<?xml version="1.0" encoding="UTF-8" standalone="yes" ?>
<!DOCTYPE bugzilla SYSTEM "https://bugs.webkit.org/page.cgi?id=bugzilla.dtd">

<bugzilla version="5.0.4.1"
          urlbase="https://bugs.webkit.org/"
          
          maintainer="admin@webkit.org"
>

    <bug>
          <bug_id>158417</bug_id>
          
          <creation_ts>2016-06-06 05:20:08 -0700</creation_ts>
          <short_desc>REGRESSION(r201449) [GTK] ARMv7 build fails with libicudata.so.55: cannot open shared object file on gtkdoc-scangobj step</short_desc>
          <delta_ts>2016-06-06 23:49:08 -0700</delta_ts>
          <reporter_accessible>1</reporter_accessible>
          <cclist_accessible>1</cclist_accessible>
          <classification_id>1</classification_id>
          <classification>Unclassified</classification>
          <product>WebKit</product>
          <component>Tools / Tests</component>
          <version>WebKit Nightly Build</version>
          <rep_platform>Unspecified</rep_platform>
          <op_sys>Unspecified</op_sys>
          <bug_status>RESOLVED</bug_status>
          <resolution>FIXED</resolution>
          
          
          <bug_file_loc></bug_file_loc>
          <status_whiteboard></status_whiteboard>
          <keywords></keywords>
          <priority>P2</priority>
          <bug_severity>Normal</bug_severity>
          <target_milestone>---</target_milestone>
          
          <blocked>154530</blocked>
          <everconfirmed>1</everconfirmed>
          <reporter name="Carlos Alberto Lopez Perez">clopez</reporter>
          <assigned_to name="Carlos Alberto Lopez Perez">clopez</assigned_to>
          <cc>bugs-noreply</cc>
    
    <cc>cfleizach</cc>
    
    <cc>cgarcia</cc>
    
    <cc>commit-queue</cc>
    
    <cc>darin</cc>
    
    <cc>gustavo</cc>
    
    <cc>jdiggs</cc>
    
    <cc>lforschler</cc>
    
    <cc>mcatanzaro</cc>
    
    <cc>ossy</cc>
          

      

      

      

          <comment_sort_order>oldest_to_newest</comment_sort_order>  
          <long_desc isprivate="0" >
    <commentid>1199541</commentid>
    <comment_count>0</comment_count>
    <who name="Carlos Alberto Lopez Perez">clopez</who>
    <bug_when>2016-06-06 05:20:08 -0700</bug_when>
    <thetext>After r201449 &lt;http://trac.webkit.org/r201449&gt; The GTK+ ARM buildbot is failing to build with the following output:

[14/293] Generating ../docs-build-no-html.stamp[K
FAILED: cd /webkit/gtk-linux-arm-release/build/WebKitBuild/Release &amp;&amp; CC=/usr/lib/ccache/cc CFLAGS=-Wno-error\  /webkit/gtk-linux-arm-release/build/Tools/gtk/generate-gtkdoc --skip-html &amp;&amp; touch docs-build-no-html.stamp
Copying template files to output directory...
Running gtkdoc-scan
Running gtkdoc-scangobj
./webkitdomgtk-4.0-scan: error while loading shared libraries: libicudata.so.55: cannot open shared object file: No such file or directory
Scan failed: 

Generating webkitdomgtk-4.0 documentation...
Traceback (most recent call last):
  File &quot;/webkit/gtk-linux-arm-release/build/Tools/gtk/generate-gtkdoc&quot;, line 201, in &lt;module&gt;
    saw_warnings = generate_documentation(webkitdom_generator)
  File &quot;/webkit/gtk-linux-arm-release/build/Tools/gtk/generate-gtkdoc&quot;, line 148, in generate_documentation
    return generate_doc(generator, arguments.skip_html)
  File &quot;/webkit/gtk-linux-arm-release/build/Tools/gtk/generate-gtkdoc&quot;, line 134, in generate_doc
    generator.generate(not skip_html)
  File &quot;/webkit/gtk-linux-arm-release/build/Tools/gtk/gtkdoc.py&quot;, line 143, in generate
    self._run_gtkdoc_scangobj()
  File &quot;/webkit/gtk-linux-arm-release/build/Tools/gtk/gtkdoc.py&quot;, line 338, in _run_gtkdoc_scangobj
    env=env, cwd=self.output_dir)
  File &quot;/webkit/gtk-linux-arm-release/build/Tools/gtk/gtkdoc.py&quot;, line 209, in _run_command
    % (args[0], process.returncode))
Exception: gtkdoc-scangobj produced a non-zero return code 127

