[Gnash-commit] [SCM] Gnash branch, master, updated. b1e5dc010068f998e62fdfc85e9bc2feea01d3cd

[Benjamin Wolsey]

This is an automated email from the git hooks/post-receive script. It was
generated because a ref change was pushed to the repository containing
the project "Gnash".

The branch, master has been updated
       via  b1e5dc010068f998e62fdfc85e9bc2feea01d3cd (commit)
      from  352e4e7c2bfcbddb71bd47e725f68f74d45af1a8 (commit)

Those revisions listed above that are new to this repository have
not appeared on any other notification email; so we list those
revisions in full, below.

- Log -----------------------------------------------------------------
http://git.savannah.gnu.org/cgit//commit/?id=b1e5dc010068f998e62fdfc85e9bc2feea01d3cd


commit b1e5dc010068f998e62fdfc85e9bc2feea01d3cd
Author: Benjamin Wolsey <address@hidden>
Date:   Wed Aug 25 13:06:49 2010 +0200

    Don't run the testcase either if ming isn't up-to-date.
    
    Hack gprocessor so it doesn't crash everywhere on Debian Lenny.

diff --git a/testsuite/misc-ming.all/Makefile.am 
b/testsuite/misc-ming.all/Makefile.am
index fe8f540..51f2d73 100644
--- a/testsuite/misc-ming.all/Makefile.am
+++ b/testsuite/misc-ming.all/Makefile.am
@@ -2249,7 +2249,6 @@ TEST_CASES = \
        BeginBitmapFillRunner \
        BitmapDataTestRunner \
        BitmapDataDrawRunner \
-       LoadBitmapTestRunner \
        LoadVarsTestRunner \
        DeviceFontTestRunner \
        EmbeddedFontTestRunner \
@@ -2397,6 +2396,12 @@ TEST_CASES += \
        $(NULL)
 endif
 
+if MING_VERSION_0_4_4
+TEST_CASES += \
+       LoadBitmapTestRunner \
+       $(NULL)
+endif
+
 check-DEJAGNU: site-update $(check_PROGRAMS)
        @runtest=$(RUNTEST); \
        if $(SHELL) -c "$$runtest --version" > /dev/null 2>&1; then \
diff --git a/utilities/processor.cpp b/utilities/processor.cpp
index 47ea485..c7b85c5 100644
--- a/utilities/processor.cpp
+++ b/utilities/processor.cpp
@@ -370,7 +370,12 @@ main(int argc, char *argv[])
 
 #ifdef RENDERER_AGG
     boost::shared_ptr<Renderer_agg_base> r(create_Renderer_agg("RGBA32"));
-    unsigned char buf[1];
+
+    // Yes, this leaks. On some systems (e.g. Debian Lenny) the data is
+    // evidently accessed after main() returns. Rather than bothering to
+    // work out why, we let this byte leak, as it's returned to the system on
+    // exit anyway.
+    unsigned char* buf = new unsigned char[1];
     r->init_buffer(buf, 1, 1, 1, 1);
 #endif
 

-----------------------------------------------------------------------

Summary of changes:
 testsuite/misc-ming.all/Makefile.am |    7 ++++++-
 utilities/processor.cpp             |    7 ++++++-
 2 files changed, 12 insertions(+), 2 deletions(-)


hooks/post-receive
-- 
Gnash