Hi,
yesterday I released a new version of the Teatime applet
http://det.cable.nu/teatime/index.rbx which uses the new functionality
of context menus for applets. Today the first Linux distributor asked
me, when 0.19 will get released, cause the build failed with 0.18.1.
So I forward this question 
Cheers, detlef
Hi,
2009/5/26 Detlef R. [email protected]:
yesterday I released a new version of the Teatime applet
http://det.cable.nu/teatime/index.rbx which uses the new functionality
of context menus for applets. Today the first Linux distributor asked
me, when 0.19 will get released, cause the build failed with 0.18.1.So I forward this question
I wanted to release 0.19 as development seriese but my computer isn’t
good condition now…
I already updated NEW file and prepared Rakefile.
Could someone (one of commiters) release 0.19 for me?
Pleaese…
I think that remaind tasks are here:
The list may not be correct… If you find some wrong step, please write
it to RELEASING and add it to the Ruby-GNOME2 repository.
ref. cairo has RELEASING document:
http://cgit.freedesktop.org/cairo/tree/RELEASING
kou
On Tue, May 26, 2009 at 12:59, Kouhei S. [email protected] wrote:
2009/5/26 Detlef R. [email protected]:
yesterday I released a new version of the Teatime applet
http://det.cable.nu/teatime/index.rbx which uses the new functionality
of context menus for applets. Today the first Linux distributor asked
me, when 0.19 will get released, cause the build failed with 0.18.1.
So I forward this question
I wanted to release 0.19 as development seriese but my computer isn’t
good condition now…
I just wanted to report that my gio bindings have been completed.
I’ll try to get them out the door soon. I have been more or less been
singlehandedly responsible for migrating from GroupWise to Exchange at
work, which has required an incredible amount of work, so this has
been on hold. I need to make sure that my sourceforge account works
and stuff (I haven’t logged in in a long time). I’m not sure if I
(pcppopper) still have commit access, either.
Le mardi 26 mai 2009 à 19:59 +0900, Kouhei S. a écrit :
- confirm packages are buildable (rake build?)
- confirm tests are passed (rake test?)
- make packages (rake dist?)
- release packages (rake release?)
- confirm released packages are correct (download packages and build them)
- tag (rake tag?)
- announce the release on the Wiki, ruby-gnome2-devel-en@ ML
I successfuly built everything except for gtkmozembed
I also built all except gtkmozembed using generated
ruby-gnome2-all-0.19.0.tar.gz
Concerning the tests I get :
Loaded suite .
Started
…E…O
Finished in 1.194534 seconds.
Error:
test_get_image(TestPage):
NoMethodError: undefined method get_image' for #<Poppler::Page:0xb65c52e0 ptr=0x85c2400> ./test/test_page.rb:6:intest_get_image’
Omission: We doesn’t have a PDF that has a thumbnail…
test_thumbnail_pixbuf(TestPage)
./test/test_page.rb:40:in `test_thumbnail_pixbuf’
Loaded suite .
Started
F…
Finished in 0.010049 seconds.
Loaded suite .
Started
…F…F…
Finished in 0.431199 seconds.
broadcast_startup_message' for #<Gdk::DisplayX11:0xb7449a00 ptr=0x9f5d0c8>"> ---Backtrace--- /home/a/rpm/BUILD/ruby-gnome2-all-0.19.0/gtk/test/test_gdk_display.rb:28:intest_broadcast_startup_message’Other modules are OK
Hi,
I’ve also run the test suite, but complete. All parts build and
installed successfully.
I got several unicode conversion errors, which are not popped up in
Pascals test, as this tests are skipped if the libuconv is not
installed.
To the poppler/test/test_page.rb that reports
I’ve tested the thumbnail extraction with poppler and Gdk::Pixbuf on
many documents and it works well. An example you can download here
http://www.datasheetdir.com/go/-xxx-x-lcvb.pdf . It contains thumbnails
for six pages. May be we should just drop this test (it works) if no one
can provide a free test PDF.
