This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
| Other format: | [Raw text] | |
On 05 Aug 2015 09:34, Andreas Schwab wrote:
> Mike Frysinger <vapier@gentoo.org> writes:
> > i'm still strongly in favor of getting rid of ChangeLogs entirely. i have
> > yet to find a scenario where the git commit itself did not contain all the
> > details that i needed, or that really the ChangeLog file itself provided
> > literally any value at all. in a historical context (CVS), it might have
> > made more sense, but that hasn't been the case for years.
>
> A simple text file is much easier to search through.
except:
- there's many more than one text file (ChangeLog*, */ChangeLog*, ...)
- `git grep foo` is easier than `find -name 'ChangeLog*' -exec grep foo {} +`
and in my experience also much faster
- it doesn't tell you anything at all about *why* (the commit message)
- it doesn't show the actual diff, just file names
- the file names in the log are often trunacted (based on the topdir)
i have no idea what you'd even search for in the ChangeLog that'd be useful.
symbol names ? super inconsistent due to the GNU style. can you describe
your workflow to illustrate their value ?
i find answers to my queries way more often using `git log` and then searching
the paged output via less. often i throw in -p or --stat when i want details
like filenames since i can actually rely on the data.
-mike
Attachment:
signature.asc
Description: Digital signature
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |