidat commondat)))))) ;; (iup:attribute obj "TITLE"))))
		      #:size  (conc cell-width btn-height)
		      #:expand  "HORIZONTAL"
		      #:fontsize btn-fontsz
		      #:action (lambda (obj)
				 (mark-for-update runs-dat)
				 (toggle-hide testnum (dboard:commondat-uidat commondat)))))) 
	  (vector-set! lftcol testnum labl)
	  (loop (+ testnum 1)(cons labl res))))))
    ;; These are the headers for each row
    (let loop ((runnum  0)
	       (keynum  0)
	       (keyvec  (make-vector nkeys))
	       (res    '()))

		    (result  '()))
		(for-each
		 (lambda (view-name)
		   (debug:print 0 *default-log-port* "Adding view " view-name)
		   (let* ((cfgtype (configf:lookup views-cfgdat view-name "type"))) ;; what type of view?
		     (if (not (string? cfgtype))
			 (debug:print-info 0 *default-log-port* "WARNING: view \"" view-name
					   "\" is missing needed sections. "
				     "\" is missing needed sections. Please consult the documenation and update ~/.mtviews.config or " *toppath* "/.mtviews.config")
					   "Please consult the documenation and update ~/.mtviews.config or "
					   *toppath* "/.mtviews.config")
			 (case (string->symbol cfgtype)
			   ;; user supplied source for a tab
			   ;;
			   ((external)
			    (let ((tab-content (dboard:add-external-tab commondat view-name views-cfgdat #f tab-num))) ;; was tabs
			   ((external) ;; was tabs
			    (let ((tab-content (dboard:add-external-tab commondat view-name views-cfgdat #f tab-num)))
			      (set! additional-tabnames (cons (cons tab-num view-name) additional-tabnames))
			      (set! tab-num (+ tab-num 1))
			      (set! result (append result (list tab-content)))))))))
		 (sort (hash-table-keys views-cfgdat) (lambda (a b)
							(let ((order-a (or (any->number (configf:lookup views-cfgdat a "order")) 999))
							      (order-b (or (any->number (configf:lookup views-cfgdat b "order")) 999)))
							  (> order-a order-b)))))
		 (sort (hash-table-keys views-cfgdat)
		       (lambda (a b)
			 (let ((order-a (or (any->number (configf:lookup views-cfgdat a "order")) 999))
			       (order-b (or (any->number (configf:lookup views-cfgdat b "order")) 999)))
			   (> order-a order-b)))))
		result))
	     (tabs (apply iup:tabs
			  #:tabchangepos-cb (lambda (obj curr prev)
					      (debug:catch-and-dump
					       (lambda ()
						 (let* ((tab-num (dboard:commondat-curr-tab-num commondat))
							(tabdat  (dboard:common-get-tabdat commondat tab-num: tab-num)))
                   
						   (dboard:tabdat-layout-update-ok-set! tabdat #f))
						 (dboard:commondat-curr-tab-num-set! commondat curr)
						 (let* ((tab-num (dboard:commondat-curr-tab-num commondat))
							(tabdat  (dboard:common-get-tabdat commondat tab-num: tab-num)))
               
						   (dboard:commondat-please-update-set! commondat #t)
						   (dboard:tabdat-layout-update-ok-set! tabdat #t)))
					       "tabchangepos"))
			  (dashboard:summary commondat stats-dat tab-num: 0)
			  runs-view
			  ;; (make-runs-view commondat runs2-dat 2)
			  (dashboard:runs-summary commondat onerun-dat tab-num: 2)
			  ;; (dashboard:new-view db data new-view-dat tab-num: 3)
			  (dashboard:run-controls commondat runcontrols-dat tab-num: 3)
			  (dashboard:run-times commondat runtimes-dat tab-num: 4)
			  ;; (dashboard:runs-summary commondat onerun-dat tab-num: 4)
			  additional-views)))
	;; (set! (iup:callback tabs tabchange-cb:) (lambda (a b c)(print "SWITCHED TO TAB: " a " " b " " c)))
	(iup:attribute-set! tabs "TABTITLE0" "Summary")
	(iup:attribute-set! tabs "TABTITLE1" "Runs")
	;; (iup:attribute-set! tabs "TABTITLE2" "Runs2")
	(iup:attribute-set! tabs "TABTITLE2" "Run Summary")
	(iup:attribute-set! tabs "TABTITLE3" "Run Control")
	(iup:attribute-set! tabs "TABTITLE4" "Run Times")
	;; (iup:attribute-set! tabs "TABTITLE3" "New View")
	;; (iup:attribute-set! tabs "TABTITLE4" "Run Control")

	;; set the tab names for user added tabs
	(for-each
	 (lambda (tab-info)
	   (iup:attribute-set! tabs (conc "TABTITLE" (car tab-info)) (cdr tab-info)))
	 additional-tabnames)
	
	(iup:attribute-set! tabs "BGCOLOR" "190 190 190")
	;; make the iup tabs object available (for changing color for example)
	(dboard:commondat-hide-not-hide-tabs-set! commondat tabs)
	;; now set up the tabdat lookup
	(dboard:common-set-tabdat! commondat 0 stats-dat)
	(dboard:common-set-tabdat! commondat 1 runs-dat)
	;;(dboard:common-set-tabdat! commondat 2 runs2-dat)
	(dboard:common-set-tabdat! commondat 2 onerun-dat)
	(dboard:common-set-tabdat! commondat 3 runcontrols-dat)
	(dboard:common-set-tabdat! commondat 4 runtimes-dat)

	(iup:vbox
	 tabs
	 ;; controls

;; Force creation of the db in case it isn't already there.
;; (tasks:open-db)

(define (dashboard:get-youngest-run-db-mod-time dbdir)
  (handle-exceptions
   exn
   (begin
     (debug:print 2 *default-log-port* "WARNING: error in accessing databases in get-youngest-run-db-mod-time: " ((condition-property-accessor 'exn 'message) exn) " db-dir="dbdir)
     (debug:print 2 *default-log-port* "WARNING: error in accessing databases in get-youngest-run-db-mod-time: "
		  ((condition-property-accessor 'exn 'message) exn) " db-dir="dbdir ", exn=" exn)
     (current-seconds)) ;; something went wrong - just print an error and return current-seconds
   (common:max (map (lambda (filen)
		      (file-modification-time filen))
		    (glob (conc dbdir "/*.db*"))))))

(define (dashboard:monitor-changed? commondat tabdat)
  (let* ((run-update-time (current-seconds))

;;  removing the tabdat-values proc 
;;
;; (define (tabdat-values tabdat)

;; runs update-rundat using the various filters from the gui
;;
(define (dashboard:do-update-rundat tabdat)
  ;; this seems like a good place to check for already running and skip if so
  ;;
  ;; (set! *updater-running* #t)
;;(if (dboard:tabdat-already-running tabdat)
;;    (begin
;;	(debug:print-info 0 *default-log-port* "Dashboard overloaded - updates will be slow, skipping update.")
;;	(dboard:tabdat-target tabdat))
  (dboard:update-rundat
   tabdat
   (hash-table-ref/default (dboard:tabdat-searchpatts tabdat) "runname" "%")
   (dboard:tabdat-numruns tabdat)
   (hash-table-ref/default (dboard:tabdat-searchpatts tabdat) "test-name" "%/%")
   ;; generate key patterns from the target stored in tabdat
   (let* ((dbkeys (dboard:tabdat-dbkeys tabdat)))
     (let ((fres   (if (dboard:tabdat-target tabdat)
                       (let ((ptparts (append (dboard:tabdat-target tabdat)(make-list (length dbkeys) "%"))))
                         (map (lambda (k v)(list k v)) dbkeys ptparts))
                       (let ((res '()))
                         (for-each (lambda (key)
                                     (if (not (equal? key "runname"))
                                         (let ((val (hash-table-ref/default (dboard:tabdat-searchpatts tabdat) key #f)))
                                           (if val (set! res (cons (list key val) res))))))
                                   dbkeys)
                         res))))
       fres))))
		       (let ((ptparts (append (dboard:tabdat-target tabdat)(make-list (length dbkeys) "%"))))
			 (map (lambda (k v)(list k v)) dbkeys ptparts))
		       (let ((res '()))
			 (for-each (lambda (key)
				     (if (not (equal? key "runname"))
					 (let ((val (hash-table-ref/default (dboard:tabdat-searchpatts tabdat) key #f)))
					   (if val (set! res (cons (list key val) res))))))
				   dbkeys)
			 res))))
       fres)))
  #;(set! *updater-running* #f))

(define (dashboard:runs-tab-updater commondat tab-num)
  (debug:catch-and-dump 
   (lambda ()
     (let* ((tabdat (dboard:common-get-tabdat commondat tab-num: tab-num))
	    (dbkeys (dboard:tabdat-dbkeys tabdat)))
       ;;(print "RA => calling runs-tab-updater with commondat " commondat " tab-num " tab-num)
       ;;(tabdat-values tabdat) ;;RA added 
       ;; (pp (dboard:tabdat->alist tabdat))
       ;; (if (dashboard:database-changed? commondat tabdat context-key: 'runs-rundat)      
       (dashboard:do-update-rundat tabdat)
       ;;(debug:print-info 13 *default-log-port* "dashboard:runs-tab-updater")
       ;;(inspect tabdat)

       (let ((uidat (dboard:commondat-uidat commondat)))
	 ;;(print "RA => Calling update-buttons with tabdat : " tabdat " uidat " uidat)
	 (update-buttons tabdat uidat (dboard:tabdat-numruns tabdat) (dboard:tabdat-num-tests tabdat)))
       ))
   "dashboard:runs-tab-updater"))

;;======================================================================
;; The heavy lifting starts here
;;======================================================================

	(dboard:commondat-uidat-set! commondat (make-dashboard-buttons commondat))
	(dboard:commondat-curr-tab-num-set! commondat 0)
	(dboard:commondat-add-updater 
	 commondat 
	 (lambda ()
	   (dashboard:runs-tab-updater commondat 1))
	 tab-num: 1)
        ;; may not want this alive (manually merged it from v1.66)
	(dboard:commondat-add-updater 
	 commondat 
	 (lambda ()
	   (dashboard:runs-tab-updater commondat 1))
	 tab-num: 2)
	(iup:callback-set! *tim*
			   "ACTION_CB"
			   (lambda (time-obj)
			     (let ((update-is-running #f))
			     (mutex-lock! (dboard:commondat-update-mutex commondat))
			     (set! update-is-running (dboard:commondat-updating commondat))
			     (if (not update-is-running)
			     (dboard:commondat-updating-set! commondat #t))
			     (mutex-unlock! (dboard:commondat-update-mutex commondat))
			     (if (not update-is-running) ;; we know that the update was not running and we now have a lock on doing an update
			     (begin
			     (dboard:common-run-curr-updaters commondat) ;; (dashboard:run-update commondat)
			     (mutex-lock! (dboard:commondat-update-mutex commondat))
			     (dboard:commondat-updating-set! commondat #f)
			     (mutex-unlock! (dboard:commondat-update-mutex commondat)))
			     (if (not *updater-running*)
				 (begin
				   ;; (mutex-lock! (dboard:commondat-update-mutex commondat))
				   ;; (set! update-is-running (dboard:commondat-updating commondat))
				   ;;(if (not update-is-running)
				   ;;	   (dboard:commondat-updating-set! commondat #t))
				   ;;(mutex-unlock! (dboard:commondat-update-mutex commondat))
				   ;;(if (not update-is-running) ;; we know that the update was not running and we now have a lock on doing an update
				   ;;	   (begin
				   (set! *updater-running* #t)
				   (dboard:common-run-curr-updaters commondat) ;; (dashboard:run-update commondat)
				   (set! *updater-running* #f)
				   ;; (mutex-lock! (dboard:commondat-update-mutex commondat))
				   ;;  (dboard:commondat-updating-set! commondat #f)
				   ;;   (mutex-unlock! (dboard:commondat-update-mutex commondat)))
				   ))
			     1))))
      
      (let ((th1 (make-thread (lambda ()
				(thread-sleep! 1)
				(dboard:common-run-curr-updaters commondat 0) ;; force update of summary tab 
				) "update buttons once"))

(declare (uses mt))

(include "common_records.scm")
(include "db_records.scm")
(include "key_records.scm")
(include "run_records.scm")

(define *rundb-mutex* (make-mutex)) ;; prevent problems opening/closing rundb's
(define *number-of-writes* 0)
(define *number-non-write-queries* 0)

;;======================================================================
;;  R E C O R D S
;;======================================================================

  (handle-exceptions
   exn
   (let ((err-status ((condition-property-accessor 'sqlite3 'status #f) exn)))
     ;; check for (exn sqlite3) ((condition-property-accessor 'exn 'message) exn)
     (if (eq? err-status 'done)
	 default
	 (begin
	   (debug:print-error 0 *default-log-port* " query " stmt " failed, params: " params ", error: " ((condition-property-accessor 'exn 'message) exn))
	   (debug:print-error 0 *default-log-port* " query " stmt " failed, params: " params ", error: " ((condition-property-accessor 'exn 'message) exn) ", exn=" exn)
	   (print-call-chain (current-error-port))
	   default)))
   (apply sqlite3:first-result db stmt params)))

;; Get/open a database
;;    if run-id => get run specific db
;;    if #f     => get main db

      dbdat))

(define (db:dbdat-get-path dbdat)
  (if (pair? dbdat)
      (cdr dbdat)
      #f))

;; mod-read:
;;     'mod   modified data
;;     'read  read data
;; Locks the mutex and depending on 'mod or 'read passed, sets the last timestamp in dbstruct
;;
;; (define (db:done-with dbstruct run-id mod-read)
(define-inline (db:generic-error-printout exn . message)
;;   (if (not (sqlite3:database? dbstruct))
;;       (begin
;; 	(mutex-lock! *rundb-mutex*)
;; 	(if (eq? mod-read 'mod)
;; 	    (dbr:dbstruct-mtime-set! dbstruct (current-milliseconds))
  (print-call-chain (current-error-port))
;; 	    (dbr:dbstruct-rtime-set! dbstruct (current-milliseconds)))
;; 	(dbr:dbstruct-inuse-set! dbstruct #f)
;; 	(mutex-unlock! *rundb-mutex*))))
  (apply debug:print-error 0 *default-log-port* message)
  (debug:print-error 0 *default-log-port* "   params: " params
		     ", error: "     ((condition-property-accessor 'exn 'message)   exn)
		     ", arguments: " ((condition-property-accessor 'exn 'arguments) exn)
		     ", location: "  ((condition-property-accessor 'exn 'location)  exn)
		     ))