https://build.webkit.org/builders/GTK%20Linux%20ARM%20Release/builds/11341/steps/compile-webkit/logs/stdio</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1199542</commentid>
    <comment_count>1</comment_count>
    <who name="Carlos Alberto Lopez Perez">clopez</who>
    <bug_when>2016-06-06 05:25:28 -0700</bug_when>
    <thetext>I can reproduce the problem on a wandboard running Debian 8 Jessie.

The issue is related to this:

1) wandboard ~/WebKit $ file WebKitBuild/Release/Documentation/webkitdomgtk-4.0/webkitdomgtk-4.0-scan
WebKitBuild/Release/Documentation/webkitdomgtk-4.0/webkitdomgtk-4.0-scan: ELF 32-bit LSB executable, ARM, EABI5 version 1 (SYSV), dynamically linked, interpreter /lib/ld-linux-armhf.so.3, for GNU/Linux 2.6.32, BuildID[sha1]=96abf72688893997831d15db58f8c814c4ce91e0, not stripped


2) wandboard ~/WebKit $ ldd WebKitBuild/Release/Documentation/webkitdomgtk-4.0/webkitdomgtk-4.0-scan | grep icu
	libicui18n.so.52 =&gt; /usr/lib/arm-linux-gnueabihf/libicui18n.so.52 (0xb47ab000)
	libicuuc.so.52 =&gt; /usr/lib/arm-linux-gnueabihf/libicuuc.so.52 (0xb467c000)
	libharfbuzz-icu.so.0 =&gt; /usr/lib/arm-linux-gnueabihf/libharfbuzz-icu.so.0 (0xb420e000)
	libicudata.so.52 =&gt; /usr/lib/arm-linux-gnueabihf/libicudata.so.52 (0xb23b3000)


3) wandboard ~/WebKit $ LD_LIBRARY_PATH=$(pwd)/WebKitBuild/DependenciesGTK/Root/lib ldd WebKitBuild/Release/Documentation/webkitdomgtk-4.0/webkitdomgtk-4.0-scan | grep icu
	libicui18n.so.52 =&gt; /usr/lib/arm-linux-gnueabihf/libicui18n.so.52 (0xb4688000)
	libicuuc.so.52 =&gt; /usr/lib/arm-linux-gnueabihf/libicuuc.so.52 (0xb4559000)
	libharfbuzz-icu.so.0 =&gt; /home/igalia/WebKit/WebKitBuild/DependenciesGTK/Root/lib/libharfbuzz-icu.so.0 (0xb40bd000)
	libicudata.so.52 =&gt; /usr/lib/arm-linux-gnueabihf/libicudata.so.52 (0xb2082000)
	libicuuc.so.55 =&gt; /home/igalia/WebKit/WebKitBuild/DependenciesGTK/Root/lib/libicuuc.so.55 (0xb1d45000)
	libicudata.so.55 =&gt; not found
	libicudata.so.55 =&gt; not found


When the ldd command is run inside the JHBuild environment (LD_LIBRARY_PATH env variable) the linker is not able to resolve the library dependencies.

My wild guess is that maybe webkitdomgtk-4.0-scan has been built outside of the JHBuild because it is linking against libicudata.so.52 instead of libicudata.so.55 ?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1199543</commentid>
    <comment_count>2</comment_count>
    <who name="Carlos Alberto Lopez Perez">clopez</who>
    <bug_when>2016-06-06 05:32:17 -0700</bug_when>
    <thetext>The libraries seem all there:

wandboard ~/WebKit $ ls -ln WebKitBuild/DependenciesGTK/Root/lib/*icu*.*
lrwxrwxrwx 1 1001 1001       26 Jun  6 10:29 WebKitBuild/DependenciesGTK/Root/lib/libharfbuzz-icu.so -&gt; libharfbuzz-icu.so.0.935.0
lrwxrwxrwx 1 1001 1001       26 Jun  6 10:29 WebKitBuild/DependenciesGTK/Root/lib/libharfbuzz-icu.so.0 -&gt; libharfbuzz-icu.so.0.935.0
-rwxr-xr-x 1 1001 1001     9344 Jun  6 10:29 WebKitBuild/DependenciesGTK/Root/lib/libharfbuzz-icu.so.0.935.0
lrwxrwxrwx 1 1001 1001       18 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicudata.so -&gt; libicudata.so.55.1
lrwxrwxrwx 1 1001 1001       18 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicudata.so.55 -&gt; libicudata.so.55.1
-rwxr-xr-x 1 1001 1001 25910236 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicudata.so.55.1
lrwxrwxrwx 1 1001 1001       18 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicui18n.so -&gt; libicui18n.so.55.1
lrwxrwxrwx 1 1001 1001       18 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicui18n.so.55 -&gt; libicui18n.so.55.1
-rwxr-xr-x 1 1001 1001  3935528 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicui18n.so.55.1
lrwxrwxrwx 1 1001 1001       16 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicuio.so -&gt; libicuio.so.55.1
lrwxrwxrwx 1 1001 1001       16 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicuio.so.55 -&gt; libicuio.so.55.1
-rwxr-xr-x 1 1001 1001    83668 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicuio.so.55.1
lrwxrwxrwx 1 1001 1001       16 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicule.so -&gt; libicule.so.55.1
lrwxrwxrwx 1 1001 1001       16 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicule.so.55 -&gt; libicule.so.55.1
-rwxr-xr-x 1 1001 1001   696304 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicule.so.55.1
lrwxrwxrwx 1 1001 1001       16 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libiculx.so -&gt; libiculx.so.55.1
lrwxrwxrwx 1 1001 1001       16 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libiculx.so.55 -&gt; libiculx.so.55.1
-rwxr-xr-x 1 1001 1001    79888 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libiculx.so.55.1
lrwxrwxrwx 1 1001 1001       18 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicutest.so -&gt; libicutest.so.55.1
lrwxrwxrwx 1 1001 1001       18 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicutest.so.55 -&gt; libicutest.so.55.1
-rwxr-xr-x 1 1001 1001    96556 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicutest.so.55.1
lrwxrwxrwx 1 1001 1001       16 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicutu.so -&gt; libicutu.so.55.1
lrwxrwxrwx 1 1001 1001       16 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicutu.so.55 -&gt; libicutu.so.55.1
-rwxr-xr-x 1 1001 1001   404264 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicutu.so.55.1
lrwxrwxrwx 1 1001 1001       16 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicuuc.so -&gt; libicuuc.so.55.1
lrwxrwxrwx 1 1001 1001       16 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicuuc.so.55 -&gt; libicuuc.so.55.1
-rwxr-xr-x 1 1001 1001  2446324 Jun  6 10:27 WebKitBuild/DependenciesGTK/Root/lib/libicuuc.so.55.1</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1199563</commentid>
    <comment_count>3</comment_count>
    <who name="Carlos Alberto Lopez Perez">clopez</who>
    <bug_when>2016-06-06 08:51:56 -0700</bug_when>
    <thetext>After some tests I was able to reduce the problem to this:

1) Save this C code example http://stuf.ro/reading-a-utf-8-file-in-c-with-icu as testicu.c
2) Compile it as follows:
$ gcc $(pkg-config --cflags --libs icu-io) testicu.c -o testicu


If you do this on the standard environment (outside of JHBuild) everything works as expected.

However, if you do that inside the internal JHBuild shell then everything goes wrong.

----

This is what happens outside of JHBuild:

$ echo &quot;Hello world!&quot; &gt; utf8_test.txt
$ gcc $(pkg-config --cflags --libs icu-io) testicu.c -o testicu

$ ldd testicu
	libicuio.so.52 =&gt; /usr/lib/arm-linux-gnueabihf/libicuio.so.52 (0xb6eb7000)
	libicui18n.so.52 =&gt; /usr/lib/arm-linux-gnueabihf/libicui18n.so.52 (0xb6d5f000)
	libicuuc.so.52 =&gt; /usr/lib/arm-linux-gnueabihf/libicuuc.so.52 (0xb6c53000)
	libicudata.so.52 =&gt; /usr/lib/arm-linux-gnueabihf/libicudata.so.52 (0xb55d7000)
	libc.so.6 =&gt; /lib/arm-linux-gnueabihf/libc.so.6 (0xb54e7000)
	libdl.so.2 =&gt; /lib/arm-linux-gnueabihf/libdl.so.2 (0xb54d4000)
	libstdc++.so.6 =&gt; /usr/lib/arm-linux-gnueabihf/libstdc++.so.6 (0xb541d000)
	libm.so.6 =&gt; /lib/arm-linux-gnueabihf/libm.so.6 (0xb53a9000)
	libgcc_s.so.1 =&gt; /lib/arm-linux-gnueabihf/libgcc_s.so.1 (0xb5380000)
	/lib/ld-linux-armhf.so.3 (0xb6ee4000)

$ ./testicu 
String size: 13

Hello world!

