Re: [gentoo-dev] [PATCH] vdr-plugin-2.eclass: make qa warning conditional
Am 09.05.24 um 15:08 schrieb Sam James: It sounds like maybe you want to turn these into log messages (einfo - https://devmanual.gentoo.org/ebuild-writing/messages/index.html), and drop the "QA notice" prefix. I changed https://github.com/gentoo/gentoo/pull/36504 accordingly and added the "Closes:" tags. Can you please review?
Re: [gentoo-dev] [PATCH] vdr-plugin-2.eclass: make qa warning conditional
Am 09.05.24 um 14:13 schrieb Sam James: Martin Dummer writes: Maybe we can agree that the qa-warnings in vdr-eclass make more sense if i change them to "eawarn" or "einfo"? Sure, make them eqawarn. Hmm - current state of vdr-plugin-2.eclass is: there are many "eqawarn" in there. In my opinion, most plugins in the vdr context will practically not develop any further anyway. It is more important to keep the current status of vdr-software in the ecosystem up to date as well as possible. So I need a practical useful approach instead of a fundamental discussion please. My point is that the QA warnings should exist, and you can worry about making them "developer-only" in future. Right now, they seem useful, and the things they flag need to be addressed. Basically yes, but here (vdr-plugins) the qawarn are used more in a way "to remind the plugin developers to adapt their sources to newer vdr build environment" or "the way i18n implemented has changed" The eclass fixes these problems with standardized quirks, the "equawarn" messages show when these quirks are applied. IMHO its not necessary to tell that to any user, only for interested packagers when they are bored and look out for some extra work. That's why I made the warnings conditional, printed out when the variable "VDR_MAINTAINTER_MODE" is set to a not-empty value. Finally, I am interested in an opinion whether this is acceptable or not, otherwise I tend to remove the warnings at all.
Re: [gentoo-dev] [PATCH] vdr-plugin-2.eclass: make qa warning conditional
Am 03.05.24 um 06:39 schrieb Sam James: What we really need is: a)https://bugs.gentoo.org/162450 to avoid scaring users; b) possibly some level of QA notice to distinguish between "check this out" (think e.g. qa-vdb LHS where it _might_ be unused, but not necessarily), and "this is definitely wrong" I am convinced we need a), I am not-at-all convinced we need b) - at least not in terms of whether bugs are reported. AFAIS https://bugs.gentoo.org/162450 is not implemented. Maybe we can agree that the qa-warnings in vdr-eclass make more sense if i change them to "eawarn" or "einfo"? In my opinion, most plugins in the vdr context will practically not develop any further anyway. It is more important to keep the current status of vdr-software in the ecosystem up to date as well as possible. So I need a practical useful approach instead of a fundamental discussion please.
Re: [gentoo-dev] Projects without members: ALSA, PHP, VDR
Please make a PR. https://github.com/gentoo/gentoo/pull/36505
Re: [gentoo-dev] Projects without members: ALSA, PHP, VDR
Am 01.05.24 um 16:03 schrieb Michał Górny: The following projects have no members right now: https://wiki.gentoo.org/wiki/Project:VDR I am already proxy maintainer for 27 vdr packages, you can add me as proxy maintainer to all remaining packages of v...@gentoo.org. Or I place a PR on github for this, let me know your preference.
[gentoo-dev] [PATCH] vdr-plugin-2.eclass: make qa warning conditional
Since Agostino's tinderbox tests now report qa warning, the group
v...@gentoo.org has 101 open bugs assigned, most of them caused by qa
warnings from vdr-plugin-2.eclass.
Many vdr plugins need small adjustments because API or makefile changes
in upstream media-video/vdr which can be easily fixed with small changes.
These warnings are only useful for the vdr plugin maintainers, so I
propose they should (only) be reported as QA-warnings when the global
variable
VDR_MAINTAINER_MODE="1"
is set in make.conf
This patch is also put to github in
https://github.com/gentoo/gentoo/pull/36504
The PR is lacking many many "Closes: " tags, which I will fill in soon.
Any comments?