;; (db:with-db dbstruct run-id sqlite3:exec "select blah fgrom blaz;")
;; r/w is a flag to indicate if the db is modified by this query #t = yes, #f = no
;;
(define (db:with-db dbstruct run-id r/w proc . params)
  (let* ((have-struct (dbr:dbstruct? dbstruct))
         (dbdat (if have-struct 
                    (db:get-db dbstruct)
                    #f))
	 (db    (if have-struct
		    (db:dbdat-get-db dbdat)
		    dbstruct))
	 (use-mutex (> *api-process-request-count* 25)))
         (dbdat     (if have-struct 
			(db:get-db dbstruct)
			#f))
	 (db        (if have-struct
			(db:dbdat-get-db dbdat)
			dbstruct))
	 (fname     (db:dbdat-get-path dbdat))
	 (use-mutex (> *api-process-request-count* 25))) ;; was 25
    (if (and use-mutex
	     (common:low-noise-print 120 "over-50-parallel-api-requests"))
	(debug:print-info 0 *default-log-port* *api-process-request-count* " parallel api requests being processed in process " (current-process-id) ", throttling access"))
    (if (common:low-noise-print 600 (conc "parallel-api-requests" *max-api-process-requests*))
	(debug:print-info 2 *default-log-port* "Parallel api request count: " *api-process-request-count* " max parallel requests: " *max-api-process-requests*))
    (handle-exceptions
    (condition-case
     exn
     (begin
       (print-call-chain (current-error-port))
       (debug:print-error 0 *default-log-port* "sqlite3 issue in db:with-db, dbstruct=" dbstruct ", run-id=" run-id ", proc=" proc ", params=" params " error: " ((condition-property-accessor 'exn 'message) exn))
       ;; there is no recovering at this time. exit
       (exit 50))
     (if use-mutex (mutex-lock! *db-with-db-mutex*))
     (let ((res (apply proc db params)))
       (if use-mutex (mutex-unlock! *db-with-db-mutex*))
       ;; (if (vector? dbstruct)(db:done-with dbstruct run-id r/w))
       (if dbdat (stack-push! (dbr:dbstruct-dbstack dbstruct) dbdat))
       res))))

       (if use-mutex (mutex-lock! *db-with-db-mutex*))
       (let ((res (apply proc db params)))
	 (if use-mutex (mutex-unlock! *db-with-db-mutex*))
	 ;; (if (vector? dbstruct)(db:done-with dbstruct run-id r/w))
	 (if dbdat (stack-push! (dbr:dbstruct-dbstack dbstruct) dbdat))
	 res))
     (exn (io-error)
	  (db:generic-error-printout exn "ERROR: i/o error with " fname ". Check permissions, disk space etc. and try again."))
     (exn (corrupt)
	  (db:generic-error-printout exn "ERROR: database " fname " is corrupt. Repair it to proceed."))
     (exn (busy)
	  (db:generic-error-printout exn "ERROR: database " fname
				     " is locked. Try copying to another location, remove original and copy back."))
     (exn (permission)(db:generic-error-printout exn "ERROR: database " fname " has some permissions problem."))
     (exn ()
	  (db:generic-error-printout exn "ERROR: Unknown error with database " fname " message: "
		       ((condition-property-accessor 'exn 'message) exn))))))
      
;;======================================================================
;; K E E P   F I L E D B   I N   dbstruct
;;======================================================================

;; (define (db:get-filedb dbstruct run-id)
;;   (let ((db (vector-ref dbstruct 2)))
;;     (if db

                (set! *db-write-access* #f)
                (dbr:dbstruct-read-only-set! dbstruct #t)))
          (dbr:dbstruct-mtdb-set!   dbstruct mtdb)
          (dbr:dbstruct-tmpdb-set!  dbstruct tmpdb)
          (dbr:dbstruct-dbstack-set! dbstruct (make-stack)) ;; BB: why a stack?  Why would the number of db's be indeterminate?  Is this a legacy of 1.db 2.db .. ?
          (stack-push! (dbr:dbstruct-dbstack dbstruct) tmpdb) ;; olddb is already a (cons db path)
          (dbr:dbstruct-refndb-set! dbstruct refndb)
          ;;	    (mutex-unlock! *rundb-mutex*)
          (if (and  (or (not dbfexists)
			(and modtimedelta
			     (> modtimedelta max-stale-tmp))) ;; if db in tmp is over ten seconds older than the file in MTRA then do a sync back
		    do-sync)
	      (begin
		(debug:print 1 *default-log-port* "filling db " (db:dbdat-get-path tmpdb) " with data \n    from " (db:dbdat-get-path mtdb) " mod time delta: " modtimedelta)
		(db:sync-tables (db:sync-all-tables-list dbstruct) #f mtdb refndb tmpdb)

    (stack-push! (dbr:dbstruct-dbstack dbstruct) tmpdb)))

(define (db:safely-close-sqlite3-db db stmt-cache #!key (try-num 3))
  (if (<= try-num 0)
      #f
      (handle-exceptions
	  exn
	  (begin
	    (thread-sleep! 3)
	    (sqlite3:interrupt! db)
	    (db:safely-close-sqlite3-db db try-num: (- try-num 1)))
	(begin
	  (print "Attempt to safely close sqlite3 db failed. Trying again. exn=" exn)
	  (thread-sleep! 3)
	  (sqlite3:interrupt! db)
	  (db:safely-close-sqlite3-db db stmtcache try-num: (- try-num 1)))
	(if (sqlite3:database? db)
	    (let* ((stmts (hash-table-ref/default stmt-cache db #f)))
	    (let* ((stmts (and stmt-cache (hash-table-ref/default stmt-cache db #f))))
	      (if stmts (map sqlite3:finalize! (hash-table-values stmts)))
	      (sqlite3:finalize! db)
	      #t)
	    #f))))

;; close all opened run-id dbs
(define (db:close-all dbstruct)
  (if (dbr:dbstruct? dbstruct)
      (handle-exceptions
	  exn
	  (begin
	    (debug:print 0 *default-log-port* "WARNING: Finalizing failed, "  ((condition-property-accessor 'exn 'message) exn))
	    (debug:print 0 *default-log-port* "WARNING: Finalizing failed, "  ((condition-property-accessor 'exn 'message) exn) ", note - exn=" exn)
	    (print-call-chain *default-log-port*))
	;; (db:sync-touched dbstruct 0 force-sync: #t) ;; NO. Do not do this here. Instead we rely on a server to be started when there are writes, even if the server itself is not going to be used as a server.
        (let ((tdbs       (map db:dbdat-get-db 
			       (stack->list (dbr:dbstruct-dbstack dbstruct))))
              (mdb        (db:dbdat-get-db (dbr:dbstruct-mtdb   dbstruct)))
              (rdb        (db:dbdat-get-db (dbr:dbstruct-refndb dbstruct)))
	      (stmt-cache (dbr:dbstruct-stmt-cache dbstruct)))
          (map (lambda (db)
		 (db:safely-close-sqlite3-db stmt-cache db))
		 (db:safely-close-sqlite3-db db stmt-cache))
	       tdbs)
          (db:safely-close-sqlite3-db stmt-cache mdb)     ;; (if (sqlite3:database? mdb) (sqlite3:finalize! mdb))
          (db:safely-close-sqlite3-db stmt-cache rdb))))) ;; (if (sqlite3:database? rdb) (sqlite3:finalize! rdb))))))
          (db:safely-close-sqlite3-db mdb stmt-cache)     ;; (if (sqlite3:database? mdb) (sqlite3:finalize! mdb))
          (db:safely-close-sqlite3-db rdb stmt-cache))))) ;; (if (sqlite3:database? rdb) (sqlite3:finalize! rdb))))))

;;   (let ((locdbs (dbr:dbstruct-locdbs dbstruct)))
;;     (if (hash-table? locdbs)
;; 	(for-each (lambda (run-id)
;; 		    (db:close-run-db dbstruct run-id))
;; 		  (hash-table-keys locdbs)))))

     ;; 
     ;;  NOPE: apply this same approach to all db files
     ;;
     (else ;; ((equal? fname "megatest.db") ;; this file can be regenerated if needed
      (handle-exceptions
       exn
       (begin
	 (print "Problems trying to repair the db, exn=" exn)
	 ;; (db:move-and-recreate-db dbdat)
	 (if (> numtries 0)
	     (db:repair-db dbdat numtries: (- numtries 1))
	     #f)
	 (debug:print 0 *default-log-port* "FATAL: file " dbpath " was found corrupted, an attempt to fix has been made but you must start over.")
	 (debug:print 0 *default-log-port*
		      "   check the following:\n"

     (let ((stmts       (make-hash-table)) ;; table-field => stmt
	   (all-stmts   '())              ;; ( ( stmt1 value1 ) ( stml2 value2 ))
	   (numrecs     (make-hash-table))
	   (start-time  (current-milliseconds))
	   (tot-count   0))
       (for-each ;; table
	(lambda (tabledat)
             
	  (let* ((tablename        (car tabledat))
		 (fields           (cdr tabledat))
		 (has-last-update  (member "last_update" fields))
		 (use-last-update  (cond
				    ((and has-last-update
					  (member "last_update" fields))
				     #t) ;; if given a number, just use it for all fields

              (thread-sleep! delay-handicap)
              (debug:print-info 0 *default-log-port* "synthetic sync delay of "delay-handicap" seconds completed")
              )
            
	    ;; first pass implementation, just insert all changed rows
	    (for-each 
	     (lambda (targdb)
               ;(print targdb " " tablename)
	       (let* ((db     (db:dbdat-get-db targdb))
                      (drp-trigger (if (member "last_update" field-names)
                                      (db:drop-trigger db tablename) 
                                       #f))
                       (is-trigger-dropped (if (member "last_update" field-names)
                                              (db:is-trigger-dropped db tablename) #f)) 
	       (let* ((db                 (db:dbdat-get-db targdb))
                      (drp-trigger        (if (member "last_update" field-names)
					      (db:drop-trigger db tablename) 
					      #f))
                      (is-trigger-dropped (if (member "last_update" field-names)
                                              (db:is-trigger-dropped db tablename)
					      #f)) 
		      (stmth  (sqlite3:prepare db full-ins)))
		 (db:delay-if-busy targdb) ;; NO WAITING
                (if (member "last_update" field-names)
                  (debug:print-info 3 *default-log-port* "is-trigger-dropped: " is-trigger-dropped)) 
		 ;; (db:delay-if-busy targdb) ;; NO WAITING
                 (if (member "last_update" field-names)
                     (debug:print-info 3 *default-log-port* "is-trigger-dropped: " is-trigger-dropped)) 
		 (for-each
		  (lambda (fromdat-lst)
		    (sqlite3:with-transaction
		     db
		     (lambda ()
		       (for-each ;; 
			(lambda (fromrow)

          ;; /tmp/bjbarcla/megatest_localdb/fullrun/.nfs.pdx.disks.icf_env_disk001.bjbarcla.gwa.issues.mtdev.matt-bisect.megatest.ext-tests.runs.all.v1.65.1.6524.dev.bb-24-justrollup-f8.rollup.fullrun/megatest.db.lock
          (delete-file* (common:get-sync-lock-filepath))
          )
	 
	 ;; clear out junk records
	 ;;
	 ((dejunk)
	  (db:delay-if-busy mtdb) ;; ok to delay on mtdb
	  ;; (db:delay-if-busy mtdb) ;; ok to delay on mtdb
	  (when (file-write-access? (db:dbdat-get-path mtdb)) (db:clean-up mtdb))
	  (db:clean-up tmpdb)
	  (db:clean-up refndb))

	 ;; sync runs, test_meta etc.
	 ;;
	 ((old2new)

(define db:trigger-list 
     (list (list "update_runs_trigger"  "CREATE TRIGGER IF NOT EXISTS update_runs_trigger AFTER UPDATE ON runs
                             FOR EACH ROW
                               BEGIN 
                                 UPDATE runs SET last_update=(strftime('%s','now'))
                                   WHERE id=old.id;
                               END;" ) 
       (list "update_run_stats_trigger"  "CREATE TRIGGER  IF NOT EXISTS update_run_stats_trigger AFTER UPDATE ON run_stats
	   (list "update_run_stats_trigger"  "CREATE TRIGGER  IF NOT EXISTS update_run_stats_trigger AFTER UPDATE ON run_stats
                             FOR EACH ROW
                               BEGIN 
                                 UPDATE run_stats SET last_update=(strftime('%s','now'))
                                   WHERE id=old.id;
                               END;" )
       (list "update_tests_trigger"  "CREATE TRIGGER  IF NOT EXISTS update_tests_trigger AFTER UPDATE ON tests
	   (list "update_tests_trigger"  "CREATE TRIGGER  IF NOT EXISTS update_tests_trigger AFTER UPDATE ON tests
                             FOR EACH ROW
                               BEGIN 
                                 UPDATE tests SET last_update=(strftime('%s','now'))
                                   WHERE id=old.id;
                               END;" )
       (list "update_teststeps_trigger"  "CREATE TRIGGER  IF NOT EXISTS update_teststeps_trigger AFTER UPDATE ON test_steps
	   (list "update_teststeps_trigger"  "CREATE TRIGGER  IF NOT EXISTS update_teststeps_trigger AFTER UPDATE ON test_steps
                             FOR EACH ROW
                               BEGIN 
                                 UPDATE test_steps SET last_update=(strftime('%s','now'))
                                   WHERE id=old.id;
                               END;" )
       (list "update_test_data_trigger"  "CREATE TRIGGER  IF NOT EXISTS update_test_data_trigger AFTER UPDATE ON test_data
	   (list "update_test_data_trigger"  "CREATE TRIGGER  IF NOT EXISTS update_test_data_trigger AFTER UPDATE ON test_data
                             FOR EACH ROW
                               BEGIN 
                                 UPDATE test_data SET last_update=(strftime('%s','now'))
                                   WHERE id=old.id;
                               END;" )))

(define (db:create-all-triggers dbstruct)
(db:with-db
   dbstruct #f #f
   (lambda (db)
(db:create-triggers db))))

(define (db:create-triggers db)
    (for-each (lambda (key)
              (sqlite3:execute db (cadr key)))
          db:trigger-list))

(define (db:drop-all-triggers dbstruct)
(db:with-db
  (db:with-db
   dbstruct #f #f
   (lambda (db)
(db:drop-triggers db))))
     (db:drop-triggers db))))

(define (db:is-trigger-dropped db tbl-name)
     (let* ((trigger-name (if (equal? tbl-name "test_steps")
				"update_teststeps_trigger" 
                                (conc "update_" tbl-name "_trigger"))))
            (sqlite3:for-each-row
		   (lambda (name)
                     ;(print name)
		     (set! res (vector name)))
		   db 
		   "select name  from sqlite_master where type = 'trigger' ;" 
		   )))
  (let* ((trigger-name (if (equal? tbl-name "test_steps")
			   "update_teststeps_trigger" 
                           (conc "update_" tbl-name "_trigger")))
	 (res          #f))
    (sqlite3:for-each-row
     (lambda (name)
       (if (equal? name trigger-name)
	   (set! res #t)))
     db 
     "SELECT name FROM sqlite_master WHERE type = 'trigger' ;" 
     )))

