git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [PATCH v3 5/5] diff: the -w option breaks --exit-code for --raw and other output modes

From
Jeff King <peff@peff.net>
Date
Aug 21, 2023, 22:23 UTC
Message-ID
<20230821222359.GA1813553@coredump.intra.peff.net>
In-Reply-To
<20230821210804.GA1809621@coredump.intra.peff.net>
On Mon, Aug 21, 2023 at 05:08:05PM -0400, Jeff King wrote:
Show 28 quoted lines
> On Mon, Aug 21, 2023 at 05:00:58PM -0400, Jeff King wrote:
> 
> > Alternatively, we could put it in the caller, like so:
> > 
> > diff --git a/diff.c b/diff.c
> > index 78f4e7518f..e7281e75eb 100644
> > --- a/diff.c
> > +++ b/diff.c
> > @@ -6528,6 +6528,7 @@ void diff_flush(struct diff_options *options)
> >  			if (check_pair_status(p))
> >  				flush_one_pair(p, options);
> >  		}
> > +		options->found_changes = !!q->nr;
> >  		separator++;
> >  	}
> >  
> > 
> > which matches the diffstat technique (I almost thought we could share
> > the code, but for the diffstat we are counting what ends up in the
> > diffstat struct; it does not clean out the original diff_queue when it
> > sees a noop pair).
> > 
> > I don't see a real reason to prefer one over the other.
> 
> Actually, on second look these are not quite the same. Yours only
> triggers if check_pair_status() is true. So something like
> --diff-filter should affect both output and exit code. Yours gets that
> right, and mine does not. Sorry for the noise. :)

Sorry, I'm dumb again. That is not where diff-filter is handled (it is handled by diffcore, duh). check_pair_status() is only checking for DIFF_STATUS_UNKNOWN. I'm not sure when that would ever be set, but it seems like we should be matching the "if it is output" logic, which is what you get by calling flush_one_pair(). So yours is definitely preferable, even if I don't understand the possible differences. ;)

-Peff
Previous: Jeff KingNext: Junio C Hamano
Message 32 of 36 in “git bug report”
  1. Paul WatsonAug 4, 2023
  2. rsbecker@nexbridge.comAug 4, 2023
  3. Paul WatsonAug 4, 2023
  4. rsbecker@nexbridge.comAug 4, 2023
  5. Paul WatsonAug 7, 2023
  6. Paul WatsonAug 8, 2023
  7. rsbecker@nexbridge.comAug 8, 2023
  8. Junio C HamanoAug 8, 2023
  9. diff: tighten interaction between -w and --exit-codeJunio C Hamano, Aug 16, 2023
  10. Jeff KingAug 17, 2023
  11. Junio C HamanoAug 17, 2023
  12. Jeff KingAug 17, 2023
  13. Junio C HamanoAug 17, 2023
  14. 0/5 fix interactions with "-w" and "--exit-code"Junio C Hamano, Aug 17, 2023
  15. 1/5 diff: --dirstat leakfixJunio C Hamano, Aug 17, 2023
  16. 2/5 diff: move the fallback "--exit-code" code downJunio C Hamano, Aug 17, 2023
  17. 3/5 diff: mode-only change should be noticed by "--patch -w --exit-code"Junio C Hamano, Aug 17, 2023
  18. Junio C HamanoAug 18, 2023
  19. 4/5 diff: teach "--stat -w --exit-code" to notice differencesJunio C Hamano, Aug 17, 2023
  20. 5/5 diff: teach "--name-status" and friends to honor "--exit-code -w"Junio C Hamano, Aug 17, 2023
  21. Junio C HamanoAug 17, 2023
  22. 0/5 fix interactions with "-w" and "--exit-code"Junio C Hamano, Aug 18, 2023
  23. 2/5 diff: mode-only change should be noticed by "--patch -w --exit-code"Junio C Hamano, Aug 18, 2023
  24. 3/5 diff: teach "--stat -w --exit-code" to notice differencesJunio C Hamano, Aug 18, 2023
  25. Jeff KingAug 21, 2023
  26. 4/5 t4040: remove test that succeeded for a wrong reasonJunio C Hamano, Aug 18, 2023
  27. 1/5 diff: move the fallback "--exit-code" code downJunio C Hamano, Aug 18, 2023
  28. Jeff KingAug 21, 2023
  29. 5/5 diff: the -w option breaks --exit-code for --raw and other output modesJunio C Hamano, Aug 18, 2023
  30. Jeff KingAug 21, 2023
  31. Jeff KingAug 21, 2023
  32. Jeff KingAug 21, 2023
  33. Junio C HamanoAug 21, 2023
  34. Junio C HamanoAug 21, 2023
  35. Jeff KingAug 22, 2023
  36. Jeff KingAug 21, 2023

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.