[RFC] Patches being used by O.S. Systems and ready to be pushed

[RFC] Patches being used by O.S. Systems and ready to be pushed

[Otavio Salvador]

Hello,

We've been using those patches here and we'd like to push them all and
avoid diverging from OE as possible.

Could people take a look on them and ack/nack each patch? If required
I'll add the needed fixes for push them all and make sure that all
regressions are solved.

Thanks in advance,

[PATCH 1/6] rt61-firmware: add 1.2

[Otavio Salvador]

---
 packages/rt2x00/rt61-firmware_1.2.bb |   14 ++++++++++++++
 1 files changed, 14 insertions(+), 0 deletions(-)
 create mode 100644 packages/rt2x00/rt61-firmware_1.2.bb

diff --git a/packages/rt2x00/rt61-firmware_1.2.bb b/packages/rt2x00/rt61-firmware_1.2.bb
new file mode 100644
index 0000000..ae683c8
--- /dev/null
+++ b/packages/rt2x00/rt61-firmware_1.2.bb
@@ -0,0 +1,14 @@
+DESCRIPTION = "Firmware for rt61 based USB wifi adaptors"
+LICENSE = "unknown"
+
+SRC_URI = "http://www.ralinktech.com.tw/data/RT61_Firmware_V${PV}.zip"
+
+S = "${WORKDIR}/RT61_Firmware_V${PV}"
+
+do_install() {
+	install -d ${D}/${base_libdir}/firmware
+	install -m 0644 rt61.bin ${D}/${base_libdir}/firmware/
+}
+
+FILES_${PN} = "${base_libdir}/firmware/"
+PACKAGE_ARCH = "all"
-- 
1.6.2.rc0

[PATCH 2/6] cups: add 1.3.8.

[Otavio Salvador]

---
 .../cups/cups-1.3.8/use_echo_only_in_init.patch    |   11 +++++++++++
 packages/cups/cups_1.3.8.bb                        |    8 ++++++++
 2 files changed, 19 insertions(+), 0 deletions(-)
 create mode 100644 packages/cups/cups-1.3.8/use_echo_only_in_init.patch
 create mode 100644 packages/cups/cups_1.3.8.bb

diff --git a/packages/cups/cups-1.3.8/use_echo_only_in_init.patch b/packages/cups/cups-1.3.8/use_echo_only_in_init.patch
new file mode 100644
index 0000000..21ff0e5
--- /dev/null
+++ b/packages/cups/cups-1.3.8/use_echo_only_in_init.patch
@@ -0,0 +1,11 @@
+--- a/init/cups.sh.in.orig	2008-10-04 16:50:46.000000000 -0300
++++ b/init/cups.sh.in	2008-10-04 16:51:39.000000000 -0300
+@@ -68,7 +68,7 @@
+ 		ECHO_ERROR=:
+ 		;;
+ 
+-	Linux*)
++	DISABLELinux*)
+ 		IS_ON=/bin/true
+ 		if test -f /etc/init.d/functions; then
+ 			. /etc/init.d/functions
diff --git a/packages/cups/cups_1.3.8.bb b/packages/cups/cups_1.3.8.bb
new file mode 100644
index 0000000..e9fd985
--- /dev/null
+++ b/packages/cups/cups_1.3.8.bb
@@ -0,0 +1,8 @@
+require cups.inc
+
+SRC_URI += "file://use_echo_only_in_init.patch;patch=1"
+PR = "r1"
+
+DEFAULT_PREFERENCE = "-1"
+
+EXTRA_OECONF += " --disable-gssapi "
-- 
1.6.2.rc0

[PATCH 3/6] package.bbclass, package_ipk.bbclass: add dynamic depends on locale packages (ipk specific)

[Otavio Salvador]

---
 classes/package.bbclass     |   10 ----------
 classes/package_ipk.bbclass |    4 ++++
 2 files changed, 4 insertions(+), 10 deletions(-)

diff --git a/classes/package.bbclass b/classes/package.bbclass
index a7cdb5f..5559592 100644
--- a/classes/package.bbclass
+++ b/classes/package.bbclass
@@ -293,16 +293,6 @@ python package_do_split_locales() {
 		bb.data.setVar('DESCRIPTION_' + pkg, '%s translation for %s' % (l, pn), d)
 
 	bb.data.setVar('PACKAGES', ' '.join(packages), d)
-
-	# Disabled by RP 18/06/07
-	# Wildcards aren't supported in debian
-	# They break with ipkg since glibc-locale* will mean that 
-	# glibc-localedata-translit* won't install as a dependency
-	# for some other package which breaks meta-toolchain
-	# Probably breaks since virtual-locale- isn't provided anywhere
-	#rdep = (bb.data.getVar('RDEPENDS_%s' % mainpkg, d, 1) or bb.data.getVar('RDEPENDS', d, 1) or "").split()
-	#rdep.append('%s-locale*' % pn)
-	#bb.data.setVar('RDEPENDS_%s' % mainpkg, ' '.join(rdep), d)
 }
 
 python populate_packages () {
diff --git a/classes/package_ipk.bbclass b/classes/package_ipk.bbclass
index 3a97729..2d28cf8 100644
--- a/classes/package_ipk.bbclass
+++ b/classes/package_ipk.bbclass
@@ -244,6 +244,10 @@ python do_package_ipk () {
 		rprovides = (bb.data.getVar("RPROVIDES", localdata, 1) or "").split()
 		rreplaces = (bb.data.getVar("RREPLACES", localdata, 1) or "").split()
 		rconflicts = (bb.data.getVar("RCONFLICTS", localdata, 1) or "").split()
+
+		if pkgname.find('-locale-') == -1:
+			rdepends.append('%s-locale*' % pkgname)
+
 		if rdepends:
 			ctrlfile.write("Depends: %s\n" % ", ".join(rdepends))
 		if rsuggests:
-- 
1.6.2.rc0

[PATCH 4/6] parted: drop versions up to 1.8.8

[Otavio Salvador]

Versions older then 1.8.8 are not maintained upstream and looks to
have no users inside of OE hence makes no sense to be kept.

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
---
 conf/checksums.ini                                 |   21 ----------
 .../parted/parted-1.8.0/memory_allocation.patch    |   42 --------------------
 packages/parted/parted_1.6.23.bb                   |   12 ------
 packages/parted/parted_1.6.25.1.bb                 |   17 --------
 packages/parted/parted_1.8.0.bb                    |   18 --------
 packages/parted/parted_1.8.2.bb                    |   16 -------
 packages/parted/parted_1.8.7.bb                    |   30 --------------
 7 files changed, 0 insertions(+), 156 deletions(-)
 delete mode 100644 packages/parted/parted-1.8.0/memory_allocation.patch
 delete mode 100644 packages/parted/parted_1.6.23.bb
 delete mode 100644 packages/parted/parted_1.6.25.1.bb
 delete mode 100644 packages/parted/parted_1.8.0.bb
 delete mode 100644 packages/parted/parted_1.8.2.bb
 delete mode 100644 packages/parted/parted_1.8.7.bb

diff --git a/conf/checksums.ini b/conf/checksums.ini
index 9845ff6..4742b61 100644
--- a/conf/checksums.ini
+++ b/conf/checksums.ini
@@ -16610,26 +16610,6 @@ sha256=211a0092cfd50923eaad75c20630c0a1818e56605d4b6fbc10702d5a34ea7107
 md5=129f8623dc4f70188a015e3cbd7eae82
 sha256=74188aff5f2e5b77515a8cf3979199a598d58da8e4f573462b59cbf94cf1c99d
 
