Megatest

;;     along with Megatest.  If not, see <http://www.gnu.org/licenses/>.

;; Always use two or four digit decimal
;; 1.01, 1.02...1.10,1.11,1.1101 ... 1.99,2.00..

(declare (unit megatest-version))

(define megatest-version 1.6522)

  -version                : print megatest version (currently " megatest-version ")

Launching and managing runs
  -run                    : run all tests or as specified by -testpatt
  -remove-runs            : remove the data for a run, requires -runname and -testpatt
                            Optionally use :state and :status, use -keep-records to remove only
                            the run data.
  -kill-runs               : kill an existing run (all incomplete tests marked abort, launched marked not-started)
  -set-state-status X,Y   : set state to X and status to Y, requires controls per -remove-runs
  -rerun FAIL,WARN...     : force re-run for tests with specificed status(s)
  -rerun-clean            : set all tests not COMPLETED+PASS,WARN,WAIVED to NOT_STARTED,n/a
                            and then run the specified testpatt with -preclean
  -rerun-all              : set all tests to NOT_STARTED,n/a and run with -preclean
  -lock                   : lock run specified by target and runname
  -unlock                 : unlock run specified by target and runname

			"-itempatt"
			"-setlog"
			"-set-toplog"
			"-runstep"
			"-logpro"
			"-m"
			"-rerun"

			"-days"
			"-rename-run"
			"-to"
			;; values and messages
			":category"
			":variable"
			":value"

			;; queries
			"-test-paths" ;; get path(s) to a test, ordered by youngest first

			"-runall"    ;; run all tests, respects -testpatt, defaults to %
			"-run"       ;; alias for -runall
			"-remove-runs"
                        "-kill-runs"
                        "-keep-records" ;; use with -remove-runs to remove only the run data
			"-rebuild-db"
			"-cleanup-db"
			"-rollup"
			"-update-meta"
			"-create-megatest-area"
			"-mark-incompletes"

					    (printf "Sending signal/term to ~A\n" pid)
					    (process-signal pid signal/term))))))
		       (process:children #f))
		      (original-exit exit-code)))))

;; for some switches always print the command to stderr
;;
(if (args:any? "-run" "-runall" "-remove-runs" "-set-state-status" "-kill-runs")
    (debug:print 0 *default-log-port* (string-intersperse (argv) " ")))

;; some switches imply homehost. Exit here if not on homehost
;;
(let ((homehost-required  (list "-cleanup-db" "-server")))
  (if (apply args:any? homehost-required)
      (if (not (common:on-homehost?))

     ((not target)
      (debug:print-error 0 *default-log-port* "Missing required parameter for " action ", you must specify -target or -reqtarg")
      (exit 1))
     ((not (or (args:get-arg ":runname")
	       (args:get-arg "-runname")))
      (debug:print-error 0 *default-log-port* "Missing required parameter for " action ", you must specify the run name pattern with -runname patt")
      (exit 2))
     ((not (or (args:get-arg "-testpatt") (eq? action 'kill-runs)))
      (debug:print-error 0 *default-log-port* "Missing required parameter for " action ", you must specify the test pattern with -testpatt")
      (exit 3))
     (else
      (if (not (car *configinfo*))
	  (begin
	    (debug:print-error 0 *default-log-port* "Attempted " action "on test(s) but run area config file not found")
	    (exit 1))
	  ;; put test parameters into convenient variables
	  (begin
	    ;; check for correct version, exit with message if not correct
	    (common:exit-on-version-changed)
	    (runs:operate-on  action
			      target
			      (common:args-get-runname)  ;; (or (args:get-arg "-runname")(args:get-arg ":runname"))
			      (common:args-get-testpatt #f) ;; (args:get-arg "-testpatt")
			      state: (common:args-get-state)
			      status: (common:args-get-status)
			      new-state-status: (args:get-arg "-set-state-status")
                              mode: mode)))
      (set! *didsomething* #t)))))

(if (args:get-arg "-kill-runs")
    (general-run-call 
     "-kill-runs"
     "kill runs"
     (lambda (target runname keys keyvals)
       (operate-on 'kill-runs mode: #f)