$ pkg-config --cflags --libs icu-io
-I/usr/include/arm-linux-gnueabihf -licuio -licui18n -licuuc -licudata 


-----------

This is what happens inside of JHBuild:

$ Tools/jhbuild/jhbuild-wrapper --gtk shell
$ echo &quot;Hello world!&quot; &gt; utf8_test.txt
$ gcc $(pkg-config --cflags --libs icu-io) testicu.c -o testicu

$ ldd testicu
	libicuio.so.55 =&gt; /home/igalia/WebKit/WebKitBuild/DependenciesGTK/Root/lib/libicuio.so.55 (0xb6f49000)
	libicui18n.so.55 =&gt; /home/igalia/WebKit/WebKitBuild/DependenciesGTK/Root/lib/libicui18n.so.55 (0xb6c56000)
	libicuuc.so.55 =&gt; /home/igalia/WebKit/WebKitBuild/DependenciesGTK/Root/lib/libicuuc.so.55 (0xb6a56000)
	libicudata.so.55 =&gt; not found
	libc.so.6 =&gt; /lib/arm-linux-gnueabihf/libc.so.6 (0xb6952000)
	libicudata.so.55 =&gt; not found
	libdl.so.2 =&gt; /lib/arm-linux-gnueabihf/libdl.so.2 (0xb693f000)
	libm.so.6 =&gt; /lib/arm-linux-gnueabihf/libm.so.6 (0xb68cb000)
	libstdc++.so.6 =&gt; /usr/lib/arm-linux-gnueabihf/libstdc++.so.6 (0xb6814000)
	libgcc_s.so.1 =&gt; /lib/arm-linux-gnueabihf/libgcc_s.so.1 (0xb67eb000)
	/lib/ld-linux-armhf.so.3 (0xb6f69000)
	libicudata.so.55 =&gt; not found
	libicudata.so.55 =&gt; not found
$ ./testicu
./testicu: error while loading shared libraries: libicudata.so.55: cannot open shared object file: No such file or directory

$ pkg-config --cflags --libs icu-io
-L/home/igalia/WebKit/WebKitBuild/DependenciesGTK/Root/lib -licuio -licui18n -licuuc -licudata</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1199567</commentid>
    <comment_count>4</comment_count>
    <who name="Carlos Alberto Lopez Perez">clopez</who>
    <bug_when>2016-06-06 09:03:36 -0700</bug_when>
    <thetext>It seems there is something wrong with the libicudata we are building

inside_internal_jhbuild $ ldd /home/igalia/WebKit/WebKitBuild/DependenciesGTK/Root/lib/libicuio.so.55.1 
	libicuuc.so.55 =&gt; /home/igalia/WebKit/WebKitBuild/DependenciesGTK/Root/lib/libicuuc.so.55 (0xb6d39000)
	libicudata.so.55 =&gt; not found
	libicui18n.so.55 =&gt; /home/igalia/WebKit/WebKitBuild/DependenciesGTK/Root/lib/libicui18n.so.55 (0xb6a32000)
	libdl.so.2 =&gt; /lib/arm-linux-gnueabihf/libdl.so.2 (0xb6a1f000)
	libm.so.6 =&gt; /lib/arm-linux-gnueabihf/libm.so.6 (0xb69ab000)
	libstdc++.so.6 =&gt; /usr/lib/arm-linux-gnueabihf/libstdc++.so.6 (0xb68f4000)
	libgcc_s.so.1 =&gt; /lib/arm-linux-gnueabihf/libgcc_s.so.1 (0xb68cb000)
	libc.so.6 =&gt; /lib/arm-linux-gnueabihf/libc.so.6 (0xb67db000)
	/lib/ld-linux-armhf.so.3 (0xb6f59000)
	libicudata.so.55 =&gt; not found
	libicudata.so.55 =&gt; not found