-[ftp://ftp.gnu.org/gnu/parted/parted-1.6.23.tar.gz]
-md5=7e46a32def60ea355c193d9225691742
-sha256=5144f413dcce1a0b94263ebf3730f2a2afe6a9557bf261f9d6f48c890423f4c8
-
-[ftp://ftp.gnu.org/gnu/parted/parted-1.6.25.1.tar.gz]
-md5=8ad8b2c8aa865d9d6a43a4ecfe021782
-sha256=b8f151450dc71e2477d44b7f9ebb77a9ccf8d08776b77edb9bddee07885f1b02
-
-[ftp://ftp.gnu.org/gnu/parted/parted-1.8.0.tar.gz]
-md5=52d3e565fc3369d3388a9e02d4b17410
-sha256=6c57feb082c23e52f38366fff4ffa3961e5dc294d4ebb61ffa31462b5ae3681e
-
-[ftp://ftp.gnu.org/gnu/parted/parted-1.8.2.tar.gz]
-md5=cce5bce8fa1d430e1ca3de3722807e01
-sha256=2f67c40e2589e7804213493f24ebf80cd10847b47c95098854db0250511c3ed1
-
-[ftp://ftp.gnu.org/gnu/parted/parted-1.8.7.tar.gz]
-md5=2ec5cb88b146f1a8ccd5b389455503c6
-sha256=bffb732ab47a8dfb528ab2513f6ae82a4f795c119decfe1f96310bf32c4dfc9e
-
 [ftp://ftp.gnu.org/gnu/parted/parted-1.8.8.tar.gz]
 md5=0d494591731082ec57cc18627728124a
 sha256=db59b77b7ef46f00d83eddedde8a4126bffa6bff5fa94a335dd5d30bf72d4286
@@ -25269,4 +25249,3 @@ sha256=f57c4e33eb2cdd87a6c2f01bfa4794340fbe61ea1a1cfc7dac3b6671e1dd22af
 [http://heanet.dl.sourceforge.net/zziplib/zziplib-0.13.49.tar.bz2]
 md5=5f7b88ebb2bcd7e8044328482d079661
 sha256=f57c4e33eb2cdd87a6c2f01bfa4794340fbe61ea1a1cfc7dac3b6671e1dd22af
-
diff --git a/packages/parted/parted-1.8.0/memory_allocation.patch b/packages/parted/parted-1.8.0/memory_allocation.patch
deleted file mode 100644
index ba37096..0000000
--- a/packages/parted/parted-1.8.0/memory_allocation.patch
+++ /dev/null
@@ -1,42 +0,0 @@
---- parted-1.8.0/libparted/exception.c_orig	2006-12-11 19:44:23.000000000 +0000
-+++ parted-1.8.0/libparted/exception.c	2006-12-11 19:47:53.000000000 +0000
-@@ -231,6 +231,9 @@
- 		     PedExceptionOption ex_opts, const char* message, ...)
- {
- 	va_list		arg_list;
-+        int result;
-+	static int size = 1000;
-+
- 
- 	if (ex)
- 		ped_exception_catch ();
-@@ -239,16 +242,22 @@
- 	if (!ex)
- 		goto no_memory;
- 
--	ex->message = (char*) malloc (8192);
--	if (!ex->message)
--		goto no_memory;
--
- 	ex->type = ex_type;
- 	ex->options = ex_opts;
- 
--	va_start (arg_list, message);
--	vsnprintf (ex->message, 8192, message, arg_list);
--	va_end (arg_list);
-+        while (1) {
-+	            ex->message = (char*) malloc (size);
-+		    if (!ex->message)
-+		              goto no_memory;
-+		    va_start (arg_list, message);
-+		    result = vsnprintf (ex->message, size, message, arg_list);
-+		    va_end (arg_list);
-+		    
-+		    if (result > -1 && result < size)
-+		                    break;
-+		    size += 10;
-+		  }
-+	            
- 
- 	return do_throw ();
- 
diff --git a/packages/parted/parted_1.6.23.bb b/packages/parted/parted_1.6.23.bb
deleted file mode 100644
index cb9b08f..0000000
--- a/packages/parted/parted_1.6.23.bb
+++ /dev/null
@@ -1,12 +0,0 @@
-DESCRIPTION = "parted, the GNU partition resizing program"
-HOMEPAGE = "http://www.gnu.org/software/parted/parted.html"
-LICENSE = "GPLv2"
-SECTION = "console/tools"
-DEPENDS = "readline e2fsprogs-libs"
-PR = "r0"
-
-SRC_URI = "${GNU_MIRROR}/parted/parted-${PV}.tar.gz"
-
-EXTRA_OECONF = "--disable-Werror"
-
-inherit autotools
diff --git a/packages/parted/parted_1.6.25.1.bb b/packages/parted/parted_1.6.25.1.bb
deleted file mode 100644
index 8681055..0000000
--- a/packages/parted/parted_1.6.25.1.bb
+++ /dev/null
@@ -1,17 +0,0 @@
-DESCRIPTION = "parted, the GNU partition resizing program"
-HOMEPAGE = "http://www.gnu.org/software/parted/parted.html"
-LICENSE = "GPLv2"
-SECTION = "console/tools"
-DEPENDS = "readline e2fsprogs-libs"
-PR = "r0"
-
-SRC_URI = "${GNU_MIRROR}/parted/parted-${PV}.tar.gz"
-
-EXTRA_OECONF = "--disable-Werror"
-#export CFLAGS = "-mstructure-size-boundary=8"
-
-inherit autotools pkgconfig
-
-do_stage() {
-	autotools_stage_all
-}
diff --git a/packages/parted/parted_1.8.0.bb b/packages/parted/parted_1.8.0.bb
deleted file mode 100644
index bff27a6..0000000
--- a/packages/parted/parted_1.8.0.bb
+++ /dev/null
@@ -1,18 +0,0 @@
-DESCRIPTION = "parted, the GNU partition resizing program"
-HOMEPAGE = "http://www.gnu.org/software/parted/parted.html"
-LICENSE = "GPLv2"
-SECTION = "console/tools"
-DEPENDS = "readline e2fsprogs-libs"
-PR = "r0"
-
-SRC_URI = "${GNU_MIRROR}/parted/parted-${PV}.tar.gz \
-           file://memory_allocation.patch;patch=0"
-
-EXTRA_OECONF = "--disable-Werror"
-#export CFLAGS = "-mstructure-size-boundary=8"
-
-inherit autotools pkgconfig
-
-do_stage() {
-	autotools_stage_all
-}
diff --git a/packages/parted/parted_1.8.2.bb b/packages/parted/parted_1.8.2.bb
deleted file mode 100644
index 1185014..0000000
--- a/packages/parted/parted_1.8.2.bb
+++ /dev/null
@@ -1,16 +0,0 @@
-DESCRIPTION = "parted, the GNU partition resizing program"
-HOMEPAGE = "http://www.gnu.org/software/parted/parted.html"
-LICENSE = "GPLv2"
-SECTION = "console/tools"
-DEPENDS = "readline e2fsprogs-libs"
-PR = "r0"
-
-SRC_URI = "${GNU_MIRROR}/parted/parted-${PV}.tar.gz"
-           
-#EXTRA_OECONF = "--disable-Werror"
-
-inherit autotools pkgconfig
-
-do_stage() {
-	autotools_stage_all
-}
diff --git a/packages/parted/parted_1.8.7.bb b/packages/parted/parted_1.8.7.bb
deleted file mode 100644
index 59e3aea..0000000
--- a/packages/parted/parted_1.8.7.bb
+++ /dev/null
@@ -1,30 +0,0 @@
-DESCRIPTION = "parted, the GNU partition resizing program"
-HOMEPAGE = "http://www.gnu.org/software/parted/parted.html"
-LICENSE = "GPLv2"
-SECTION = "console/tools"
-DEPENDS = "readline e2fsprogs-libs"
-PR = "r2"
-
-SRC_URI = "${GNU_MIRROR}/parted/parted-${PV}.tar.gz \
-           file://syscalls.h \
-           file://syscalls.patch;patch=1 \
-           file://cross-gross-hack.patch;patch=1"
-           
-EXTRA_OECONF = "--disable-Werror"
-
-inherit autotools pkgconfig
-
-do_configure_prepend() {
-	cp ${WORKDIR}/syscalls.h ${S}/libparted/arch/
-}
-
-do_configure() {
-        gnu-configize
-        libtoolize --force
-        autoconf
-        oe_runconf
-}
-
-do_stage() {
-	autotools_stage_all
-}
-- 
1.6.2.rc0

[PATCH 5/6] sun-jre6: add 1.6.0.10

[Otavio Salvador]

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
---
 packages/sun-jre/sun-jre6_1.6.0.10.bb |   89 +++++++++++++++++++++++++++++++++
 1 files changed, 89 insertions(+), 0 deletions(-)
 create mode 100644 packages/sun-jre/sun-jre6_1.6.0.10.bb