       (operate-on 'set-state-status
		   state: (common:args-get-state)
		   status: (common:args-get-status)
		   new-state-status: (args:get-arg "-set-state-status")


                   mode: #f) ;; kill-req on running,launched,remotehoststart things, 
       )))

(if (args:get-arg "-remove-runs")
    (general-run-call 
     "-remove-runs"
     "remove runs"
     (lambda (target runname keys keyvals)
       (operate-on 'remove-runs mode: (if (args:get-arg "-keep-records")

  -version                   : print megatest version (currently " megatest-version ")
			     
Run management:		     
   run                       : initiate or resume a run, already completed and in-progress
                               tests are not affected.
   rerun-clean               : clean and rerun all not completed pass/fail tests
   rerun-all                 : clean and rerun entire run
   kill-run                  : kill all tests in run
   remove                    : remove runs
   set-ss                    : set state/status
   archive                   : compress and move test data to archive disk
   kill                      : stop tests or entire runs
   db                        : database utilities

Queries:

    ))

;; alist to map actions to old megatest commands
(define *action-keys*
  '((run         . "-run")
    (rerun-clean . "-rerun-clean")
    (rerun-all   . "-rerun-all")
    (kill-run    . "-kill-runs")
    (sync        . "")
    (archive     . "-archive")
    (set-ss      . "-set-state-status")
    (remove      . "-remove-runs")))

;; manually keep this list updated from the keys to
;; the case *action* near the end of this file.

	;    (hash-table-set! args:arg-hash "-log" logf)) ;; fake out future queries of -log
	(print *default-log-port* "Sending log output to " logf)
	(set! *default-log-port* oup)
)))

(if *action*
    (case (string->symbol *action*)
      ((run remove rerun rerun-clean rerun-all set-ss archive kill list kill-run)
       (let* ((mtconfdat (simple-setup (args:get-arg "-start-dir")))
	      (mtconf    (car mtconfdat))
	      (area      (args:get-arg "-area")) ;; look up the area to dispatch to from [areas] section
	      (areasec   (if area (configf:lookup mtconf "areas" area) #f))
	      (areadat   (if areasec (common:val->alist areasec) #f))
	      (area-path (if areadat (alist-ref 'path areadat) #f))
	      (pktsdirs  (configf:lookup mtconf "setup" "pktsdirs"))

	 (states       (if state  (string-split state  ",") '()))
	 (statuses     (if status (string-split status ",") '()))
	 (state-status (if (string? new-state-status) (string-split new-state-status ",") '(#f #f)))
	 (rp-mutex     (make-mutex))
	 (bup-mutex    (make-mutex))
         (keep-records (args:get-arg "-keep-records"))) ;; used in conjunction with -remove-runs to keep the records, TODO: consolidate this with "mode".

    (let* ((write-access-actions '(remove-runs set-state-status archive run-wait kill-runs))
           (dbfile             (conc  *toppath* "/megatest.db"))
           (readonly-mode      (not (file-write-access? dbfile))))
      (when (and readonly-mode
                 (member action write-access-actions))
        (debug:print-error 0 *default-log-port* "megatest.db is readonly.  Cannot proceed with action ["action"] in which write-access isrequired .")
        (exit 1)))
    

			       '()))
		(lasttpath "/does/not/exist/I/hope")
		(worker-thread #f))
	   (debug:print-info 4 *default-log-port* "runs:operate-on run=" run ", header=" header)
	   (if (not (null? tests))
	       (begin
		 (case action
                   ((kill-runs)
                    (tasks:kill-runner target run-name "%")
                    (debug:print 1 *default-log-port* "Killing tests for run: " runkey " " (db:get-value-by-header run header "runname"))
                    )
		   ((remove-runs)
		    ;; (if (tasks:need-server run-id)(tasks:start-and-wait-for-server tdbdat run-id 10))
		    ;; seek and kill in flight -runtests with % as testpatt here
		    ;; (if (equal? testpatt "%")
		    (tasks:kill-runner target run-name testpatt)