inside_internal_jhbuild $ ls -l WebKitBuild/DependenciesGTK/Root/lib/libicudata.so*
lrwxrwxrwx 1 igalia igalia       18 Jun  6 12:52 WebKitBuild/DependenciesGTK/Root/lib/libicudata.so -&gt; libicudata.so.55.1
lrwxrwxrwx 1 igalia igalia       18 Jun  6 12:52 WebKitBuild/DependenciesGTK/Root/lib/libicudata.so.55 -&gt; libicudata.so.55.1
-rwxr-xr-x 1 igalia igalia 25910236 Jun  6 12:52 WebKitBuild/DependenciesGTK/Root/lib/libicudata.so.55.1

$ ldd WebKitBuild/DependenciesGTK/Root/lib/libicudata.so.55.1
	not a dynamic executable

$ file WebKitBuild/DependenciesGTK/Root/lib/libicudata.so.55.1
WebKitBuild/DependenciesGTK/Root/lib/libicudata.so.55.1: ELF 32-bit LSB shared object, ARM, EABI5 version 1 (SYSV), dynamically linked, BuildID[sha1]=1abd350642e527e20bb600a32865f3b5b2abde4a, not stripped


ldd outputs not a dynamic executable for the libicudata we build. But it works for the icudata from the system

$ ldd /usr/lib/arm-linux-gnueabihf/libicudata.so.52.1 
	libc.so.6 =&gt; /lib/arm-linux-gnueabihf/libc.so.6 (0xb5790000)
	/lib/ld-linux-armhf.so.3 (0xb6f10000)</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1199571</commentid>
    <comment_count>5</comment_count>
    <who name="Carlos Alberto Lopez Perez">clopez</who>
    <bug_when>2016-06-06 09:09:48 -0700</bug_when>
    <thetext>(In reply to comment #4)
&gt; 
&gt; $ ldd WebKitBuild/DependenciesGTK/Root/lib/libicudata.so.55.1
&gt; 	not a dynamic executable
&gt; 
&gt; $ file WebKitBuild/DependenciesGTK/Root/lib/libicudata.so.55.1
&gt; WebKitBuild/DependenciesGTK/Root/lib/libicudata.so.55.1: ELF 32-bit LSB
&gt; shared object, ARM, EABI5 version 1 (SYSV), dynamically linked,
&gt; BuildID[sha1]=1abd350642e527e20bb600a32865f3b5b2abde4a, not stripped
&gt; 
&gt; 
&gt; ldd outputs not a dynamic executable for the libicudata we build. But it
&gt; works for the icudata from the system
&gt; 
&gt; $ ldd /usr/lib/arm-linux-gnueabihf/libicudata.so.52.1 
&gt; 	libc.so.6 =&gt; /lib/arm-linux-gnueabihf/libc.so.6 (0xb5790000)
&gt; 	/lib/ld-linux-armhf.so.3 (0xb6f10000)


This debian bug report has some interesting comments about the same (or very similar) issue: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=653457#10</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1199637</commentid>
    <comment_count>6</comment_count>
    <who name="Carlos Alberto Lopez Perez">clopez</who>
    <bug_when>2016-06-06 11:22:32 -0700</bug_when>
    <thetext>(In reply to comment #5)
&gt; (In reply to comment #4)
&gt; &gt; 
&gt; &gt; $ ldd WebKitBuild/DependenciesGTK/Root/lib/libicudata.so.55.1
&gt; &gt; 	not a dynamic executable
&gt; &gt; 
&gt; &gt; $ file WebKitBuild/DependenciesGTK/Root/lib/libicudata.so.55.1
&gt; &gt; WebKitBuild/DependenciesGTK/Root/lib/libicudata.so.55.1: ELF 32-bit LSB
&gt; &gt; shared object, ARM, EABI5 version 1 (SYSV), dynamically linked,
&gt; &gt; BuildID[sha1]=1abd350642e527e20bb600a32865f3b5b2abde4a, not stripped
&gt; &gt; 
&gt; &gt; 
&gt; &gt; ldd outputs not a dynamic executable for the libicudata we build. But it
&gt; &gt; works for the icudata from the system
&gt; &gt; 
&gt; &gt; $ ldd /usr/lib/arm-linux-gnueabihf/libicudata.so.52.1 
&gt; &gt; 	libc.so.6 =&gt; /lib/arm-linux-gnueabihf/libc.so.6 (0xb5790000)
&gt; &gt; 	/lib/ld-linux-armhf.so.3 (0xb6f10000)
&gt; 
&gt; 
&gt; This debian bug report has some interesting comments about the same (or very
&gt; similar) issue: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=653457#10

Indeed, very same issue.

