Megatest

;; THESE ARE ALL CALLED ON THE CLIENT SIDE!!!
;;

;; generate entries for ~/.megatestrc with the following
;;
;;  grep define ../rmt.scm | grep rmt: |perl -pi -e 's/\(define\s+\((\S+)\W.*$/\1/'|sort -u









;;======================================================================
;;  S U P P O R T   F U N C T I O N S
;;======================================================================

;; if a server is either running or in the process of starting call client:setup
;; else return #f to let the calling proc know that there is no server available
;;
(define (rmt:get-connection-info areapath #!key (area-dat #f)) ;; TODO: push areapath down.
  (let* ((runremote (or area-dat *runremote*)))
    (if runremote
	(let* ((cinfo  (remote-conndat runremote))
	       (run-id 0))
	  (if cinfo
	      cinfo
	      (if (server:check-if-running areapath)
		  (client:setup areapath)
		  #f)))
	#f)))

(define *send-receive-mutex* (make-mutex)) ;; should have separate mutex per run-id

;; RA => e.g. usage (rmt:send-receive 'get-var #f (list varname))
;;
(define (rmt:send-receive cmd rid params #!key (attemptnum 1)(area-dat #f)) ;; start attemptnum at 1 so the modulo below works as expected

  ;; do all the prep locked under the rmt-mutex
  (mutex-lock! *rmt-mutex*)

  ;; 1. check if server is started IFF cmd is a write OR if we are not on the homehost, store in runremote
  ;; 2. check the age of the connections. refresh the connection if it is older than timeout-20 seconds.
  ;; 3. do the query, if on homehost use local access
  ;;
  (let* ((start-time (current-seconds)) ;; snapshot time so all use cases get same value
         (areapath *toppath*);; TODO - resolve from dbstruct to be compatible with multiple areas
         (dbfile (conc *toppath* "/megatest.db"))
         (readonly-mode (not (file-write-access? dbfile))) ;; TODO: use dbstruct or runremote to figure this out in future
	 (runremote  (or area-dat *runremote*)))
    ;;(print "BB> readonly-mode is "readonly-mode" dbfile is "dbfile)
    (cond
     ;; give up if more than 15 attempts
     ((> attemptnum 15)
      (debug:print 0 *default-log-port* "ERROR: 15 tries to start/connect to server. Giving up.")
      (exit 1))

     ;; readonly mode, read request-  handle it - case 20
     ((and readonly-mode
           (member cmd api:read-only-queries)) 
      (mutex-unlock! *rmt-mutex*)
      (debug:print-info 12 *default-log-port* "rmt:send-receive, case 20")
      (rmt:open-qry-close-locally cmd 0 params)
      )

     ;; readonly mode, write request.  Do nothing, return #f
     (readonly-mode
      (mutex-unlock! *rmt-mutex*)
      (debug:print-info 12 *default-log-port* "rmt:send-receive, case 21")
      (debug:print 0 *default-log-port* "WARNING: write transaction requested on a readonly area.  cmd="cmd" params="params)
      #f
      )

     ;; reset the connection if it has been unused too long
     ((and runremote
           (remote-conndat runremote)
	   (let ((expire-time (+ (- start-time (remote-server-timeout runremote))(random 30)))) ;; add 30 seconds of noise so that not all running tests expire at the same time causing a storm of server starts
	     (< (http-transport:server-dat-get-last-access (remote-conndat runremote)) expire-time)))
      (debug:print-info 12 *default-log-port* "rmt:send-receive, case  8")
      (remote-conndat-set! runremote #f)
      (mutex-unlock! *rmt-mutex*)
      (rmt:send-receive cmd rid params attemptnum: attemptnum))
     ;; ensure we have a record for our connection for given area
     ((not runremote)                     
      (set! *runremote* (make-remote))
      (mutex-unlock! *rmt-mutex*)
      (debug:print-info 12 *default-log-port* "rmt:send-receive, case  1")
      (rmt:send-receive cmd rid params attemptnum: attemptnum))
     ;; ensure we have a homehost record
     ((not (pair? (remote-hh-dat runremote)))  ;; not on homehost
      (thread-sleep! 0.1) ;; since we shouldn't get here, delay a little
      (remote-hh-dat-set! runremote (common:get-homehost))
      (mutex-unlock! *rmt-mutex*)
      (debug:print-info 12 *default-log-port* "rmt:send-receive, case  2")
      (rmt:send-receive cmd rid params attemptnum: attemptnum))
     ;; on homehost and this is a read
     ((and (not (remote-force-server runremote))      ;; honor forced use of server
	   (cdr (remote-hh-dat runremote))     ;; on homehost
           (member cmd api:read-only-queries)) ;; this is a read
      (mutex-unlock! *rmt-mutex*)
      (debug:print-info 12 *default-log-port* "rmt:send-receive, case  3")
      (rmt:open-qry-close-locally cmd 0 params))