diff --git a/packages/sun-jre/sun-jre6_1.6.0.10.bb b/packages/sun-jre/sun-jre6_1.6.0.10.bb
new file mode 100644
index 0000000..d72cfa0
--- /dev/null
+++ b/packages/sun-jre/sun-jre6_1.6.0.10.bb
@@ -0,0 +1,89 @@
+
+DESCRIPTION = "Sun J2SE Runtime Environment"
+
+PR="r0"
+
+LICENSE = "Sun Binary Code License Agreement"
+
+SP = "jdk-6u10-dlj-linux-i586.bin"
+SD = "jdk1.6.0_10"
+S = ${WORKDIR}
+
+SRC_URI = "http://download.java.net/dlj/binaries/jdk-6u10-dlj-linux-i586.bin"
+
+PRIVATE_LIBS = "${@base_read_file('${WORKDIR}/${SD}.shlibs')}"
+
+do_install() {
+	export MORE=10000
+	sh ${WORKDIR}/${SP} --accept-license >/dev/null
+	chmod -R go=u-w ${SD}
+	chmod -R u+w ${SD}
+
+	mkdir -p ${D}${libdir}/sun-java6-${PV}
+	mv ${S}/${SD}/jre/* ${D}${libdir}/sun-java6-${PV}
+
+	mkdir -p ${D}${libdir}/mozilla/plugins
+	cd ${D}${libdir}/mozilla/plugins
+	ln -s ${libdir}/sun-java6-${PV}/plugin/i386/ns7/libjavaplugin_oji.so .
+
+	cd ${D}${libdir}/sun-java6-${PV}
+
+	for p in `find . -name '*.pack'`; do
+		./bin/unpack200 $p $(echo $p | sed 's,.pack$,.jar,g') || exit 1
+		rm $p
+	done
+
+	for i in	bin/ControlPanel	\
+			bin/javaws	\
+			bin/jcontrol	\
+			bin/keytool	\
+			bin/orbd	\
+			bin/pack200	\
+			bin/policytool	\
+			bin/rmid	\
+			bin/rmiregistry	\
+			bin/servertool	\
+			bin/tnameserv	\
+			bin/unpack200
+	do
+		rm $i
+	done
+
+	rm -rf javaws lib/javaws*
+
+	rm -rf plugin/desktop plugin/i386/ns7-gcc29
+
+	rm -f lib/fontconfig.*.bfc lib/fontconfig.*.properties.src
+
+	rm -f lib/javaws.jar lib/ext/ldapsec.jar lib/cmm/PYCC.pf \
+		lib/audio/soundbank.gm
+
+	rm -rf lib/i386/server lib/i386/motif21 lib/i386/client/classes.jsa \
+		lib/i386/libjavaplugin_nscp_gcc29.so \
+		lib/i386/headless
+
+	rm -rf fonts oblique-fonts man
+
+	rm -rf lib/locale lib/zi lib/management
+	rm -rf lib/desktop LICENSE README COPYRIGHT THIRDPARTYLICENSEREADME.txt \
+	    Welcome.html
+
+	# Find out what shared libs we contain and spit it out to a file
+	find . -name "*.so" -exec basename {} \; | sort | uniq \
+						> ${WORKDIR}/${SD}.shlibs
+}
+
+PACKAGES =+ "${PN}-plugin"
+
+FILES_${PN} = "/usr/"
+FILES_${PN}-dbg += "/usr/*/.debug /usr/*/*/.debug /usr/*/*/*/.debug /usr/*/*/*/*/.debug /usr/*/*/*/*/*/.debug"
+FILES_${PN}-plugin = "${libdir}/firefox-3.0.1/plugins/* ${libdir}/sun-java6-${PV}/plugin/i386/ns7/libjavaplugin_oji.so"
+
+RDEPENDS_${PN} += "libx11-locale"
+RDEPENDS_${PN}-plugin += "${PN}"
+
+pkg_postinst_${PN} () {
+	grep -v JAVA_HOME /etc/environment > /etc/environment.new
+	mv /etc/environment.new /etc/environment
+	echo "JAVA_HOME=\"${libdir}/sun-java6-${PV}\"" >> /etc/environment
+}
-- 
1.6.2.rc0

[PATCH 6/6] flash-plugin: add 10.0.12.36

[Otavio Salvador]

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
---
 conf/checksums.ini                              |    5 ++
 packages/adobe-flash/flash-plugin_10.0.12.36.bb |   57 +++++++++++++++++++++++
 2 files changed, 62 insertions(+), 0 deletions(-)
 create mode 100644 packages/adobe-flash/flash-plugin_10.0.12.36.bb

diff --git a/conf/checksums.ini b/conf/checksums.ini
index 4742b61..a4e7ff3 100644
--- a/conf/checksums.ini
+++ b/conf/checksums.ini
@@ -5494,6 +5494,10 @@ sha256=ce4f7d11b3c04a7368c916ca4abc284dd0c0256f461dfb7f07df1ab445e7a5c0
 md5=2bfc127cdda02834d0491ab531a20960
 sha256=ce4f7d11b3c04a7368c916ca4abc284dd0c0256f461dfb7f07df1ab445e7a5c0
 
+[http://fpdownload.macromedia.com/get/flashplayer/current//flash-plugin-10.0.12.36-release.i386.rpm]
+md5=8845d4536fcd3a329690c93bce9bbf8e
+sha256=c7345faf10b269b3556f2e1373c4afc35f08891f8bbd54e45d49151a8c8e4ac0
+
 [http://downloads.sourceforge.net/flex/flex-2.5.31.tar.bz2]
 md5=363dcc4afc917dc51306eb9d3de0152f
 sha256=701353279a17655d78e3b3678ad78d0375f5bf45877ad8b3507d589c42427f26
@@ -25249,3 +25253,4 @@ sha256=f57c4e33eb2cdd87a6c2f01bfa4794340fbe61ea1a1cfc7dac3b6671e1dd22af
 [http://heanet.dl.sourceforge.net/zziplib/zziplib-0.13.49.tar.bz2]
 md5=5f7b88ebb2bcd7e8044328482d079661
 sha256=f57c4e33eb2cdd87a6c2f01bfa4794340fbe61ea1a1cfc7dac3b6671e1dd22af
+
diff --git a/packages/adobe-flash/flash-plugin_10.0.12.36.bb b/packages/adobe-flash/flash-plugin_10.0.12.36.bb
new file mode 100644
index 0000000..3d618f2
--- /dev/null
+++ b/packages/adobe-flash/flash-plugin_10.0.12.36.bb
@@ -0,0 +1,57 @@
+DESCRIPTION = "Adobe Flash 9 plugin for Mozilla/Firefox web browsers"
+LICENSE = "AdobeFlash"
+
+PR="r1"
+
+DEPENDS = "rpm2cpio-native"
+
+ADOBE_MIRROR = "http://fpdownload.macromedia.com/get/flashplayer/current/"
+
+# Adobe provides two versions of flash player packages
+# One "unversioned" download at :
+# ${ADOBE_MIRROR}/install_flash_player_9_linux.tar.gz
+# And the other, versioned, which is encapsulated in RPM
+# We use the versioned one to make sure we can keep track of the version
+# and thus do repeatable builds at the cost of future download failures 
+# when adobe updates.
+
+SRC_URI ="${ADOBE_MIRROR}/flash-plugin-${PV}-release.i386.rpm"
+
+# SRC_URI ="${ADOBE_MIRROR}/install_flash_player_9_linux.tar.gz"
+# S = "${WORKDIR}/install_flash_player_9_linux"
+
+do_configure() {
+	rpm=${WORKDIR}/flash-plugin-${PV}-release.i386.rpm
+	if [ -f "$rpm" ]; then
+		rpm2cpio.pl $rpm | cpio -id
+		find . -type f -exec mv -i {} "`pwd`" \;
+	fi
+}
+
+# Is this necessary ?
+# do_compile () {
+# 	WE_DO_NOT_COMPILE="yes"
+# }
+
+do_install() {
+	mkdir -p ${D}${libdir}/mozilla/plugins/
+	install -m 0755 libflashplayer.so ${D}${libdir}/mozilla/plugins/
+	# AFAIK this is not needed
+	# install -m 0644 flashplayer.xpt ${D}${libdir}/mozilla/plugins/
+}
+
+FILES_${PN} = "${libdir}/* ${libdir}/X11/*"
+
+pkg_postinst() {
+	test -n "$D" && exit 1
+	mkdir -p ${libdir}/X11/fonts/misc
+	cat >> ${libdir}/X11/fonts/misc/fonts.alias <<_PUNT
+-*-*-medium-r-normal-*-10-*              -misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso8859-1
+-*-helvetica-medium-r-normal-*-10-*      -misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso8859-1
+-*-serf-medium-r-normal-*-10-*           -misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso8859-1
+-*-*-*-*-*-*-*-*-*-*-*-*-big5-*          -misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso8859-1
+-*-*-*-*-*-*-*-*-*-*-*-*-gb2312.1980-*   -misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso8859-1
+-*-*-medium-r-normal-*-14-*-*-*-*-*-*-*  -misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso8859-1
+-*-fixed-*--14-*                         -misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso8859-1
+_PUNT
+}
-- 
1.6.2.rc0

Re: [PATCH 1/6] rt61-firmware: add 1.2

[Koen Kooi]

On 12-02-09 17:38, Otavio Salvador wrote:
> ---
>   packages/rt2x00/rt61-firmware_1.2.bb |   14 ++++++++++++++
>   1 files changed, 14 insertions(+), 0 deletions(-)
>   create mode 100644 packages/rt2x00/rt61-firmware_1.2.bb
>
> diff --git a/packages/rt2x00/rt61-firmware_1.2.bb b/packages/rt2x00/rt61-firmware_1.2.bb
> new file mode 100644
> index 0000000..ae683c8
> --- /dev/null
> +++ b/packages/rt2x00/rt61-firmware_1.2.bb
> @@ -0,0 +1,14 @@
> +DESCRIPTION = "Firmware for rt61 based USB wifi adaptors"
> +LICENSE = "unknown"
> +
> +SRC_URI = "http://www.ralinktech.com.tw/data/RT61_Firmware_V${PV}.zip"
> +
> +S = "${WORKDIR}/RT61_Firmware_V${PV}"
> +
> +do_install() {
> +	install -d ${D}/${base_libdir}/firmware
> +	install -m 0644 rt61.bin ${D}/${base_libdir}/firmware/
> +}
> +
> +FILES_${PN} = "${base_libdir}/firmware/"
> +PACKAGE_ARCH = "all"

ACK

Re: [PATCH 2/6] cups: add 1.3.8.