(define (db:drop-triggers db)
  (for-each (lambda (key) 
              (sqlite3:execute db (conc "drop trigger " (car key))))
          db:trigger-list))
  (for-each
   (lambda (key) 
     (sqlite3:execute db (conc "drop trigger if exists " (car key))))
   db:trigger-list))

(define  (db:drop-trigger db tbl-name)
      (let* ((trigger-name (if (equal? tbl-name "test_steps")
				"update_teststeps_trigger" 
                                (conc "update_" tbl-name "_trigger"))))
       (for-each (lambda (key) 
             (if (equal? (car key) trigger-name)
             (sqlite3:execute db (conc "drop trigger " trigger-name))))
      db:trigger-list)))
  (let* ((trigger-name (if (equal? tbl-name "test_steps")
			   "update_teststeps_trigger" 
                           (conc "update_" tbl-name "_trigger"))))
    (for-each
     (lambda (key) 
       (if (equal? (car key) trigger-name)
           (sqlite3:execute db (conc "drop trigger if exists " trigger-name))))
     db:trigger-list)))

(define  (db:create-trigger db tbl-name)
      (let* ((trigger-name (if (equal? tbl-name "test_steps")
                              "update_teststeps_trigger" 
                              (conc "update_" tbl-name "_trigger"))))
       (for-each (lambda (key) 
             (if (equal? (car key) trigger-name)

	;; archive disk areas, cached info from [archive-disks]
	(sqlite3:execute db "CREATE TABLE IF NOT EXISTS archive_disks (
                                id INTEGER PRIMARY KEY,
                                archive_area_name TEXT,
                                disk_path TEXT,
                                last_df INTEGER DEFAULT -1,
                                last_df_time TIMESTAMP DEFAULT (strftime('%s','now')),
                                creation_time TIMESTAMP DEFAULT (strftime('%','now')));")
                                creation_time TIMESTAMP DEFAULT (strftime('%s','now')));")
	;; individual bup (or tar) data chunks
	(sqlite3:execute db "CREATE TABLE IF NOT EXISTS archive_blocks (
                                id INTEGER PRIMARY KEY,
                                archive_disk_id INTEGER,
                                disk_path TEXT,
                                last_du INTEGER DEFAULT -1,
                                last_du_time TIMESTAMP DEFAULT (strftime('%s','now')),
                                creation_time TIMESTAMP DEFAULT (strftime('%','now')));")
                                creation_time TIMESTAMP DEFAULT (strftime('%s','now')));")
	;; tests allocated to what chunks. reusing a chunk for a test/item_path is very efficient
	;; NB// the per run/test recording of where the archive is stored is done in the test
	;;      record. 
	(sqlite3:execute db "CREATE TABLE IF NOT EXISTS archive_allocations (
                                id INTEGER PRIMARY KEY,
                                archive_block_id INTEGER,
                                testname TEXT,
                                item_path TEXT,
                                creation_time TIMESTAMP DEFAULT (strftime('%','now')));")
                                creation_time TIMESTAMP DEFAULT (strftime('%s','now')));")
	;; move this clean up call somewhere else
	(sqlite3:execute db "DELETE FROM tasks_queue WHERE state='done' AND creation_time < ?;" (- (current-seconds)(* 24 60 60))) ;; remove older than 24 hrs
	(sqlite3:execute db (conc "CREATE INDEX IF NOT EXISTS runs_index ON runs (runname" (if havekeys "," "") keystr ");"))
	;; (sqlite3:execute db "CREATE VIEW runs_tests AS SELECT * FROM runs INNER JOIN tests ON runs.id=tests.run_id;")
	(sqlite3:execute db "CREATE TABLE IF NOT EXISTS extradat (id INTEGER PRIMARY KEY, run_id INTEGER, key TEXT, val TEXT);")
	(sqlite3:execute db "CREATE TABLE IF NOT EXISTS metadat (id INTEGER PRIMARY KEY, var TEXT, val TEXT,
                                  CONSTRAINT metadat_constraint UNIQUE (var));")

                              id           INTEGER PRIMARY KEY,
                              test_id      INTEGER,
                              state        TEXT DEFAULT 'new',
                              status       TEXT DEFAULT 'n/a',
                              archive_type TEXT DEFAULT 'bup',
                              du           INTEGER,
                              archive_path TEXT);")))
         (print "creating trigges from init") 
         ;; (print "creating trigges from init") 
        (db:create-triggers db)    
     db)) ;; )

;;======================================================================
;; A R C H I V E S
;;======================================================================

       
       (debug:print-info 18 *default-log-port* "Found " (length oldlaunched) " old LAUNCHED items, " (length toplevels) " old LAUNCHED toplevel tests and " (length incompleted) " tests marked RUNNING but apparently dead.")
       (if (and (null? incompleted)
                (null? oldlaunched)
                (null? toplevels))
           #f
           #t)))))

;; given a launch delay (minimum time from last launch) return amount of time to wait
;;
;; (define (db:launch-delay-left dbstruct run-id launch-delay)
  


(define (db:get-status-from-final-status-file run-dir)
  (let (
       (infile (conc run-dir "/.final-status")))
  (let ((infile (conc run-dir "/.final-status")))

       ;; first verify we are able to write the output file
       (if (not (file-read-access? infile))
          (begin 
	        (debug:print 0 *default-log-port* "ERROR: cannot read " infile)
    ;; first verify we are able to write the output file
    (if (not (file-read-access? infile))
        (begin 
	  (debug:print 0 *default-log-port* "ERROR: cannot read " infile)
          (debug:print 0 *default-log-port* "ERROR: run-dir is " run-dir)
          #f
          )
          (with-input-from-file infile read-lines)
       )
  )
)




;;  select end_time-now from
;;      (select testname,item_path,event_time+run_duration as
        (with-input-from-file infile read-lines)
	)))

;; check duration against test-run.dat file if it exists and update the value in
;; the db if necessary
;;
(define (db:adjust-run-duration dbstruct test-id run-dir event-time run-duration)
  (let* ((datf             (conc run-dir ".mt_data/test-run.dat"))
	 (modt             (if (and (file-exists? datf)
				    (file-read-access? datf))
			       (file-modification-time datf)
			       #f)) ;; (+ event-time run-duration))))
	 (alt-run-duration (if modt
			       (- modt event-time)
			       #f)))
    (if (and alt-run-duration
	     (> alt-run-duration run-duration))
	(begin
	  (debug:print 0 *default-log-port* "Test " test-id " run duration mismatch. Setting to " alt-run-duration)
	  (db:with-db
	   dbstruct #f #f
	   (lambda (db)
	     (sqlite3:execute db "UPDATE tests SET run_duration=? WHERE id=?;" alt-run-duration test-id)
;;                          end_time,strftime('%s','now') as now from tests where state in
;;      ('RUNNING','REMOTEHOSTSTART','LAUNCHED'));

	     #t)))
	#f))) ;; #f = we did NOT adjust the time
	      
(define (db:find-and-mark-incomplete dbstruct run-id ovr-deadtime)
  (let* ((incompleted '())
	 (oldlaunched '())
	 (toplevels   '())
          ;; The default running-deadtime is 720 seconds = 12 minutes.
          ;; "(running-deadtime-default (+ server-start-allowance (* 2 launch-monitor-period)))" = 200 + (2 * (200 + 30 + 30))
         (deadtime-trim (or ovr-deadtime (configf:lookup-number *configdat* "setup" "deadtime")))

         )
    (debug:print-info 4  *default-log-port* "running-deadtime = " running-deadtime)
    (debug:print-info 4  *default-log-port* "deadtime-trim = " deadtime-trim)

    (db:with-db 
     dbstruct #f #f
     (lambda (db)
       (let* ((stmth1 (db:get-cache-stmth
		       dbstruct db
		       "SELECT id,rundir,uname,testname,item_path,event_time,run_duration FROM tests 
                           WHERE run_id=? AND (strftime('%s','now') - event_time) > (run_duration + ?)
                                          AND state IN ('RUNNING');"))
	      (stmth2 (db:get-cache-stmth
		       dbstruct db
		       "SELECT id,rundir,uname,testname,item_path,event_time,run_duration FROM tests 
                           WHERE run_id=? AND (strftime('%s','now') - event_time) > (run_duration + ?)
                                          AND state IN ('REMOTEHOSTSTART');"))
	      (stmth3 (db:get-cache-stmth
		       dbstruct db
		       "SELECT id,rundir,uname,testname,item_path FROM tests
                           WHERE run_id=? AND (strftime('%s','now') - event_time) > 86400
                                          AND state IN ('LAUNCHED');")))
       ;; in RUNNING or REMOTEHOSTSTART for more than 10 minutes
       ;;
       ;; HOWEVER: this code in run:test seems to work fine
       ;;              (> (- (current-seconds)(+ (db:test-get-event_time testdat)
       ;;                     (db:test-get-run_duration testdat)))
       ;;                    600) 
       ;; (db:delay-if-busy dbdat)
       (sqlite3:for-each-row 
        (lambda (test-id run-dir uname testname item-path event-time run-duration)
          (if (and (equal? uname "n/a")
                   (equal? item-path "")) ;; this is a toplevel test
              ;; what to do with toplevel? call rollup?
              (begin
                (set! toplevels   (cons (list test-id run-dir uname testname item-path run-id) toplevels))
                (debug:print-info 0 *default-log-port* "Found old toplevel test in RUNNING state, test-id=" test-id))
              (begin
                (set! incompleted (cons (list test-id run-dir uname testname item-path run-id) incompleted))
                (debug:print-info 0 *default-log-port* "Found old test in RUNNING state, test-id=" test-id" exceeded running-deadtime "running-deadtime" now="(current-seconds)" event-time="event-time" run-duration="run-duration))))
	 ;; in RUNNING or REMOTEHOSTSTART for more than 10 minutes
	 ;;
	 ;; HOWEVER: this code in run:test seems to work fine
	 ;;              (> (- (current-seconds)(+ (db:test-get-event_time testdat)
	 ;;                     (db:test-get-run_duration testdat)))
	 ;;                    600) 
	 ;; (db:delay-if-busy dbdat)
	 (sqlite3:for-each-row 
	  (lambda (test-id run-dir uname testname item-path event-time run-duration)
	    (if (not (db:adjust-run-duration dbstruct test-id run-dir event-time run-duration))
		(if (and (equal? uname "n/a")
			 (equal? item-path "")) ;; this is a toplevel test
		    ;; what to do with toplevel? call rollup?
		    (begin
		      (set! toplevels   (cons (list test-id run-dir uname testname item-path run-id) toplevels))
		      (debug:print-info 0 *default-log-port* "Found old toplevel test in RUNNING state, test-id=" test-id))
		    (begin
		      (set! incompleted (cons (list test-id run-dir uname testname item-path run-id) incompleted))
		      (debug:print-info 0 *default-log-port* "Found old test in RUNNING state, test-id="
        db
        
        "SELECT id,rundir,uname,testname,item_path,event_time,run_duration FROM tests WHERE run_id=? AND (strftime('%s','now') - event_time) > (run_duration + ?) AND state IN ('RUNNING');"
        run-id running-deadtime) ;; default time 720 seconds

					test-id" exceeded running-deadtime "running-deadtime" now="(current-seconds)
					" event-time="event-time" run-duration="run-duration)))))
	  stmth1
	  run-id running-deadtime) ;; default time 720 seconds
	    
       
       (sqlite3:for-each-row 
        (lambda (test-id run-dir uname testname item-path event-time run-duration)
          (if (and (equal? uname "n/a")
                   (equal? item-path "")) ;; this is a toplevel test
              ;; what to do with toplevel? call rollup?
              (begin
                (set! toplevels   (cons (list test-id run-dir uname testname item-path run-id) toplevels))
                (debug:print-info 0 *default-log-port* "Found old toplevel test in RUNNING state, test-id=" test-id))
              (begin
                (debug:print-info 0 *default-log-port* "Found old test in REMOTEHOSTSTART state, test-id=" test-id" exceeded running-deadtime "running-deadtime" now="(current-seconds)" event-time="event-time" run-duration="run-duration)
                (set! incompleted (cons (list test-id run-dir uname testname item-path run-id) incompleted)))))
	 (sqlite3:for-each-row 
	  (lambda (test-id run-dir uname testname item-path event-time run-duration)
	    (if (not (db:adjust-run-duration dbstruct test-id run-dir event-time run-duration))
		(if (and (equal? uname "n/a")
			 (equal? item-path "")) ;; this is a toplevel test
		    ;; what to do with toplevel? call rollup?
		    (begin
		      (set! toplevels   (cons (list test-id run-dir uname testname item-path run-id) toplevels))
		      (debug:print-info 0 *default-log-port* "Found old toplevel test in RUNNING state, test-id=" test-id))
		    (begin
		      (debug:print-info 0 *default-log-port* "Found old test in REMOTEHOSTSTART state, test-id=" test-id
					" exceeded running-deadtime "running-deadtime" now="(current-seconds)" event-time="event-time
					" run-duration="run-duration)
		      (set! incompleted (cons (list test-id run-dir uname testname item-path run-id) incompleted))))))
        db
        "SELECT id,rundir,uname,testname,item_path,event_time,run_duration FROM tests WHERE run_id=? AND (strftime('%s','now') - event_time) > (run_duration + ?) AND state IN ('REMOTEHOSTSTART');"
        run-id remotehoststart-deadtime) ;; default time 230 seconds

       ;; in LAUNCHED for more than one day. Could be long due to job queues TODO/BUG: Need override for this in config
       ;;
       ;; (db:delay-if-busy dbdat)
       (sqlite3:for-each-row
        (lambda (test-id run-dir uname testname item-path)
          (if (and (equal? uname "n/a")
                   (equal? item-path "")) ;; this is a toplevel test
              ;; what to do with toplevel? call rollup?
              (set! toplevels   (cons (list test-id run-dir uname testname item-path run-id) toplevels))
              (begin
                (debug:print-info 0 *default-log-port* "Found old test in LAUNCHED state, test-id=" test-id" 1 day since event_time marked")
	  stmth2
	  run-id remotehoststart-deadtime) ;; default time 230 seconds
	 
	 ;; in LAUNCHED for more than one day. Could be long due to job queues TODO/BUG: Need override for this in config
	 ;;
	 ;; (db:delay-if-busy dbdat)
	 (sqlite3:for-each-row
	  (lambda (test-id run-dir uname testname item-path)
	    (if (and (equal? uname "n/a")
		     (equal? item-path "")) ;; this is a toplevel test
		;; what to do with toplevel? call rollup?
		(set! toplevels   (cons (list test-id run-dir uname testname item-path run-id) toplevels))
		(begin
		  (debug:print-info 0 *default-log-port* "Found old test in LAUNCHED state, test-id=" test-id
				    " 1 day since event_time marked")
                (set! oldlaunched (cons (list test-id run-dir uname testname item-path run-id) oldlaunched)))))
        db
        "SELECT id,rundir,uname,testname,item_path FROM tests WHERE run_id=? AND (strftime('%s','now') - event_time) > 86400 AND state IN ('LAUNCHED');"
        run-id)
	  stmth3
	  run-id)
       
       (debug:print-info 18 *default-log-port* "Found " (length oldlaunched) " old LAUNCHED items, " (length toplevels) " old LAUNCHED toplevel tests and " (length incompleted) " tests marked RUNNING but apparently dead.")
	 (debug:print-info 18 *default-log-port* "Found " (length oldlaunched) " old LAUNCHED items, "
			   (length toplevels) " old LAUNCHED toplevel tests and "
			   (length incompleted) " tests marked RUNNING but apparently dead."))