     ;; on homehost and this is a write, we already have a server, but server has died
     ((and (cdr (remote-hh-dat runremote))         ;; on homehost
           (not (member cmd api:read-only-queries))  ;; this is a write
           (remote-server-url runremote)           ;; have a server
           (not (server:check-if-running *toppath*)))  ;; server has died.
      (set! *runremote* (make-remote))
      (mutex-unlock! *rmt-mutex*)
      (debug:print-info 12 *default-log-port* "rmt:send-receive, case  4.1")
      (rmt:send-receive cmd rid params attemptnum: attemptnum))

     ;; on homehost and this is a write, we already have a server
     ((and (not (remote-force-server runremote))     ;; honor forced use of server
	   (cdr (remote-hh-dat runremote))           ;; on homehost
           (not (member cmd api:read-only-queries))  ;; this is a write
           (remote-server-url runremote))            ;; have a server
      (mutex-unlock! *rmt-mutex*)
      (debug:print-info 12 *default-log-port* "rmt:send-receive, case  4")
      (rmt:open-qry-close-locally cmd 0 params))






     ;;  on homehost, no server contact made and this is a write, passively start a server 
     ((and (not (remote-force-server runremote)) ;; honor forced use of server
	   (cdr (remote-hh-dat runremote))       ;; new
           (not (remote-server-url runremote))
	   (not (member cmd api:read-only-queries)))
      (debug:print-info 12 *default-log-port* "rmt:send-receive, case  5")
      (let ((server-url  (server:check-if-running *toppath*))) ;; (server:read-dotserver->url *toppath*))) ;; (server:check-if-running *toppath*))) ;; Do NOT want to run server:check-if-running - very expensive to do for every write call
	(if server-url
	    (remote-server-url-set! runremote server-url) ;; the string can be consumed by the client setup if needed

	    (server:kind-run *toppath*)))


      (mutex-unlock! *rmt-mutex*)
      (debug:print-info 12 *default-log-port* "rmt:send-receive, case  5.1")
      (rmt:open-qry-close-locally cmd 0 params))







     ((and (not (cdr (remote-hh-dat runremote)))        ;; not on a homehost 
           (not (remote-conndat runremote)))            ;; and no connection
      (debug:print-info 12 *default-log-port* "rmt:send-receive, case  6  hh-dat: " (remote-hh-dat runremote) " conndat: " (remote-conndat runremote))
      (mutex-unlock! *rmt-mutex*)
      (server:start-and-wait *toppath*)
      (if (common:force-server?)(remote-force-server-set! runremote #t))
      (remote-conndat-set! runremote (rmt:get-connection-info *toppath*)) ;; calls client:setup which calls client:setup-http
      (rmt:send-receive cmd rid params attemptnum: attemptnum)) ;; TODO: add back-off timeout as
     ;; all set up if get this far, dispatch the query
     ((and (not (remote-force-server runremote))
	   (cdr (remote-hh-dat runremote))) ;; we are on homehost
      (mutex-unlock! *rmt-mutex*)
      (debug:print-info 12 *default-log-port* "rmt:send-receive, case  7")
      (rmt:open-qry-close-locally cmd (if rid rid 0) params))

     ;; not on homehost, do server query
     (else
      (mutex-unlock! *rmt-mutex*)
      (debug:print-info 12 *default-log-port* "rmt:send-receive, case  9")
      (mutex-lock! *rmt-mutex*)
      (let* ((conninfo (remote-conndat runremote))
	     (dat      (case (remote-transport runremote)
			 ((http) (condition-case ;; handling here has caused a lot of problems. However it is needed to deal with attemtped communication to servers that have gone away
                                  (http-transport:client-api-send-receive 0 conninfo cmd params)
                                  ((commfail)(vector #f "communications fail"))
                                  ((exn)(vector #f "other fail" (print-call-chain)))))
			 (else
			  (debug:print 0 *default-log-port* "ERROR: transport " (remote-transport runremote) " not supported")
			  (exit))))
	     (success  (if (vector? dat) (vector-ref dat 0) #f))
	     (res      (if (vector? dat) (vector-ref dat 1) #f)))
	(if (vector? conninfo)(http-transport:server-dat-update-last-access conninfo)) ;; refresh access time
	;; (mutex-unlock! *rmt-mutex*)
        (debug:print-info 12 *default-log-port* "rmt:send-receive, case  9. conninfo=" conninfo " dat=" dat " runremote = "runremote)
	(if success
	    (case (remote-transport runremote)
	      ((http)
	       (mutex-unlock! *rmt-mutex*)
	       res)
	      (else
	       (debug:print 0 *default-log-port* "ERROR: transport " (remote-transport runremote) " is unknown")
	       (mutex-unlock! *rmt-mutex*)
	       (exit 1)))
	    (begin
	      (debug:print 0 *default-log-port* "WARNING: communication failed. Trying again, try num: " attemptnum)
	      (remote-conndat-set!    runremote #f)
	      (remote-server-url-set! runremote #f)
              (debug:print-info 12 *default-log-port* "rmt:send-receive, case  9.1")
	      (mutex-unlock! *rmt-mutex*)
	      (server:start-and-wait *toppath*)
	      (rmt:send-receive cmd rid params attemptnum: (+ attemptnum 1)))))))))