[Koen Kooi]

On 12-02-09 17:38, Otavio Salvador wrote:
> ---
>   .../cups/cups-1.3.8/use_echo_only_in_init.patch    |   11 +++++++++++
>   packages/cups/cups_1.3.8.bb                        |    8 ++++++++
>   2 files changed, 19 insertions(+), 0 deletions(-)
>   create mode 100644 packages/cups/cups-1.3.8/use_echo_only_in_init.patch
>   create mode 100644 packages/cups/cups_1.3.8.bb
>
> diff --git a/packages/cups/cups-1.3.8/use_echo_only_in_init.patch b/packages/cups/cups-1.3.8/use_echo_only_in_init.patch
> new file mode 100644
> index 0000000..21ff0e5
> --- /dev/null
> +++ b/packages/cups/cups-1.3.8/use_echo_only_in_init.patch
> @@ -0,0 +1,11 @@
> +--- a/init/cups.sh.in.orig	2008-10-04 16:50:46.000000000 -0300
> ++++ b/init/cups.sh.in	2008-10-04 16:51:39.000000000 -0300
> +@@ -68,7 +68,7 @@
> + 		ECHO_ERROR=:
> + 		;;
> +
> +-	Linux*)
> ++	DISABLELinux*)
> + 		IS_ON=/bin/true
> + 		if test -f /etc/init.d/functions; then
> + 			. /etc/init.d/functions
> diff --git a/packages/cups/cups_1.3.8.bb b/packages/cups/cups_1.3.8.bb
> new file mode 100644
> index 0000000..e9fd985
> --- /dev/null
> +++ b/packages/cups/cups_1.3.8.bb
> @@ -0,0 +1,8 @@
> +require cups.inc
> +
> +SRC_URI += "file://use_echo_only_in_init.patch;patch=1"
> +PR = "r1"
> +
> +DEFAULT_PREFERENCE = "-1"
> +
> +EXTRA_OECONF += " --disable-gssapi "

ACK

Re: [PATCH 3/6] package.bbclass, package_ipk.bbclass: add dynamic depends on locale packages (ipk specific)

[Koen Kooi]

On 12-02-09 17:38, Otavio Salvador wrote:
> ---
>   classes/package.bbclass     |   10 ----------
>   classes/package_ipk.bbclass |    4 ++++
>   2 files changed, 4 insertions(+), 10 deletions(-)

I'd like a bigger explanation what does wants to do and what the impact 
of it is.

regards,

Koen




>
> diff --git a/classes/package.bbclass b/classes/package.bbclass
> index a7cdb5f..5559592 100644
> --- a/classes/package.bbclass
> +++ b/classes/package.bbclass
> @@ -293,16 +293,6 @@ python package_do_split_locales() {
>   		bb.data.setVar('DESCRIPTION_' + pkg, '%s translation for %s' % (l, pn), d)
>
>   	bb.data.setVar('PACKAGES', ' '.join(packages), d)
> -
> -	# Disabled by RP 18/06/07
> -	# Wildcards aren't supported in debian
> -	# They break with ipkg since glibc-locale* will mean that
> -	# glibc-localedata-translit* won't install as a dependency
> -	# for some other package which breaks meta-toolchain
> -	# Probably breaks since virtual-locale- isn't provided anywhere
> -	#rdep = (bb.data.getVar('RDEPENDS_%s' % mainpkg, d, 1) or bb.data.getVar('RDEPENDS', d, 1) or "").split()
> -	#rdep.append('%s-locale*' % pn)
> -	#bb.data.setVar('RDEPENDS_%s' % mainpkg, ' '.join(rdep), d)
>   }
>
>   python populate_packages () {
> diff --git a/classes/package_ipk.bbclass b/classes/package_ipk.bbclass
> index 3a97729..2d28cf8 100644
> --- a/classes/package_ipk.bbclass
> +++ b/classes/package_ipk.bbclass
> @@ -244,6 +244,10 @@ python do_package_ipk () {
>   		rprovides = (bb.data.getVar("RPROVIDES", localdata, 1) or "").split()
>   		rreplaces = (bb.data.getVar("RREPLACES", localdata, 1) or "").split()
>   		rconflicts = (bb.data.getVar("RCONFLICTS", localdata, 1) or "").split()
> +
> +		if pkgname.find('-locale-') == -1:
> +			rdepends.append('%s-locale*' % pkgname)
> +
>   		if rdepends:
>   			ctrlfile.write("Depends: %s\n" % ", ".join(rdepends))
>   		if rsuggests:

Re: [PATCH 4/6] parted: drop versions up to 1.8.8

[Koen Kooi]

On 12-02-09 17:38, Otavio Salvador wrote:
> Versions older then 1.8.8 are not maintained upstream and looks to
> have no users inside of OE hence makes no sense to be kept.

> --- a/conf/checksums.ini
> +++ b/conf/checksums.ini
> @@ -16610,26 +16610,6 @@ sha256=211a0092cfd50923eaad75c20630c0a1818e56605d4b6fbc10702d5a34ea7107
>   md5=129f8623dc4f70188a015e3cbd7eae82
>   sha256=74188aff5f2e5b77515a8cf3979199a598d58da8e4f573462b59cbf94cf1c99d
>
> -[ftp://ftp.gnu.org/gnu/parted/parted-1.6.23.tar.gz]
> -md5=7e46a32def60ea355c193d9225691742
> -sha256=5144f413dcce1a0b94263ebf3730f2a2afe6a9557bf261f9d6f48c890423f4c8
> -
> -[ftp://ftp.gnu.org/gnu/parted/parted-1.6.25.1.tar.gz]
> -md5=8ad8b2c8aa865d9d6a43a4ecfe021782
> -sha256=b8f151450dc71e2477d44b7f9ebb77a9ccf8d08776b77edb9bddee07885f1b02
> -
> -[ftp://ftp.gnu.org/gnu/parted/parted-1.8.0.tar.gz]
> -md5=52d3e565fc3369d3388a9e02d4b17410
> -sha256=6c57feb082c23e52f38366fff4ffa3961e5dc294d4ebb61ffa31462b5ae3681e
> -
> -[ftp://ftp.gnu.org/gnu/parted/parted-1.8.2.tar.gz]
> -md5=cce5bce8fa1d430e1ca3de3722807e01
> -sha256=2f67c40e2589e7804213493f24ebf80cd10847b47c95098854db0250511c3ed1
> -
> -[ftp://ftp.gnu.org/gnu/parted/parted-1.8.7.tar.gz]
> -md5=2ec5cb88b146f1a8ccd5b389455503c6
> -sha256=bffb732ab47a8dfb528ab2513f6ae82a4f795c119decfe1f96310bf32c4dfc9e

There's no point in deleting things from checksums.ini, they get added 
back once someone uses the .ini in TMPDIR to add checksums.
Otherwise, the patch looks ok.

regards,

Koen

Re: [PATCH 5/6] sun-jre6: add 1.6.0.10

[Koen Kooi]

On 12-02-09 17:38, Otavio Salvador wrote:
> Signed-off-by: Otavio Salvador<otavio@ossystems.com.br>
> ---

NACK:

> +PACKAGES =+ "${PN}-plugin"
> +
> +FILES_${PN} = "/usr/"
> +FILES_${PN}-dbg += "/usr/*/.debug /usr/*/*/.debug /usr/*/*/*/.debug /usr/*/*/*/*/.debug /usr/*/*/*/*/*/.debug"
> +FILES_${PN}-plugin = "${libdir}/firefox-3.0.1/plugins/* ${libdir}/sun-java6-${PV}/plugin/i386/ns7/libjavaplugin_oji.so"

it would be better to install the plugin to /usr/lib/mozilla/plugins and 
have a postinst that symlinks it to ${libdir}/firefox-*, 
${libdir}/fennec-*, ${libdir}/iceweasle when those are present. That way 
other browser implement NPAPI (webkit...) can use the one in 
mozilla/plugins.

regards,

Koen

Re: [PATCH 6/6] flash-plugin: add 10.0.12.36

[Koen Kooi]

On 12-02-09 17:38, Otavio Salvador wrote:
> Signed-off-by: Otavio Salvador<otavio@ossystems.com.br>

> +	cat>>  ${libdir}/X11/fonts/misc/fonts.alias<<_PUNT

Won't that append to the file each time the package gets upgraded?

regards,

Koen

Re: [PATCH 3/6] package.bbclass, package_ipk.bbclass: add dynamic depends on locale packages (ipk specific)

[Phil Blundell]

On Thu, 2009-02-12 at 14:38 -0200, Otavio Salvador wrote:
> +
> +		if pkgname.find('-locale-') == -1:
> +			rdepends.append('%s-locale*' % pkgname)
> +

This looks like it will add locale dependencies to all the subpackages
(e.g. foo-doc, foo-dev and the like) as well as the main "foo" package,
which is a bit different to how the old code used to work.  But I guess
that's harmless enough: the worst effect will be to bloat the Packages
file a bit, and I doubt anybody will care too much.  So, thumbs up from
me.

p.

Re: [PATCH 3/6] package.bbclass, package_ipk.bbclass: add dynamic depends on locale packages (ipk specific)

[Mike (mwester)]

Phil Blundell wrote:
> On Thu, 2009-02-12 at 14:38 -0200, Otavio Salvador wrote:
>> +
>> +		if pkgname.find('-locale-') == -1:
>> +			rdepends.append('%s-locale*' % pkgname)
>> +
> 
> This looks like it will add locale dependencies to all the subpackages
> (e.g. foo-doc, foo-dev and the like) as well as the main "foo" package,
> which is a bit different to how the old code used to work.  But I guess
> that's harmless enough: the worst effect will be to bloat the Packages
> file a bit, and I doubt anybody will care too much.  So, thumbs up from
> me.