       ;; These are defunct tests, do not do all the overhead of set-state-status. Force them to INCOMPLETE.
	 ;; These are defunct tests, do not do all the overhead of set-state-status. Force them to INCOMPLETE.
       ;;
       ;; (db:delay-if-busy dbdat)
	 ;; (db:delay-if-busy dbdat)
       (let* (;; (min-incompleted (filter (lambda (x)
              ;;      		      (let* ((testpath (cadr x))
              ;;      			     (tdatpath (conc testpath "/testdat.db"))
              ;;      			     (dbexists (common:file-exists? tdatpath)))
              ;;      			(or (not dbexists) ;; if no file then something wrong - mark as incomplete
              ;;      			    (> (- (current-seconds)(file-modification-time tdatpath)) 600)))) ;; no change in 10 minutes to testdat.db - she's dead Jim
              ;;      		    incompleted))
              (min-incompleted-ids (map car incompleted)) ;; do 'em all
              (all-ids             (append min-incompleted-ids (map car oldlaunched))))
         (if (> (length all-ids) 0)
             (begin
	       ;; (launch:is-test-alive "localhost" 435)
               (debug:print 0 *default-log-port* "WARNING: Marking test(s); " (string-intersperse (map conc all-ids) ", ") " as DEAD")
               (for-each
	 (let* ((min-incompleted-ids (map car incompleted)) ;; do 'em all
		(all-ids             (append min-incompleted-ids (map car oldlaunched))))
	   (if (> (length all-ids) 0)
	       (begin
		 ;; (launch:is-test-alive "localhost" 435)
		 (debug:print 0 *default-log-port* "WARNING: Marking test(s); " (string-intersperse (map conc all-ids) ", ")
			      " as DEAD")
		 (for-each
                  (lambda (test-id)
                    (let* (;; (run-dir (db:test-get-rundir-from-test-id dbstruct run-id test-id))
			   (tinfo   (db:get-test-info-by-id dbstruct run-id test-id))
			   (run-dir (db:test-get-rundir     tinfo))
			   (host    (db:test-get-host       tinfo))
			   (pid     (db:test-get-process_id tinfo))
			   (result (db:get-status-from-final-status-file run-dir)))
		      (if (and (list? result) (> (length result) 1) (equal? "PASS" (cadr result)) (equal? "COMPLETED" (car result))) 
			  (begin
			    (debug:print 0 *default-log-port* "INFO: test " test-id " actually passed, so marking PASS not DEAD")
			    (db:set-state-status-and-roll-up-items dbstruct run-id test-id 'foo "COMPLETED" "PASS"
								   "Test stopped responding but it has PASSED; marking it PASS in the DB."))
			  (let ((is-alive (launch:is-test-alive host pid)))
			    (db:set-state-status-and-roll-up-items
			     dbstruct run-id test-id 'foo "COMPLETED" "PASS"
			     "Test stopped responding but it has PASSED; marking it PASS in the DB."))
			  (let ((is-alive (and (not (eq? pid 0))  ;; 0 is default in re-used field "attemptnum" where pid stored.
					       (launch:is-test-alive host pid))))
			    (if is-alive
				(debug:print 0 *default-log-port* "INFO: test " test-id " on host " host " has a process on pid " pid ", NOT setting to DEAD.")
				(debug:print 0 *default-log-port* "INFO: test " test-id " on host " host
					     " has a process on pid " pid ", NOT setting to DEAD.")
				(begin
				  (debug:print 0 *default-log-port* "INFO: test " test-id " final state/status is not COMPLETED/PASS. It is " result)
				  (db:set-state-status-and-roll-up-items dbstruct run-id test-id 'foo "COMPLETED" "DEAD"
									 "Test stopped responding while in RUNNING or REMOTEHOSTSTART; presumed dead.")))))))
                     ;; call end of eud of run detection for posthook - from merge, is it needed?
                     ;; (launch:end-of-run-check run-id)
				  (debug:print 0 *default-log-port* "INFO: test " test-id
					       " final state/status is not COMPLETED/PASS. It is " result)
				  (db:set-state-status-and-roll-up-items
				   dbstruct run-id test-id 'foo "COMPLETED" "DEAD"
				   "Test stopped responding while in RUNNING or REMOTEHOSTSTART; presumed dead.")))))))
		  ;; call end of eud of run detection for posthook - from merge, is it needed?
		  ;; (launch:end-of-run-check run-id)
		  all-ids)
	       ;;call end of eud of run detection for posthook
	       (launch:end-of-run-check run-id)
	       )))))))
		 ;;call end of eud of run detection for posthook
		 (launch:end-of-run-check run-id)
		 )))))))


;; ALL REPLACED BY THE BLOCK ABOVE
;;
;; 	    (sqlite3:execute 
;; 	     db
;; 	     (conc "UPDATE tests SET state='INCOMPLETE' WHERE run_id=? AND id IN (" 
;; 		   (string-intersperse (map conc all-ids) ",")
;; 		   ");")
;;              run-id))))
;; 
;;     ;; Now do rollups for the toplevel tests
;;     ;;
;;     ;; (db:delay-if-busy dbdat)
;;     (for-each
;;      (lambda (toptest)
;;        (let ((test-name (list-ref toptest 3)))
;; ;;	     (run-id    (list-ref toptest 5)))
;; 	 (db:top-test-set-per-pf-counts dbstruct run-id test-name)))
;;      toplevels)))

;; BUG: Probably broken - does not explicitly use run-id in the query
;;
(define (db:top-test-set-per-pf-counts dbstruct run-id test-name)
  (db:general-call dbstruct 'top-test-set-per-pf-counts (list test-name test-name test-name test-name test-name test-name test-name test-name test-name test-name test-name test-name test-name test-name test-name test-name test-name)))

;; Clean out old junk and vacuum the database

			  (string->number res)
			  #f)))
          (if newres
              newres
              res))
        res)))

(define (db:no-sync-close-db db)
  (db:safely-close-sqlite3-db db))
(define (db:no-sync-close-db db stmt-cache)
  (db:safely-close-sqlite3-db db stmt-cache))

;; transaction protected lock aquisition
;; either:
;;    fails    returns  (#f . lock-creation-time)
;;    succeeds (returns (#t . lock-creation-time)
;; use (db:no-sync-del! db keyname) to release the lock
;;
(define (db:no-sync-get-lock db-in keyname)
  (let ((db (db:no-sync-db db-in)))
    (sqlite3:with-transaction
     db
     (lambda ()
       (handle-exceptions
	   exn
	   (let ((lock-time (current-seconds)))
	     (sqlite3:execute db "INSERT INTO no_sync_metadat (var,val) VALUES(?,?);" keyname lock-time)
	     `(#t . ,lock-time))
	 (let ((lock-time (current-seconds)))
	   (debug:print-info 2 *default-log-port* "db:no-sync-get-lock keyname=" keyname ", lock-time=" lock-time ", exn=" exn)
	   (sqlite3:execute db "INSERT INTO no_sync_metadat (var,val) VALUES(?,?);" keyname lock-time)
	   `(#t . ,lock-time))
	 `(#f . ,(sqlite3:first-result db "SELECT val FROM no_sync_metadat WHERE var=?;" keyname)))))))



;; use a global for some primitive caching, it is just silly to
;; re-read the db over and over again for the keys since they never
;; change

		       (lambda (key)
			 (set! res (cons key res)))
		       db
		       "SELECT fieldname FROM keys ORDER BY id DESC;")))
	(set! *db-keys* res)
	res)))

;; extract index number given a header/data structure
(define (db:get-index-by-header header field)
  (list-index (lambda (x)(equal? x field)) header))

;; look up values in a header/data structure
(define (db:get-value-by-header row header field)
  (if (or (null? header) (not row))
      #f
      (let loop ((hed (car header))
                 (tal (cdr header))
                 (n   0))
        (if (equal? hed field)
            (handle-exceptions
             exn
             (begin
               (debug:print 0 *default-log-port* "WARNING: attempt to read non-existant field, row=" row " header=" header " field=" field)
               (debug:print 0 *default-log-port* "WARNING: attempt to read non-existant field, row="
			    row " header=" header " field=" field ", exn=" exn)
               #f)
             (vector-ref row n))
	    (if (null? tal) #f (loop (car tal)(cdr tal)(+ n 1)))))))

;; Accessors for the header/data structure
;; get rows and header from 
(define (db:get-header vec)(vector-ref vec 0))

(define-record simple-run target id runname state status owner event_time)
(define-record-printer (simple-run x out)
  (fprintf out "#,(simple-run ~S ~S ~S ~S)"
	   (simple-run-target x) (simple-run-id x) (simple-run-runname x) (time->string (seconds->local-time (simple-run-event_time x) ))))

;; simple get-runs
;;
(define (db:simple-get-runs dbstruct runpatt count offset target)
(define (db:simple-get-runs dbstruct runpatt count offset target last-update)
    (let* ((res       '())
	   (keys       (db:get-keys dbstruct))
	   (runpattstr (db:patt->like "runname" runpatt))
	   (remfields  (list "id" "runname" "state" "status" "owner" "event_time"))
	   (targstr    (string-intersperse keys "||'/'||"))
	   (keystr     (conc targstr " AS target,"
			     (string-intersperse remfields ",")))
	   (qrystr     (conc "SELECT " keystr " FROM runs WHERE (" runpattstr ") " ;; runname LIKE ? "
			     ;; Generate: " AND x LIKE 'keypatt' ..."
			     " AND target LIKE '" target "'"
			     " AND state != 'deleted' ORDER BY event_time DESC "
			     " AND state != 'deleted' "
			     (if (number? last-update)
				 (conc " AND last_update >= " last-update)
				 "")
			     " ORDER BY event_time DESC "
			     (if (number? count)
				 (conc " LIMIT " count)
				 "")
			     (if (number? offset)
				 (conc " OFFSET " offset)
				 ""))))
				 "")))
	   )
    (debug:print-info 11 *default-log-port* "db:get-runs START qrystr: " qrystr " target: " target " offset: " offset " limit: " count)
    (db:with-db dbstruct #f #f
		(lambda (db)		
		  (sqlite3:for-each-row
		   (lambda (target id runname state status owner event_time)
		     (set! res (cons (make-simple-run target id runname state status owner event_time) res)))
		   db

						     (conc " ORDER BY " sort-by " ")
						     " ")))
				(if sort-order sort-order " ")
				(if limit  (conc " LIMIT " limit)   " ")
				(if offset (conc " OFFSET " offset) " ")
				";"
				)))

    (debug:print-info 8 *default-log-port* "db:get-tests-for-run run-id=" run-id ", qry=" qry)
    (let* ((res (db:with-db dbstruct run-id #f
			    (lambda (db)
			      ;; (let* ((stmth (db:get-cache-stmth dbstruct db qry))) ;; due to use of last-update we can't efficiently cache this query
				(reverse
				 (sqlite3:fold-row
				  (lambda (res . row)

	  -1 "-" "-"))

;;
;; 1. cache tests-match-qry
;; 2. compile qry and store in hash
;; 3. convert for-each-row to fold
;;
(define (db:get-tests-for-run-state-status dbstruct run-id testpatt)
#;(define (db:get-tests-for-run-state-status dbstruct run-id testpatt)
  (db:with-db
   dbstruct run-id #f
   (lambda (db)
     (let* ((res            '())
	    (stmt-cache      (dbr:dbstruct-stmt-cache dbstruct))
	    (stmth           (let* ((sh (db:hoh-get stmt-cache db testpatt)))
			       (or sh

	(sqlite3:fold-row
	 (lambda (res id testname item-path state status)
	   ;;  id,run_id,testname,state,status,event_time,host,cpuload,diskfree,uname,rundir,item_path,run_duration,final_logf,comment
	   (cons (vector id run-id testname state status -1         ""     -1      -1       ""    "-"  item-path -1           "-"         "-") res))
	 '()
	 stmth
	 run-id))))))

(define (db:get-tests-for-run-state-status dbstruct run-id testpatt #!optional (last-update 0))
  (let* ((res            '())
	 (tests-match-qry (tests:match->sqlqry testpatt))
	 (qry             (conc "SELECT id,testname,item_path,state,status,event_time,run_duration FROM tests WHERE run_id=? " 
				" AND last_update > ? "
				(if tests-match-qry (conc " AND (" tests-match-qry ") ") "")
				)))
    (debug:print-info 8 *default-log-port* "db:get-tests-for-run qry=" qry)
    (db:with-db dbstruct run-id #f
		(lambda (db)
		  (sqlite3:fold-row
		   (lambda (res id testname item-path state status event-time run-duration)
		     ;;            id,run_id,testname,state,status,event_time,host,cpuload,diskfree,uname,rundir,item_path,run_duration,final_logf,comment
		     (cons (vector id run-id testname state status event-time  ""     -1      -1       ""    "-"  item-path run-duration  "-"         "-") res))
		   '()
		   db 
		   qry
		   run-id
		   (or last-update 0))))))

(define (db:get-testinfo-state-status dbstruct run-id test-id)
  (let ((res            #f))
    (db:with-db dbstruct run-id #f
		(lambda (db)
		  (sqlite3:for-each-row
		   (lambda (run-id testname item-path state status)

  (db:with-db
   dbstruct #f #f
   (lambda (db)
     (sqlite3:execute db "UPDATE tests SET state='DELETED',status='n/a',comment='' WHERE id=?;" test-id))))

;; 
(define (db:delete-old-deleted-test-records dbstruct)
  (let (;; (run-ids  (db:get-all-run-ids dbstruct))
	(targtime (- (current-seconds)(* 30 24 60 60)))) ;; one month in the past
  (let ((targtime (- (current-seconds)
		     (or (configf:lookup-number *configdat* "setup" "keep-deleted-records")
			 (* 30 24 60 60))))) ;; one month in the past
    (db:with-db
     dbstruct
     0
     #t
     (lambda (db)
       (sqlite3:with-transaction
	db

;; NEW BEHAVIOR: Look only at single run with run-id
;; 
;; (define (db:get-running-stats dbstruct run-id)
(define (db:get-count-tests-running-for-run-id dbstruct run-id fastmode)
  (let* ((qry (if fastmode
		  "SELECT count(id) FROM tests WHERE state in ('RUNNING','LAUNCHED','REMOTEHOSTSTART') AND run_id=? LIMIT 1;"
		   "SELECT count(id) FROM tests WHERE state in ('RUNNING','LAUNCHED','REMOTEHOSTSTART') AND run_id=?;")))
		  "SELECT count(id) FROM tests WHERE state in ('RUNNING','LAUNCHED','REMOTEHOSTSTART') AND run_id=?;")))
    (db:with-db
     dbstruct
     run-id
     #f
     (lambda (db)
       (let* ((stmth (db:get-cache-stmth dbstruct db qry)))
	 (sqlite3:first-result stmth run-id))))))

;; For a given testname how many items are running? Used to determine
;; probability for regenerating html
;; 
;;
(define (db:get-count-tests-running-for-testname dbstruct run-id testname)
  (db:with-db
   dbstruct
   run-id
   #f
   (lambda (db)
     (let* ((stmt "SELECT count(id) FROM tests WHERE state in ('RUNNING','LAUNCHED','REMOTEHOSTSTART') AND run_id=? AND NOT (uname = 'n/a' AND item_path = '') AND testname=?;")
	    (stmth (db:get-cache-stmth dbstruct db stmt)))
     (sqlite3:first-result
       (sqlite3:first-result
      db
      "SELECT count(id) FROM tests WHERE state in ('RUNNING','LAUNCHED','REMOTEHOSTSTART') AND run_id=? AND NOT (uname = 'n/a' AND item_path = '') AND testname=?;" run-id testname))))

	stmth run-id testname)))))

(define (db:get-not-completed-cnt dbstruct run-id)
(db:with-db
   dbstruct
   run-id
   #f
   (lambda (db)

  (let* ((run-ids (db:get-all-run-ids mtdb)))
    (for-each 
     (lambda (run-id)
       (let ((testrecs (db:get-all-tests-info-by-run-id mtdb run-id)))
	 (db:prep-megatest.db-adj-test-ids (db:dbdat-get-db mtdb) run-id testrecs)))
     run-ids)))

;; Get test data using test_id, run-id is not used
;; Get test data using test_id, run-id is not used - but it will be!
;; 
(define (db:get-test-info-by-id dbstruct run-id test-id)
  (db:with-db
   dbstruct
   #f ;; run-id
   #f
   (lambda (db)

                                     db
                                     "SELECT state,status,count(id) FROM tests WHERE run_id=?  GROUP BY state,status;"
                                     run-id )))))
   test-count-recs))


;; BBnote: db:get-all-state-status-counts-for-test returns dbr:counts object aggregating state and status of items of a given test, *not including rollup state/status*
;;
;; NOTE: This is called within a transaction
;;
(define (db:get-all-state-status-counts-for-test dbstruct run-id test-name item-path item-state-in item-status-in)
  (let* ((test-info   (db:get-test-info dbstruct run-id test-name item-path))
         (item-state  (or item-state-in (db:test-get-state test-info))) 
         (item-status (or item-status-in (db:test-get-status test-info)))
         (other-items-count-recs (db:with-db
                                  dbstruct #f #f
                                  (lambda (db)

      (if dbdat
	  (let* ((dbpath (db:dbdat-get-path dbdat))
		 (db     (db:dbdat-get-db   dbdat)) ;; we'll return this so (db:delay--if-busy can be called inline
		 (dbfj   (conc dbpath "-journal")))
	    (if (handle-exceptions
		 exn
		 (begin
		   (debug:print-info 0 *default-log-port* "WARNING: failed to test for existance of " dbfj)
		   (debug:print-info 0 *default-log-port* "WARNING: failed to test for existance of " dbfj ", exn=" exn)
		   (thread-sleep! 1)
		   (db:delay-if-busy count (- count 1))) 
		 (common:file-exists? dbfj))
		(case count
		  ((6)
		   (thread-sleep! 0.2)
		   (db:delay-if-busy count: 5))

		 (newr  (if (and patt repl)
			    (begin
                              (handle-exceptions
                               exn
                               (begin
                                  (debug:print 0 *default-log-port*
                                  "WARNING: itemmap has problem \"" itemmap "\", patt: " patt ", repl: " repl)
                                  "WARNING: itemmap has problem \"" itemmap "\", patt: " patt ", repl: " repl ", exn=" exn)
                                 res)
                              (string-substitute patt repl res))


                              )
			    (begin
                              (debug:print 0 *default-log-port*

;;======================================================================
;; Copyright 2017, Matthew Welland.
;; 
;; This file is part of Megatest.
;; 
;;     Megatest is free software: you can redistribute it and/or modify
;;     it under the terms of the GNU General Public License as published by
;;     the Free Software Foundation, either version 3 of the License, or
;;     (at your option) any later version.
;; 
;;     Megatest is distributed in the hope that it will be useful,
;;     but WITHOUT ANY WARRANTY; without even the implied warranty of
;;     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
;;     GNU General Public License for more details.
;; 
;;     You should have received a copy of the GNU General Public License
;;     along with Megatest.  If not, see <http://www.gnu.org/licenses/>.

;;======================================================================

(declare (unit dbmod))

(module dbmod
	*
	
(import scheme chicken data-structures extras)
(import (prefix sqlite3 sqlite3:) posix typed-records srfi-18)

(define (just-testing)
  (print "JUST TESTING"))

;; (define (debug:print . params) #f)
;; (define (debug:print-info . params) #f)
;; 
;; (define (set-functions dbgp dbgpinfo)
;;   (set! debug:print dbgp)
;;   (set! debug:print-info dbgpinfo))

)

(import (prefix iup iup:))
(use canvas-draw)
(import canvas-draw-iup)
(use regex typed-records matchable)

(declare (unit dcommon))

(declare (uses megatest-version))
(declare (uses gutils))
(declare (uses db))
;; (declare (uses synchash))

(include "megatest-version.scm")
(include "common_records.scm")
(include "db_records.scm")
(include "key_records.scm")
(include "run_records.scm")

;; yes, this is non-ideal 
(define dashboard:update-summary-tab #f)

	     (numrows      1)
	     (numcols      1))
	(iup:attribute-set! stats-matrix "CLEARVALUE" "CONTENTS")
	(iup:attribute-set! stats-matrix "NUMCOL" max-col )
	(iup:attribute-set! stats-matrix "NUMLIN" (if (< max-row max-visible) max-visible max-row)) ;; min of 20
	(iup:attribute-set! stats-matrix "NUMCOL_VISIBLE" max-col-vis)
	(iup:attribute-set! stats-matrix "NUMLIN_VISIBLE" (if (> max-row max-visible) max-visible max-row))
	(print "row-indices: " row-indices " col-indices: " col-indices)
	;;(print "row-indices: " row-indices " col-indices: " col-indices)
	;; Row labels
	(for-each (lambda (ind)
		    (let* ((name (car ind))
			   (num  (cadr ind))
			   (key  (conc num ":0")))
		      (if (not (equal? (iup:attribute stats-matrix key) name))
			  (begin

    (dboard:commondat-please-update-set! commondat #f)
    recalc))

(define (dashboard:get-youngest-run-db-mod-time dbdir)
  (handle-exceptions
   exn
   (begin
     (debug:print 2 *default-log-port* "WARNING: error in accessing databases in get-youngest-run-db-mod-time: " ((condition-property-accessor 'exn 'message) exn) " db-dir="dbdir)
     (debug:print 2 *default-log-port* "WARNING: error in accessing databases in get-youngest-run-db-mod-time: " ((condition-property-accessor 'exn 'message) exn)
		  " db-dir="dbdir ", exn=" exn)
     (current-seconds)) ;; something went wrong - just print an error and return current-seconds
   (common:max (map (lambda (filen)
		      (file-modification-time filen))
		    (glob (conc dbdir "/*.db*"))))))

