I've transfered ownership of `debbuild` after running out of ideas for that
tool, so `rpmbuild` is of even less interest to me. Feel free to close this PR
or elaborate on it as you deem appropriate.
--
You are receiving this because you are subscribed to this thread.
Reply to this email
Closed #953.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/pull/953#event-2849137933___
Rpm-maint mailing list
It's easier (and less error-prone) to change incompatible behavior [in a single
spot](https://github.com/ascherer/cwebbin/blob/master/cwebbin.spec#L54), e.g.,
when switching from GNU/Linux to macOS.
--
You are receiving this because you are subscribed to this thread.
Reply to this email
I've been using the double-underscore names in [my
specfiles](https://gist.github.com/ascherer/474764be05e3d8031f0e1cf375b61b4c)
for years, and secondary versions like `%{__ln_s}` and `%{__mkdir_p}` always
came quite handy.
--
You are receiving this because you are subscribed to this thread.
BSDs sed _requires_ an `extension` for the `-i` option, even an
empty (`-i `) extension to suppress creation of a backup
for in-place editing, while GNU seds `-i` defaults to no
backup without an express `suffix`. With `%{__sed_i}` it should be easier
to write portable specfiles for both
You can view, comment on, or merge this pull request online at:
https://github.com/rpm-software-management/rpm/pull/634
-- Commit Summary --
* Fix double point.
-- File Changes --
M doc/rpmbuild.8 (6)
-- Patch Links --
https://github.com/rpm-software-management/rpm/pull/634.patch
@ascherer pushed 1 commit.
17bbc985df9bb538dcb7e6aea8316a4f9f0dd0fe Simplify Lua backwards compatibility.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
Re-establish string `"_LOADED"` for pre-5.3.4 Lua.
You can view, comment on, or merge this pull request online at:
https://github.com/rpm-software-management/rpm/pull/609
-- Commit Summary --
* Fix issue #608.
-- File Changes --
M lib/rpmliblua.c (6)
-- Patch Links --
It seems [you're at
it](https://github.com/rpm-software-management/rpm/pull/169). KYFC!
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
Even trying to install `lua5.3` and associated dev stuff in parallel fails,
because Ubutun 16.04 only provides Lua 5.3.1 in its standard repos.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
It seems that `"_LOADED"` was replaced by macro `LUA_LOADED_TABLE` in [lua
5.3.4](https://github.com/lua/lua/commit/b2aa2ba046502bdcdfa3de4af898810667c1843a).
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
f13de84bab9cd5b1be30c44002285743c8c3f7a9 is the first bad commit; see [the
associated log
files](https://gist.github.com/ascherer/0235fd0c7c16bbe8a03d87eb88c27aeb) for
details.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on
I'll `bisect`: compilation worked at 7ffc4d1 in September (when I submitted my
last PR).
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
Lua 5.2.4 Copyright (C) 1994-2015 Lua.org, PUC-Rio
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
Trying to do the following on Kubuntu 16.04.5
```
git clone --depth=1 g...@github.com:ascherer/rpm.git
cd rpm
sudo add-apt-repository ppa:bitcoin/bitcoin
sudo apt update
sudo apt install autoconf-archive fakechroot libarchive-dev:amd64 libdb4.8
libdb4.8++ libdb4.8++-dev libdb4.8-dev
No, I'm just being pedantic. :smiley:
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/pull/579#issuecomment-435932580___
Rpm-maint
The `%patch` macro uses only single-character options; for consistency, these
are passed to the `patch` command as well.
You can view, comment on, or merge this pull request online at:
https://github.com/rpm-software-management/rpm/pull/579
-- Commit Summary --
* Pass %patch options to
@ascherer pushed 1 commit.
27409eb Add short rationale for shortcut '--scm=SCM'.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
See issue #463.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/570#issuecomment-429056223___
Rpm-maint mailing list
Aside: I haven't noticed (m)any incompatibilities with `git`, `git_am`,
`quilt`, `gendiff`, and the default `patch` backends with any of my specfiles,
where `%autosetup`/`%autopatch` is used.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or
It's much easier to type. As the co-author and current maintainer of
`debbuild`, I often run tests on my own packages with various SCM programs.
