darcs

Patch 1797 add --replay=SEED option to darcs-test (and 20 more)

Title add --replay=SEED option to darcs-test (and 20 more)
Superseder Nosy List bfrk
Related Issues
Status accepted Assigned To
Milestone

Created on 2019-02-14.07:53:25 by bfrk, last changed 2019-06-14.12:09:42 by ganesh.

Files
File name Status Uploaded Type Edit Remove
add-__replay_seed-option-to-darcs_test.dpatch bfrk, 2019-02-14.07:53:23 application/x-darcs-patch
patch-preview.txt bfrk, 2019-02-14.07:53:23 text/x-darcs-patch
unnamed bfrk, 2019-02-14.07:53:23 text/plain
See mailing list archives for discussion on individual patches.
Messages
msg20652 (view) Author: bfrk Date: 2019-02-14.07:53:23
These are all minor cleanups that fell out of my recent work on integrating
RepoPatchV3 and that make sense independently.

21 patches for repository http://darcs.net/screened:

patch d0ff6c3564223a94dddfd8219b8848b29b52c501
Author: Ben Franksen <ben.franksen@online.de>
Date:   Wed Jan 30 10:52:32 CET 2019
  * add --replay=SEED option to darcs-test

patch 4099459a4dc96cdc995e6265d3d63fd568d5b333
Author: Ben Franksen <ben.franksen@online.de>
Date:   Sun Jan 27 11:45:45 CET 2019
  * improve test for conflict chain resolution
  
  Apart from fixing a bug (the old version tested the wrong file), the test is
  now tolerant with respect to the order in which the alternatives to the
  baseline are offered. Interestingly, it turned out that this test actually
  succeeds for the darcs-1 patch format.

patch 8d36a9db27dcb1954a31a13f4a80e3f72b14a906
Author: Ben Franksen <ben.franksen@online.de>
Date:   Mon Jan 28 21:27:16 CET 2019
  * mark test for resolution of conflict chains as not failing
  
  We don't have many tests for conflict resolution and this is a pretty good
  one, so we just skip it when the format is darcs-2.

patch 39a97dee15c916a5eee4201ed3a7fa22acf6d360
Author: Ben Franksen <ben.franksen@online.de>
Date:   Tue Jan 29 22:54:25 CET 2019
  * harness: better diagnosis when permutivity fails

patch dbc2db4b5ea2333f7ab8bf6adf62cc93f44e3042
Author: Ben Franksen <ben.franksen@online.de>
Date:   Sat Feb  9 12:53:02 CET 2019
  * rollback of rename conflictedEffect to isConflicted
  
  Contrary to what I stated there, conflictedEffect does have a lot to do with
  the effect of a patch. It is just in the wrong class which I am going to fix
  in another patch.

patch 83caff0b028b44a64f34129108a01d4425593ee9
Author: Ben Franksen <ben.franksen@online.de>
Date:   Tue Feb  5 11:52:45 CET 2019
  * better diagnostics when property mergeEitherWay fails

patch 6b71c18a134c23f7ea1b201a4a8aebf31d1cce0f
Author: Ben Franksen <ben.franksen@online.de>
Date:   Sun Feb 10 21:12:44 CET 2019
  * remove obsolete directory src/Darcs/Patch/Bracketed

patch a21884f915519d0ec83fa6bc9154ab6a5875645a
Author: Ben Franksen <ben.franksen@online.de>
Date:   Sun Feb 10 23:25:29 CET 2019
  * small layout fix in Darcs.Util.Hash

patch 1554fee44f2dcd273f4375eeea8883ecbcd2480d
Author: Ben Franksen <ben.franksen@online.de>
Date:   Tue Feb 12 20:10:02 CET 2019
  * fix crash when a hash (given with -h) is not found
  
  It now throws a new type of exception (MatchFailure) instead of calling error.

patch da6404d35d6a607c5f2dd95f70f279ba57caa313
Author: Ben Franksen <ben.franksen@online.de>
Date:   Tue Feb 12 22:27:10 CET 2019
  * break overlong lines in D.R.Repair

patch f322850570ce4316ce7b96f17bac9528901d8399
Author: Ben Franksen <ben.franksen@online.de>
Date:   Wed Feb 13 08:50:30 CET 2019
  * tests/sshlib: skip test if remote darcs doesn't understand our patch format

patch 36558c67de91477e60d913c040bbd1d409575731
Author: Ben Franksen <ben.franksen@online.de>
Date:   Sun Feb 10 23:26:16 CET 2019
  * remove test repos before creating them (apply, apply-hunks)

patch 8b474154e4d52fa3f3d327aafb94e1efd65a295f
Author: Ben Franksen <ben.franksen@online.de>
Date:   Wed Feb 13 08:55:29 CET 2019
  * tests/apply-hunks.sh: remove step6
  
  The repo temp2 is not used and obviously irrelevant for the test.