Applying the above Debian patch to the icu sources and rebuilding fixes the build and al liking problems.

wandboard ~/WebKit $ LD_LIBRARY_PATH=$(pwd)/WebKitBuild/DependenciesGTK/Root/lib ldd WebKitBuild/Release/bin/WebKitWebProcess |grep icu
	libicui18n.so.55 =&gt; /home/igalia/WebKit/WebKitBuild/DependenciesGTK/Root/lib/libicui18n.so.55 (0xb454d000)
	libicuuc.so.55 =&gt; /home/igalia/WebKit/WebKitBuild/DependenciesGTK/Root/lib/libicuuc.so.55 (0xb4317000)
	libharfbuzz-icu.so.0 =&gt; /home/igalia/WebKit/WebKitBuild/DependenciesGTK/Root/lib/libharfbuzz-icu.so.0 (0xb395e000)
	libicudata.so.55 =&gt; /home/igalia/WebKit/WebKitBuild/DependenciesGTK/Root/lib/libicudata.so.55 (0xb1298000)


I&apos;m going to upload a patch ASAP.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1199650</commentid>
    <comment_count>7</comment_count>
      <attachid>280612</attachid>
    <who name="Carlos Alberto Lopez Perez">clopez</who>
    <bug_when>2016-06-06 11:44:20 -0700</bug_when>
    <thetext>Created attachment 280612
Patch</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1199663</commentid>
    <comment_count>8</comment_count>
      <attachid>280612</attachid>
    <who name="Michael Catanzaro">mcatanzaro</who>
    <bug_when>2016-06-06 12:13:51 -0700</bug_when>
    <thetext>Comment on attachment 280612
Patch

Looks like we run about 10 sed scripts to do this not only here but also to a bunch of different Makefiles when building the Fedora package... so this solution is much simpler than what we have.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1199693</commentid>
    <comment_count>9</comment_count>
      <attachid>280612</attachid>
    <who name="WebKit Commit Bot">commit-queue</who>
    <bug_when>2016-06-06 13:32:41 -0700</bug_when>
    <thetext>Comment on attachment 280612
Patch

Clearing flags on attachment: 280612

Committed r201726: &lt;http://trac.webkit.org/changeset/201726&gt;</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1199694</commentid>
    <comment_count>10</comment_count>
    <who name="WebKit Commit Bot">commit-queue</who>
    <bug_when>2016-06-06 13:32:46 -0700</bug_when>
    <thetext>All reviewed patches have been landed.  Closing bug.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>1199957</commentid>
    <comment_count>11</comment_count>
    <who name="Carlos Garcia Campos">cgarcia</who>
    <bug_when>2016-06-06 23:49:08 -0700</bug_when>
    <thetext>Wow, this was tricky, thank you very much Carlos!</thetext>
  </long_desc>
      
          <attachment
              isobsolete="0"
              ispatch="1"
              isprivate="0"
          >
            <attachid>280612</attachid>
            <date>2016-06-06 11:44:20 -0700</date>
            <delta_ts>2016-06-06 13:32:41 -0700</delta_ts>
            <desc>Patch</desc>
            <filename>bug-158417-20160606204544.patch</filename>
            <type>text/plain</type>
            <size>2901</size>
            <attacher name="Carlos Alberto Lopez Perez">clopez</attacher>
            