With the exception of a single (private) specfile, where several non-automatic
`git commit`commands are hard-coded, non of my other specfiles uses the
Try `%{quote:Bodoni Specimen.pdf}`. (Off the top of my hat.)
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
… and for `git`, `git_am`, `gendiff`. I don't have `hg`, `quilt` installed to
check these. What else?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
Why?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/pull/564#issuecomment-427672948___
Rpm-maint mailing list
In addition to `--define="__scm "` and `-D "__scm "`, rpmbuild now
acknowledges the short form `-S ` for selecting the versioning system from
the command line.
You can view, comment on, or merge this pull request online at:
https://github.com/rpm-software-management/rpm/pull/564
-- Commit
Hand over to PR #564.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/100#issuecomment-427557546___
Rpm-maint mailing list
Closed #100.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/100#event-1888486062___
Rpm-maint mailing list
And reuse the already exported environment.
You can view, comment on, or merge this pull request online at:
https://github.com/rpm-software-management/rpm/pull/563
-- Commit Summary --
* Remove redundant 'cd %{_builddir}'.
-- File Changes --
M build/parsePrep.c (9)
M macros.in (2)
Closed #558.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/pull/558#event-1882474077___
Rpm-maint mailing list
Done.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/pull/557#issuecomment-426715690___
Rpm-maint mailing list
@ascherer pushed 1 commit.
1338c02 Compactify tar option string.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
@ascherer pushed 1 commit.
cc31bc5 Correctly inject gem spec options.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
Don't worry. I'll smash my commits after I'm back at my own machine, where I
can actually compile and check this stuff.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
@ascherer pushed 1 commit.
6e2b98a DRY: Uniform error handling.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
@ascherer pushed 1 commit.
001509e Quit processing if tar fails.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
@ascherer pushed 2 commits.
6036ec0 Source: file extraction may use other filters.
0b4473e Use default option macros for extractors.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
@pmatilai I'll try to provide a more generic approach. Combining #557 and #558
surely is feasible.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
@ascherer pushed 1 commit.
c288ac1 Add %autosetup option -tN.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
Closed #548.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/548#event-1874148855___
Rpm-maint mailing list
Hand over to PR #558.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/548#issuecomment-425521671___
Rpm-maint mailing list
The numeric value `N` from `%setup -pN` will be passed to `tar`'s
`--strip-component=N` option,
which eliminates top `N` parts of paths in the tarball.
You can view, comment on, or merge this pull request online at:
https://github.com/rpm-software-management/rpm/pull/558
-- Commit Summary --
Closed #549.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/549#event-1873855343___
Rpm-maint mailing list
Hand over to PR #557.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/549#issuecomment-425478667___
Rpm-maint mailing list
`%patch` is supplied with `%_default_patch_flags` from `macros.in`.
A similar macro `%_default_tar_flags` may be useful when `%__tar` is invoked by
`%setup` "behind the scenes". A possible setting in `macros.in` is
`--keep-old-files` (`-k`).
You can view, comment on, or merge this pull request
As it stands, `%___build_shell` might as well be renamed to/replaced with
`%_buildshell`.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
I believe that's what the `-n` option of `%setup` is for. You can (or have to)
set/repeat the desired `%buildsubdir` on every invocation of `%setup`. (IIRC,
multiple `%setup`s are deemed esoteric/byzantine anyway.)
--
You are receiving this because you are subscribed to this thread.
Reply to
`%__build_shell` is perfectly set up to work without it.
Note: You can use `rpmbuild --define="_buildshell /bin/bash"` to use a
different shell interpreter.
You can view, comment on, or merge this pull request online at:
https://github.com/rpm-software-management/rpm/pull/550
-- Commit
`%patch` is supplied with `%_default_patch_flags` from `macros.in`.
A similar macro `%_default_tar_flags` may be useful when `%__tar` is invoked by
`%setup` "behind the scenes". A possible setting in `macros.in` might be
`--keep-old-files` (`-k`).