How much bloat?  Opkg already trips the OOM killer on the NSLU2 (32MB of
RAM) when it loads the current Packages file and then tries to do
something; we can work around it for now, but if this doubles the size
in-memory, we may simply not have enough RAM to do whatever it is that
opkg does with all that in-memory.

I have grave concerns over anything that adds unnecessary entries that
cannot be turned off.

Mike (mwester)

Re: [PATCH 5/6] sun-jre6: add 1.6.0.10

[Ken Gilmer]

Also, I may be missing something here but isn't this binary-only package
for i586 architecture?  Not seeing in the recipe where the architecture
is constrained.

thx
ken

On Thu, 2009-02-12 at 18:29 +0100, Koen Kooi wrote:
> On 12-02-09 17:38, Otavio Salvador wrote:
> > Signed-off-by: Otavio Salvador<otavio@ossystems.com.br>
> > ---
> 
> NACK:
> 
> > +PACKAGES =+ "${PN}-plugin"
> > +
> > +FILES_${PN} = "/usr/"
> > +FILES_${PN}-dbg += "/usr/*/.debug /usr/*/*/.debug /usr/*/*/*/.debug /usr/*/*/*/*/.debug /usr/*/*/*/*/*/.debug"
> > +FILES_${PN}-plugin = "${libdir}/firefox-3.0.1/plugins/* ${libdir}/sun-java6-${PV}/plugin/i386/ns7/libjavaplugin_oji.so"
> 
> it would be better to install the plugin to /usr/lib/mozilla/plugins and 
> have a postinst that symlinks it to ${libdir}/firefox-*, 
> ${libdir}/fennec-*, ${libdir}/iceweasle when those are present. That way 
> other browser implement NPAPI (webkit...) can use the one in 
> mozilla/plugins.
> 
> regards,
> 
> Koen
> 
> 
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel

Re: [PATCH 6/6] flash-plugin: add 10.0.12.36

[Phil Blundell]

On Thu, 2009-02-12 at 14:38 -0200, Otavio Salvador wrote:
> +pkg_postinst() {
> +	test -n "$D" && exit 1
> +	mkdir -p ${libdir}/X11/fonts/misc
> +	cat >> ${libdir}/X11/fonts/misc/fonts.alias <<_PUNT
> +-*-*-medium-r-normal-*-10-*              -misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso8859-1
> +-*-helvetica-medium-r-normal-*-10-*      -misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso8859-1
> +-*-serf-medium-r-normal-*-10-*           -misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso8859-1
> +-*-*-*-*-*-*-*-*-*-*-*-*-big5-*          -misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso8859-1
> +-*-*-*-*-*-*-*-*-*-*-*-*-gb2312.1980-*   -misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso8859-1
> +-*-*-medium-r-normal-*-14-*-*-*-*-*-*-*  -misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso8859-1
> +-*-fixed-*--14-*                         -misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso8859-1
> +_PUNT
> +}

This seems altogether pretty weird.  I'm not quite sure what you are
trying to accomplish but I am fairly sure that this can't be the right
way to do it :-)

For one thing, the fonts.alias file will be appended to every time the
package is installed (or, at least, configured) but won't be removed
when it is uninstalled.  For another thing, these changes will be wiped
out when font-alias is upgraded or reinstalled.  And, finally, you seem
to be aliasing a whole load of fonts of various different sizes (and
even different encodings!) to the same thing, which doesn't seem like it
can be generally good news.

If this is a workaround for some craving that the flash plugin has for
server-side fonts, and the intention is to allow you to run it without
crashing if you don't have the fonts in question, then it would probably
be better to put those bits in some appropriate package of their own so
that distribution people (and/or end users) can decide what they want to
do about it.  I guess there might be an argument for distributing a
dummy-server-fonts package that just ships one font and aliases
everything else to it for the purpose of keeping legacy applications
happy.

p.

Re: [PATCH 5/6] sun-jre6: add 1.6.0.10

[Koen Kooi]

On 12-02-09 19:39, Ken Gilmer wrote:
> Also, I may be missing something here but isn't this binary-only package
> for i586 architecture?  Not seeing in the recipe where the architecture
> is constrained.

If you use insane.bbclass you'll find out at package QA time :) Sadly 
some people tend to turn off QA :(
If it really is binary only, compatible_host should be set.

regards,

Koen

Re: [PATCH 6/6] flash-plugin: add 10.0.12.36

[Koen Kooi]

On 12-02-09 21:27, Phil Blundell wrote:
> I guess there might be an argument for distributing a
> dummy-server-fonts package that just ships one font and aliases
> everything else to it for the purpose of keeping legacy applications
> happy.

The xorg-minimal-fonts tries to do that.

regards,

Koen

Re: [PATCH 3/6] package.bbclass, package_ipk.bbclass: add dynamic depends on locale packages (ipk specific)

[Koen Kooi]

On 12-02-09 19:38, Mike (mwester) wrote:
> Phil Blundell wrote:
>> On Thu, 2009-02-12 at 14:38 -0200, Otavio Salvador wrote:
>>> +
>>> +		if pkgname.find('-locale-') == -1:
>>> +			rdepends.append('%s-locale*' % pkgname)
>>> +
>> This looks like it will add locale dependencies to all the subpackages
>> (e.g. foo-doc, foo-dev and the like) as well as the main "foo" package,
>> which is a bit different to how the old code used to work.  But I guess
>> that's harmless enough: the worst effect will be to bloat the Packages
>> file a bit, and I doubt anybody will care too much.  So, thumbs up from
>> me.
>
> How much bloat?  Opkg already trips the OOM killer on the NSLU2 (32MB of
> RAM) when it loads the current Packages file and then tries to do
> something; we can work around it for now, but if this doubles the size
> in-memory, we may simply not have enough RAM to do whatever it is that
> opkg does with all that in-memory.
>
> I have grave concerns over anything that adds unnecessary entries that
> cannot be turned off.

You can make them approx 50% smaller by doing:

echo -n "Stripping source lines from Package files"
for i in `find .. -name Packages` ; do grep -v ^Source: $i|gzip 
-c9>$i.gz ;gunzip -c $i.gz>$i ; touch $i.sig ; done

And I have devices with 8MB ram that run opkg, so I echo Mike's comments :)

regards,

Koen

Re: [PATCH 3/6] package.bbclass, package_ipk.bbclass: add dynamic depends on locale packages (ipk specific)

[Phil Blundell]

On Thu, 2009-02-12 at 12:38 -0600, Mike (mwester) wrote:
> Phil Blundell wrote:
> > This looks like it will add locale dependencies to all the subpackages
> > (e.g. foo-doc, foo-dev and the like) as well as the main "foo" package,
> > which is a bit different to how the old code used to work.  But I guess
> > that's harmless enough: the worst effect will be to bloat the Packages
> > file a bit, and I doubt anybody will care too much.  So, thumbs up from
> > me.
> 
> How much bloat? 

For a source package "foo", you'd be talking about adding one copy of
the literal string "foo-locale-*" to the Depends line for each of its
subpackages (foo-doc, foo-dev, foo-dbg and the like).  The exact amount
obviously depends on the length of the package name, but I guess you
could call it about 16 bytes per binary package as a rough
approximation.

Exactly what the real-world impact of that would be will depend on your
distro policy.  If you already strip out the -dev and -doc packages from
the main Packages feed then obviously the impact will be virtually nil
because most of the affected packages won't be there in the first place.
At the other extreme, if your Packages files currently contain
everything from the control data (including Source:, Description: and so
on) then they are probably so huge already that another sixteen bytes
per package is negligible in percentage terms.

If those 16*n bytes did turn out to cause a problem for someone then it
probably wouldn't be that hard to reinstate the original behaviour of
only adding this dependency on the primary binary.

p.

Re: [PATCH 6/6] flash-plugin: add 10.0.12.36

[Phil Blundell]

On Thu, 2009-02-12 at 22:36 +0100, Koen Kooi wrote:
> On 12-02-09 21:27, Phil Blundell wrote:
> > I guess there might be an argument for distributing a
> > dummy-server-fonts package that just ships one font and aliases
> > everything else to it for the purpose of keeping legacy applications
> > happy.
> 
> The xorg-minimal-fonts tries to do that.

Yeah, sort of like xorg-minimal-fonts but more so.  That package seems
to ship rather more than one font, including some slightly
fringe-looking ISO8859-xx encoding variants, giving you about 500k of
data which seems like rather a lot if it's just going to be a
placeholder.  Also, its fonts.alias file is rather more selective than I
was thinking of and I'm not sure it would handle the specific cases that
are mentioned in the flash-plugin package.

p.