              <data encoding="base64">U3VidmVyc2lvbiBSZXZpc2lvbjogMjAxNzE3CmRpZmYgLS1naXQgYS9Ub29scy9DaGFuZ2VMb2cg
Yi9Ub29scy9DaGFuZ2VMb2cKaW5kZXggNTJmMDVhNjI5N2RiYTlhM2U1ZmI4N2IxYjAwYTI1ODY2
OTgwNGJmMy4uN2EwYTNkYTdjYzRmZDI0N2U3NjJhOWFiNTk1NzQ0MTRhM2IwNDk2ZSAxMDA2NDQK
LS0tIGEvVG9vbHMvQ2hhbmdlTG9nCisrKyBiL1Rvb2xzL0NoYW5nZUxvZwpAQCAtMSwzICsxLDI5
IEBACisyMDE2LTA2LTA2ICBDYXJsb3MgQWxiZXJ0byBMb3BleiBQZXJleiAgPGNsb3BlekBpZ2Fs
aWEuY29tPgorCisgICAgICAgIFJFR1JFU1NJT04ocjIwMTQ0OSkgW0dUS10gQVJNdjcgYnVpbGQg
ZmFpbHMgd2l0aCBsaWJpY3VkYXRhLnNvLjU1OiBjYW5ub3Qgb3BlbiBzaGFyZWQgb2JqZWN0IGZp
bGUgb24gZ3RrZG9jLXNjYW5nb2JqIHN0ZXAuCisgICAgICAgIGh0dHBzOi8vYnVncy53ZWJraXQu
b3JnL3Nob3dfYnVnLmNnaT9pZD0xNTg0MTcKKworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkg
KE9PUFMhKS4KKworICAgICAgICBUaGUgdG9vbGNoYWluIG9uIGFybWhmIChBUk12Nykgc2VlbXMg
dW5hYmxlIHRvIHByb3Blcmx5IGhhbmRsZQorICAgICAgICBhIHNoYXJlZCBsaWJyYXJ5IHRoYXQg
ZG9lc24ndCBsaW5rIHdpdGggYW55dGhpbmcuCisKKyAgICAgICAgQW5kIGxpYmljdWRhdGEgaXMg
YnVpbHQgYnkgZGVmYXVsdCBpbiB0aGlzIHdheSBiZWNhdXNlIGl0IG9ubHkKKyAgICAgICAgY29u
dGFpbnMgc3RhdGljIGRhdGEuCisKKyAgICAgICAgVGhhdCBtYWtlcyBpY3UgdW51c2FibGUgb24g
YXJtaGYgYmVjYXVzZSB0aGUgbGlua2VyIHdpbGwgYmUKKyAgICAgICAgdW5hYmxlIHRvIHJlc29s
dmUgdGhlIGxpYmljdWRhdGEgZGVwZW5kZW5jaWVzLgorCisgICAgICAgIE1vc3QgKGlmIG5vdCBh
bGwpIGRpc3RyaWJ1dGlvbnMgd29ya2Fyb3VuZCB0aGlzIGJ5IGxpbmtpbmcKKyAgICAgICAgbGli
aWN1ZGF0YSB3aXRoIGxpYmM2LCB3aGljaCBpcyBhbHJlYWR5IGEgbmVlZGVkIGRlcGVuZGVuY3kg
Zm9yCisgICAgICAgIGFueSBvZiB0aGUgb3RoZXIgaWN1IHNoYXJlZCBsaWJyYXJpZXMuCisKKyAg
ICAgICAgU28gaW1wb3J0IGhlcmUgdGhlIERlYmlhbiBwYXRjaCBmaXhpbmcgdGhpcyBpc3N1ZS4g
Rm9yIGZ1cnRoZXIKKyAgICAgICAgZGV0YWlscyBjaGVjayBodHRwczovL2J1Z3MuZGViaWFuLm9y
Zy82NTM0NTcKKworICAgICAgICAqIGd0ay9qaGJ1aWxkLm1vZHVsZXM6CisgICAgICAgICogZ3Rr
L3BhdGNoZXMvaWN1ZGF0YS1zdGRsaWJzLnBhdGNoOiBBZGRlZC4KKwogMjAxNi0wNi0wNCAgQWxl
eGV5IFByb3NrdXJ5YWtvdiAgPGFwQGFwcGxlLmNvbT4KIAogICAgICAgICBSRUdSRVNTSU9OIChy
MjAxMjYzKTogU29tZSB0ZXN0cyBoYXZlIGJlY29tZSBmbGFreSB0aW1lb3V0cy4KZGlmZiAtLWdp
dCBhL1Rvb2xzL2d0ay9qaGJ1aWxkLm1vZHVsZXMgYi9Ub29scy9ndGsvamhidWlsZC5tb2R1bGVz