--
You are receiving this because you are
`%patch` provides option `-pN` that gets promoted to `patch` with the same name.
I propose a similar option `-pN` for `%setup` that gets promoted to `tar` as
its option `--strip-component=N` (available both in Linux and BSD tar).
--
You are receiving this because you are subscribed to this
@ignatenkobrain what exactly are you trying to set up in the `BUILD` arena?
After downloading ca. 80 MB (which most certainly is not a 'minimal' example),
I find that the two tarballs contain top-level directories `nipy-data-0.2/` and
`nipy-templates-0.2/` respectively. So, with your example
Has anybody tried to use
```
%autosetup -c -T -a 0 -N
%autosetup -T -a 1
```
?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
Maybe `%autosetup` should be "internalized" in `parsePrep.c` alongside
`%setup`? For the most part – after groking and converting some options – it
would call `doSetupMacro()` (plus `%autopatch`).
Moreover, it would be possible to issue a more useful diagnostic if the user
requests an unknown
This reminds me of the [weirdness of `%patch -P 1 2
3`](https://github.com/rpm-software-management/rpm/pull/19). Would it useful to
make matters with `%setup` more complicated by permitting options `-b` and `-a`
to have multiple numeric arguments? :wink:
BTW, `debbuild`'s `%autosetup`
You can use `%autosetup -N -S git` to initialize `%__scm` and then
`%autopatch`/`%apply_patch` continues appropriately. I believe I have a private
specfile somewhere that makes use of such a recipe (plus some manual `git
commit`invocations in between).
--
You are receiving this because you
@Conan-Kudo you may want to try the new `%_debmacrodir` from debbuild 18.8.1
for your macro bundle.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
@pmatilai Maybe I should simply delete the empty `%build` stage in this
specfile.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
After neutralizing the TeX macros, I can't reproduce the patch error.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
Closed #484.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/484#event-1773221029___
Rpm-maint mailing list
I just learned about `make check` (and `fakechroot`) and can report that the
modified `rpmbuild` produces the exactly same 70,986 bytes of log output as the
original `rpmbuild` as of `HEAD`.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or
@ascherer pushed 1 commit.
74daeff Addendum: Clean up logic in 'doScript()'.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
Yep, I just wrote a [gist with an escape
function](https://gist.github.com/ascherer/c5813ade3df9207573be0227555371d0) on
my machine at work. I'll incorporate this in the real code at home later and
rephrase the PR accordingly.
--
You are receiving this because you are subscribed to this
Don't rush me! I'll try to add the '%' masking ASAP. :o) Maybe in the meantime
someone can direct me to the canonical way inside the rpm codebase?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
@ascherer pushed 1 commit.
c77f6d3 Activate empty default for %___build_body.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
Thanks, now `sgb.soec` works fine with **rpmbuild 4.14.90**.
I propose to apply this (squashed) PR for inclusion.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
The following specfile works fine with **rpm 4.12.0.1**, but fails with **rpm
4.12.0.1**.
```
Name: tetex-anschrift
Summary: Eigene Briefgestaltung mit diversen LaTeX-Klassen
Version: 1.4
Release: 8
Packager: Andreas Scherer
License: LPPL
URL: localhost://home/andreas/TeX/Briefe/anschrift
And finally `sudo make install`:
```
make install-recursive
make[1]: Verzeichnis „/opt/github/rpm“ wird betreten
Making install in po
make[2]: Verzeichnis „/opt/github/rpm/po“ wird betreten
/bin/mkdir -p /usr/local/share
installing ar.gmo as /usr/local/share/locale/ar/LC_MESSAGES/rpm.mo
Àfter `sed "s/lua >=/lua5.2 >=/g" -i configure` the following `./configure`
writes this:
```
checking for a BSD-compatible install... /usr/bin/install -c
checking whether build environment is sane... yes
checking for a thread-safe mkdir -p... /bin/mkdir -p
checking for gawk... gawk
checking
Here's the output from `./autogen.sh --noconfigure`:
```
Copying file ABOUT-NLS
Copying file config.rpath
Creating directory m4
Copying file m4/codeset.m4
Copying file m4/gettext.m4
Copying file m4/glibc2.m4
Copying file m4/glibc21.m4
Copying file m4/iconv.m4
Copying file m4/intdiv0.m4
Copying
Why is it not build resp. installed by the recipe above?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
What is `/usr/local/lib/rpm/elfdeps` and how to I compile it from the sources
of rpm 4.14.90?