-- Martin
commit 00a0a3237729ec4886e07d9869fdfd6e0cd5f985
Author: Martin Dummer
Date: Wed May 1 14:49:37 2024 +0200
vdr-plugin-2.eclass: make qa warning conditional
many vdr plugins need small adjustments because API or makefile changes
in upstream media-video/vdr which can be easily fixed with small
changes
These fixes should (only) be reported as QA-warnings when the
global variable
VDR_MAINTAINER_MODE="1"
is set in make.conf
Signed-off-by: Martin Dummer
diff --git a/eclass/vdr-plugin-2.eclass b/eclass/vdr-plugin-2.eclass
index 8f56511032c8..bc9a88db59a5 100644
--- a/eclass/vdr-plugin-2.eclass
+++ b/eclass/vdr-plugin-2.eclass
@@ -160,7 +160,6 @@ vdr_create_header_checksum_file() {
# Plugins failed on compile with wrong path of libsi includes,
# this can be fixed by 'function + space separated list of files'
fix_vdr_libsi_include() {
- eqawarn "QA Notice: Fixing include of libsi-headers"
local f
for f; do
sed -i "${f}" \
@@ -244,7 +243,7 @@ vdr_gettext_missing() {
# plugins without converting to gettext
local GETTEXT_MISSING=$( grep xgettext Makefile )
- if [[ -z ${GETTEXT_MISSING} ]]; then
+ if [[ -z ${GETTEXT_MISSING} && -n ${VDR_MAINTAINER_MODE} ]]; then
eqawarn "QA Notice: Plugin isn't converted to gettext handling!"
fi
}
@@ -306,11 +305,15 @@ vdr_i18n() {
if [[ -n ${I18N_OBJECT} ]]; then
if [[ "${KEEP_I18NOBJECT:-no}" = "yes" ]]; then
- eqawarn "QA Notice: Forced to keep i18n.o"
+ if [[ -n ${VDR_MAINTAINER_MODE} ]]; then
+ eqawarn "QA Notice: Forced to keep i18n.o"
+ fi
else
sed -i "s:i18n.o::g" Makefile \
|| die "sed failed to remove i18n from Makefile"
- eqawarn "QA Notice: OBJECT i18n.o found, removed per sed"
+ if [[ -n ${VDR_MAINTAINER_MODE} ]]; then
+ eqawarn "QA Notice: OBJECT i18n.o found, removed per sed"
+ fi
fi
fi
@@ -318,7 +321,9 @@ vdr_i18n() {
if [[ -n ${I18N_STRING} ]]; then
sed -i
"s:^extern[[:space:]]*const[[:space:]]*tI18nPhrase://static const
tI18nPhrase:" i18n.h \
|| die "sed failed to replace tI18nPhrase"
- eqawarn "QA Notice: obsolete tI18nPhrase found, disabled per
sed, please recheck"
+ if [[ -n ${VDR_MAINTAINER_MODE} ]]; then
+ eqawarn "QA Notice: obsolete tI18nPhrase found, disabled
per sed, please recheck"
+ fi
fi
}
@@ -337,7 +342,9 @@ vdr_remove_i18n_include() {
|| die "sed failed to remove i18n_include"
done
- eqawarn "QA Notice: removed i18n.h include in ${@}"
+ if [[ -n ${VDR_MAINTAINER_MODE} ]]; then
+ eqawarn "QA Notice: removed i18n.h include in ${@}"
+ fi
}
# @FUNCTION: vdr-plugin-2_print_enable_command
@@ -568,7 +575,9 @@ vdr-plugin-2_src_install() {
DESTDIR="${D%/}" \
|| die "emake install (makefile target) failed"
else
- eqawarn "QA Notice: Plugin use still the old Makefile handling"
+ if [[ -n ${VDR_MAINTAINER_MODE} ]]; then
+ eqawarn "QA Notice: Plugin use still the old Makefile handling"
+ fi
insinto "${VDR_PLUGIN_DIR}"
doins libvdr-*.so.*
fi
Re: [gentoo-dev] [PATCH 1/2] */*: Replace dependency to virtual/linuxtv-dvb-headers by sys-kernel/linux-headers
Am 25.02.24 um 11:10 schrieb Matthias Schwarzott: Am 13.02.24 um 09:19 schrieb Matthias Schwarzott: virtual/linuxtv-dvb-headers has been important in the past when linux-headers was not yet up-to-date. Now it just pulls in >=sys-kernel/linux-headers-3.7. Even that could be dropped as it is part of @system. Bug: https://bugs.gentoo.org/924398 Hi, +1 from me... but please take note that media-video/vdr has been upgraded in the meantime: [I] media-video/vdr Available versions: 2.2.0-r7 2.6.4 (~)2.6.6 so you must rework your patch a little bit. CU Martin
[gentoo-dev] Re: The Gentoo VDR Project is empty
Hi, as I already have taken care about many packages from the vdr project, I will continue to do so. Is there any progress on the discussion about [gentoo-project] RFC: "Trusted contributor model"? Meanwhile I continue to wait for an answer on my email to recruit...@gentoo.org from november 2019. Bye / Martin Am 06.11.22 um 08:30 schrieb Joonas Niilola: Hey all, with the only member of VDR project gone, the project is currently essentially empty. https://wiki.gentoo.org/wiki/Project:VDR Please join the project to keep it alive. We may have some help through the proxy-maint project, but someone with immediate commit access is always better. If no one joins the project, the packages (and eclass) are bound to be reassigned. -- juippis
[gentoo-dev] Re: [gentoo-project] RFC: "Trusted contributor model"
Am 22.07.22 um 21:10 schrieb Joonas Niilola: ..., via Github PR or other means. I'd say pushing these experimental packages is rather rare (for proxied people at least), and can also be added without KEYWORDS for example. I agree here. And I think the proxy maintainers can open PRs like now which can be merged by a bot (after the QA bot has completed successfully). This bot can have a ACL of person + package which is maintained separately. For special cases like changes to package.mask the regular PR flow can be used. My proposal here: better some automation for 90% purpose than a long discussion with many proposals and finally no automation - because none of the proposals has 100% satisfaction. Martin
Re: [gentoo-dev] Need help: Many patches for unmaintained packages
Am 31.12.21 um 06:51 schrieb Jonas Stein: I prepared a public Bugzilla filter "m-n PATCH" https://bugs.gentoo.org/buglist.cgi?cmdtype=dorem=run=m-n%20PATCH_id=85842 Hello Jonas, when I try to open that link I receive an error message "The search named /m-n PATCH/ has not been made visible to you." Can you check the settings again please? Martin
Re: [gentoo-dev] [PATCH] vdr-plugin-2.eclass: add EAPI8 support, EAPI7+ fixes
if there are no more points to discuss, could someone please add a review to https://github.com/gentoo/gentoo/pull/21447 ? One more finding: repoman says BDEPEND: consider using 'virtual/pkgconfig' instead of 'dev-util/pkgconfig' I will fix this.
Re: [gentoo-dev] [PATCH] vdr-plugin-2.eclass: add EAPI8 support, EAPI7+ fixes
Am 28.06.21 um 08:18 schrieb Ulrich Mueller: On Mon, 28 Jun 2021, Martin Dummer wrote: --- a/eclass/vdr-plugin-2.eclass +++ b/eclass/vdr-plugin-2.eclass @@ -76,7 +76,7 @@ inherit flag-o-matic toolchain-funcs unpacker This should also inherit strip-linguas.eclass, because strip-linguas is called in vdr_linguas_support(). Okay, I understand. I did not notice the call to "strip-linguas", and I'm wondering how this could have worked all the years in the past. In EAPIs 5 and 6 it inherited it indirectly via eutils. But indeed, in EAPI 7 I don't see how it could have worked. Ulrich Hi, if there are no more points to discuss, could someone please add a review to https://github.com/gentoo/gentoo/pull/21447 ? Martin
Re: [gentoo-dev] [PATCH] vdr-plugin-2.eclass: add EAPI8 support, EAPI7+ fixes
Am 27.06.21 um 20:57 schrieb Ulrich Mueller:
> [Either your mailer or something on the way has mangled whitespace in
> your patch, which makes it somewhat hard to read.]
Okay... not good, I will add the patchfile as attachment now.
>> --- a/eclass/vdr-plugin-2.eclass
>> +++ b/eclass/vdr-plugin-2.eclass
>> @@ -76,7 +76,7 @@
>> inherit flag-o-matic toolchain-funcs unpacker
> This should also inherit strip-linguas.eclass, because strip-linguas is
> called in vdr_linguas_support().
>
> Also, while at it, could you remove the useless IUSE="" assignment in
> line 87?
Okay, I understand. I did not notice the call to "strip-linguas", and
I'm wondering how this could have worked all the years in the past.
Unfortunately "pkgcheck" does not report this.
All changes are applied.
Bye
Martin
From a34b8547accafbc8e658d05fd389f5714c35dcb8 Mon Sep 17 00:00:00 2001
From: Martin Dummer
Date: Sun, 27 Jun 2021 17:05:23 +0200
Subject: [PATCH] vdr-plugin-2.eclass: add EAPI8 support, EAPI7+ fixes
- add EAPI8 support, no changes necessary
- add BDEPEND=pkgconfig
- remove outdated version numbers from (R)DEPEND
- fix calling of pkg-config
- add many "|| die" which should have been here already for EAPI7
Signed-off-by: Martin Dummer
---
eclass/vdr-plugin-2.eclass | 57 +-
1 file changed, 32 insertions(+), 25 deletions(-)
diff --git a/eclass/vdr-plugin-2.eclass b/eclass/vdr-plugin-2.eclass
index 3f3acc41208..6bd3a1fbff0 100644
--- a/eclass/vdr-plugin-2.eclass
+++ b/eclass/vdr-plugin-2.eclass
@@ -73,10 +73,10 @@
[[ ${EAPI} == [5] ]] && inherit multilib
[[ ${EAPI} == [56] ]] && inherit eutils
-inherit flag-o-matic toolchain-funcs unpacker
+inherit flag-o-matic strip-linguas toolchain-funcs unpacker
case ${EAPI:-0} in
- 5|6|7)
+ 5|6|7|8)
;;
*) die "EAPI ${EAPI} unsupported."
;;
@@ -84,8 +84,6 @@ esac
EXPORT_FUNCTIONS pkg_setup src_unpack src_prepare src_compile src_install pkg_postinst pkg_postrm pkg_config
-IUSE=""
-
# Name of the plugin stripped from all vdrplugin-, vdr- and -cvs pre- and postfixes
VDRPLUGIN="${PN/#vdrplugin-/}"
VDRPLUGIN="${VDRPLUGIN/#vdr-/}"
@@ -97,10 +95,11 @@ DESCRIPTION="vdr Plugin: ${VDRPLUGIN} (based on vdr-plugin-2.eclass)"
S="${WORKDIR}/${VDRPLUGIN}-${PV}"
# depend on headers for DVB-driver and vdr-scripts
-DEPEND=">=media-tv/gentoo-vdr-scripts-0.4.2
+BDEPEND="dev-util/pkgconfig"
+DEPEND="media-tv/gentoo-vdr-scripts
virtual/linuxtv-dvb-headers"
-RDEPEND=">=media-tv/gentoo-vdr-scripts-0.4.2
- >=app-eselect/eselect-vdr-0.0.2"
+RDEPEND="media-tv/gentoo-vdr-scripts
+ app-eselect/eselect-vdr"
if [[ "${GENTOO_VDR_CONDITIONAL:-no}" = "yes" ]]; then
IUSE="${IUSE} vdr"
@@ -151,7 +150,7 @@ vdr_create_header_checksum_file() {
local CHKSUM="header-md5-vdr"
if [[ -f ${VDR_CHECKSUM_DIR}/header-md5-vdr ]]; then
- cp "${VDR_CHECKSUM_DIR}/header-md5-vdr" "${CHKSUM}"
+ cp "${VDR_CHECKSUM_DIR}/header-md5-vdr" "${CHKSUM}" || die "Could not copy header-md5-vdr"
elif type -p md5sum >/dev/null 2>&1; then
(
cd "${VDR_INCLUDE_DIR}"
@@ -179,7 +178,8 @@ fix_vdr_libsi_include() {
for f; do
sed -i "${f}" \
-e '/#include/s:"\(.*libsi.*\)":<\1>:' \
- -e '/#include/s:<.*\(libsi/.*\)>::'
+ -e '/#include/s:<.*\(libsi/.*\)>::' \
+ || die "sed failed while fixing include of libsi-headers"
done
}
@@ -192,7 +192,7 @@ fix_vdr_libsi_include() {
vdr_patchmakefile() {
einfo "Patching Makefile"
[[ -e Makefile ]] || die "Makefile of plugin can not be found!"
- cp Makefile "${WORKDIR}"/Makefile.before
+ cp Makefile "${WORKDIR}"/Makefile.before || die "Failed to copy Makefile"
# plugin makefiles use VDRDIR in strange ways
# assumptions:
@@ -216,14 +216,16 @@ vdr_patchmakefile() {
-e '/VDRINCDIR.*=/!s:$(VDRDIR)/include:$(VDRINCDIR):' \
\
-e 's:-I$(DVBDIR)/include::' \
- -e 's:-I$(DVBDIR)::'
+ -e 's:-I$(DVBDIR)::' \
+ || die "sed failed to set \$VDRDIR"
if ! grep -q APIVERSION Makefile; then
ebegin " Converting to APIVERSION"
sed -i Makefile \
-e 's:^APIVERSION = :APIVERSION ?= :' \
-e 's:$(LIBDIR)/$@.$(VDRVERSION):$(LIBDIR)/$@.$(APIVERSION):' \
- -e '/VDRVERSION =/a\APIVERSION = $(shell sed -ne '"'"'/define APIVERSION/s/^.*"\\(.*\\)".*$$/\\1/p'"'"' $(VDRDIR)/config.h)'
+ -e '/VDRVERSION =/a\APIVERSION = $(shell sed -ne '"'"'/define APIVERSION/s/^.*"\\(.*\\)".*$$/\\1/p'"'"' $(VDRDIR)/config.h)' \
+ || die "sed failed to change APIVERSION"
eend $?
fi
@@ -231,13 +233,15 @@ v
[gentoo-dev] [PATCH] vdr-plugin-2.eclass: add EAPI8 support, EAPI7+ fixes
please review and comment.