Cheers, detlef
Loaded suite test
Started
…/home/ich/sonstiges/prog/src/ruby-gnome2-dev/glib/test/test_iochannel.rb:
line 39 GLib-CRITICAL **:g_io_channel_read_to_end: assertion
channel->is_readable' failed /home/ich/sonstiges/prog/src/ruby-gnome2-dev/glib/test/test_iochannel.rb: line 54 GLib-CRITICAL **:g_io_channel_read_to_end: assertion channel->is_readable’
failed
/home/ich/sonstiges/prog/src/ruby-gnome2-dev/glib/test/test_iochannel.rb:
line 59 GLib-WARNING **:Invalid file descriptor.
…EF…F…FFFF…F…F
F…O…
Finished in 0.905522 seconds.
Error:
test_ucs4_to_utf16(TestGLibUnicode):
GLib::ConvertError: Character out of range for UTF-16
/home/ich/sonstiges/prog/src/ruby-gnome2-dev/glib/test/test_unicode.rb:321:in
to_utf16' /home/ich/sonstiges/prog/src/ruby-gnome2-dev/glib/test/test_unicode.rb:321:in test_ucs4_to_utf16’
Failure:
test_ucs4_to_utf8(TestGLibUnicode)
[/home/ich/sonstiges/prog/src/ruby-gnome2-dev/glib/test/test_unicode.rb:330]:
<“ã‚ã„ã†ãˆãŠ”> expected but was <“������������������������������”>.
Failure:
test_unichar_to_utf8(TestGLibUnicode)
[/home/ich/sonstiges/prog/src/ruby-gnome2-dev/glib/test/test_unicode.rb:336]:
<“ã‚”> expected but was <“������”>.
Failure:
test_unicode_canonical_decomposition(TestGLibUnicode)
[/home/ich/sonstiges/prog/src/ruby-gnome2-dev/glib/test/test_unicode.rb:170]:
<“a\000\000\000\001\003\000\000”> expected but was <“\000\000\000a\000
\000\003\001”>.
Failure:
test_unicode_canonical_ordering(TestGLibUnicode)
[/home/ich/sonstiges/prog/src/ruby-gnome2-dev/glib/test/test_unicode.rb:162]:
<“a\000\000\000#\003\000\000\b\003\000\000e\000\000\000#\003\000\000\004
\003\000\000\001\003\000\000”> expected but was <“a\000\000\000\b\003
\000\000#\003\000\000e\000\000\000\004\003\000\000\001\003\000\000#\003
\000\000”>.
Failure:
test_utf16_to_ucs4(TestGLibUnicode)
[/home/ich/sonstiges/prog/src/ruby-gnome2-dev/glib/test/test_unicode.rb:307]:
<“B0\000\000D0\000\000F0\000\000H0\000\000J0\000\000”> expected but was
<“\000\000B0\000\000D0\000\000F0\000\000H0\000\000J0”>.
diff:
B0D0F0H0J0
? ++
Failure:
test_utf16_to_utf8(TestGLibUnicode)
[/home/ich/sonstiges/prog/src/ruby-gnome2-dev/glib/test/test_unicode.rb:313]:
<“ã‚ã„ã†ãˆãŠ”> expected but was <“䈰ä°ä˜°ä °ä¨°”>.
Failure:
test_utf8_get_char(TestGLibUnicode)
[/home/ich/sonstiges/prog/src/ruby-gnome2-dev/glib/test/test_unicode.rb:194]:
<1110441984> expected but was <12354>.
Failure:
test_utf8_to_ucs4(TestGLibUnicode)
[/home/ich/sonstiges/prog/src/ruby-gnome2-dev/glib/test/test_unicode.rb:294]:
<“B0\000\000D0\000\000F0\000\000H0\000\000J0\000\000”> expected but was
<“\000\0000B\000\0000D\000\0000F\000\0000H\000\0000J”>.
diff:
only_win32' /home/ich/sonstiges/prog/src/ruby-gnome2-dev/glib/test/test_win32.rb:7:in test_get_package_install_directory_of_module’Loaded suite .
Started
…F…F…
Finished in 1.010254 seconds.
broadcast_startup_message' for #<Gdk::DisplayX11:0x486ba300 ptr=0x10210080>"> ---Backtrace--- /home/ich/sonstiges/prog/src/ruby-gnome2-dev/gtk/test/test_gdk_display.rb:28:in test_broadcast_startup_message’Loaded suite .
Started
…
Finished in 0.011747 seconds.
Loaded suite .
Started
.
Finished in 0.069847 seconds.
Loaded suite .
Started
…F…P…
…
Finished in 12,114583 seconds.
