sirikata / berkelium Goto Github PK

This project forked from danielrh/berkelium

Chromium Embedded in GL/DX (NOTE: No longer maintained!)

C++ 86.45% CSS 1.46% Shell 5.00% C 7.09%

berkelium's Introduction

What is Sirikata?
-----------------
Sirikata is a BSD licensed open source platform for games and virtual
worlds. We aim to provide a set of libraries and protocols which can
be used to deploy a virtual world, as well as fully featured sample
implementations of services for hosting and deploying these worlds.

Where can I find out more?
--------------------------
The most up to date information can always be found at
http://www.sirikata.com.  The source code is available from GitHub at
http://github.com/sirikata/sirikata/.

berkelium's People

Contributors

Stargazers

Watchers

Forkers

mattdev gioggiolo brownman sori tito pgsharpe senseproject romainpi klaim marcellarius drieks afrancois lauyoume agoblinking kubami mansu pierotofy rentouch makelife bygreencn zhang44444 jaz303 petroskataras cruzrr mokerjoke qqonline ingmr nondescriptpointer insaneraph nijil gongfupanada redwolf6323 chunminy meingottwalter lautitia jetmc daocalendar matiasnamendola ricardogil cexer ntvis hellbender3069 smartfire samarthp plumpmath zhangwenquan linecode xpcrts

berkelium's Issues

Ir works on ARM? (raspberry pi 3)

Hi,

Ir works on ARM? (raspberry pi 3).

Thanks.

Temp stuff added to AppData/.../Temp at each instance but not removed.

[Originally from https://github.com/sirikata/sirikata/issues/498]

Hi,

I see that each time I launch one of the binaries (glut_demo.exe, glut_input.exe, ...), Berkelium adds this stuff to the C:\Users***\AppData\Local\Temp :

a directory named berkeliumyyyyARBITRARY_NUMBER that seems to contain the user directory of chromium (cookies, history, cache, ...).
a plugin named plugin_ARBITRARY_NUMBER
So the temp directory grows rapidly because none gets released.

I noticed the same behavior when I use berkelium in my own runtime.
I am on Win7 64.

Thank you for providing this library, it could become a major step for game UI.

Cheers,

Jounayd Id Salah.

can't make tests

When compiling the berkelium library (using make), the following error occurs:

/usr/bin/python: can't open file '/home/jarrett/projects/berkelium/externals/cxxtest/cxxtestgen.py': [Errno 2] No such file or directory
make[2]: *** [cxxtest_tests.cc] Error 2
make[1]: *** [CMakeFiles/tests.dir/all] Error 2
make: *** [all] Error 2

Looks like the cxxtest folder has nothing in it. On github it looks like it 'links' to another project...

berkelium segfaults on RPi, while the build chrome works correctly

Hi,
I have built berkelium for the RPi as stated in my last issue. But I am running into a problem. When I test Berkelium it runs normal like chrome, but when I visit a website it segfaults out. If I do the same thing with the chrome that is in ./build/chromium/out/Release the website shows without any problems.

Do you have any idea about this, or is there a way to get more information then just the segmentation fault?

With regards,
Hellbender

python script action_makenames.py fails during chromium compilation

During chromium compilation on Linux (Debian testing, 64bit, gcc 4.7.1, cmake 2.8.9), the python script action_makenames.py fails with the following error:

...
  ACTION webcore_bindings_sources_HTMLNames out/Release/obj/gen/webkit/HTMLNames.cpp
  ACTION webcore_bindings_sources_SVGNames out/Release/obj/gen/webkit/SVGNames.cpp
  ACTION webcore_bindings_sources_MathMLNames out/Release/obj/gen/webkit/MathMLNames.cpp
  ACTION webcore_bindings_sources_XLinkNames out/Release/obj/gen/webkit/XLinkNames.cpp
  ACTION webcore_bindings_sources_XMLNSNames out/Release/obj/gen/webkit/XMLNSNames.cpp
Unknown parameter math for tags/attrs
Traceback (most recent call last):
  File "scripts/action_makenames.py", line 174, in <module>
    sys.exit(main(sys.argv))
  File "scripts/action_makenames.py", line 156, in main
    assert returnCode == 0
AssertionError
make: *** [out/Release/obj/gen/webkit/MathMLNames.cpp] Error 1
make: *** Waiting for unfinished jobs....
Unknown parameter a for tags/attrs
Traceback (most recent call last):
  File "scripts/action_makenames.py", line 174, in <module>
    sys.exit(main(sys.argv))
  File "scripts/action_makenames.py", line 156, in main
    assert returnCode == 0
AssertionError
Unknown parameter a interfaceName for tags/attrs
make: *** [out/Release/obj/gen/webkit/SVGNames.cpp] Error 1
Traceback (most recent call last):
  File "scripts/action_makenames.py", line 174, in <module>
    sys.exit(main(sys.argv))
  File "scripts/action_makenames.py", line 156, in main
    assert returnCode == 0
AssertionError
make: *** [out/Release/obj/gen/webkit/HTMLNames.cpp] Error 1
Unknown parameter actuate for tags/attrs
Traceback (most recent call last):
  File "scripts/action_makenames.py", line 174, in <module>
    sys.exit(main(sys.argv))
  File "scripts/action_makenames.py", line 156, in main
    assert returnCode == 0
AssertionError
make: *** [out/Release/obj/gen/webkit/XLinkNames.cpp] Error 1
Unknown parameter xmlns for tags/attrs
Traceback (most recent call last):
  File "scripts/action_makenames.py", line 174, in <module>
    sys.exit(main(sys.argv))
  File "scripts/action_makenames.py", line 156, in main
    assert returnCode == 0
AssertionError
make: *** [out/Release/obj/gen/webkit/XMLNSNames.cpp] Error 1
Failed to install: chromium

It looks like the python script is called a perl script, and the perl script is dying on line 209:

die "Unknown parameter $parameter for tags/attrs\n" if !defined($parameters{$parameter});

The 'unknown parameter's are:

math
a
a interfaceName
actuate
xmlns

I'm not sure where these parameters are coming from.

Anyone have any idea how to correct this?

Patches for Ubuntu 12.04 with a Clang 3.2 trunk 157126 and newer Gtk

Hi,

At the moment, I'm too lazy to play with creating proper patches, creating a git clone, making a pull request, etc. Also, I'm extremely low on disk space, so I'd rather not play too much having built Chromium and Berkelium successfully.

That said, two patches need to be applied to Chromium source code so it can be built with Clang 3.2 trunk, revision 157126.

1. Some assembly is not specific enough.

See http://code.google.com/p/seccompsandbox/issues/detail?id=14

This patch, from aforementioned link but adjusted for comments in that same issue, needs to be applied to seccompsandbox:

Index: fault_handler_i386.S
===================================================================
--- fault_handler_i386.S    (revision 153)
+++ fault_handler_i386.S    (working copy)
@@ -178,9 +178,9 @@
         // callers might be confused by this and will need fixing for running
         // inside of the seccomp sandbox.
      20:lea  playground$sa_segv, %eax
-        cmp  $0, 0(%eax)         // SIG_DFL
+        cmpl $0, 0(%eax)         // SIG_DFL
         jz   21f
-        cmp  $1, 0(%eax)         // SIG_IGN
+        cmpl  $1, 0(%eax)         // SIG_IGN
         jnz  22f                 // can't really ignore synchronous signals

Or even better! See this patch that also fixed x86_64. http://code.google.com/p/seccompsandbox/source/detail?r=154

2. Header include missing

Glib's internationalization header needs to be included.

Index: build/chromium/src/ui/base/l10n/l10n_util.cc
===================================================================
--- build/chromium/src/ui/base/l10n/l10n_util.cc    (revision 87952)
+++ build/chromium/src/ui/base/l10n/l10n_util.cc    (working copy)
@@ -6,6 +6,8 @@

 #if defined(TOOLKIT_USES_GTK)
 #include <glib/gutils.h>
+#include <glib/gi18n.h>
+
 #endif

 #include <algorithm>

That seems to be about it. Now, hopefully someone can spare some time to properly integrate these patches into the build process. :-)

patch unexpectedly ends in middle of line.

Hi to all,

I'm trying to build the berkelium source code in my Mac machine. When it's running the chromium_libpng-1.5.patch, i'm getting the error like this,

Hunk #4 succeeded at 343 with fuzz 1.
patch unexpectedly ends in middle of line
patch unexpectedly ends in middle of line

Could you please anyone make me to know about the problem??