I will open a github PR for code changes later.
From d5f6916d7fdd194e0c2ab11eea0d48ecb01d122c Mon Sep 17 00:00:00 2001
From: Martin Dummer
Date: Sun, 27 Jun 2021 17:05:23 +0200
Subject: [PATCH] vdr-plugin-2.eclass: add EAPI8 support, EAPI7+ fixes
- add EAPI8 support, no changes necessary
- add BDEPEND=pkgconfig
- remove outdated version numbers from (R)DEPEND
- fix calling of pkg-config
- add many "|| die" which should have been here already for EAPI7
Signed-off-by: Martin Dummer
---
eclass/vdr-plugin-2.eclass | 53 ++
1 file changed, 31 insertions(+), 22 deletions(-)
diff --git a/eclass/vdr-plugin-2.eclass b/eclass/vdr-plugin-2.eclass
index 3f3acc41208..41e43ff34ee 100644
--- a/eclass/vdr-plugin-2.eclass
+++ b/eclass/vdr-plugin-2.eclass
@@ -76,7 +76,7 @@
inherit flag-o-matic toolchain-funcs unpacker
case ${EAPI:-0} in
- 5|6|7)
+ 5|6|7|8)
;;
*) die "EAPI ${EAPI} unsupported."
;;
@@ -97,10 +97,11 @@ DESCRIPTION="vdr Plugin: ${VDRPLUGIN} (based on
vdr-plugin-2.eclass)"
S="${WORKDIR}/${VDRPLUGIN}-${PV}"
# depend on headers for DVB-driver and vdr-scripts
-DEPEND=">=media-tv/gentoo-vdr-scripts-0.4.2
+BDEPEND="dev-util/pkgconfig"
+DEPEND="media-tv/gentoo-vdr-scripts
virtual/linuxtv-dvb-headers"
-RDEPEND=">=media-tv/gentoo-vdr-scripts-0.4.2
- >=app-eselect/eselect-vdr-0.0.2"
+RDEPEND="media-tv/gentoo-vdr-scripts
+ app-eselect/eselect-vdr"
if [[ "${GENTOO_VDR_CONDITIONAL:-no}" = "yes" ]]; then
IUSE="${IUSE} vdr"
@@ -151,7 +152,7 @@ vdr_create_header_checksum_file() {
local CHKSUM="header-md5-vdr"
if [[ -f ${VDR_CHECKSUM_DIR}/header-md5-vdr ]]; then
- cp "${VDR_CHECKSUM_DIR}/header-md5-vdr" "${CHKSUM}"
+ cp "${VDR_CHECKSUM_DIR}/header-md5-vdr" "${CHKSUM}" ||
die "Could not copy header-md5-vdr"
elif type -p md5sum >/dev/null 2>&1; then
(
cd "${VDR_INCLUDE_DIR}"
@@ -179,7 +180,8 @@ fix_vdr_libsi_include() {
for f; do
sed -i "${f}" \
-e '/#include/s:"\(.*libsi.*\)":<\1>:' \
- -e '/#include/s:<.*\(libsi/.*\)>::'
+ -e '/#include/s:<.*\(libsi/.*\)>::' \
+ || die "sed failed while fixing include of
libsi-headers"
done
}
@@ -192,7 +194,7 @@ fix_vdr_libsi_include() {
vdr_patchmakefile() {
einfo "Patching Makefile"
[[ -e Makefile ]] || die "Makefile of plugin can not be found!"
- cp Makefile "${WORKDIR}"/Makefile.before
+ cp Makefile "${WORKDIR}"/Makefile.before || die "Failed to copy
Makefile"
# plugin makefiles use VDRDIR in strange ways
# assumptions:
@@ -216,14 +218,16 @@ vdr_patchmakefile() {
-e '/VDRINCDIR.*=/!s:$(VDRDIR)/include:$(VDRINCDIR):' \
\
-e 's:-I$(DVBDIR)/include::' \
- -e 's:-I$(DVBDIR)::'
+ -e 's:-I$(DVBDIR)::' \
+ || die "sed failed to set \$VDRDIR"
if ! grep -q APIVERSION Makefile; then
ebegin " Converting to APIVERSION"
sed -i Makefile \
-e 's:^APIVERSION = :APIVERSION ?= :' \
-e
's:$(LIBDIR)/$@.$(VDRVERSION):$(LIBDIR)/$@.$(APIVERSION):' \
- -e '/VDRVERSION =/a\APIVERSION = $(shell sed -ne
'"'"'/define APIVERSION/s/^.*"\\(.*\\)".*$$/\\1/p'"'"' $(VDRDIR)/config.h)'
+ -e '/VDRVERSION =/a\APIVERSION = $(shell sed -ne
'"'"'/define APIVERSION/s/^.*"\\(.*\\)".*$$/\\1/p'"'"'
$(VDRDIR)/config.h)' \
+ || die "sed failed to change APIVERSION"
eend $?