Re: [PATCH 5/6] sun-jre6: add 1.6.0.10

[Otavio Salvador]

[-- Attachment #1: Type: text/plain, Size: 1457 bytes --]

Koen Kooi <k.kooi@student.utwente.nl> writes:

> On 12-02-09 17:38, Otavio Salvador wrote:
>> Signed-off-by: Otavio Salvador<otavio@ossystems.com.br>
>> ---
>
> NACK:
>
>> +PACKAGES =+ "${PN}-plugin"
>> +
>> +FILES_${PN} = "/usr/"
>> +FILES_${PN}-dbg += "/usr/*/.debug /usr/*/*/.debug /usr/*/*/*/.debug /usr/*/*/*/*/.debug /usr/*/*/*/*/*/.debug"
>> +FILES_${PN}-plugin = "${libdir}/firefox-3.0.1/plugins/* ${libdir}/sun-java6-${PV}/plugin/i386/ns7/libjavaplugin_oji.so"
>
> it would be better to install the plugin to /usr/lib/mozilla/plugins
> and have a postinst that symlinks it to ${libdir}/firefox-*,
> ${libdir}/fennec-*, ${libdir}/iceweasle when those are present. That
> way other browser implement NPAPI (webkit...) can use the one in
> mozilla/plugins.

I've not added the symbolic links since it is quite easy to make firefox
to use that dir. In fact, it has already been done at
d09c754ec52a3f09f1cad6a34d12b9ef7283bc17 and can be done for others I
think.

I've fixed the style issues and the plugin installation as you've
spot. Please check and ack if it is all OK.



-- 
        O T A V I O    S A L V A D O R
---------------------------------------------
 E-mail: otavio@debian.org      UIN: 5906116
 GNU/Linux User: 239058     GPG ID: 49A5F855
 Home Page: http://otavio.ossystems.com.br
---------------------------------------------
"Microsoft sells you Windows ... Linux gives
 you the whole house."

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0005-sun-jre6-add-1.6.0.10.patch --]
[-- Type: text/x-diff, Size: 3449 bytes --]

From 3fe02f9bc58405b518731064ad7fbe82d0bb0f1b Mon Sep 17 00:00:00 2001
From: Otavio Salvador <otavio@ossystems.com.br>
Date: Tue, 10 Feb 2009 15:53:16 -0200
Subject: [PATCH 5/6] sun-jre6: add 1.6.0.10

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
---
 packages/sun-jre/sun-jre6_1.6.0.10.bb |   71 +++++++++++++++++++++++++++++++++
 1 files changed, 71 insertions(+), 0 deletions(-)
 create mode 100644 packages/sun-jre/sun-jre6_1.6.0.10.bb

diff --git a/packages/sun-jre/sun-jre6_1.6.0.10.bb b/packages/sun-jre/sun-jre6_1.6.0.10.bb
new file mode 100644
index 0000000..f67894c
--- /dev/null
+++ b/packages/sun-jre/sun-jre6_1.6.0.10.bb
@@ -0,0 +1,71 @@
+DESCRIPTION = "Sun J2SE Runtime Environment"
+LICENSE = "Sun Binary Code License Agreement"
+RDEPENDS_${PN} += "libx11-locale"
+RDEPENDS_${PN}-plugin += "${PN}"
+PR = "r1"
+PRIVATE_LIBS = "${@base_read_file('${WORKDIR}/${SD}.shlibs')}"
+
+SRC_URI = "http://download.java.net/dlj/binaries/jdk-6u10-dlj-linux-i586.bin"
+
+SP = "jdk-6u10-dlj-linux-i586.bin"
+SD = "jdk1.6.0_10"
+S = "${WORKDIR}"
+
+do_install() {
+    export MORE=10000
+    sh ${WORKDIR}/${SP} --accept-license >/dev/null
+    chmod -R go=u-w ${SD}
+    chmod -R u+w ${SD}
+    mkdir -p ${D}${libdir}/sun-java6-${PV}
+    mv ${S}/${SD}/jre/* ${D}${libdir}/sun-java6-${PV}
+    mkdir -p ${D}${libdir}/mozilla/plugins
+    cd ${D}${libdir}/mozilla/plugins
+    ln -s ${libdir}/sun-java6-${PV}/plugin/i386/ns7/libjavaplugin_oji.so .
+    cd ${D}${libdir}/sun-java6-${PV}
+    for p in `find . -name '*.pack'`; do
+    ./bin/unpack200 $p $(echo $p | sed 's,.pack$,.jar,g') || exit 1
+    rm $p
+    done
+    for i in bin/ControlPanel \
+             bin/javaws       \
+             bin/jcontrol     \
+             bin/keytool      \
+             bin/orbd         \
+             bin/pack200      \
+             bin/policytool   \
+             bin/rmid         \
+             bin/rmiregistry  \
+             bin/servertool   \
+             bin/tnameserv    \
+             bin/unpack200
+    do
+    rm $i
+    done
+    rm -rf javaws lib/javaws*
+    rm -rf plugin/desktop plugin/i386/ns7-gcc29
+    rm -f lib/fontconfig.*.bfc lib/fontconfig.*.properties.src
+    rm -f lib/javaws.jar lib/ext/ldapsec.jar lib/cmm/PYCC.pf \
+        lib/audio/soundbank.gm
+    rm -rf lib/i386/server lib/i386/motif21 lib/i386/client/classes.jsa \
+        lib/i386/libjavaplugin_nscp_gcc29.so \
+        lib/i386/headless
+    rm -rf fonts oblique-fonts man
+    rm -rf lib/locale lib/zi lib/management
+    rm -rf lib/desktop LICENSE README COPYRIGHT THIRDPARTYLICENSEREADME.txt \
+        Welcome.html
+    # Find out what shared libs we contain and spit it out to a file
+    find . -name "*.so" -exec basename {} \; | sort | uniq > ${WORKDIR}/${SD}.shlibs
+}
+
+PACKAGES =+ "${PN}-plugin"
+FILES_${PN} = "/usr/"
+FILES_${PN}-dbg += "/usr/*/.debug /usr/*/*/.debug /usr/*/*/*/.debug /usr/*/*/*/*/.debug /usr/*/*/*/*/*/.debug"
+FILES_${PN}-plugin = "${libdir}/mozilla/plugins/* ${libdir}/sun-java6-${PV}/plugin/i386/ns7/libjavaplugin_oji.so"
+
+pkg_postinst_${PN} () {
+                grep -v JAVA_HOME /etc/environment > /etc/environment.new
+                mv /etc/environment.new /etc/environment
+                echo "JAVA_HOME=\"${libdir}/sun-java6-${PV}\"" >> /etc/environment
+}
+
+COMPATIBLE_HOST = "i.86.*-linux"
-- 
1.6.2.rc0

Re: [PATCH 6/6] flash-plugin: add 10.0.12.36

[Otavio Salvador]

[-- Attachment #1: Type: text/plain, Size: 799 bytes --]

Koen Kooi <k.kooi@student.utwente.nl> writes:

> On 12-02-09 17:38, Otavio Salvador wrote:
>> Signed-off-by: Otavio Salvador<otavio@ossystems.com.br>
>
>> +	cat>>  ${libdir}/X11/fonts/misc/fonts.alias<<_PUNT
>
> Won't that append to the file each time the package gets upgraded?

From what I can tell it is indeed not required; I've droped it and if I
discover any regression due that I can readd it later.

Please ack of rest looks OK.


-- 
        O T A V I O    S A L V A D O R
---------------------------------------------
 E-mail: otavio@debian.org      UIN: 5906116
 GNU/Linux User: 239058     GPG ID: 49A5F855
 Home Page: http://otavio.ossystems.com.br
---------------------------------------------
"Microsoft sells you Windows ... Linux gives
 you the whole house."

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0006-flash-plugin-add-10.0.12.36.patch --]
[-- Type: text/x-diff, Size: 3687 bytes --]

From 154836cb8881eb56197b8fa1dec956d0408463a9 Mon Sep 17 00:00:00 2001
From: Otavio Salvador <otavio@ossystems.com.br>
Date: Tue, 10 Feb 2009 15:47:04 -0200
Subject: [PATCH 6/6] flash-plugin: add 10.0.12.36

Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
---
 conf/bitbake.conf                               |    1 +
 conf/checksums.ini                              |    5 +++
 packages/adobe-flash/flash-plugin_10.0.12.36.bb |   35 +++++++++++++++++++++++
 3 files changed, 41 insertions(+), 0 deletions(-)
 create mode 100644 packages/adobe-flash/flash-plugin_10.0.12.36.bb

diff --git a/conf/bitbake.conf b/conf/bitbake.conf
index 81d4c93..a86317c 100644
--- a/conf/bitbake.conf
+++ b/conf/bitbake.conf
@@ -494,6 +494,7 @@ FREESMARTPHONE_GIT = "git://git.freesmartphone.org"
 GENTOO_MIRROR = "http://distro.ibiblio.org/pub/linux/distributions/gentoo/distfiles"
 APACHE_MIRROR = "http://www.apache.org/dist"
 KERNELORG_MIRROR = "http://kernel.org"
+ADOBE_MIRROR = "http://fpdownload.macromedia.com/get/flashplayer/current/"
 