Failure:
test_create_clock(TestClock) [./test/test_clock.rb:3]:
exception expected but none was thrown.
Pending: Async is correct behavior?
test_state(TestElement)
./test/test_element.rb:60:in `test_state’
Loaded suite .
Started
F…
Finished in 0.142499 seconds.
Loaded suite .
Started
…O
Finished in 1.396346 seconds.
Loaded suite .
Started
…/test/test_context.rb: line 16
GnomePrint-CRITICAL **:gnome_print_lineto: assertion pc->haspage' failed ./test/test_context.rb: line 22 GnomePrint-CRITICAL **:gnome_print_lineto: assertion gp_gc_has_currentpoint (pc->gc)’ failed …
Finished in 0.126727 seconds.
Hi,
2009/5/26 Pascal T. [email protected]:
I also built all except gtkmozembed using generated
ruby-gnome2-all-0.19.0.tar.gz
Thanks for confirming.
- Error:
test_get_image(TestPage):
NoMethodError: undefined methodget_image' for #<Poppler::Page:0xb65c52e0 ptr=0x85c2400> ./test/test_page.rb:6:intest_get_image’
This test requires cairo supported Ruby/Poppler.
We should check Ruby/Poppler supports cairo with
Poppler.cairo_available?
If you can, could you fix the test to work on non cairo supported
Ruby/Poppler
environment too.
Tue May 26 14:33:06 +0200 2009 running test
for /home/a/rpm/BUILD/ruby-gnome2-all-0.19.0/pango
- Failure:
test_end_index(TestAttribute) [./test/test-attribute.rb:15]:
<-1> expected but was
<4294967295>.
This will be fixed with your patch.
Tue May 26 14:33:06 +0200 2009 running test
for /home/a/rpm/BUILD/ruby-gnome2-all-0.19.0/gtk
- Failure:
test_broadcast_startup_message(TestGdkDisplay)
[/home/a/rpm/BUILD/ruby-gnome2-all-0.19.0/gtk/test/test_gdk_display.rb:27]:
Exception raised:
Class:
Message: <"undefined methodbroadcast_startup_message' for #<Gdk::DisplayX11:0xb7449a00 ptr=0x9f5d0c8>"> ---Backtrace--- /home/a/rpm/BUILD/ruby-gnome2-all-0.19.0/gtk/test/test_gdk_display.rb:28:intest_broadcast_startup_message’
/home/a/rpm/BUILD/ruby-gnome2-all-0.19.0/gtk/test/test_gdk_display.rb:27:in `test_broadcast_startup_message’
Ah. This test requires GLib fix.
I submitted a patch for that but it doesn’t checked yet.
http://bugzilla.gnome.org/show_bug.cgi?id=528020
We can omit the test. Could you add omit(…) in the test?
- Failure:
test_program_name(TestGtkAboutDialog)
[/home/a/rpm/BUILD/ruby-gnome2-all-0.19.0/gtk/test/test_gtk_about_dialog.rb:10]:
<“run-test.rb”> expected but was
<“ruby”>.
This test will work well with “cd gtk && test/run-test.rb”.
kou
Hi,
2009/5/27 Detlef R. [email protected]:
I’ve also run the test suite, but complete. All parts build and
installed successfully.
Thanks for confirming!
I got several unicode conversion errors, which are not popped up in
Pascals test, as this tests are skipped if the libuconv is not
installed.
It seems that endian is different.
To the poppler/test/test_page.rb that reports
- Omission: We doesn’t have a PDF that has a thumbnail…
I’ve tested the thumbnail extraction with poppler and Gdk::Pixbuf on
many documents and it works well. An example you can download here
http://www.datasheetdir.com/go/-xxx-x-lcvb.pdf . It contains thumbnails
for six pages. May be we should just drop this test (it works) if no one
can provide a free test PDF.
Does someone have a free test PDF?
If nobody have it, I’ll ask Calros, the maintainer of Poppler glib
bindings,
about it later.
kou
Hi,
2009/5/27 Detlef R. [email protected]:
I got several unicode conversion errors, which are not popped up in
Pascals test, as this tests are skipped if the libuconv is not
installed.It seems that endian is different.
My dev computer is a PowerPc mac (big-enidan)
OK. Maybe uconv doesn’t support big-endian.
We should use NKF.nkf instead of uconv.
e.g.