(define (dboard:get-last-db-update tabdat context)
  (hash-table-ref/default (dboard:tabdat-last-db-update tabdat) context 0))

~~~~~~~~~~~~

Chicken scheme and a number of "eggs" are required for building
Megatest. See the script installall.sh in the utils directory of the
source distribution for an automated way to install everything
needed for building Megatest on Linux.

Megatest. In the v1.66 and beyond assistance to create the build
system is built into the Makefile.

.Installation steps (overview)
-------------------------------------
./configure
make chicken
setup.sh make -j install
-------------------------------------

Or install the needed build system manually:

. Chicken scheme from http://call-cc.org
. IUP from http://webserver2.tecgraf.puc-rio.br/iup/
. CD from http://webserver2.tecgraf.puc-rio.br/cd/
. IM from https://webserver2.tecgraf.puc-rio.br/im/
. ffcall from http://webserver2.tecgraf.puc-rio.br/iup/
. Nanomsg from https://nanomsg.org/ (NOTE: Plan is to eliminate nanomsg dependency).
. Needed eggs (look at the eggs lists in the Makefile)

Then follow these steps:

.Installation steps (self-built chicken scheme build system)
-------------------------------------
./configure
make -j install
-------------------------------------

</script>
</head>
<body class="book">
<div id="header">
<h1>The Megatest Users Manual</h1>
<span id="author">Matt Welland</span><br>
<span id="email" class="monospaced">&lt;<a href="mailto:matt@kiatoa.com">matt@kiatoa.com</a>&gt;</span><br>
<span id="revnumber">version 1.0,</span>
<span id="revdate">April 2012</span>
<span id="revnumber">version 1.5,</span>
<span id="revdate">June 2020</span>
<div id="toc">
  <div id="toctitle">Table of Contents</div>
  <noscript><p><b>JavaScript must be enabled in your browser to display the table of contents.</b></p></noscript>
</div>
</div>
<div id="content">
<div class="sect1">
<h2 id="_preface">Preface</h2>
<div class="sectionbody">
<div class="paragraph"><p>This book is organised as three sub-books; getting started, writing tests and reference.</p></div>
<div class="listingblock">
<div class="title">License</div>
<div class="content monospaced">
<pre>    Copyright 2006-2017, Matthew Welland.
<pre>    Copyright 2006-2020, Matthew Welland.

    This document is part of Megatest.

    Megatest is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation, either version 3 of the License, or
    (at your option) any later version.

<h2 id="_why_megatest">Why Megatest?</h2>
<div class="sectionbody">
<div class="paragraph"><p>The Megatest project was started for two reasons, the first was an
immediate and pressing need for a generalized tool to manage a suite
of regression tests and the second was the fact that I had written or
maintained several such tools at different companies over the years. I
thought a single open source tool, flexible enough to meet the needs
of any team doing continuous integrating and or running a complex
of any team doing continuous integration and or running a complex
suite of tests for release qualification would solve some problems for
me and for others.</p></div>
<div class="literalblock">
<div class="content monospaced">
<pre>-- Matt Welland, original author of the Megatest tool suite.</pre>
</div></div>
</div>

<div class="imageblock">
<div class="content">
<img src="megatest-system-architecture.png" alt="Static">
</div>
</div>
</div>
</div>
</div>
<div class="sect1">
<h2 id="_road_map">Road Map</h2>
<div class="sectionbody">
</div>
</div>
<div class="sect1">
<h2 id="_todo_road_map">TODO / Road Map</h2>
<div class="sectionbody">
<div class="paragraph"><p>Note: This road-map is a wish list and not a formal plan. Items are in
rough priority but are subject to change. Development is driven by
user requests, developer "itch" and bug reports. Please contact
<a href="mailto:matt@kiatoa.com">matt@kiatoa.com</a> with requests or bug reports. Requests from inside
Intel generally take priority.</p></div>
<div class="paragraph"><p>Dashboard and runs</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
Multi-area dashboard view
</p>
</li>
</ol></div>
<div class="paragraph"><p>Tests Support</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
Add variable $MT_RUNPATH = $MT_LINKTREE/$MT_TARGET/$MT_RUNNAME
</p>
</li>
<li>
<p>
Improve [script], especially indent handling
</p>
</li>
</ol></div>
<div class="paragraph"><p>Scalability</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
Overflow database methodology - combine the best of the v1.63
  multi-db approach and the current db-in-tmp approach (currently
  slowness can be seen when number of tests in a db goes over 50-100k,
  with the overflow db it will be able to handle 1000&#8217;s of runs with
  50-100k tests per run). High priority - goal is to complete this by
  20Q3.
</p>
</li>
</ol></div>
<div class="paragraph"><p>Mtutils/CI</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
Enable mtutil calls from dashboard (for remote control)
</p>
</li>
<li>
<p>
Logs browser (esp. for surfacing mtutil related activities)
</p>
</li>
<li>
<p>
Embed ftfplan for distributed automation, completed activities trigger QA runs which trigger deployment etc.
</p>
</li>
<li>
<p>
Jenkins junit XML support [DONE]
</p>
</li>
<li>
<p>
Add output flushing in teamcity support
</p>
</li>
</ol></div>
<div class="paragraph"><p>Build system</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
./configure &#8658; ubuntu, sles11, sles12, rh7 [WIP]
</p>
</li>
<li>
<p>
Switch to using simple runs query everywhere
</p>
</li>
<li>
<p>
Add end_time to runs and add a rollup call that sets state, status and end_time
</p>
</li>
</ol></div>
<div class="paragraph"><p>Code refactoring/quality/performance</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
Switch to scsh-process pipeline management for job execution/control
</p>
</li>
<li>
<p>
Use call-with-environment-variables where possible.
</p>
</li>
</ol></div>
<div class="paragraph"><p>Migration to inmem db and or overflow db</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
Re-work the dbstruct data structure?
</p>
<div class="olist loweralpha"><ol class="loweralpha">
<li>
<p>
[ run-id.db inmemdb last-mod last-read last-sync inuse ]
</p>
</li>
</ol></div>
</li>
</ol></div>
<div class="paragraph"><p>Some ideas for Megatest 2.0</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
Aggressive megatest.config and runconfig.config caching.
</p>
<div class="olist loweralpha"><ol class="loweralpha">
<li>
<p>
Cache the configs in $MT_RUNPATH
</p>
</li>
<li>
<p>
Following invocations of –run, -rerun* will calculate the new config but only overwrite the cached file IF changed
</p>
</li>
</ol></div>
</li>
<li>
<p>
If the cached file changes ALL existing tests go from COMPLETED &#8594; STALE, I’m not sure what to do about RUNNING tests
</p>
</li>
<li>
<p>
!VARS in runconfigs are not exported to the environment. They are accessed via rget as if the ! was not there.
</p>
</li>
<li>
<p>
Per test copy commands (example is incomplete).
</p>
</li>
</ol></div>
<div class="listingblock">
<div class="content monospaced">
<pre>[testcopy]
%/iind% unison SRC DEST
% cp –r SRC DEST</pre>
</div></div>
<div class="paragraph"><p>Add ability to move runs to other Areas (overlaps with overflow db system)</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
allow shrinking megatest.db data by moving runs to an alternate
  Megatest area with same keys.
</p>
</li>
<li>
<p>
add param -destination [area|path]. when specified runs are copied to new
  area and removed from local db.
</p>
</li>
<li>
<p>
the data move would involve these steps
</p>
<div class="olist loweralpha"><ol class="loweralpha">
<li>
<p>
copy the run data to destination area megatest.db
</p>
</li>
<li>
<p>
mark the run records as deleted, do not remove the run data on disk
</p>
</li>
</ol></div>
</li>
<li>
<p>
accessing the data would be by running dashboard in the satellite area
</p>
</li>
<li>
<p>
future versions of Megatest dashboard should support displaying areas in a
  merged way.
</p>
</li>
<li>
<p>
some new controls would be supported in the config
</p>
<div class="olist loweralpha"><ol class="loweralpha">
<li>
<p>
[setup] &#8658; allow-runs [no|yes]  &#8656;= used to disallow runs
</p>
</li>
<li>
<p>
[setup] &#8658; auto-migrate=[areaname|path]  &#8656;= used to automatically
   migrate data to a satellite area.
</p>
</li>
</ol></div>
</li>
</ol></div>
<div class="paragraph"><p>Eliminate ties to homehost (part of overflow db system)</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
Server creates captain pkt
</p>
</li>
<li>
<p>
Create a lock in the db
</p>
</li>
<li>
<p>
Relinquish db when done
</p>
</li>
</ol></div>
<div class="paragraph"><p>Tasks - better management of run manager processes etc.</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
adjutant queries tasks table for next action <span class="red">[Migrate into mtutil]</span>
</p>
<div class="olist loweralpha"><ol class="loweralpha">
<li>
<p>
Task table used for tracking runner process <span class="red">[Replaced by mtutil]</span>
</p>
</li>
<li>
<p>
Task table used for jobs to run <span class="red">[Replaced by mtutil]</span>
</p>
</li>
<li>
<p>
Task table used for queueing runner actions (remove runs,
   cleanRunExecute, etc)  <span class="red">[Replaced by mtutil</span>]
</p>
</li>
</ol></div>
</li>
<li>
<p>
adjutant (server/task dispatch/execution manager)
</p>
</li>
</ol></div>
<div class="paragraph"><p>Stale propagation</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
Mark dependent tests for clean/rerun -rerun-downstream
</p>
</li>
<li>
<p>
On run start check for defunct tests in RUNNING, LAUNCHED or REMOTEHOSTSTART and correct or notify
</p>
</li>
<li>
<p>
Fix: refresh of gui sometimes fails on last item (race condition?)
</p>
</li>
</ol></div>
<div class="paragraph"><p>Bin list</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
Rerun step and or subsequent steps from gui [DONE?]
</p>
</li>
<li>
<p>
Refresh test area files from gui
</p>
</li>
<li>
<p>
Clean and re-run button
</p>
</li>
<li>
<p>
Clean up STATE and STATUS handling.
</p>
<div class="olist loweralpha"><ol class="loweralpha">
<li>
<p>
Dashboard and Test control panel are reverse order - choose and fix
</p>
</li>
<li>
<p>
Move seldom used states and status to drop down selector
</p>
</li>
</ol></div>
</li>
<li>
<p>
Access test control panel when clicking on Run Summary tests
</p>
</li>
<li>
<p>
Feature: -generate-index-tree
</p>
</li>
<li>
<p>
Change specifing of state and status to use STATE1/STATUS1,STATE2/STATUS2
</p>
</li>
<li>
<p>
rest api available for use with Perl, Ruby etc. scripts
</p>
</li>
<li>
<p>
megatest.config setup entries for:
</p>
<div class="olist loweralpha"><ol class="loweralpha">
<li>
<p>
run launching (e.g. /bin/sh %CMD% &gt; /dev/null)
</p>
</li>
<li>
<p>
browser "konqueror %FNAME%
</p>
</li>
</ol></div>
</li>
<li>
<p>
refdb: Add export of csv, json and sexp
</p>
</li>
<li>
<p>
Convert to using call-with-environment-variables where possible. Should allow handling of parallel runs in same process.
</p>
</li>
<li>
<p>
Re-work text interface wizards. Several bugs on record. Possibly convert to gui based.
</p>
</li>
<li>
<p>
Add to testconfig requirements section; launchlimiter scriptname, calls scriptname to check if ok to launch test
</p>
</li>
<li>
<p>
Refactor Run Summary view, currently very clumsy
</p>
</li>
<li>
<p>
Add option to show steps in Run Summary view
</p>
</li>
<li>
<p>
Refactor guis for resizeablity
</p>
</li>
<li>
<p>
Add filters to Run Summary view and Run Control view
</p>
</li>
<li>
<p>
Add to megatest.config or testconfig; rerunok STATE/STATUS,STATE/STATUS&#8230;
</p>
</li>
<li>
<p>
Launch gates for diskspace; /path/one&gt;1G,/path/two&gt;200M,/tmp&gt;5G,#{scheme <strong>toppath</strong>}&gt;1G
</p>
</li>
<li>
<p>
Tool tips
</p>
</li>
<li>
<p>
Filters on Run Summary, Summary and Run Control panel
</p>
</li>
<li>
<p>
Built in log viewer (partially implemented)
</p>
</li>
<li>
<p>
Refactor the test control panel
   Help and documentation
</p>
</li>
<li>
<p>
Complete the user manual (I’ve been working on this lately).
</p>
</li>
<li>
<p>
Online help in the gui
   Streamlined install
</p>
</li>
<li>
<p>
Deployed or static build
</p>
</li>
<li>
<p>
Added option to compile IUP (needed for VMs)
</p>
</li>
<li>
<p>
Server side run launching
</p>
</li>
<li>
<p>
Wizards for creating tests, regression areas (current ones are text only and limited).
</p>
</li>
<li>
<p>
Fully functional built in web service (currently you can browse runs but it is very simplistic).
</p>
</li>
<li>
<p>
Gui panels for editing megatest.config and runconfigs.config
</p>
</li>
<li>
<p>
Fully isolated tests (no use of NFS to see regression area files)
</p>
</li>
<li>
<p>
Windows version
</p>
</li>
</ol></div>
</div>
</div>
<div class="sect1">
<h2 id="_installation">Installation</h2>
<div class="sectionbody">
<div class="sect2">
<h3 id="_dependencies">Dependencies</h3>
<div class="paragraph"><p>Chicken scheme and a number of "eggs" are required for building
Megatest. See the script installall.sh in the utils directory of the
source distribution for an automated way to install everything
needed for building Megatest on Linux.</p></div>
<div class="paragraph"><p>Megatest. In the v1.66 and beyond assistance to create the build
system is built into the Makefile.</p></div>
<div class="listingblock">
<div class="title">Installation steps (overview)</div>
<div class="content monospaced">
<pre>./configure
make chicken
setup.sh make -j install</pre>
</div></div>
<div class="paragraph"><p>Or install the needed build system manually:</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
Chicken scheme from <a href="http://call-cc.org">http://call-cc.org</a>
</p>
</li>
<li>
<p>
IUP from <a href="http://webserver2.tecgraf.puc-rio.br/iup/">http://webserver2.tecgraf.puc-rio.br/iup/</a>
</p>
</li>
<li>
<p>
CD from <a href="http://webserver2.tecgraf.puc-rio.br/cd/">http://webserver2.tecgraf.puc-rio.br/cd/</a>
</p>
</li>
<li>
<p>
IM from <a href="https://webserver2.tecgraf.puc-rio.br/im/">https://webserver2.tecgraf.puc-rio.br/im/</a>
</p>
</li>
<li>
<p>
ffcall from <a href="http://webserver2.tecgraf.puc-rio.br/iup/">http://webserver2.tecgraf.puc-rio.br/iup/</a>
</p>
</li>
<li>
<p>
Nanomsg from <a href="https://nanomsg.org/">https://nanomsg.org/</a> (NOTE: Plan is to eliminate nanomsg dependency).
</p>
</li>
<li>
<p>
Needed eggs (look at the eggs lists in the Makefile)
</p>
</li>
</ol></div>
<div class="paragraph"><p>Then follow these steps:</p></div>
<div class="listingblock">
<div class="title">Installation steps (self-built chicken scheme build system)</div>
<div class="content monospaced">
<pre>./configure
make -j install</pre>
</div></div>
</div>
</div>
</div>
<div class="sect1">
<h2 id="_getting_started">Getting Started</h2>
<div class="sectionbody">
<div class="listingblock">

# script that takes path as parameter and returns number of bytes available:
free-space-script check-space.sh</pre>
</div></div>
</div>
<div class="sect2">
<h3 id="_trim_trailing_spaces">Trim trailing spaces</h3>
<div class="admonitionblock">
<table><tr>
<td class="icon">
<img src="/nfs/pdx/disks/ice.disk.002/icfadm/pkgs/asciidoc/8.6.7/images/icons/note.png" alt="Note">
</td>
<td class="content">As of Megatest version v1.6548 trim-trailing-spaces defaults to yes.</td>
</tr></table>
</div>
<div class="listingblock">
<div class="content monospaced">
<pre>[configf:settings trim-trailing-spaces yes]</pre>
<pre>[configf:settings trim-trailing-spaces no]
#                 |&lt;== next line padded with spaces to here
DEFAULT_INDENT
[configf:settings trim-trailing-spaces no]</pre>
</div></div>
<div class="paragraph"><p>The variable DEFAULT_INDENT would be a string of 3 spaces</p></div>
</div>
<div class="sect2">
<h3 id="_job_submission_control">Job Submission Control</h3>
<div class="sect3">
<h4 id="_submit_jobs_to_host_types_based_on_test_name_2">Submit jobs to Host Types based on Test Name</h4>
<div class="listingblock">
<div class="title">In megatest.config</div>

<h3 id="_wait_on_other_tests">Wait on Other Tests</h3>
<div class="listingblock">
<div class="content monospaced">
<pre># A normal waiton waits for the prior tests to be COMPLETED
# and PASS, CHECK or WAIVED
waiton test1 test2</pre>
</div></div>
<div class="admonitionblock">
<table><tr>
<td class="icon">
<img src="/nfs/pdx/disks/ice.disk.002/icfadm/pkgs/asciidoc/8.6.7/images/icons/note.png" alt="Note">
</td>
<td class="content">Dynamic waiton lists must be capable of being calculated at the
beginning of a run. This is because Megatest walks the tree of waitons
to create the list of tests to execute.</td>
</tr></table>
</div>
<div class="listingblock">
<div class="title">This works</div>
<div class="content monospaced">
<pre>waiton [system somescript.sh]</pre>
</div></div>
<div class="listingblock">
<div class="title">This does NOT work (the full context for the test is not available so #{shell &#8230;} is NOT enabled to evaluate.</div>
<div class="content monospaced">
<pre>waiton #{shell somescript.sh}</pre>
</div></div>
<div class="listingblock">
<div class="title">This does NOT work</div>
<div class="content monospaced">
<pre>waiton [system somescript_that_depends_on_a_prior_test.sh]</pre>
</div></div>
</div>
<div class="sect2">
<h3 id="_mode">Mode</h3>
<div class="paragraph"><p>The default (i.e. if mode is not specified) is normal. All pre-dependent tests
must be COMPLETED and PASS, CHECK or WAIVED before the test will start</p></div>
<div class="listingblock">
<div class="content monospaced">

<h3 id="_disks">Disks</h3>
<div class="paragraph"><p>A disks section in testconfig will override the disks section in
megatest.config. This can be used to allocate disks on a per-test or per item
basis.</p></div>
</div>
<div class="sect2">
<h3 id="_controlled_waiver_propagation">Controlled waiver propagation</h3>
<div class="paragraph"><p>If test is FAIL and previous test in run with same MT_TARGET is WAIVED then apply the following rules from the testconfig:
If a waiver check is specified in the testconfig apply the check and if it passes then set this FAIL to WAIVED</p></div>
<div class="paragraph"><p>Waiver check has two parts, 1) a list of waiver, rulename, filepatterns and 2) the rulename script spec (note that "diff" and "logpro" are predefined)</p></div>
<div class="paragraph"><p>If test is FAIL and previous test in run with same MT_TARGET is WAIVED
or if the test/itempath is listed under the matching target in the
waivers roll forward file (see below for file spec) then apply the
following rules from the testconfig: If a waiver check is specified in
the testconfig apply the check and if it passes then set this FAIL to
WAIVED</p></div>
<div class="paragraph"><p>Waiver check has two parts, 1) a list of waiver, rulename,
filepatterns and 2) the rulename script spec (note that "diff" and
"logpro" are predefined)</p></div>
<div class="listingblock">
<div class="content monospaced">
<pre>###### EXAMPLE FROM testconfig #########
# matching file(s) will be diff'd with previous run and logpro applied
# if PASS or WARN result from logpro then WAIVER state is set
#
[waivers]
# logpro_file    rulename      input_glob
waiver_1         logpro        lookittmp.log

[waiver_rules]

# This builtin rule is the default if there is no &lt;waivername&gt;.logpro file
# diff   diff %file1% %file2%

# This builtin rule is applied if a &lt;waivername&gt;.logpro file exists
# logpro diff %file1% %file2% | logpro %waivername%.logpro %waivername%.html</pre>
</div></div>
<div class="sect3">
<h4 id="_waiver_roll_forward_files">Waiver roll-forward files</h4>
<div class="paragraph"><p>To transfer waivers from one Megatest area to another it is possible
to dump waivers into a file and reference that file in another area.</p></div>
<div class="listingblock">
<div class="title">Dumping the waivers</div>
<div class="content monospaced">
<pre>megatest -list-waivers -runname %-a &gt; mywaivers.dat</pre>
</div></div>
<div class="listingblock">
<div class="title">Referencing the saved waivers</div>
<div class="content monospaced">
<pre># In megatest.config, all files listed will be loaded - recomended to use
# variables to select directorys to minimize what gets loaded.
[setup]
waivers-dirs /path/to/waiver/files /another/path/to/waiver/files</pre>
</div></div>
<div class="listingblock">
<div class="title">Waiver files format</div>
<div class="content monospaced">
<pre>[the/target/here]
# comments are fine
testname1/itempath A comment about why it was waived
testname2          A comment for a non-itemized test</pre>