Thanks,
Vijay

building berkelium results in linker error

System:
Debian testing 64bit
gcc 4.7.1
cmake 2.8.9

Error is:

Linking CXX shared library liblibberkelium_d.so
/usr/bin/ld: BFD (GNU Binutils for Debian) 2.22 internal error, aborting at ../../bfd/elf64-x86-64.c line 3387 in elf_x86_64_relocate_section

/usr/bin/ld: Please report this bug.

collect2: error: ld returned 1 exit status
make[2]: *** [liblibberkelium_d.so] Error 1
make[1]: *** [CMakeFiles/libberkelium.dir/all] Error 2
make: *** [all] Error 2

anyone have any ideas?

chromium compile errors on linux

Several compile errors arise when compiling chromium on linux (Debian testing, 64bit, gcc 4.7.1, cmake 2.8.9).

This is the first error received (this same type of error seems to appear in other files as well):

In file included from ui/base/ui_base_paths.cc:9:0:
./base/file_util.h: In member function ‘void file_util::ScopedFDClose::operator()(int*) const’:
./base/file_util.h:380:11: error: ‘close’ was not declared in this scope
./base/file_util.h:380:11: error: invalid type in declaration before ‘;’ token
In file included from app/app_paths.cc:9:0:
./base/file_util.h: In member function ‘void file_util::ScopedFDClose::operator()(int*) const’:
./base/file_util.h:380:11: error: ‘close’ was not declared in this scope
./base/file_util.h:380:11: error: invalid type in declaration before ‘;’ token
make: *** [out/Release/obj.target/app_base/ui/base/ui_base_paths.o] Error 1
make: *** [out/Release/obj.target/app_base/app/app_paths.o] Error 1
In file included from ui/base/l10n/l10n_util.cc:17:0:
./base/file_util.h: In member function ‘void file_util::ScopedFDClose::operator()(int*) const’:
./base/file_util.h:380:11: error: ‘close’ was not declared in this scope
./base/file_util.h:380:11: error: invalid type in declaration before ‘;’ token
make: *** [out/Release/obj.target/app_base/ui/base/l10n/l10n_util.o] Error 1
In file included from chrome/utility/utility_main.cc:7:0:
./base/file_util.h: In member function ‘void file_util::ScopedFDClose::operator()(int*) const’:
./base/file_util.h:380:11: error: ‘close’ was not declared in this scope
./base/file_util.h:380:11: error: invalid type in declaration before ‘;’ token
In file included from chrome/utility/utility_thread.cc:9:0:
./base/file_util.h: In member function ‘void file_util::ScopedFDClose::operator()(int*) const’:
./base/file_util.h:380:11: error: ‘close’ was not declared in this scope
./base/file_util.h:380:11: error: invalid type in declaration before ‘;’ token
make: *** [out/Release/obj.target/utility/chrome/utility/utility_main.o] Error 1
make: *** [out/Release/obj.target/utility/chrome/utility/utility_thread.o] Error 1
Failed to install: chromium

Solution:
Need to include unistd.h in files where this error occurs (POSIX specific error I think).
so, these files need this line:
#include <unistd.h>

Related to this error (chromium build using gcc 4.7): http://code.google.com/p/chromium/issues/detail?id=120724

I don't have push priveleges, so I can't push back my changes (I also don't quite get the whole push request thing.. :)

berkelium doesn't build on raspbian ( raspberry pi hardfp )

I am working on getting berkelium and chromium compiled for the raspberry pi, but I run into two problems.

With using ./util/build_chromium it halts when the gyp script has to run an external shell script. for example when it is trying to find the architecture of python. It says the script exited with exit status 0 ( which i think is good ) but the process exits on it. When I go into the chromium src directory and run 'make' everything runs fine. I experienced this same problem when checking out the latest chromium and trying to build that. Once again all shell scripts made the process fail.
the cxxtests won't build because it can't find sources. After i patched the common.gypi to hardcoded select arm as the architecture, chromium and berkelium build. But the cxx tests never build so i have to comment them out of the files to get the build process to finish.

I will add the failure texts to this issue as soon as I get them ( building an a raspberry pi natively is a waiting game, but it works ) for clearer information

Is a Win64 version possible?

Hi as the title states, I was wondering if a a Win64 version is possible? I did read somewhere that in fact the chrome side of things are still 32bit only but I'm not sure if that is still the case. Thanks for any info.