@pmatilai pushed 1 commit.
3e63622a0aa8054836f40b6b093740505339bf26 Add %{xdg:...} macro for evaluating
XDG base directory paths
--
View it on GitHub:
https://github.com/rpm-software-management/rpm/pull/3660/files/d5f7e9aa1637df31bce382ec64df7b7af593935b..3e63622a0aa8054836f40b6b093740505339bf
dmnks left a comment (rpm-software-management/rpm#3660)
> Maybe, but somebody might try hard to find some use case for such macro.
I don't see any issue with this macro being documented. Whether or not it's
useful for packaging is irrelevant, it's still a macro.
> And I don't see the macro to b
voxik left a comment (rpm-software-management/rpm#3663)
To me it seems that the example above does not clearly demonstrates what the
`%{trim}` really does. E.g. to me it seems that the `\` were omitted just as
and side effect and the `%{trim}` usage should be more localized.
--
Reply to this e
allhailwesttexas created an issue (rpm-software-management/rpm#3664)
**Describe the bug**
rpmspec has stopped reporting debug packages between 4.19.1.1 and 4.20.0. We
use this output to track which packages are expected to be built from a
specfile, which is now inconsistent between Fedora 40 (4.
Closed #3661 as not planned.
--
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/3661#event-16820953001
You are receiving this because you are subscribed to this thread.
Message ID:
___
Rpm-maint
@ferdnyc commented on this pull request.
> +Typographic conventions:
+- Put option aliases on separate lines
+- Put option arguments with pre-determined values into <> (see below)
+
+*-d*
+*--do-this*
+ Do Y instead of default action.
+
+*-o* _ARG_,
+*--output*=_ARG_
+ Use file _ARG_
pmatilai left a comment (rpm-software-management/rpm#2207)
Yeah, I've seen those at one point or another.
spec-helper is equally inflexible as rpm's `%__os_install_post` method, and has
these inconsistently named environment variable disablers. It's what more or
less inspired the current disab
pmatilai left a comment (rpm-software-management/rpm#3661)
I don't think macro expansion eats backslashes, but the spec parser has them
for breakfast, lunch and dinner. Basically: always escape `\` if it's subjected
to the spec parser in any way at all.
Note that "%global macro using %{expand:
voxik left a comment (rpm-software-management/rpm#3660)
I wonder what is utility of this macro for general packager community 🤔 in
distributions, RPM does not manage the user home directory, even if it did, the
user is always `root` for package installation. Or are user installed RPMs a
thing?
voxik left a comment (rpm-software-management/rpm#3660)
> Not everything in rpm exists solely for the packagers.
Maybe, but somebody might try hard to find some use case for such macro. And I
don't see the macro to be "### File and path operations". It does not do any
operation or does it?
--
pmatilai left a comment (rpm-software-management/rpm#3660)
Not everything in rpm exists solely for the packagers.
--
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/pull/3660#issuecomment-2728899095
You are receiving this because you are subscrib
gpg may prompt for a password which will hold up any package from building
You can view, comment on, or merge this pull request online at:
https://github.com/rpm-software-management/rpm/pull/3662
-- Commit Summary --
* Never gpg sign git commits in macros
-- File Changes --
M macros.in
@pmatilai commented on this pull request.
> @@ -136,6 +136,7 @@ various common operations.
| `%{suffix:...}` | expand to suffix part of a file name |
| `%{url2path:...}` | convert url to a local path |
| `%{uncompress:...}` | expand to a command for outputting argument file to
stdout,
pmatilai left a comment (rpm-software-management/rpm#3660)
XDG_CACHE_HOME is just a weird oversight, I'll add it to the patch.
Annoyingly
https://specifications.freedesktop.org/basedir-spec/basedir-spec-latest.html is
messed up by some migration status tracker now, but looking at
https://web.a
pmatilai left a comment (rpm-software-management/rpm#2207)
Maybe a combination of the Suse drop-in mechanism and existing macro definition
practise could work though: run a discovery on the drop-in directory contents,
defining macros for each script present. And then auto-generate the
`%__os_po
Merged #3660 into master.
--
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/pull/3660#event-16797025145
You are receiving this because you are subscribed to this thread.
Message ID:
___
Rpm-maint mail
Closed #2259 as duplicate.
--
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/2259#event-16800815503
You are receiving this because you are subscribed to this thread.
Message ID:
___
Rpm-maint m
pmatilai left a comment (rpm-software-management/rpm#2207)
Uh, and that's pretty much exactly what @fweimer 's patch to redhat-rpm-config
in https://src.fedoraproject.org/rpms/redhat-rpm-config/pull-request/318 does.
No wonder it sounded so familiar when describing it above - I had seen the PR
ferdnyc left a comment (rpm-software-management/rpm#3661)
@pmatilai
> I don't know if it's still in the Fedora guidelines, but "always use
> %global" was bad advice that spread far and wide.
It basically is; there's a bit of hemming and hawing, someone added a
disclaimer about the difference
It trims the body by removing all leading and trailing whitespaces from each
line of the string
The new operation can be useful for writing human-readable, multi-line macros.
For example, the macros from
[ghc-rpm-macros-extra](https://src.fedoraproject.org/rpms/ghc-rpm-macros/blob/rawhide/f/mac
20 matches
Mail list logo