</div></div>
</div>
</div>
<div class="sect2">
<h3 id="_ezsteps">Ezsteps</h3>
<div class="listingblock">
<div class="title">Example ezsteps with logpro rules</div>
<div class="content monospaced">
<pre>[ezsteps]

<td class="tableblock halign-left valign-top" ><p class="tableblock monospaced"></p></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock monospaced">( key1 key2 &#8230; )</p></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock monospaced"></p></td>
</tr>
</tbody>
</table>
</div>
</div>
<div class="sect1">
<h2 id="_test_plan">Test Plan</h2>
<div class="sectionbody">
<div class="sect2">
<h3 id="_tests">Tests</h3>
<div class="paragraph"><p>itemwait|33</p></div>
<div class="paragraph"><p>rerun-downstream-item|20</p></div>
<div class="paragraph"><p>rerunclean|20</p></div>
<div class="paragraph"><p>fullrun|18</p></div>
<div class="paragraph"><p>goodtests|18</p></div>
<div class="paragraph"><p>kill-rerun|17</p></div>
<div class="paragraph"><p>items-runconfigvars|16</p></div>
<div class="paragraph"><p>ro_test|16</p></div>
<div class="paragraph"><p>runconfig-tests|16</p></div>
<div class="paragraph"><p>env-pollution|13</p></div>
<div class="paragraph"><p>itemmap|11</p></div>
<div class="paragraph"><p>testpatt_envvar|10</p></div>
<div class="paragraph"><p>toprun|10</p></div>
<div class="paragraph"><p>chained-waiton|8</p></div>
<div class="paragraph"><p>skip-on-fileexists|8</p></div>
<div class="paragraph"><p>killrun_preqfail|7</p></div>
<div class="paragraph"><p>subrun|6</p></div>
<div class="paragraph"><p>dependencies|5</p></div>
<div class="paragraph"><p>itemwait-simple|4</p></div>
<div class="paragraph"><p>rollup|4</p></div>
<div class="paragraph"><p>end-of-run|3</p></div>
<div class="paragraph"><p>killrun|3</p></div>
<div class="paragraph"><p>listener|3</p></div>
<div class="paragraph"><p>test2|3</p></div>
<div class="paragraph"><p>testpatt|3</p></div>
<div class="paragraph"><p>env-pollution-usecacheno|2</p></div>
<div class="paragraph"><p>set-values|2
envvars|1
listruns-tests|1
subrun-usecases|1</p></div>
</div>
</div>
</div>
<div class="sect1">
<h2 id="_megatest_internals">Megatest Internals</h2>
<div class="sectionbody">
<div class="imageblock graphviz">
<div class="content">
<img src="server.png" alt="server.png">
</div>
</div>
</div>
</div>
<div class="sect1">
<h2 id="_road_map">Road Map</h2>
<div class="sectionbody">
<div class="paragraph"><p>Note 1: This road-map continues to evolve and subject to change without notice.</p></div>
<div class="sect2">
<h3 id="_here_is_a_smattering_of_ideas_for_megatest_2_0">Here is a smattering of ideas for Megatest 2.0</h3>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
Add variable $MT_RUNPATH = $MT_LINKTREE/$MT_TARGET/$MT_RUNNAME
</p>
</li>
</ol></div>
<div class="listingblock">
<div class="content monospaced">
<pre>[db]
api legacy|new</pre>
</div></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
One big lesson from the 1.63-1.65 generation was that the main.db, 1.db … model was really good at scaling. I’d like to combine that model with the current also-very-good model. Obviously this is a disruptive change. I think making the old model the default and the new model an option for at least one generation would be fair.
</p>
</li>
<li>
<p>
Rigorous megatest.config and runconfig.config caching.
</p>
<div class="olist loweralpha"><ol class="loweralpha">
<li>
<p>
Cache the configs in $MT_RUNPATH
</p>
</li>
<li>
<p>
Following invocations of –run, -rerun* will calculate the new config but only overwrite the cached file IF changed
</p>
</li>
</ol></div>
</li>
<li>
<p>
If the cached file changes ALL existing tests go from COMPLETED &#8594; STALE, I’m not sure what to do about RUNNING tests
</p>
</li>
<li>
<p>
!VARS in runconfigs are not exported to the environment. They are accessed via rget as if the ! was not there.
</p>
</li>
<li>
<p>
Per test copy commands (crude example below is not correct).
</p>
</li>
</ol></div>
<div class="listingblock">
<div class="content monospaced">
<pre>[testcopy]
%/iind% unison SRC DEST
% cp –r SRC DEST</pre>
</div></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
Test management via pkts (optional?)
</p>
<div class="olist loweralpha"><ol class="loweralpha">
<li>
<p>
Control pkt types: run, kill, rerunclean, clean, archive, status?
</p>
</li>
<li>
<p>
Status pkt types: ack, step, status_change
</p>
</li>
</ol></div>
</li>
<li>
<p>
Add nanomsg as a transport option
.
</p>
</li>
</ol></div>
</div>
<div class="sect2">
<h3 id="_rfc_m01_add_ability_to_move_runs_to_other_areas">RFC M01: Add ability to move runs to other Areas</h3>
<div class="paragraph"><p><strong>Purpose</strong>: allow shrinking megatest.db data by moving runs to an alternate
 Megatest area with same keys.</p></div>
<div class="paragraph"><p><strong>Method</strong>: extend db sync to take a different megatest area as a destination.</p></div>
<div class="paragraph"><p><strong>Design</strong>:</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
add param -destination [area|path]. when specified runs are copied to new
  area and removed from local db.
</p>
</li>
<li>
<p>
the data move would involve these steps
</p>
<div class="olist loweralpha"><ol class="loweralpha">
<li>
<p>
copy the run data to destination area megatest.db
</p>
</li>
<li>
<p>
mark the run records as deleted, do not remove the run data on disk
</p>
</li>
</ol></div>
</li>
<li>
<p>
accessing the data would be by running dashboard in the satellite area
</p>
</li>
<li>
<p>
future versions of Megatest dashboard should support displaying areas in a
  merged way.
</p>
</li>
<li>
<p>
some new controls would be supported in the config
</p>
<div class="olist loweralpha"><ol class="loweralpha">
<li>
<p>
[setup] &#8658; allow-runs [no|yes]  &#8656;= used to disallow runs
</p>
</li>
<li>
<p>
[setup] &#8658; auto-migrate=[areaname|path]  &#8656;= used to automatically
   migrate data to a satellite area.
</p>
</li>
</ol></div>
</li>
</ol></div>
<div class="paragraph"><p><strong>Branch</strong>: This work is taking place on branch v1.65-reduce-records</p></div>
</div>
<div class="sect2">
<h3 id="_rfc_m02_move_data_into_completed_runs_db">RFC M02: Move data into completed-runs.db</h3>
<div class="paragraph"><p><strong>Purpose</strong>: shrink megatest.db data to enable lower load and higher performance.</p></div>
<div class="paragraph"><p><strong>Method</strong>: add a completed-runs.db and automatically move runs data from megatest.db to that db</p></div>
<div class="paragraph"><p><strong>Design</strong>:</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
completed-runs.db is a full megatest database with complete schema
</p>
</li>
<li>
<p>
the data move would involve these steps
</p>
<div class="olist loweralpha"><ol class="loweralpha">
<li>
<p>
copy the run data to completed-runs.db
</p>
</li>
<li>
<p>
remove the run data, first from /tmp/…/megatest.db and /tmp/…/megatest_ref.db, followed by megatest.db
</p>
</li>
</ol></div>
</li>
<li>
<p>
accessing the data would be unchanged for most operations.
</p>
</li>
<li>
<p>
a mode -full-db will be added which when specified would attach the completed-runs.db to megatest.db before doing the query
</p>
</li>
<li>
<p>
mechanisms for moving runs to/from the megatest.db would be added
</p>
<div class="olist loweralpha"><ol class="loweralpha">
<li>
<p>
-reduce-records &#8658; move runs to completed-runs.db
</p>
</li>
<li>
<p>
-restore-records &#8658; move runs from completed-runs.db to megatest.db
</p>
</li>
</ol></div>
</li>
</ol></div>
<div class="paragraph"><p><strong>Branch</strong>: This work is taking place on branch v1.65-reduce-records</p></div>
</div>
<div class="sect2">
<h3 id="_rfc_m03_automatic_homehost_migrations">RFC M03: Automatic homehost migrations</h3>
<div class="paragraph"><p><strong>Purpose</strong>: Automatically migrate homehost.</p></div>
<div class="paragraph"><p><strong>Method</strong>: Check that there are no tests running, launched or remotehoststart in past ½ hour then if not on homehost migrate the db to current host</p></div>
<div class="paragraph"><p><strong>Design</strong>:</p></div>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
Check that the system is quiescent, i.e. that there are no runs in flight or recently run
</p>
</li>
<li>
<p>
Create a lock
</p>
</li>
<li>
<p>
Migrate the /tmp cache db to the current host
</p>
</li>
<li>
<p>
Update the .homehost file
</p>
</li>
<li>
<p>
Remove the lock
</p>
</li>
</ol></div>
<div class="paragraph"><p><strong>Branch</strong>: This work not yet started</p></div>
</div>
<div class="sect2">
<h3 id="_architecture_refactor">Architecture Refactor</h3>
<div class="sect3">
<h4 id="_goals">Goals</h4>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
Reduce load on the file system. Sqlite3 files on network filesystem can be
  a burden. <span class="green">[DONE]</span>
</p>
</li>
<li>
<p>
Reduce number of servers and frequency of start/stop. This is mostly an
  issue of clutter but also a reduction in "moving parts". <span class="green">[DONE]</span>
</p>
</li>
<li>
<p>
Coalesce activities to a single home host where possible. Give the user
  feedback that they have started the dashboard on a host other than the
  home host. <span class="green">[DONE]</span>
</p>
</li>
<li>
<p>
Reduce number of processes involved in managing running tests.
</p>
</li>
</ol></div>
</div>
<div class="sect3">
<h4 id="_changes_needed">Changes Needed</h4>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
ACID compliant db will be on /tmp and synced to megatest.db with a five
  second max delay. <span class="green">[DONE]</span>
</p>
</li>
<li>
<p>
Read/writes to db for processes on homehost will go direct to /tmp
  megatest.db file. <span class="green">[DONE]</span>
</p>
</li>
<li>
<p>
Read/wites fron non-homehost processes will go through one server. Bulk
  reads (e.g. for dashboard or list-runs) will be cached on the current host
  in /tmp and synced from the home megatest.db in the testsuite area. <span class="green">[DONE]</span>
</p>
</li>
<li>
<p>
Db syncs rely on the target db file timestame minus some margin. <span class="green">[DONE]</span>
</p>
</li>
<li>
<p>
Since bulk reads do not use the server we can switch to simple RPC for the
  network transport. <span class="green">[DONE]</span>
</p>
</li>
<li>
<p>
Test running manager process extended to manage multiple running tests.
</p>
</li>
</ol></div>
</div>
</div>
<div class="sect2">
<h3 id="_current_items">Current Items</h3>
<div class="sect3">
<h4 id="_ww05_migrate_to_inmem_db">ww05 - migrate to inmem-db</h4>
<div class="olist arabic"><ol class="arabic">
<li>
<p>
Switch to inmem db with fast sync to on disk db&#8217;s <span class="green">[DONE]</span>
</p>
</li>
<li>
<p>
Server polls tasks table for next action
</p>
<div class="olist loweralpha"><ol class="loweralpha">
<li>
<p>
Task table used for tracking runner process <span class="red">[Replaced by mtutil]</span>
</p>
</li>
<li>
<p>
Task table used for jobs to run <span class="red">[Replaced by mtutil]</span>
</p>
</li>
<li>
<p>
Task table used for queueing runner actions (remove runs,
   cleanRunExecute, etc)  <span class="red">[Replaced by mtutil</span>]
</p>
</li>
</ol></div>
</li>
</ol></div>
<div class="paragraph"><p>shifting, note that the preceding blank line is needed.</p></div>
</div>
</div>
</div>
</div>
<div class="sect1">
<h2 id="_index">Index</h2>
<div class="sectionbody">
</div>
</div>
</div>
<div id="footnotes"><hr></div>
<div id="footer">
<div id="footer-text">
Version 1.0<br>
Last updated 2020-07-22 14:21:46 PDT
Version 1.5<br>
Last updated 2020-09-08 08:39:29 PDT
</div>
</div>
</body>
</html>

The Megatest Users Manual
=========================
Matt Welland <matt@kiatoa.com>
v1.0, April 2012
v1.5, June 2020
:doctype: book


[preface]
Preface
-------

This book is organised as three sub-books; getting started, writing tests and reference.

.License
----------------------------
    Copyright 2006-2017, Matthew Welland.
    Copyright 2006-2020, Matthew Welland.

    This document is part of Megatest.

    Megatest is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation, either version 3 of the License, or
    (at your option) any later version.

-------------

The Megatest project was started for two reasons, the first was an
immediate and pressing need for a generalized tool to manage a suite
of regression tests and the second was the fact that I had written or
maintained several such tools at different companies over the years. I
thought a single open source tool, flexible enough to meet the needs
of any team doing continuous integrating and or running a complex
of any team doing continuous integration and or running a complex
suite of tests for release qualification would solve some problems for
me and for others.

 -- Matt Welland, original author of the Megatest tool suite.

Megatest Design Philosophy
--------------------------

which can launch jobs on local and remote Linux hosts. Currently
megatest uses the network filesystem to call home to your master
sqlite3 database. Megatest has been used with the Intel Netbatch and
lsf (also known as openlava) batch systems and it should be
straightforward to use it with other similar systems.

include::overview.txt[]

include::plan.txt[]

include::installation.txt[]

include::getting_started.txt[]

include::study_plan.txt[]

// :leveloffset: 0

include::writing_tests.txt[]

include::howto.txt[]

include::reference.txt[]

include::testplan.txt[]

Megatest Internals
------------------

["graphviz", "server.png"]
----------------------------------------------------------------------
include::server.dot[]
----------------------------------------------------------------------

include::plan.txt[]
// include::plan.txt[]
// to allow the getting_started.txt to be a stand-alone document use level
shifting, note that the preceding blank line is needed.
// shifting, note that the preceding blank line is needed.
// :leveloffset: 2


// [appendix]
// Example Appendix
// ================
// One or more optional appendixes go here at section level zero.

//     but WITHOUT ANY WARRANTY; without even the implied warranty of
//     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
//     GNU General Public License for more details.
// 
//     You should have received a copy of the GNU General Public License
//     along with Megatest.  If not, see <http://www.gnu.org/licenses/>.

// Copyright 2006-2012, Matthew Welland.
// Copyright 2006-2020, Matthew Welland.

TODO / Road Map
---------------
Note 1: This road-map continues to evolve and subject to change without notice.

Note: This road-map is a wish list and not a formal plan. Items are in
rough priority but are subject to change. Development is driven by
user requests, developer "itch" and bug reports. Please contact
matt@kiatoa.com with requests or bug reports. Requests from inside
Intel generally take priority.

Dashboard and runs
Here is a smattering of ideas for Megatest 2.0
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

. Multi-area dashboard view

Tests Support

. Add variable $MT_RUNPATH = $MT_LINKTREE/$MT_TARGET/$MT_RUNNAME
. Improve [script], especially indent handling
----------------
[db]
api legacy|new
----------------
. One big lesson from the 1.63-1.65 generation was that the main.db, 1.db … model was really good at scaling. I’d like to combine that model with the current also-very-good model. Obviously this is a disruptive change. I think making the old model the default and the new model an option for at least one generation would be fair.
.   Rigorous megatest.config and runconfig.config caching.

Scalability

. Overflow database methodology - combine the best of the v1.63
  multi-db approach and the current db-in-tmp approach (currently
  slowness can be seen when number of tests in a db goes over 50-100k,
  with the overflow db it will be able to handle 1000's of runs with
  50-100k tests per run). High priority - goal is to complete this by
  20Q3.

Mtutils/CI

. Enable mtutil calls from dashboard (for remote control)
. Logs browser (esp. for surfacing mtutil related activities)
. Embed ftfplan for distributed automation, completed activities trigger QA runs which trigger deployment etc.
. Jenkins junit XML support [DONE]
. Add output flushing in teamcity support

Build system

. ./configure => ubuntu, sles11, sles12, rh7 [WIP]
. Switch to using simple runs query everywhere
. Add end_time to runs and add a rollup call that sets state, status and end_time

Code refactoring/quality/performance

. Switch to scsh-process pipeline management for job execution/control
. Use call-with-environment-variables where possible.

Migration to inmem db and or overflow db

. Re-work the dbstruct data structure?
.. [ run-id.db inmemdb last-mod last-read last-sync inuse ]

Some ideas for Megatest 2.0

.   Aggressive megatest.config and runconfig.config caching.
..  Cache the configs in $MT_RUNPATH
..  Following invocations of –run, -rerun* will calculate the new config but only overwrite the cached file IF changed
. If the cached file changes ALL existing tests go from COMPLETED -> STALE, I’m not sure what to do about RUNNING tests
. !VARS in runconfigs are not exported to the environment. They are accessed via rget as if the ! was not there.
. Per test copy commands (crude example below is not correct).
. Per test copy commands (example is incomplete).
----------------
[testcopy]
%/iind% unison SRC DEST
% cp –r SRC DEST
----------------
.	Test management via pkts (optional?)
..	Control pkt types: run, kill, rerunclean, clean, archive, status?
..	Status pkt types: ack, step, status_change
.	Add nanomsg as a transport option
.	


RFC M01: Add ability to move runs to other Areas
Add ability to move runs to other Areas (overlaps with overflow db system)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

*Purpose*: allow shrinking megatest.db data by moving runs to an alternate
 Megatest area with same keys.
. allow shrinking megatest.db data by moving runs to an alternate
  Megatest area with same keys.

*Method*: extend db sync to take a different megatest area as a destination.

*Design*:

. add param -destination [area|path]. when specified runs are copied to new
  area and removed from local db.
. the data move would involve these steps
.. copy the run data to destination area megatest.db
.. mark the run records as deleted, do not remove the run data on disk
. accessing the data would be by running dashboard in the satellite area
. future versions of Megatest dashboard should support displaying areas in a
  merged way.
. some new controls would be supported in the config
.. [setup] => allow-runs [no|yes]  <== used to disallow runs
.. [setup] => auto-migrate=[areaname|path]  <== used to automatically
   migrate data to a satellite area.

*Branch*: This work is taking place on branch v1.65-reduce-records

Eliminate ties to homehost (part of overflow db system)
RFC M02: Move data into completed-runs.db
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

*Purpose*: shrink megatest.db data to enable lower load and higher performance.

. Server creates captain pkt
*Method*: add a completed-runs.db and automatically move runs data from megatest.db to that db

*Design*:

. completed-runs.db is a full megatest database with complete schema
. the data move would involve these steps
.. copy the run data to completed-runs.db
.. remove the run data, first from /tmp/…/megatest.db and /tmp/…/megatest_ref.db, followed by megatest.db
. accessing the data would be unchanged for most operations.
. a mode -full-db will be added which when specified would attach the completed-runs.db to megatest.db before doing the query
. mechanisms for moving runs to/from the megatest.db would be added
.. -reduce-records => move runs to completed-runs.db
.. -restore-records => move runs from completed-runs.db to megatest.db

*Branch*: This work is taking place on branch v1.65-reduce-records

RFC M03: Automatic homehost migrations
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

*Purpose*: Automatically migrate homehost.

*Method*: Check that there are no tests running, launched or remotehoststart in past ½ hour then if not on homehost migrate the db to current host

*Design*:

. Check that the system is quiescent, i.e. that there are no runs in flight or recently run
. Create a lock
. Create a lock in the db
. Migrate the /tmp cache db to the current host
. Update the .homehost file
. Remove the lock

. Relinquish db when done
*Branch*: This work not yet started

Architecture Refactor
~~~~~~~~~~~~~~~~~~~~~

Tasks - better management of run manager processes etc.
Goals
^^^^^

. Reduce load on the file system. Sqlite3 files on network filesystem can be
  a burden. [green]#[DONE]#
. Reduce number of servers and frequency of start/stop. This is mostly an
  issue of clutter but also a reduction in "moving parts". [green]#[DONE]#
. Coalesce activities to a single home host where possible. Give the user
  feedback that they have started the dashboard on a host other than the
  home host. [green]#[DONE]#
. Reduce number of processes involved in managing running tests.

Changes Needed
^^^^^^^^^^^^^^

. ACID compliant db will be on /tmp and synced to megatest.db with a five
  second max delay. [green]#[DONE]#
. Read/writes to db for processes on homehost will go direct to /tmp
  megatest.db file. [green]#[DONE]#
. Read/wites fron non-homehost processes will go through one server. Bulk
  reads (e.g. for dashboard or list-runs) will be cached on the current host
  in /tmp and synced from the home megatest.db in the testsuite area. [green]#[DONE]#
. Db syncs rely on the target db file timestame minus some margin. [green]#[DONE]#
. Since bulk reads do not use the server we can switch to simple RPC for the
  network transport. [green]#[DONE]#
. Test running manager process extended to manage multiple running tests.

Current Items
~~~~~~~~~~~~~

ww05 - migrate to inmem-db
^^^^^^^^^^^^^^^^^^^^^^^^^^

. Switch to inmem db with fast sync to on disk db's [green]#[DONE]#
. Server polls tasks table for next action
. adjutant queries tasks table for next action [red]#[Migrate into mtutil]#
.. Task table used for tracking runner process [red]#[Replaced by mtutil]#
.. Task table used for jobs to run [red]#[Replaced by mtutil]#
.. Task table used for queueing runner actions (remove runs,
   cleanRunExecute, etc)  [red]#[Replaced by mtutil#]
. adjutant (server/task dispatch/execution manager)  

Stale propagation

 . Mark dependent tests for clean/rerun -rerun-downstream
 . On run start check for defunct tests in RUNNING, LAUNCHED or REMOTEHOSTSTART and correct or notify
 . Fix: refresh of gui sometimes fails on last item (race condition?)
// ww32
// ~~~~
// 
// . Rerun step and or subsequent steps from gui
// . Refresh test area files from gui
// . Clean and re-run button
// . Clean up STATE and STATUS handling.
// .. Dashboard and Test control panel are reverse order - choose and fix
// .. Move seldom used states and status to drop down selector
// . Access test control panel when clicking on Run Summary tests
// . Feature: -generate-index-tree
// . Change specifing of state and status to use STATE1/STATUS1,STATE2/STATUS2
// 
 
Bin list

 . Rerun step and or subsequent steps from gui [DONE?]
 . Refresh test area files from gui
 . Clean and re-run button
 . Clean up STATE and STATUS handling.
 .. Dashboard and Test control panel are reverse order - choose and fix
 .. Move seldom used states and status to drop down selector
 . Access test control panel when clicking on Run Summary tests
 . Feature: -generate-index-tree
 . Change specifing of state and status to use STATE1/STATUS1,STATE2/STATUS2
 
// ww33
// ~~~~
// 
// . http api available for use with Perl, Ruby etc. scripts
// . megatest.config setup entries for:
// .. run launching (e.g. /bin/sh %CMD% > /dev/null)
// .. browser "konqueror %FNAME%
 . rest api available for use with Perl, Ruby etc. scripts
 . megatest.config setup entries for:
 .. run launching (e.g. /bin/sh %CMD% > /dev/null)
 .. browser "konqueror %FNAME%
// 
// ww34
// ~~~~
// 
// . Mark dependent tests for clean/rerun -rerun-downstream
// . On run start check for defunct tests in RUNNING, LAUNCHED or REMOTEHOSTSTART and correct or notify
// . Fix: refresh of gui sometimes fails on last item (race condition?)
// 
// ww35
// ~~~~
// 
// . refdb: Add export of csv, json and sexp
// . Convert to using call-with-environment-variables where possible. Should allow handling of parallel runs in same process.
// . Re-work text interface wizards. Several bugs on record. Possibly convert to gui based.
// . Add to testconfig requirements section; launchlimiter scriptname, calls scriptname to check if ok to launch test
// . Refactor Run Summary view, currently very clumsy
// . Add option to show steps in Run Summary view
// 
// ww36
// ~~~~

 . refdb: Add export of csv, json and sexp
 . Convert to using call-with-environment-variables where possible. Should allow handling of parallel runs in same process.
// 
// . Refactor guis for resizeablity
// . Add filters to Run Summary view and Run Control view
// . Add to megatest.config or testconfig; rerunok STATE/STATUS,STATE/STATUS...
// . Launch gates for diskspace; /path/one>1G,/path/two>200M,/tmp>5G,#{scheme *toppath*}>1G
 . Re-work text interface wizards. Several bugs on record. Possibly convert to gui based.
 . Add to testconfig requirements section; launchlimiter scriptname, calls scriptname to check if ok to launch test
 . Refactor Run Summary view, currently very clumsy
 . Add option to show steps in Run Summary view
  . Refactor guis for resizeablity
 . Add filters to Run Summary view and Run Control view
 . Add to megatest.config or testconfig; rerunok STATE/STATUS,STATE/STATUS...
 . Launch gates for diskspace; /path/one>1G,/path/two>200M,/tmp>5G,#{scheme *toppath*}>1G
// 
// Bin List
// ~~~~~~~~
// 
// .	Quality improvements
// ..	Server stutters occasionally
// ..	Large number of items or tests still has some issues.
// ..	Code refactoring
// ..	Replace remote process with true API using json (supports Web app also)
// .	Streamline the gui
// ..	Everything resizable
// ..	Less clutter
// ..	Tool tips
// ..	Filters on Run Summary, Summary and Run Control panel
// ..	Built in log viewer (partially implemented)
// ..	Refactor the test control panel
// .	Help and documentation
// ..	Complete the user manual (I’ve been working on this lately).
// ..	Online help in the gui
// .	Streamlined install
// ..	Deployed version (download a location independent ready to run binary bundle)
 . Tool tips
 . Filters on Run Summary, Summary and Run Control panel
 . Built in log viewer (partially implemented)
 . Refactor the test control panel
   Help and documentation
 . Complete the user manual (I’ve been working on this lately).
 . Online help in the gui
   Streamlined install
 . Deployed or static build
// ..	Install Makefile (in progress, needed for Mike to install on VMs)
// ..	Added option to compile IUP (needed for VMs)
// .	Server side run launching
 . Added option to compile IUP (needed for VMs)
 . Server side run launching
// .	Support for re-running, cleaning etc. of individual steps (ezsteps makes this very easy to implement).
// .	Launch process needs built in daemonizing (easy to do, just need to test it thoroughly).
// .	Wizards for creating tests, regression areas (current ones are text only and limited).
// .	Fully functional built in web service (currently you can browse runs but it is very simplistic).
 . Wizards for creating tests, regression areas (current ones are text only and limited).
 . Fully functional built in web service (currently you can browse runs but it is very simplistic).
// .	Wildcards in runconfigs: e.g. [p1271/9/%/%]
// .	Gui panels for editing megatest.config and runconfigs.config
// .	Fully isolated tests (no use of NFS to see regression area files)
// .	Windows version
 . Gui panels for editing megatest.config and runconfigs.config
 . Fully isolated tests (no use of NFS to see regression area files)
 . Windows version

# script that takes path as parameter and returns number of bytes available:
free-space-script check-space.sh
-------------------

Trim trailing spaces
~~~~~~~~~~~~~~~~~~~~

NOTE: As of Megatest version v1.6548 trim-trailing-spaces defaults to yes.

------------------
[configf:settings trim-trailing-spaces yes]
[configf:settings trim-trailing-spaces no]
#                 |<== next line padded with spaces to here
DEFAULT_INDENT
[configf:settings trim-trailing-spaces no]
------------------

The variable DEFAULT_INDENT would be a string of 3 spaces

Job Submission Control
~~~~~~~~~~~~~~~~~~~~~~

Submit jobs to Host Types based on Test Name
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

~~~~~~~~~~~~~~~~~~~

-------------------
# A normal waiton waits for the prior tests to be COMPLETED 
# and PASS, CHECK or WAIVED
waiton test1 test2
-------------------

NOTE: Dynamic waiton lists must be capable of being calculated at the
beginning of a run. This is because Megatest walks the tree of waitons
to create the list of tests to execute.

.This works
-------------------
waiton [system somescript.sh]
-------------------

.This does NOT work (the full context for the test is not available so #{shell ...} is NOT enabled to evaluate.
-------------------
waiton #{shell somescript.sh}
-------------------

.This does NOT work
-------------------
waiton [system somescript_that_depends_on_a_prior_test.sh]
-------------------

Mode
~~~~

The default (i.e. if mode is not specified) is normal. All pre-dependent tests
must be COMPLETED and PASS, CHECK or WAIVED before the test will start