fi
@@ -231,13 +235,15 @@ vdr_patchmakefile() {
# Do not overwrite CXXFLAGS, add LDFLAGS if missing
sed -i Makefile \
-e '/^CXXFLAGS[[:space:]]*=/s/=/?=/' \
- -e '/LDFLAGS/!s:-shared:$(LDFLAGS) -shared:'
+ -e '/LDFLAGS/!s:-shared:$(LDFLAGS) -shared:' \
+ || die "sed failed to fix compile-flags"
# Disabling file stripping, the package manager takes care of it
sed -i Makefile \
-e '/@.*strip/d' \
-e '/strip \$(LIBDIR)\/\$@/d' \
- -e 's/STRIP.*=.*$/STRIP = true/'
+ -e 's/STRIP.*=.*$/STRIP = true/' \
+ || die "sed failed to fix file stripping"
# Use a file instead of a variable as single-stepping via ebuild
Re: [gentoo-dev] Please review changes for vdr-plugin-2.eclass
Am 17.05.21 um 18:11 schrieb Martin Dummer:
> So please take a view on the intended changes to vdr-plugin-2.eclass -
> add documentation to satisfy pkgcheck.
>
> It is placed here:
>
> https://github.com/gentoo/gentoo/pull/20845
>
Due to very little feedback I post here the intended changes to
vdr-plugin-2.eclass on the dev mailing list.
Please take a look. If some devs with github accounts could add a review
or approval on the pull request, this would be nice.
==
commit b19db356742510f880ab1b2d9840043c95cca768
Author: Martin Dummer
Date: 2021-05-17 01:15:33 +0200
vdr-plugin-2.eclass: fix pkgcheck findings in documentation
Add documentation to eclass functions and variables as found by
pkgcheck,
no functional changes.
pkgcheck reported:
EclassDocMissingFunc: vdr-plugin-2: undocumented functions: has_vdr,
vdr-plugin-2_print_enable_command, vdr-plugin-2_src_util,
vdr_create_header_checksum_file, vdr_create_plugindb_file,
vdr_detect_po_dir,
vdr_gettext_missing, vdr_i18n, vdr_linguas_support, vdr_patchmakefile
EclassDocMissingVar: vdr-plugin-2: undocumented variables:
COMMON_DEPEND, VDRPLUGIN
Signed-off-by: Martin Dummer
diff --git a/eclass/vdr-plugin-2.eclass b/eclass/vdr-plugin-2.eclass
index e2aee26c423..187114438df 100644
--- a/eclass/vdr-plugin-2.eclass
+++ b/eclass/vdr-plugin-2.eclass
@@ -14,6 +14,12 @@
# @DESCRIPTION:
# Eclass for easing maintenance of vdr plugin ebuilds
+# @ECLASS-VARIABLE: VDRPLUGIN
+# @INTERNAL
+# @DESCRIPTION:
+# The name of the vdr plugin, plain name without "vdr-" or "plugin"
prefix or suffix.
+# This variable is derived from ${PN}
+
# @ECLASS-VARIABLE: VDR_CONFD_FILE
# @DEFAULT_UNSET
# @DESCRIPTION:
@@ -54,28 +60,6 @@
# PO_SUBDIR="bla foo/bla"
# @CODE
-# @FUNCTION: fix_vdr_libsi_include
-# @DESCRIPTION:
-# Plugins failed on compile with wrong path of libsi includes,
-# this can be fixed by 'function + space separated list of files'
-#
-# Example:
-# @CODE
-# fix_vdr_libsi_include bla.c foo.c
-# @CODE
-
-# @FUNCTION: vdr_remove_i18n_include
-# @DESCRIPTION:
-# Compile will fail if plugin still use the old i18n language handling,
-# most parts are fixed by vdr-plugin-2.eclass internal functions itself.