Uconv.u8tou4 → NKF.nkf(“-W -w16B”, string) on big-endian
Uconv.u8tou4 → NKF.nkf(“-W -w16L”, string) on little-endian
Because I am not in this test stuff and can’t really test cause I have
cairo installed, short question: OK to commit?
It seems that two “_” exists. (typo?)
Others are OK for me.
assert_kind_of(Cairo::ImageSurface, page.get_image(mapping.image_id)) assert_kind_of(Cairo::ImageSurface, mapping.image)
else
kou
Hi,
Kouhei S. schrieb:
It seems that two “_” exists. (typo?)
outch thanks for pointing this out!
Cheer, detlef
Hi,
Kouhei S. schrieb:
Hi,
2009/5/27 Detlef R. [email protected]:
I got several unicode conversion errors, which are not popped up in
Pascals test, as this tests are skipped if the libuconv is not
installed.It seems that endian is different.
My dev computer is a PowerPc mac (big-enidan)
[copied from other mail]
- Error:
test_get_image(TestPage):
NoMethodError: undefined methodget_image' for #<Poppler::Page:0xb65c52e0 ptr=0x85c2400> ./test/test_page.rb:6:intest_get_image’
This test requires cairo supported Ruby/Poppler.
We should check Ruby/Poppler supports cairo with
Poppler.cairo_available? If you can, could you fix the test to work on
non cairo supported Ruby/Poppler environment too.
Because I am not in this test stuff and can’t really test cause I have
cairo installed, short question: OK to commit?
ruby-gnome2/poppler/test/test_page.rb
ruby-gnome2-dev/poppler/test/test_page.rb
— ruby-gnome2/poppler/test/test_page.rb 2009-05-27
15:11:27.000000000 +0200 +++
ruby-gnome2-dev/poppler/test/test_page.rb 2009-05-23
03:36:31.000000000 +0200 @@ -2,7 +2,7 @@ def test_get_image
document = Poppler::Document.new(image_pdf)
page, mapping = find_first_image_mapping(document)
Cheers, detlef
Le mercredi 27 mai 2009 à 20:56 +0900, Kouhei S. a écrit :
Ah. This test requires GLib fix.
I submitted a patch for that but it doesn’t checked yet.
Bug 528020 – language bindings friendly gdk_x11_display_broadcast_startup_message()We can omit the test. Could you add omit(…) in the test?
Done
- Failure:
test_program_name(TestGtkAboutDialog)
[/home/a/rpm/BUILD/ruby-gnome2-all-0.19.0/gtk/test/test_gtk_about_dialog.rb:10]:
<“run-test.rb”> expected but was
<“ruby”>.This test will work well with “cd gtk && test/run-test.rb”.
hmm no it does not, I still get “ruby”
Hi,
Pascal T. schrieb:
- Failure:
test_program_name(TestGtkAboutDialog)
[/home/a/rpm/BUILD/ruby-gnome2-all-0.19.0/gtk/test/test_gtk_about_dialog.rb:10]:
<“run-test.rb”> expected but was
<“ruby”>.This test will work well with “cd gtk && test/run-test.rb”.
hmm no it does not, I still get “ruby”
this comes from gtk/src/gtkmain.c in the gtk_m_init function. There is
the following code block:
gargv = ALLOCA_N(char *, gargc + 1);
gargv[0] = RVAL2CSTR(rb_argv0);
for (i = 0; i < gargc; i++) {
if (TYPE(RARRAY_PTR(argary)[i]) == T_STRING) {
gargv[i+1] = RVAL2CSTR(RARRAY_PTR(argary)[i]);
}
else {
gargv[i+1] = "";
}
}
8<----------snip----------
is_initialized = gtk_init_check(&gargc, &gargv);
The value of rb_argv0 is pushed in the first position of the parameters
of the gtk_init_check call. At this point rb_argv0 is “ruby”.
If my traces are right, the gtk_m_init function is the first function in
ruby-gtks code space. So, for me, the rb_argv0 value seems unusable for
this task. AFAIK $0 always contains the programs name. So with this
little patch we get, what we expect.