 # You can use the mirror of your country to get faster downloads by putting
 #  export DEBIAN_MIRROR = "ftp://ftp.de.debian.org/debian/pool"
diff --git a/conf/checksums.ini b/conf/checksums.ini
index 4742b61..a4e7ff3 100644
--- a/conf/checksums.ini
+++ b/conf/checksums.ini
@@ -5494,6 +5494,10 @@ sha256=ce4f7d11b3c04a7368c916ca4abc284dd0c0256f461dfb7f07df1ab445e7a5c0
 md5=2bfc127cdda02834d0491ab531a20960
 sha256=ce4f7d11b3c04a7368c916ca4abc284dd0c0256f461dfb7f07df1ab445e7a5c0
 
+[http://fpdownload.macromedia.com/get/flashplayer/current//flash-plugin-10.0.12.36-release.i386.rpm]
+md5=8845d4536fcd3a329690c93bce9bbf8e
+sha256=c7345faf10b269b3556f2e1373c4afc35f08891f8bbd54e45d49151a8c8e4ac0
+
 [http://downloads.sourceforge.net/flex/flex-2.5.31.tar.bz2]
 md5=363dcc4afc917dc51306eb9d3de0152f
 sha256=701353279a17655d78e3b3678ad78d0375f5bf45877ad8b3507d589c42427f26
@@ -25249,3 +25253,4 @@ sha256=f57c4e33eb2cdd87a6c2f01bfa4794340fbe61ea1a1cfc7dac3b6671e1dd22af
 [http://heanet.dl.sourceforge.net/zziplib/zziplib-0.13.49.tar.bz2]
 md5=5f7b88ebb2bcd7e8044328482d079661
 sha256=f57c4e33eb2cdd87a6c2f01bfa4794340fbe61ea1a1cfc7dac3b6671e1dd22af
+
diff --git a/packages/adobe-flash/flash-plugin_10.0.12.36.bb b/packages/adobe-flash/flash-plugin_10.0.12.36.bb
new file mode 100644
index 0000000..d6cde60
--- /dev/null
+++ b/packages/adobe-flash/flash-plugin_10.0.12.36.bb
@@ -0,0 +1,35 @@
+# Adobe provides two versions of flash player packages
+#
+# One "unversioned" download at :
+#
+#     ${ADOBE_MIRROR}/install_flash_player_9_linux.tar.gz
+#
+# And the other, versioned, which is encapsulated in RPM.
+#
+# We use the versioned one to make sure we can keep track of the
+# version and thus do repeatable builds at the cost of future download
+# failures when adobe updates.
+DESCRIPTION = "Adobe Flash 9 plugin for Mozilla/Firefox web browsers"
+LICENSE = "AdobeFlash"
+DEPENDS = "rpm2cpio-native"
+PR = "r1"
+
+SRC_URI = "${ADOBE_MIRROR}/flash-plugin-${PV}-release.i386.rpm"
+
+do_configure() {
+        rpm=${WORKDIR}/flash-plugin-${PV}-release.i386.rpm
+        if [ -f "$rpm" ]; then
+                rpm2cpio.pl $rpm | cpio -id
+                find . -type f -exec mv -i {} "`pwd`" \;
+        fi
+}
+do_install() {
+        mkdir -p ${D}${libdir}/mozilla/plugins/
+        install -m 0755 libflashplayer.so ${D}${libdir}/mozilla/plugins/
+        # AFAIK this is not needed
+        # install -m 0644 flashplayer.xpt ${D}${libdir}/mozilla/plugins/
+}
+
+FILES_${PN} = "${libdir}/* ${libdir}/X11/*"
+
+COMPATIBLE_HOST = "i.86.*-linux"
-- 
1.6.2.rc0

Re: [PATCH 4/6] parted: drop versions up to 1.8.8

[Otavio Salvador]

Koen Kooi <k.kooi@student.utwente.nl> writes:

> On 12-02-09 17:38, Otavio Salvador wrote:
>> Versions older then 1.8.8 are not maintained upstream and looks to
>> have no users inside of OE hence makes no sense to be kept.
>
>> --- a/conf/checksums.ini
>> +++ b/conf/checksums.ini
>> @@ -16610,26 +16610,6 @@ sha256=211a0092cfd50923eaad75c20630c0a1818e56605d4b6fbc10702d5a34ea7107
>>   md5=129f8623dc4f70188a015e3cbd7eae82
>>   sha256=74188aff5f2e5b77515a8cf3979199a598d58da8e4f573462b59cbf94cf1c99d
>>
>> -[ftp://ftp.gnu.org/gnu/parted/parted-1.6.23.tar.gz]
>> -md5=7e46a32def60ea355c193d9225691742
>> -sha256=5144f413dcce1a0b94263ebf3730f2a2afe6a9557bf261f9d6f48c890423f4c8
>> -
>> -[ftp://ftp.gnu.org/gnu/parted/parted-1.6.25.1.tar.gz]
>> -md5=8ad8b2c8aa865d9d6a43a4ecfe021782
>> -sha256=b8f151450dc71e2477d44b7f9ebb77a9ccf8d08776b77edb9bddee07885f1b02
>> -
>> -[ftp://ftp.gnu.org/gnu/parted/parted-1.8.0.tar.gz]
>> -md5=52d3e565fc3369d3388a9e02d4b17410
>> -sha256=6c57feb082c23e52f38366fff4ffa3961e5dc294d4ebb61ffa31462b5ae3681e
>> -
>> -[ftp://ftp.gnu.org/gnu/parted/parted-1.8.2.tar.gz]
>> -md5=cce5bce8fa1d430e1ca3de3722807e01
>> -sha256=2f67c40e2589e7804213493f24ebf80cd10847b47c95098854db0250511c3ed1
>> -
>> -[ftp://ftp.gnu.org/gnu/parted/parted-1.8.7.tar.gz]
>> -md5=2ec5cb88b146f1a8ccd5b389455503c6
>> -sha256=bffb732ab47a8dfb528ab2513f6ae82a4f795c119decfe1f96310bf32c4dfc9e
>
> There's no point in deleting things from checksums.ini, they get added
> back once someone uses the .ini in TMPDIR to add checksums.
> Otherwise, the patch looks ok.

I think people are suppose to clean the checksums.ini before adding the
missing ones instead of blindly adding all them. Besides, keeping it in
ini just makes it bigger so I'd prefer to drop them and NACK the patches
that add them back.

-- 
        O T A V I O    S A L V A D O R
---------------------------------------------
 E-mail: otavio@debian.org      UIN: 5906116
 GNU/Linux User: 239058     GPG ID: 49A5F855
 Home Page: http://otavio.ossystems.com.br
---------------------------------------------
"Microsoft sells you Windows ... Linux gives
 you the whole house."

Re: [PATCH 3/6] package.bbclass, package_ipk.bbclass: add dynamic depends on locale packages (ipk specific)

[Otavio Salvador]

Koen Kooi <k.kooi@student.utwente.nl> writes:

> On 12-02-09 17:38, Otavio Salvador wrote:
>> ---
>>   classes/package.bbclass     |   10 ----------
>>   classes/package_ipk.bbclass |    4 ++++
>>   2 files changed, 4 insertions(+), 10 deletions(-)
>
> I'd like a bigger explanation what does wants to do and what the
> impact of it is.

It adds dependencies on locales packages for the included languages.

-- 
        O T A V I O    S A L V A D O R
---------------------------------------------
 E-mail: otavio@debian.org      UIN: 5906116
 GNU/Linux User: 239058     GPG ID: 49A5F855
 Home Page: http://otavio.ossystems.com.br
---------------------------------------------
"Microsoft sells you Windows ... Linux gives
 you the whole house."

Re: [PATCH 3/6] package.bbclass, package_ipk.bbclass: add dynamic depends on locale packages (ipk specific)

[Otavio Salvador]

"Mike (mwester)" <mwester@dls.net> writes:

>> This looks like it will add locale dependencies to all the subpackages
>> (e.g. foo-doc, foo-dev and the like) as well as the main "foo" package,
>> which is a bit different to how the old code used to work.  But I guess
>> that's harmless enough: the worst effect will be to bloat the Packages
>> file a bit, and I doubt anybody will care too much.  So, thumbs up from
>> me.
>
> How much bloat?  Opkg already trips the OOM killer on the NSLU2 (32MB of
> RAM) when it loads the current Packages file and then tries to do
> something; we can work around it for now, but if this doubles the size
> in-memory, we may simply not have enough RAM to do whatever it is that
> opkg does with all that in-memory.
>
> I have grave concerns over anything that adds unnecessary entries that
> cannot be turned off.

Personally I don't think any bloat needs to be pushed and with that in
mind I think we could then rework it to be enabled by default but
disabled using a variable.

What people think? I'd also be interested in a BitBake expert comment
about it all.

-- 
        O T A V I O    S A L V A D O R
---------------------------------------------
 E-mail: otavio@debian.org      UIN: 5906116
 GNU/Linux User: 239058     GPG ID: 49A5F855
 Home Page: http://otavio.ossystems.com.br
---------------------------------------------
"Microsoft sells you Windows ... Linux gives
 you the whole house."

