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

Re: [PATCH 1/3] describe: debug is incompatible with contains

From
Michael J Gruber <git@drmicha.warpmail.net>
Date
Mar 17, 2017, 10:54 UTC
Message-ID
<a76bd1ae-74a8-a85c-852a-e36514cd05d5@drmicha.warpmail.net>
In-Reply-To
<xmqqzigme474.fsf@gitster.mtv.corp.google.com>
Junio C Hamano venit, vidit, dixit 15.03.2017 20:21:
Show 12 quoted lines
> Michael J Gruber <git@drmicha.warpmail.net> writes:
> 
>> `git describe --contains` calls into `git name-rev` which does not have
>> any searching to do and thus does not display any debug information.
>>
>> Say so in the documentation and catch the incompatible arguments.
> 
> I am not sure if this is worth it.  Those who are really doing the
> debugging would be staring at the code while running it anyway, so
> it is not like this new error condition would help anybody from
> wasting time scratching her head before viewing the source and
> realize that the underlying name-rev does not honor the option.

The story was: I tried to understand why git describe --contains did not do what I expected. The documentation said that --debug would output candidates, but it did not do anything.

So, instead of learning how git describe --contains works, I got side-tracked into understanding why git describe --contains does not do what the documentation says. That was a waste of time that we can avoid.

"viewing the source" should not be necessary to understand what is going on, should it?

Show 8 quoted lines
> If "--debug" is truly valuable, "name-rev" can gain "--debug" later
> and then we can pass it down if we want.
> 
> Also, it is not like "--debug" is incompatible.  It is just the
> "--contains" codepath is overly silent and does not give any useful
> information when run in the debug mode.  "incompatible" is more like
> "would not work correctly when both are given", which is not the
> case here.

Well, thee notion of giving debug output is certainly not incompatible. But if the "--debug" does not output anything with "--contains" then it is not working, which I would call incompatible (implementation, not concept).

Show 33 quoted lines
>>
>> Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net>
>> ---
>>  Documentation/git-describe.txt | 2 +-
>>  builtin/describe.c             | 2 ++
>>  2 files changed, 3 insertions(+), 1 deletion(-)
>>
>> diff --git a/Documentation/git-describe.txt b/Documentation/git-describe.txt
>> index 8755f3af7b..0f9adb6e9a 100644
>> --- a/Documentation/git-describe.txt
>> +++ b/Documentation/git-describe.txt
>> @@ -69,7 +69,7 @@ OPTIONS
>>  --debug::
>>  	Verbosely display information about the searching strategy
>>  	being employed to standard error.  The tag name will still
>> -	be printed to standard out.
>> +	be printed to standard out. This is incompatible with --contains.
>>  
>>  --long::
>>  	Always output the long format (the tag, the number of commits
>> diff --git a/builtin/describe.c b/builtin/describe.c
>> index 76c18059bf..01a6d159a0 100644
>> --- a/builtin/describe.c
>> +++ b/builtin/describe.c
>> @@ -462,6 +462,8 @@ int cmd_describe(int argc, const char **argv, const char *prefix)
>>  
>>  	if (longformat && abbrev == 0)
>>  		die(_("--long is incompatible with --abbrev=0"));
>> +	if (contains && debug)
>> +		die(_("--debug is incompatible with --contains"));
>>  
>>  	if (contains) {
>>  		struct string_list_item *item;
Previous: Junio C HamanoNext: Michael J Gruber
Message 43 of 46 in “describe --contains sanity”
  1. 0/3 describe --contains sanityMichael J Gruber, Mar 15, 2017
  2. 3/3 name-rev: Allow lightweight tags and branch refsMichael J Gruber, Mar 15, 2017
  3. Junio C HamanoMar 15, 2017
  4. Junio C HamanoMar 15, 2017
  5. Junio C HamanoMar 15, 2017
  6. 0/2 Teach name-rev to pay more attention to lightweight tagsJunio C Hamano, Mar 15, 2017
  7. 2/2 name-rev: favor describing with tags and use committer date to tiebreakJunio C Hamano, Mar 15, 2017
  8. Lars SchneiderMar 17, 2017
  9. Junio C HamanoMar 17, 2017
  10. Junio C HamanoMar 17, 2017
  11. Lars SchneiderMar 17, 2017
  12. Junio C HamanoMar 17, 2017
  13. Junio C HamanoMar 17, 2017
  14. 2/3 name-rev: favor describing with tags and use committer date to tiebreakMichael J Gruber, Mar 29, 2017
  15. 0/3 name-rev sanityMichael J Gruber, Mar 29, 2017
  16. Junio C HamanoMar 29, 2017
  17. Junio C HamanoMar 29, 2017
  18. Michael J GruberMar 30, 2017
  19. Junio C HamanoMar 30, 2017
  20. 3/4 name-rev: provide debug outputMichael J Gruber, Mar 31, 2017
  21. Junio C HamanoMar 31, 2017
  22. Junio C HamanoMar 31, 2017
  23. Michael J GruberMar 31, 2017
  24. Junio C HamanoMar 31, 2017
  25. Junio C HamanoMar 31, 2017
  26. Michael J GruberApr 3, 2017
  27. Junio C HamanoApr 3, 2017
  28. Junio C HamanoMay 20, 2017
  29. Junio C HamanoMar 31, 2017
  30. 4/4 describe: pass --debug down to name-revMichael J Gruber, Mar 31, 2017
  31. 0/4 name-rev sanityMichael J Gruber, Mar 31, 2017
  32. 1/4 name-rev: refactor logic to see if a new candidate is a better nameMichael J Gruber, Mar 31, 2017
  33. 2/4 name-rev: favor describing with tags and use committer date to tiebreakMichael J Gruber, Mar 31, 2017
  34. 1/3 name-rev: refactor logic to see if a new candidate is a better nameMichael J Gruber, Mar 29, 2017
  35. 3/3 name-rev: provide debug outputMichael J Gruber, Mar 29, 2017
  36. Michael J GruberMar 17, 2017
  37. Junio C HamanoMar 17, 2017
  38. 1/2 name-rev: refactor logic to see if a new candidate is a better nameJunio C Hamano, Mar 15, 2017
  39. Stefan BellerMar 16, 2017
  40. Jacob KellerMar 16, 2017
  41. 1/3 describe: debug is incompatible with containsMichael J Gruber, Mar 15, 2017
  42. Junio C HamanoMar 15, 2017
  43. Michael J GruberMar 17, 2017
  44. 2/3 git-prompt: add a describe style for any tagsMichael J Gruber, Mar 15, 2017
  45. Junio C HamanoMar 15, 2017
  46. Michael J GruberMar 17, 2017

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.