— ruby-gnome2/gtk/src/rbgtkmain.c 2009-05-27
22:36:18.000000000 +0200 +++ ruby-gnome2-dev/gtk/src/rbgtkmain.c
2009-05-28 00:44:50.000000000 +0200 @@ -93,6 +93,7 @@
gint i, gargc;
VALUE argary;
gchar** gargv;
VALUE progname;
if (_initialized)
return self;
@@ -108,9 +109,10 @@
Check_Type(argary, T_ARRAY);
gargc = RARRAY_LEN(argary);
}
progname = rb_eval_string(“$0”);
gargv[0] = RVAL2CSTR(progname);
for (i = 0; i < gargc; i++) {
if (TYPE(RARRAY_PTR(argary)[i]) == T_STRING) {
I know, that the use of rb_eval_string may not be the right™ way,
but I don’t wanted to bother with this strange func pointers for hooked
variables. If someone wants to change this: feel free to do so.
Cheers, detlef
Hi,
2009/5/28 Detlef R. [email protected]:
Check_Type(argary, T_ARRAY); if (TYPE(RARRAY_PTR(argary)[i]) == T_STRING) {I know, that the use of rb_eval_string may not be the right™ way,
but I don’t wanted to bother with this strange func pointers for hooked
variables. If someone wants to change this: feel free to do so.
You can use rb_gv_get(“$0”).
kou
Hi,
2009/5/28 Pascal T. [email protected]:
/home/a/rpm/BUILD/ruby-gnome2-all-0.19.0/gtk/test/test_gdk_display.rb:27:in `test_broadcast_startup_message’
Ah. This test requires GLib fix.
I submitted a patch for that but it doesn’t checked yet.
Bug 528020 – language bindings friendly gdk_x11_display_broadcast_startup_message()We can omit the test. Could you add omit(…) in the test?
Done
Thanks!
I want Pascal or Detlef to release 0.19.0. Could you release it, please?
–
kou
Hi,
Kouhei S. schrieb:
You can use rb_gv_get(“$0”).
thanks, I didn’t knew this function.
A grep over the sources shows, that rb_argv0 is used in several other
places as the program name, too. May be they also need to be fixed.
If no one is faster than me (wink 
) I will take a look at it in
the next week.
Cheers, detlef
Hi,
Kouhei S. schrieb:
I want Pascal or Detlef to release 0.19.0. Could you release it, please?
I’m very sorry, but the time I can spend for ruby-gnome until next
Monday goes straight to zero. The only thing, I will commit the program
name fix in the next few minutes.
Cheers, detlef
Le jeudi 28 mai 2009 à 23:21 +0200, Detlef R. a écrit :
name fix in the next few minutes.
I can’t do it tonight but will try tomorrow
Le jeudi 28 mai 2009 à 23:27 +0200, Pascal T. a écrit :
Monday goes straight to zero. The only thing, I will commit the program
name fix in the next few minutes.I can’t do it tonight but will try tomorrow
I guess I’m not allowed to do so, rake release fails :
misc/release.rb:77:in find_target_release_link': can't find package ID (RuntimeError) from misc/release.rb:98:ingo_edit_release_page’
And I don’t see any link on the webpage to manage the files
Hi,
In 1243591164.12871.1.camel@plop
“Re: [ruby-gnome2-devel-en] 0.19” on Fri, 29 May 2009 11:59:23 +0200,
Pascal T. [email protected] wrote:
I guess I’m not allowed to do so, rake release fails :
misc/release.rb:77:in
find_target_release_link': can't find package ID (RuntimeError) from misc/release.rb:98:ingo_edit_release_page’And I don’t see any link on the webpage to manage the files
I’m sorry. I’ve update your permission.
Please try again!
kou
Le vendredi 29 mai 2009 à 22:51 +0900, Kouhei S. a écrit :
from misc/release.rb:98:in `go_edit_release_page’
And I don’t see any link on the webpage to manage the files
I’m sorry. I’ve update your permission.
Please try again!
rake release failed on submitting news :
misc/release.rb:207:in submit_news': undefined methodsummary=’ for
nil:NilClass (NoMethodError)
from misc/release.rb:235:in `main’
from misc/release.rb:239
rake aborted!
The tarballs are however available and checked and the release is tagged
I will update the wiki later and post a real announce here (except if
someone want to do it before 
)
This forum is not affiliated to the Ruby language, Ruby on Rails framework, nor any Ruby applications discussed here.
Sponsor our Newsletter | Privacy Policy | Terms of Service | Remote Ruby Jobs