Re: [PATCH 3/6] package.bbclass, package_ipk.bbclass: add dynamic depends on locale packages (ipk specific)

[Otavio Salvador]

Phil Blundell <pb@reciva.com> writes:

> On Thu, 2009-02-12 at 12:38 -0600, Mike (mwester) wrote:
>> Phil Blundell wrote:
>> > This looks like it will add locale dependencies to all the subpackages
>> > (e.g. foo-doc, foo-dev and the like) as well as the main "foo" package,
>> > which is a bit different to how the old code used to work.  But I guess
>> > that's harmless enough: the worst effect will be to bloat the Packages
>> > file a bit, and I doubt anybody will care too much.  So, thumbs up from
>> > me.
>> 
>> How much bloat? 
>
> For a source package "foo", you'd be talking about adding one copy of
> the literal string "foo-locale-*" to the Depends line for each of its
> subpackages (foo-doc, foo-dev, foo-dbg and the like).  The exact amount
> obviously depends on the length of the package name, but I guess you
> could call it about 16 bytes per binary package as a rough
> approximation.
>
> Exactly what the real-world impact of that would be will depend on your
> distro policy.  If you already strip out the -dev and -doc packages from
> the main Packages feed then obviously the impact will be virtually nil
> because most of the affected packages won't be there in the first place.
> At the other extreme, if your Packages files currently contain
> everything from the control data (including Source:, Description: and so
> on) then they are probably so huge already that another sixteen bytes
> per package is negligible in percentage terms.
>
> If those 16*n bytes did turn out to cause a problem for someone then it
> probably wouldn't be that hard to reinstate the original behaviour of
> only adding this dependency on the primary binary.

Now I have two good points and I'm uncertain where to go :-)

Push it and sort the regressions out depending how hard is it or make it
optional?

-- 
        O T A V I O    S A L V A D O R
---------------------------------------------
 E-mail: otavio@debian.org      UIN: 5906116
 GNU/Linux User: 239058     GPG ID: 49A5F855
 Home Page: http://otavio.ossystems.com.br
---------------------------------------------
"Microsoft sells you Windows ... Linux gives
 you the whole house."

Re: [PATCH 4/6] parted: drop versions up to 1.8.8

[Koen Kooi]

On 13-02-09 00:51, Otavio Salvador wrote:

> I think people are suppose to clean the checksums.ini before adding the
> missing ones instead of blindly adding all them. Besides, keeping it in
> ini just makes it bigger so I'd prefer to drop them and NACK the patches
> that add them back.

They get added back by the process we recommend, so NACKing patches on 
that is bogus. And besides, people might have the 'old' stuff in 
overlays and you can't put checksums.ini in an overlay.

regards,

Koen

Re: [PATCH 3/6] package.bbclass, package_ipk.bbclass: add dynamic depends on locale packages (ipk specific)

[Koen Kooi]

On 12-02-09 17:38, Otavio Salvador wrote:

> +
> +		if pkgname.find('-locale-') == -1:
	if not '-locale-' and not '-dbg' and not '-dev' in pkgname


regards,

Koen

Re: [PATCH 6/6] flash-plugin: add 10.0.12.36

[Koen Kooi]

On 13-02-09 00:49, Otavio Salvador wrote:
> Koen Kooi<k.kooi@student.utwente.nl>  writes:
>
>> On 12-02-09 17:38, Otavio Salvador wrote:
>>> Signed-off-by: Otavio Salvador<otavio@ossystems.com.br>
>>> +	cat>>   ${libdir}/X11/fonts/misc/fonts.alias<<_PUNT
>> Won't that append to the file each time the package gets upgraded?
>
>  From what I can tell it is indeed not required; I've droped it and if I
> discover any regression due that I can readd it later.
>
> Please ack of rest looks OK.

ACK

Re: [PATCH 5/6] sun-jre6: add 1.6.0.10

[Koen Kooi]

On 13-02-09 00:48, Otavio Salvador wrote:
> Koen Kooi<k.kooi@student.utwente.nl>  writes:
>
>> On 12-02-09 17:38, Otavio Salvador wrote:
>>> Signed-off-by: Otavio Salvador<otavio@ossystems.com.br>
>>> ---
>> NACK:
>>
>>> +PACKAGES =+ "${PN}-plugin"
>>> +
>>> +FILES_${PN} = "/usr/"
>>> +FILES_${PN}-dbg += "/usr/*/.debug /usr/*/*/.debug /usr/*/*/*/.debug /usr/*/*/*/*/.debug /usr/*/*/*/*/*/.debug"
>>> +FILES_${PN}-plugin = "${libdir}/firefox-3.0.1/plugins/* ${libdir}/sun-java6-${PV}/plugin/i386/ns7/libjavaplugin_oji.so"
>> it would be better to install the plugin to /usr/lib/mozilla/plugins
>> and have a postinst that symlinks it to ${libdir}/firefox-*,
>> ${libdir}/fennec-*, ${libdir}/iceweasle when those are present. That
>> way other browser implement NPAPI (webkit...) can use the one in
>> mozilla/plugins.
>
> I've not added the symbolic links since it is quite easy to make firefox
> to use that dir. In fact, it has already been done at
> d09c754ec52a3f09f1cad6a34d12b9ef7283bc17 and can be done for others I
> think.

Please do :)

> I've fixed the style issues and the plugin installation as you've
> spot. Please check and ack if it is all OK.

ACK

Re: [PATCH 4/6] parted: drop versions up to 1.8.8

[Philip Balister]

[-- Attachment #1: Type: text/plain, Size: 775 bytes --]

Koen Kooi wrote:
> On 13-02-09 00:51, Otavio Salvador wrote:
> 
>> I think people are suppose to clean the checksums.ini before adding the
>> missing ones instead of blindly adding all them. Besides, keeping it in
>> ini just makes it bigger so I'd prefer to drop them and NACK the patches
>> that add them back.
> 
> They get added back by the process we recommend, so NACKing patches on 
> that is bogus. And besides, people might have the 'old' stuff in 
> overlays and you can't put checksums.ini in an overlay.

Just leave the old checksums in. After recent issues highlighted on the 
list about deleting stuff, the more stuff we do not delete the better.

I'm thinking about suggesting we need 2 or 3 people to agree on recipe 
deletion ....

Philip

[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/x-pkcs7-signature, Size: 3303 bytes --]

Re: [PATCH 3/6] package.bbclass, package_ipk.bbclass: add dynamic depends on locale packages (ipk specific)

[Tom Rini]

On Thu, Feb 12, 2009 at 11:36:50PM +0100, Koen Kooi wrote:
> On 12-02-09 19:38, Mike (mwester) wrote:
>> Phil Blundell wrote:
>>> On Thu, 2009-02-12 at 14:38 -0200, Otavio Salvador wrote:
>>>> +
>>>> +		if pkgname.find('-locale-') == -1:
>>>> +			rdepends.append('%s-locale*' % pkgname)
>>>> +
>>> This looks like it will add locale dependencies to all the subpackages
>>> (e.g. foo-doc, foo-dev and the like) as well as the main "foo" package,
>>> which is a bit different to how the old code used to work.  But I guess
>>> that's harmless enough: the worst effect will be to bloat the Packages
>>> file a bit, and I doubt anybody will care too much.  So, thumbs up from
>>> me.
>>
>> How much bloat?  Opkg already trips the OOM killer on the NSLU2 (32MB of
>> RAM) when it loads the current Packages file and then tries to do
>> something; we can work around it for now, but if this doubles the size
>> in-memory, we may simply not have enough RAM to do whatever it is that
>> opkg does with all that in-memory.
>>
>> I have grave concerns over anything that adds unnecessary entries that
>> cannot be turned off.
>
> You can make them approx 50% smaller by doing:
>
> echo -n "Stripping source lines from Package files"
> for i in `find .. -name Packages` ; do grep -v ^Source: $i|gzip  
> -c9>$i.gz ;gunzip -c $i.gz>$i ; touch $i.sig ; done

How hard would it be to make that happen more easily, with the normal
tools?  Perhaps some flag to generate just the minimum information?

-- 
Tom Rini

Re: [PATCH 3/6] package.bbclass, package_ipk.bbclass: add dynamic depends on locale packages (ipk specific)

[Koen Kooi]

On 13-02-09 04:29, Tom Rini wrote:
> On Thu, Feb 12, 2009 at 11:36:50PM +0100, Koen Kooi wrote:
>> On 12-02-09 19:38, Mike (mwester) wrote:

>>> I have grave concerns over anything that adds unnecessary entries that
>>> cannot be turned off.
>> You can make them approx 50% smaller by doing:
>>
>> echo -n "Stripping source lines from Package files"
>> for i in `find .. -name Packages` ; do grep -v ^Source: $i|gzip
>> -c9>$i.gz ;gunzip -c $i.gz>$i ; touch $i.sig ; done
>
> How hard would it be to make that happen more easily, with the normal
> tools?  Perhaps some flag to generate just the minimum information?

The problem, for me, is that I want a packages file will all the 
information, put that in an sqlite DB for the feedbrowser and strips 
things out. The feedbrowser only knows about http urls, so access to the 
ipk files itself isn't possible.
I suspect there's an easy solution for that, but I haven't given it a 
lot of thought :)

regards,

Koen