Am I really the only one who was annoyed by this behaviour?  Reviewer: you may
want to note that nobody else has commented on the ticket I filed for
issue1551, so it's not like there is some sort of consensus on this. I think it's
quite clear, but maybe somebody would disagree?  Anyway, if you think it's also
obviously the right way to behave, then I bet you can just push it in.

This behaviour has annoyed me for a while because I sometimes to use the long
comment editor to look over my patch names and (for example) make sure they fit
under 72 characters.  What tends to happen is that I accidentally hit 'n' and
then have to go record my patch again, redoing my cherry-picking (hmm, there's
another point to be made there)

The way I see it, for darcs send this confirmation makes sense because there is
no way back.  For darcs record and amend-record, however, it does not seem very
useful.

2 patches for repository http://darcs.net/releases/branch-2.5:

Sun Jul 18 10:23:27 BST 2010  Eric Kow <kowey@darcs.net>
  * Resolve issue1551: only use unchanged file content prompt in darcs send.

Sun Jul 18 10:23:34 BST 2010  Eric Kow <kowey@darcs.net>
  * Haddock promptYorn.

Attachments

• resolve-issue1551_-only-use-unchanged-file-content-prompt-in-darcs-send_.dpatch
• unnamed

2 patches for repository http://darcs.net:

Minor hlint-esque tweak of the issue1551 patch

Sun Jul 18 10:34:48 BST 2010  Eric Kow <kowey@darcs.net>
  * Resolve issue1551: only use unchanged file content prompt in darcs send.

Sun Jul 18 10:23:34 BST 2010  Eric Kow <kowey@darcs.net>
  * Haddock promptYorn.

Attachments

• resolve-issue1551_-only-use-unchanged-file-content-prompt-in-darcs-send_.dpatch
• unnamed

Hi Eric,

Eric Kow <bugs@darcs.net> writes:
> Sun Jul 18 10:23:27 BST 2010  Eric Kow <kowey@darcs.net>
>   * Resolve issue1551: only use unchanged file content prompt in darcs send.
this patch is failing the issue1465_ortryrunning.sh test. I don't have
time to investigate why, so please look into that when you have a bit of
time.

> Sun Jul 18 10:23:34 BST 2010  Eric Kow <kowey@darcs.net>
>   * Haddock promptYorn.
I am pushing this one.

Yours,
   Petr.

I can confirm this really makes 1465 fail on Linux (but not on MacOS X)
Problem seems to be darcs record succeeding where the test wants it to
fail, so I wonder if on Linux it's been "failing" for the wrong reason
(the confirmation prompt) the whole time.

3 patches for repository http://darcs.net:

Phew, this test failure confused me!  My patch removes the confirmation prompt
from darcs record if you don't make any modifications (*).

The result was that the issue1465 test would fail. Basically issue1465 seems
to be about checking Darcs' fall-through behaviour on editors.

Darcs tries exactly 5 editors in order
1. first of DARCS_EDITOR, DARCSEDITOR, VISUAL, EDITOR that is defined; or vi
2. emacs
3. emacs -nw
4. nano
5. edit

I think what Trent wants in that test is basically to make sure that we only
fall through from one editor (eg. from 1. to 2.) for legimitate reasons. A
legitimate reason would be the editor not existing. An illegitimate reason
would be the user hitting control-C or something (false in his test).

Two things about the test.  First, it tests for darcs record failing, but the
failure seems to come from Darcs's issue1551 follow-up confirmation prompt.
Take the prompt away and the expected failure goes away (hence the test
complaining).  Second, on MacOS X it was passing for the wrong reason (the
env -u error was confused for darcs legimitately failing).

My modifications remove the test for darcs failure while trying to preserve the
core idea of testing the fall-through behaviour.  One glaring omission in this
modification: what's Darcs supposed to do if the user hits ^-C or something?
Right now, it just goes ahead and records the patch. Shouldn't it fail?

(*) Because
    a. confirmation prompts are annoying and
    b. I claim that editing long-comment just to see it in an
        editor is a legitimate workflow
    
    However
    c. we still have to keep it in darcs send because there
    is no last resort

Sun Jul 18 10:34:48 BST 2010  Eric Kow <kowey@darcs.net>
  * Resolve issue1551: only use unchanged file content prompt in darcs send.

Wed Oct 20 15:16:27 BST 2010  Eric Kow <kowey@darcs.net>
  * Improve issue1465 test (portability, tidiness).
  
  - No env -u on MacOS X.
  - No need to clean up directories (new harness)
    Keeping them around makes debugging easier.

Wed Oct 20 18:24:22 BST 2010  Eric Kow <kowey@darcs.net>
  * Make issue1465 test a bit more straightforward.

Attachments

• resolve-issue1551_-only-use-unchanged-file-content-prompt-in-darcs-send_.dpatch
• unnamed

This patch bundle (with 3 patches) was just applied to the repository http://darcs.net/screened.
This message was brought to you by DarcsWatch
http://darcswatch.nomeata.de/repo_http:__darcs.net_screened.html#bundle-fa1b4fc56a81b19c05f1cbb3d60431475954c045

As far as the issue1551 resolution goes, I'm happy with the code change 
and I agree with the principle. However I think it's worth giving people 
a bit of time to object as it's a behaviour change.

Also, after this patch:

> Wed Oct 20 18:24:22 BST 2010  Eric Kow <kowey@darcs.net>
>   * Make issue1465 test a bit more straightforward.

the issue1465 test stops working on my machine. Let me know if you can't 
reproduce and I'll try to investigate further.

for the problem with the issue1465 test, look at patch433. I this patch 
and patch433 can/should go in at the same time.

I'll accept this assuming I can verify that patch433 does indeed fix the 
problem with the issue1465 test.

This patch bundle (with 3 patches) was just applied to the repository http://darcs.net/.
This message was brought to you by DarcsWatch
http://darcswatch.nomeata.de/repo_http:__darcs.net_.html#bundle-fa1b4fc56a81b19c05f1cbb3d60431475954c045

This patch bundle (with 2 patches) was just applied to the repository http://darcs.net/.
This message was brought to you by DarcsWatch
http://darcswatch.nomeata.de/repo_http:__darcs.net_.html#bundle-5e0f542cbb84292a79db22ebb854c256fed443b4

This patch bundle (with 3 patches) was just applied to the repository http://darcs.net/reviewed.
This message was brought to you by DarcsWatch
http://darcswatch.nomeata.de/repo_http:__darcs.net_reviewed.html#bundle-fa1b4fc56a81b19c05f1cbb3d60431475954c045

This patch bundle (with 2 patches) was just applied to the repository http://darcs.net/reviewed.
This message was brought to you by DarcsWatch
http://darcswatch.nomeata.de/repo_http:__darcs.net_reviewed.html#bundle-5e0f542cbb84292a79db22ebb854c256fed443b4