;; (define (rmt:update-db-stats run-id rawcmd params duration)
;;   (mutex-lock! *db-stats-mutex*)
;;   (handle-exceptions
;;    exn
;;    (begin
;;      (debug:print 0 *default-log-port* "WARNING: stats collection failed in update-db-stats")
;;      (debug:print 0 *default-log-port* " message: " ((condition-property-accessor 'exn 'message) exn))
;;      (print "exn=" (condition->list exn))
;;      #f) ;; if this fails we don't care, it is just stats
;;    (let* ((cmd      (conc "run-id=" run-id " " (if (eq? rawcmd 'general-call) (car params) rawcmd)))
;; 	  (stat-vec (hash-table-ref/default *db-stats* cmd #f)))
;;      (if (not (vector? stat-vec))
;; 	 (let ((newvec (vector 0 0)))
;; 	   (hash-table-set! *db-stats* cmd newvec)
;; 	   (set! stat-vec newvec)))
;;      (vector-set! stat-vec 0 (+ (vector-ref stat-vec 0) 1))
;;      (vector-set! stat-vec 1 (+ (vector-ref stat-vec 1) duration))))
;;   (mutex-unlock! *db-stats-mutex*))


(define (rmt:print-db-stats)
  (let ((fmtstr "~40a~7-d~9-d~20,2-f")) ;; "~20,2-f"
    (debug:print 18 *default-log-port* "DB Stats\n========")
    (debug:print 18 *default-log-port* (format #f "~40a~8a~10a~10a" "Cmd" "Count" "TotTime" "Avg"))
    (for-each (lambda (cmd)
		(let ((cmd-dat (hash-table-ref *db-stats* cmd)))

(define (rmt:open-qry-close-locally cmd run-id params #!key (remretries 5))
  (let* ((qry-is-write   (not (member cmd api:read-only-queries)))
	 (db-file-path   (db:dbfile-path)) ;;  0))
	 (dbstruct-local (db:setup))  ;; make-dbr:dbstruct path:  dbdir local: #t)))
	 (read-only      (not (file-write-access? db-file-path)))
	 (start          (current-milliseconds))
	 (resdat         (if (not (and read-only qry-is-write))
			     (let ((v (api:execute-requests dbstruct-local (vector (symbol->string cmd) params))))
			       (handle-exceptions ;; there has been a long history of receiving strange errors from values returned by the client when things go wrong..
				exn               ;;  This is an attempt to detect that situation and recover gracefully
				(begin
				  (debug:print0 *default-log-port* "ERROR: bad data from server " v " message: "  ((condition-property-accessor 'exn 'message) exn))
				  (vector #t '())) ;; should always get a vector but if something goes wrong return a dummy
				(if (and (vector? v)
					 (> (vector-length v) 1))
				    (let ((newvec (vector (vector-ref v 0)(vector-ref v 1))))
				      newvec)           ;; by copying the vector while inside the error handler we should force the detection of a corrupted record
				    (vector #t '()))))  ;; we could also check that the returned types are valid
			     (vector #t '())))
	 (success        (vector-ref resdat 0))
	 (res            (vector-ref resdat 1))
	 (duration       (- (current-milliseconds) start)))
    (if (and read-only qry-is-write)
        (debug:print 0 *default-log-port* "ERROR: attempt to write to read-only database ignored. cmd=" cmd))
    (if (not success)

	      #f))
	(begin
	  ;; (rmt:update-db-stats run-id cmd params duration)
	  ;; mark this run as dirty if this was a write, the watchdog is responsible for syncing it
	  (if qry-is-write
	      (let ((start-time (current-seconds)))
		(mutex-lock! *db-multi-sync-mutex*)
/		(set! *db-last-access* start-time)  ;; THIS IS PROBABLY USELESS? (we are on a client)
                (mutex-unlock! *db-multi-sync-mutex*)))))
    res))