		    ;; (debug:print 0 *default-log-port* "not attempting to kill any run launcher processes as testpatt is " testpatt))
		    (debug:print 1 *default-log-port* "Removing tests for run: " runkey " " (db:get-value-by-header run header "runname")))
		   ((set-state-status)
		    ;; (if (tasks:need-server run-id)(tasks:start-and-wait-for-server tdbdat run-id 10))
		    (debug:print 1 *default-log-port* "Modifying state and staus for tests for run: " runkey " " (db:get-value-by-header run header "runname")))
		   ((print-run)
		    (debug:print 1 *default-log-port* "Printing info for run " runkey ", run=" run ", tests=" tests ", header=" header)

                                      ((started)
                                       ;; if last visit was within last second, sleep 1 second
                                       (if (< (- now last-visit) 1.0)
                                           (thread-sleep! 1.0))
                                       (hash-table-set! backgrounded-remove-last-visit test-fulln (current-seconds))
                                       ;; send to back of line, loop
                                       (let ((newtal (append tal (list test))))
                                         (loop (car newtal)(cdr newtal)))
                                       )
                                      ((done)
                                       ;; drop this one; if remaining, loop, else finish
                                       (hash-table-set! backgrounded-remove-last-visit test-fulln (current-seconds))
                                       (let ((subrun-remove-succeeded (hash-table-ref/default backgrounded-remove-result test-fulln 'exception)))
                                         (cond
                                          ((eq? subrun-remove-succeeded 'exception)

                                            (loop new-test-dat tal)
                                            (loop (car tal)(append tal (list new-test-dat)))))
                                      (begin
                                        (runs:remove-test-directory new-test-dat mode) ;; 'remove-all)
                                        (if (not (null? tal))
                                            (loop (car tal)(cdr tal)))))))
				(rmt:update-run-stats run-id (rmt:get-raw-run-stats run-id)))
                               ((kill-runs)
                                ;; RUNNING -> KILLREQ
                                ;; LAUNCHED,RUNNING,REMOTEHOSTSTART -> NOT STARTED
                                (cond
                                 ((and has-subrun (member test-state (list "RUNNING" "LAUNCHED" "REMOTEHOSTSTART" "KILLREQ")))
                                  (common:send-thunk-to-background-thread
                                   (lambda ()
                                     (let* ((subrun-remove-succeeded
                                             (subrun:kill-subrun run-dir keep-records)))
                                       #t)))
                                  (if (not (null? tal))
				    (loop (car tal)(cdr tal)))
                                  )
                                 ((member test-state (list "RUNNING" "LAUNCHED" "REMOTEHOSTSTART" "KILLREQ"))
                                  (mt:test-set-state-status-by-id run-id (db:test-get-id test) "KILLREQ" "n/a" #f)
                                  (if (not (null? tal))
				      (loop (car tal)(cdr tal)))
                                  )))
			       ((set-state-status)
                                (let* ((new-state (car state-status))
                                       (new-status (cadr state-status))
                                       (test-id (db:test-get-id test))
                                       (test-run-dir (db:test-get-rundir new-test-dat))
                                       (has-subrun (and (subrun:subrun-test-initialized? test-run-dir)
                                                      (not (subrun:subrun-removed? test-run-dir)))))

              (subrun:exec-sub-megatest test-run-dir action-switches-str "remove")))
        (if remove-result
            (begin
              (subrun:set-subrun-removed test-run-dir)
              #t)
            #f))
      #t))

(define (subrun:kill-subrun test-run-dir )
  (if (and (not (subrun:subrun-removed? test-run-dir)) (subrun:subrun-test-initialized? test-run-dir))
      (let* ((action-switches-str
              (conc "-kill-runs" ))
             (kill-result
              (subrun:exec-sub-megatest test-run-dir action-switches-str "kill")))
        kill-result)
      #t))

(define (subrun:launch-cmd test-run-dir)
  (if (subrun:subrun-removed? test-run-dir)
      (subrun:unset-subrun-removed test-run-dir))      

  (let* ((log-prefix "run")
         (switches (subrun:selector+log-switches test-run-dir log-prefix))