-# Remove unneeded i18.n includes from files, if they are still wrong there,
-# this can be fixed by 'function + space separated list of files"
-#
-# Example:
-# @CODE
-# vdr_remove_i18n_include bla.n foo.n
-# @CODE
-
# Applying your own local/user patches:
# This is done by using the
# (EAPI = 5) epatch_user() function of the eutils.eclass,
@@ -112,12 +96,10 @@ DESCRIPTION="vdr Plugin: ${VDRPLUGIN} (based on
vdr-plugin-2.eclass)"
# Works in most cases
S="${WORKDIR}/${VDRPLUGIN}-${PV}"
-# depend on headers for DVB-driver
-COMMON_DEPEND=">=media-tv/gentoo-vdr-scripts-0.4.2"
-
-DEPEND="${COMMON_DEPEND}
+# depend on headers for DVB-driver and vdr-scripts
+DEPEND=">=media-tv/gentoo-vdr-scripts-0.4.2
virtual/linuxtv-dvb-headers"
-RDEPEND="${COMMON_DEPEND}
+RDEPEND=">=media-tv/gentoo-vdr-scripts-0.4.2
>=app-eselect/eselect-vdr-0.0.2"
if [[ "${GENTOO_VDR_CONDITIONAL:-no}" = "yes" ]]; then
@@ -126,9 +108,13 @@ if [[ "${GENTOO_VDR_CONDITIONAL:-no}" = "yes" ]]; then
RDEPEND="vdr? ( ${RDEPEND} )"
fi
+# @FUNCTION: vdr_create_plugindb_file
+# @INTERNAL
+# @USAGE: [more plugin names]
+# @DESCRIPTION:
# New method of storing plugindb
-# Called from src_install
-# file maintained by normal portage-methods
+# Called from src_install
+# file maintained by normal portage-methods
vdr_create_plugindb_file() {
local NEW_VDRPLUGINDB_DIR=/usr/share/vdr/vdrplugin-rebuild/
local DB_FILE="${NEW_VDRPLUGINDB_DIR}/${CATEGORY}-${PF}"
@@ -151,6 +137,13 @@ vdr_create_plugindb_file() {
} > "${D%/}/${DB_FILE}"
}
+# @FUNCTION: vdr_create_header_checksum_file
+# @USAGE:
+# @INTERNAL
+# @DESCRIPTION:
+# Create a file with md5 checksums of .h header files of the plugin
being compiled.
+# The checksum files are used to detect if a plugin is compiled against
the current
+# vdr binary.
vdr_create_header_checksum_file() {
# Danger: Not using $ROOT here, as compile will also not use it !!!
# If vdr in $ROOT and / differ, plugins will not run anyway
@@ -175,6 +168,11 @@ vdr_create_header_checksum_file() {
done
}
+# @FUNCTION: fix_vdr_libsi_include
+# @USAGE: [more filenames]
+# @DESCRIPTION:
+# Plugins failed on compile with wrong path of libsi includes,
+# this can be fixed by 'function + space separated list of files'
fix_vdr_libsi_include() {
eqawarn "Fixing include of libsi-headers"
local f
@@ -185,6 +183,12 @@ fix_vdr_libsi_inc
[gentoo-dev] Please review changes for vdr-plugin-2.eclass
Hello Devs, Jörg / "hd_brummy" says every changes on eclasses should be reviewed here before being committed. So please take a view on the intended changes to vdr-plugin-2.eclass - add documentation to satisfy pkgcheck. It is placed here: https://github.com/gentoo/gentoo/pull/20845 Thanks Martin
Re: [gentoo-dev] Packages up for grabs
Hi, I can take > [Bv] net-misc/teamviewer //madmartin
[gentoo-dev] Last-rites: sys-block/rts_pstor
Hello developers, can someone please insert the lines below into the tree? (If I file a github PR, it will probably not able to merge because packages.mask changes too quickly) # Martin Dummer (2020-11-20) # Does not compile with kernels >=5.5, no upstream development # since years, for most hardware the in-kernel module # rtsx_pci_sdmmc should be preferred over this driver. # Bugs #712484 #717184 and #741909. Removal in 30 days. sys-block/rts_pstor Thanks, Martin
Re: [gentoo-dev] [virtualization project] Help needed with libvirt virtualization stack
I can try to keep app-emulation/virt-manager app-emulation/virt-viewer up-to-date. virt-manager-3.0.0 is out since a week. Greetings /martin 0xCE5F3E9B6DE05D12.asc Description: application/pgp-keys signature.asc Description: OpenPGP digital signature
Re: [gentoo-dev] Last rites: next batch of py2 packages
On 2020-09-19 16:36, Azamat Hackimov wrote: > Hello. > > сб, 19 сент. 2020 г. в 15:51, Michał Górny : >> sys-firmware/nvidia-firmware >> > Created PR https://github.com/gentoo/gentoo/pull/17600 > Excellent! Thats an important firmware package, I would have started a PR also in a few minutes! 0xCE5F3E9B6DE05D12.asc Description: application/pgp-keys
Re: [gentoo-dev] Packages up for grabs
On 2020-04-14 10:36, Joonas Niilola wrote: > Hey, > > here's a list of packages recently dropped to maintainer-needed due to > retirement of multiple inactive proxied maintainers. > > (b) = open bugs, > (v) = new version is available. > app-backup/rear (b,v) > Hi, In the meantime I decided to take over "rear", too. Version bump PR is filed, but I cannot find other open bugs than "maintainer needed" ?!? Bye Martin 0xCE5F3E9B6DE05D12.asc Description: application/pgp-keys signature.asc Description: OpenPGP digital signature