CmluZGV4IDAzNDljYjhlNDRjNGVkZDM0NjY4Mzk3NmJmNjhmM2FiMDBiOTVjNzguLjc0MjIyYjI4
ZTMzYjEzODE4OGQ4MDliNmI5YjFmNzUzZTE4NGEyOGMgMTAwNjQ0Ci0tLSBhL1Rvb2xzL2d0ay9q
aGJ1aWxkLm1vZHVsZXMKKysrIGIvVG9vbHMvZ3RrL2poYnVpbGQubW9kdWxlcwpAQCAtNDk2LDYg
KzQ5Niw3IEBACiAgICAgPGJyYW5jaCBtb2R1bGU9ImljdTRjLTU1XzEtc3JjLnRneiIgdmVyc2lv
bj0iNTUuMSIgY2hlY2tvdXRkaXI9ImljdSIKICAgICAgICAgICAgIHJlcG89IndlYmtpdGd0ay1q
aGJ1aWxkLW1pcnJvciIKICAgICAgICAgICAgIGhhc2g9InNoYTI1NjplMTZiMjJjYmVmZGQzNTRi
ZWMxMTQ1NDFmNzg0OWExMmY4ZmMyMDE1MzIwY2E1MjgyZWU0ZmQ3ODc1NzE0NTdiIj4KKyAgICAg
IDxwYXRjaCBmaWxlPSJpY3VkYXRhLXN0ZGxpYnMucGF0Y2giIHN0cmlwPSIxIi8+CiAgICAgPC9i
cmFuY2g+CiAgIDwvYXV0b3Rvb2xzPgogCmRpZmYgLS1naXQgYS9Ub29scy9ndGsvcGF0Y2hlcy9p
Y3VkYXRhLXN0ZGxpYnMucGF0Y2ggYi9Ub29scy9ndGsvcGF0Y2hlcy9pY3VkYXRhLXN0ZGxpYnMu
cGF0Y2gKbmV3IGZpbGUgbW9kZSAxMDA2NDQKaW5kZXggMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAw
MDAwMDAwMDAwMDAwMDAwMC4uNWU5MmJkZWMxYjk1NTVkNGFiNjMyN2U1NTI1MmZlZjUyNTM2YTAy
ZgotLS0gL2Rldi9udWxsCisrKyBiL1Rvb2xzL2d0ay9wYXRjaGVzL2ljdWRhdGEtc3RkbGlicy5w
YXRjaApAQCAtMCwwICsxLDE1IEBACitJbmRleDogaWN1LTUyfm0xL3NvdXJjZS9jb25maWcvbWgt
bGludXgKKz09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09
PT09PT09PT09PT09PT09PT0KKy0tLSBpY3UtNTJ+bTEub3JpZy9zb3VyY2UvY29uZmlnL21oLWxp
bnV4CTIwMTMtMDktMTQgMTg6NTM6MjMuMjg0MDQwNDY3IC0wNDAwCisrKysgaWN1LTUyfm0xL3Nv
dXJjZS9jb25maWcvbWgtbGludXgJMjAxMy0wOS0xNCAxODo1MzoyMy4yODQwNDA0NjcgLTA0MDAK
K0BAIC0yMSw3ICsyMSw5IEBACisgTERfUlBBVEhfUFJFID0gLVdsLC1ycGF0aCwKKyAKKyAjIyBU
aGVzZSBhcmUgdGhlIGxpYnJhcnkgc3BlY2lmaWMgTERGTEFHUworLUxERkxBR1NJQ1VEVD0tbm9k
ZWZhdWx0bGlicyAtbm9zdGRsaWIKKysjTERGTEFHU0lDVURUPS1ub2RlZmF1bHRsaWJzIC1ub3N0
ZGxpYgorKyMgRGViaWFuIGNoYW5nZTogbGlua2luZyBpY3VkYXRhIGFzIGRhdGEgb25seSBjYXVz
ZXMgdG9vIG1hbnkgcHJvYmxlbXMuCisrTERGTEFHU0lDVURUPQorIAorICMjIENvbXBpbGVyIHN3
aXRjaCB0byBlbWJlZCBhIGxpYnJhcnkgbmFtZQorICMgVGhlIGluaXRpYWwgdGFiIGluIHRoZSBu
ZXh0IGxpbmUgaXMgdG8gcHJldmVudCBpY3UtY29uZmlnIGZyb20gcmVhZGluZyBpdC4K
</data>

          </attachment>
      

    </bug>

</bugzilla>