This missing executable (?) breaks all my packages.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
@ascherer pushed 1 commit.
efb1355 Purge scriptlet files for production.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
@ascherer pushed 1 commit.
bcff142 Purge test stage scriptlets.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
@ascherer pushed 1 commit.
1080ad3 Successful run with modified rpm 4.14.90 (almost).
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
@ascherer pushed 3 commits.
3f29960 Failed run with modified rpm 4.14.90.
ffcf9ef Reshuffle code to concat stage script parts.
665c32b Clean out the diagnostics and old comments.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
@ascherer pushed 1 commit.
de4dd0a Failed run with modified rpm 4.14.90.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
Successo! It looks like I have installed enough development stuff to make
```
./autogen.sh --noconfigure
sed "s/lua >=/lua5.2 >=/g" -i configure
./configure
make
sudo make install
```
complete. Now I seem to have
```
LD_LIBRARY_PATH=/usr/local/lib rpmbuild --version
```
respond with "4.14.90" and
@ascherer pushed 1 commit.
6ba0bec Add header for 'rasprintf()'.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
@ascherer pushed 1 commit.
3383d3d Next step of my 'gedankenexperiment'.
--
You are receiving this because you are subscribed to this thread.
View it on GitHub:
So I guess I should investigate `rpmPopMacro(NULL, '___build_body')` and
`rpmPushMacro(NULL, '___build_body', sb)` (approximately), iff the global
context is correct.
Probably I should reestablish the build environment in order to compile
`rpmbuild` locally. (The last time was on Mint, but on
Thanks @n3npq for your comments. My considerations were indeed triggered by
today's work on `debbuild`, which (a) didn't use `%___build_post` at all, and
(b) would use the inspecific value **0** in all situations, even errors. After
inspecting the PR for `debbuild`, I changed `debbuild`'s
Closed #479.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/479#event-1757060262___
Rpm-maint mailing list
See the following pull request #480.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/479#issuecomment-408390409___
Rpm-maint
Wouldn't it be nice (or would it) to use the full power of macro
`%___build_template` if only it included the `%___build_body` and the vital
`%___build_post` part already?
Unfortunately, I'm not familiar with the internal macro engine of RPM to finish
this myself, but I hope for someone more
Don't abort macro
[`%__build_template`](https://github.com/rpm-software-management/rpm/blob/master/macros.in#L834)
after `%___build_pre`, but set `%___build_body` "somewhere in the middle".
(See the [dynamic solution in
At least with **rpmbuild 4.12.0.1** on Kubuntu 16.04 it seems to be impossible
to comment parts of script sections. This code
```
%build
# Don't expand %autopatch here
```
_does_ invoke macro `%autopatch` quite inappropriately.
--
You are receiving this because you are subscribed to this
Closed #127.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/issues/127#event-924250381___
Rpm-maint mailing list
In the particular case of [`%autosetup` in
`sgb.spec`](https://github.com/ascherer/sgb/blob/local/sgb.spec#L52), using
`-N` or `-p1` depending on the CLI option `--without patches` is utterly
redundant, because [the set of patches is
I have [at least one
example](https://github.com/ascherer/sgb/blob/local/sgb.spec#L53) where the
conditional macro `%{?with_patches:-p1}` does _not_ produce the expected `-p1`
option correctly.
This occurs with **rpmbuild 4.12** on **Kubuntu 16.04**.