patch 36553aaae8718a2115cc1a4709db049dd29160b1
Author: Ben Franksen <ben.franksen@online.de>
Date:   Wed Feb 13 09:01:46 CET 2019
  * renamed apply-hunks.sh to issue701.sh

patch 210a3d0043efa1d7a490759b0fde0044739fe1f6
Author: Ben Franksen <ben.franksen@online.de>
Date:   Wed Feb 13 09:12:34 CET 2019
  * fix shebang line in tests/conflict-fight-failure.sh

patch 356b7343226d2abe711d8cdc650085252d602b7b
Author: Ben Franksen <ben.franksen@online.de>
Date:   Wed Oct 24 19:58:01 CEST 2018
  * tests/conflict-fight-failure: adapt number of iterations to patch format

patch 7052c3cfc21cdc93b820273960f6ed4eed779f9f
Author: Ben Franksen <ben.franksen@online.de>
Date:   Wed Feb 13 09:03:40 CET 2019
  * simplify and clean up conflict-fight test scripts
  
  The option --ignore-times is the default when running test scripts and we
  also have a default author configured. Also don't remove test repos when
  done (this is bad for manual inspection of the test repo).

patch 895762fa0c810724218f1767e106e3496b87d270
Author: Ben Franksen <ben.franksen@online.de>
Date:   Tue Feb 12 18:20:42 CET 2019
  * make test for issue1978 independent of patch format

patch 30ad28b1b6fd10fe537774934b12ac8d2cf3fc68
Author: Ben Franksen <ben.franksen@online.de>
Date:   Fri Feb  8 17:49:54 CET 2019
  * replace use of fromAnonymousPrim with call to constructor
  
  This further minimizes calls to fromAnonymousPrim which is good because with
  V3 this is now an unsafe method.

patch 23843596a4faba7cfe37649c102dbfd1a422906f
Author: Ben Franksen <ben.franksen@online.de>
Date:   Sun Sep 16 12:46:10 CEST 2018
  * use mergeNoConflicts in the definition of standardResolution

patch 0e4e365438c1bb7a668e70f4598f3b6727184cc2
Author: Ben Franksen <ben.franksen@online.de>
Date:   Mon Oct 22 00:07:21 CEST 2018
  * tests/repair.sh: group remove of test repo with creation
Attachments
msg20686 (view) Author: ganesh Date: 2019-06-03.19:26:42
>   * add --replay=SEED option to darcs-test

OK

>   * improve test for conflict chain resolution
>   * mark test for resolution of conflict chains as not failing

OK

>   * harness: better diagnosis when permutivity fails

OK

>   * rollback of rename conflictedEffect to isConflicted

OK

>   * better diagnostics when property mergeEitherWay fails

OK

>   * remove obsolete directory src/Darcs/Patch/Bracketed

OK

>   * small layout fix in Darcs.Util.Hash

> -  get = do a <- get ; b <- get ; c <- get ; d <- get ; e <- get ;
return (SHA1 a b c d e)
> +  get = do a <- get; b <- get; c <- get; d <- get; e <- get; return
(SHA1 a b c d e)

OK (though we have Applicative these days :-))

>   * fix crash when a hash (given with -h) is not found

OK

>   * break overlong lines in D.R.Repair

OK

>   * tests/sshlib: skip test if remote darcs doesn't understand our patch format

OK

>   * remove test repos before creating them (apply, apply-hunks)

OK

>   * tests/apply-hunks.sh: remove step6

OK

>   * renamed apply-hunks.sh to issue701.sh

OK

>   * fix shebang line in tests/conflict-fight-failure.sh

OK

>   * tests/conflict-fight-failure: adapt number of iterations to patch format

OK (a comment about the rationale for the counts might be nice)

>   * simplify and clean up conflict-fight test scripts

OK

>   * make test for issue1978 independent of patch format

OK

>   * replace use of fromAnonymousPrim with call to constructor
>   
>   This further minimizes calls to fromAnonymousPrim which is good because with
>   V3 this is now an unsafe method.

OK [The patch description is wrong - talks about fromAnonymousPrim
rather than fromPrim - perhaps commuted with a rename in your pending code?]

>   * use mergeNoConflicts in the definition of standardResolution

OK

>   * tests/repair.sh: group remove of test repo with creation

OK
msg20714 (view) Author: bfrk Date: 2019-06-11.15:12:46
>>   * small layout fix in Darcs.Util.Hash
> 
>> -  get = do a <- get ; b <- get ; c <- get ; d <- get ; e <- get ;
> return (SHA1 a b c d e)
>> +  get = do a <- get; b <- get; c <- get; d <- get; e <- get; return
> (SHA1 a b c d e)
> 
> OK (though we have Applicative these days :-))

Oh my goodness. I have no idea why i haven't made this change in the
first place, probably didn't want to be distracted...

>>   * tests/conflict-fight-failure: adapt number of iterations to patch format
> 
> OK (a comment about the rationale for the counts might be nice)

Right, though I guess you'll notice soon enough once you play with the
numbers...

>>   * replace use of fromAnonymousPrim with call to constructor
>>   
>>   This further minimizes calls to fromAnonymousPrim which is good because with
>>   V3 this is now an unsafe method.
> 
> OK [The patch description is wrong - talks about fromAnonymousPrim
> rather than fromPrim - perhaps commuted with a rename in your pending code?]

Yes, definitely. Sorry about that.
History
Date User Action Args
2019-02-14 07:53:25bfrkcreate
2019-02-14 07:58:53bfrksetstatus: needs-screening -> needs-review
2019-06-03 19:26:42ganeshsetmessages: + msg20686
2019-06-03 19:27:06ganeshsetstatus: needs-review -> accepted-pending-tests
2019-06-11 15:12:47bfrksetmessages: + msg20714
2019-06-14 12:09:42ganeshsetstatus: accepted-pending-tests -> accepted