(define (rmt:send-receive-no-auto-client-setup connection-info cmd run-id params)
  (let* ((run-id   (if run-id run-id 0))
	 (res  	   (handle-exceptions
		    exn

;; hand off a call to one of the db:queries statements
;; added run-id to make looking up the correct db possible 
;;
(define (rmt:general-call stmtname run-id . params)
  (rmt:send-receive 'general-call run-id (append (list stmtname run-id) params)))


;; given a hostname, return a pair of cpu load and update time representing latest intelligence from tests running on that host
(define (rmt:get-latest-host-load hostname)
  (rmt:send-receive 'get-latest-host-load 0 (list hostname)))

;; (define (rmt:sync-inmem->db run-id)
;;   (rmt:send-receive 'sync-inmem->db run-id '()))

(define (rmt:sdb-qry qry val run-id)
  ;; add caching if qry is 'getid or 'getstr
  (rmt:send-receive 'sdb-qry run-id (list qry val)))

;; NOT COMPLETED
(define (rmt:runtests user run-id testpatt params)
  (rmt:send-receive 'runtests run-id testpatt))

;;======================================================================
;;  T E S T   M E T A 
;;======================================================================

(define (rmt:get-tests-tags)
  (rmt:send-receive 'get-tests-tags #f '()))

;;======================================================================
;;  K E Y S 
;;======================================================================

;; These require run-id because the values come from the run!
;;
(define (rmt:get-key-val-pairs run-id)
  (rmt:send-receive 'get-key-val-pairs run-id (list run-id)))

(define (rmt:get-keys)
  (if *db-keys* *db-keys* 
     (let ((res (rmt:send-receive 'get-keys #f '())))
       (set! *db-keys* res)
       res)))

(define (rmt:get-keys-write) ;; dummy query to force server start
  (let ((res (rmt:send-receive 'get-keys-write #f '())))
    (set! *db-keys* res)
    res))

;; we don't reuse run-id's (except possibly *after* a db cleanup) so it is safe
;; to cache the resuls in a hash
;;
(define (rmt:get-key-vals run-id)
  (or (hash-table-ref/default *keyvals* run-id #f)
      (let ((res (rmt:send-receive 'get-key-vals #f (list run-id))))

  (rmt:send-receive 'delete-test-records run-id (list run-id test-id)))

;; This is not needed as test steps are deleted on test delete call
;;
;; (define (rmt:delete-test-step-records run-id test-id)
;;   (rmt:send-receive 'delete-test-step-records run-id (list run-id test-id)))

(define (rmt:test-set-state-status run-id test-id state status msg)
  (rmt:send-receive 'test-set-state-status run-id (list run-id test-id state status msg)))

(define (rmt:test-toplevel-num-items run-id test-name)
  (rmt:send-receive 'test-toplevel-num-items run-id (list run-id test-name)))

;; (define (rmt:get-previous-test-run-record run-id test-name item-path)
;;   (rmt:send-receive 'get-previous-test-run-record run-id (list run-id test-name item-path)))

  (rmt:send-receive 'get-count-tests-running-for-testname run-id (list run-id testname)))

(define (rmt:get-count-tests-running-in-jobgroup run-id jobgroup)
  (rmt:send-receive 'get-count-tests-running-in-jobgroup run-id (list run-id jobgroup)))

;; state and status are extra hints not usually used in the calculation
;;
(define (rmt:set-state-status-and-roll-up-items run-id test-name item-path state status comment)
  (rmt:send-receive 'set-state-status-and-roll-up-items run-id (list run-id test-name item-path state status comment)))

(define (rmt:update-pass-fail-counts run-id test-name)
  (rmt:general-call 'update-pass-fail-counts run-id test-name test-name test-name))

(define (rmt:top-test-set-per-pf-counts run-id test-name)
  (rmt:send-receive 'top-test-set-per-pf-counts run-id (list run-id test-name)))

(define (rmt:update-run-event_time run-id)
  (rmt:send-receive 'update-run-event_time #f (list run-id)))

(define (rmt:get-runs-by-patt  keys runnamepatt targpatt offset limit fields last-runs-update) ;; fields of #f uses default
  (rmt:send-receive 'get-runs-by-patt #f (list keys runnamepatt targpatt offset limit fields last-runs-update)))

(define (rmt:find-and-mark-incomplete run-id ovr-deadtime)
  ;; (if (rmt:send-receive 'have-incompletes? run-id (list run-id ovr-deadtime))
  (rmt:send-receive 'mark-incomplete run-id (list run-id ovr-deadtime))) ;; )

(define (rmt:get-main-run-stats run-id)
  (rmt:send-receive 'get-main-run-stats #f (list run-id)))

(define (rmt:get-var varname)
  (rmt:send-receive 'get-var #f (list varname)))