--
You are receiving this because you
If one does _not_ use the `-S SCM` option of the `%autosetup` macro in a
specfile, it is possible to activate a SCM of choice by invoking `rpmbuild`
with the somewhat convoluted option
```
--define="__scm SCM"
```
It would be even more convenient to have a shortcut option `-S SCM` on the
I can't find any good reason. Case closed.
---
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/pull/55#issuecomment-187211753___
Rpm-maint mailing list
Rpm-maint@lists.rpm.org
Replacing `%__scm_apply_quilt(qp:m:)` according to
http://mywiki.wooledge.org/Bashism from
```
%{__quilt} import %{-p:-p%{-p*}} %{1} && %{__quilt} push %{-q}
```
to
```
mkfifo debbuild.fifo; cat >debbuild.fifo &\
%{__quilt} import %{-p:-p%{-p*}} -P %{-m*} debbuild.fifo && %{__quilt} push
%{-q}\
Another try with `<(cat)` instead of `%{1}`: Set **bash** as executing shell in
the specfile
```
%prep
%define _buildshell /bin/shell
%autosetup -S svn
```
This produces the correct commands
```
+ /bin/cat
/home/andreas/Programming/RPM/SOURCES/0001-Fix-intermediate-bugs.patch
+ /usr/bin/svn
I suggest to stick with `%{1}` (for both `%__scm_apply_svn` and
`%__scm_apply_quilt`) and go without compressed patches (well, `quilt` _does_
this internally).
---
Reply to this email directly or view it on GitHub:
@lkardos Thanks for the hint; that syntax was new to me. In a first attempt, it
indeed works from the console with `quilt` and `.zip`ped patches. However,
`<(cat)` seems to be **bash** syntax. Both **debbuild** and **rpmbuild** use
`#!/bin/sh` as the executing shell interpreter, which – at
Keyword "process substitution": http://mywiki.wooledge.org/Bashism
---
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/pull/55#issuecomment-176393161___
Rpm-maint mailing list
Rpm-maint@lists.rpm.org
Personally, I don't use the Perl macros, but I know of [at least this
project](https://github.com/pld-linux/perl-Pod-Simple/blob/master/perl-Pod-Simple.spec#L5)
that makes use of `macros.perl`. (I suggested to [use
`%_rpmconfigdir`](https://github.com/pld-linux/perl-Pod-Simple/pull/1) for the
What error do you get? I just checked, installed _this_ branch and edited
[MMIX specfile](https://github.com/ascherer/mmix/blob/local/mmix.spec#L36)
accordingly.
```
debbuild -bp SPECS/mmix.spec
```
runs perfectly with `-S git`, `-S git_am`, `-S patch`, and `-S quilt`, but
**not** with `-S
> @@ -1120,6 +1122,23 @@ done \
> %{__patch} %{-p:-p%{-p*}} %{-q:-s}\
> %{__bzr} commit %{-q} -m %{-m*}
>
> +# Subversion
> +%__scm_setup_svn(q)\
> +%{__svnadmin} create .svnrepos\
> +%{__svn} mkdir %{-q} -m "Create directory structure."
> file://`pwd`/.svnrepos/trunk\
> +%{__svn} checkout
> @@ -1120,6 +1122,23 @@ done \
> %{__patch} %{-p:-p%{-p*}} %{-q:-s}\
> %{__bzr} commit %{-q} -m %{-m*}
>
> +# Subversion
> +%__scm_setup_svn(q)\
> +%{__svnadmin} create .svnrepos\
> +%{__svn} mkdir %{-q} -m "Create directory structure."
> file://`pwd`/.svnrepos/trunk\
> +%{__svn} checkout
> @@ -1120,6 +1122,23 @@ done \
> %{__patch} %{-p:-p%{-p*}} %{-q:-s}\
> %{__bzr} commit %{-q} -m %{-m*}
>
> +# Subversion
> +%__scm_setup_svn(q)\
> +%{__svnadmin} create .svnrepos\
> +%{__svn} mkdir %{-q} -m "Create directory structure."
> file://`pwd`/.svnrepos/trunk\
> +%{__svn} checkout
1 - 100 of 110 matches
Mail list logo