Re: [gentoo-dev] Packages up for grabs
Hello Joonas, I will take On 14.04.20 10:36, Joonas Niilola wrote: > app-admin/passwordsafe (b,v) if my abiulities are sufficient CU Martin 0xCE5F3E9B6DE05D12.asc Description: application/pgp-keys signature.asc Description: OpenPGP digital signature
Re: [gentoo-dev] Last rites: dev-util/electron, app-editors/atom
Am 30.03.20 um 08:55 schrieb Michał Górny: > # Michał Górny (2020-03-30) > # Unmaintained. Humongous number of vulnerabilities. Python 2 only. > # Needs really active maintainer. app-editors/atom is the only revdep. > # Removal in 30 days. Bug #699866. > dev-util/electron > app-editors/atom > If these two have to leave, you may last-rite app-eselect/eselect-electron as well Martin
Re: [gentoo-dev] New acct-* package policy
Am 21.12.19 um 18:32 schrieb Michał Górny: > Please commit and push uid-gid.txt [3] change to data/api.git *before* > your packages. This makes sure that any potential collisions are caught > as merge conflicts before they hit users. The CI also catches UID/GID > collisions. One remark from a "proxy-maintainer" view: A proxy-maintainer (like me) cannot commit or add github pull requests for https://gitweb.gentoo.org/data/api.git/tree/files/uid-gid.tx and therefore cannot allocate a valid UID/GID for valid new ebuilds in pull requests. So for proxy-maintained packages which need acct-user/acct-group ebuilds this must be - either done like before - announcement on the gentoo-dev ML, then committed by ??? - or done by a member of the proxy-maintainers group. Maybe you find the time to refine that exception in the documentation. Bye Martin signature.asc Description: OpenPGP digital signature
[gentoo-dev] RFC: UID/GID assignment for www-misc/vdradmin-am (453/453)
Hello gentoo-dev. I would like to ask to assign uid/gid 453 for the package www-misc/vdradmin-am Martin (proxy-maintainer) signature.asc Description: OpenPGP digital signature
[gentoo-dev] RFC: UID/GID assignment for media-video/vdr (452/452)
Hello gentoo-dev. I would like to ask to assign uid/gid 452 for the package media-video/vdr and its many plugins media-plugins/vdr-* Martin (proxy-maintainer) signature.asc Description: OpenPGP digital signature
Re: [gentoo-dev] Re: EAPI 2 must die
Am 06.06.19 um 17:28 schrieb Anthony G. Basile: > Didn't we have some "archive" for old ebuilds? Maybe we can move > it there. What about an overlay for this purpose? Its like in real life they come into life and leave the same way... Despite that, I usually dig in the git repository when I look for deleted files et.al. like described here https://stackoverflow.com/questions/7203515/git-how-to-find-a-deleted-file-in-the-project-commit-history signature.asc Description: OpenPGP digital signature
[gentoo-dev] Question about developer/bug assigned person's mail adress (as seen in Bug #453778)
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi, I have a question: Bug https://bugs.gentoo.org/show_bug.cgi?id=453778 is assigned to Dustin Polke gentoobugsie.20.dsuraw...@spamgourmet.com While the bug is easy to fix and close, no action happens. So I tried to find out whats the meaning of this strange email address. After a visit to https://spamgourmet.com/ this mail address means that only the first 20 emails send to this address are forwarded to the registered user's real mail address. Any further mails are discarded. Now my question: is this a good behaviour for a developer? What's your opinion? Martin -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.19 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iEYEARECAAYFAlFbPnoACgkQrl/4JlwVxHi9xwCg7/03GDd+1Wjf5Ju0PcTnq3ke l4cAoPZYLxhMN4ibzig0SACAvtF2WZ/k =aBHl -END PGP SIGNATURE-
Re: [gentoo-dev] Packages up for grabs
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 If I manage one day to achieve the gentoo dev status then I am willing to pick up maintainership of app-laptop/nvidiabl but until then? -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.19 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iEYEARECAAYFAlFNe0AACgkQrl/4JlwVxHgrpQCgvt/KKljLgGa1WLJJ4zUw4J4X 8XQAoMj9Ovu1a9MA2dunN4y0qS+EGXfE =AS1h -END PGP SIGNATURE-
