Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
are called p5-*.
I hope that's all of them.
|
|
requested by Thomas Klausner.
|
|
Changelog:
pdf2djvu (0.7.14) unstable; urgency=low
* Ignore links with no action. (Previously such links would cause segfault).
Thanks to Krzysztof Szafran for the bug report.
* Update German translation. Thanks to Chris Leick.
* Update Russian translation. Thanks to Kyrill Detinov.
* Update Ukrainian translation. Thanks to Serhij Dubyk.
-- Jakub Wilk <jwilk@jwilk.net> Tue, 18 Sep 2012 17:27:28 +0200
|
|
|
|
|
|
_XOPEN_SOURCE_EXTENDED=1 in C99 code.
Fixes package on Solaris.
|
|
When run 'make package' as root, avoid /usr/bin/acroread symlink creation.
|
|
|
|
|
|
use the right libusb
|
|
- Documentation updates (STR #4112, STR #4130, STR #4134)
- Fixes for libusb-based USB backend (STR #4128)
- The lpq command did not show the owner or title of jobs unless passed
a username on the command-line (STR #4135)
- Localized empty strings contained the message catalog metadata
(STR #4119)
- Fixed a crash in the libusb-based USB backend (STR #4099)
- The cups-lpd mini-daemon no longer handled jobs with multiple copies
(STR #4118)
- Multiple libusb backend fixes (STR #4098, STR #4100)
- The IPP backend no longer tries to get the job status for printers
that do not implement the required operation (STR #4083)
- Sending a document in an unsupported format to an IPP printer now
automatically cancels the job (STR #4093)
- Fix some error reporting issues when printing from /dev/null and
other unusual situations (STR #4015)
- The scheduler now sets the CUPS_MAX_MESSAGE environment variable for
filters (STR #4074)
- Fixed a build issue when using older versions of autoconf (STR #4084)
- The IPP backend now treats the client-error-not-possible status code
as a job history issue, allowing IPP printing to Windows to work
(STR #4047)
- The IPP backend incorrectly included the document-format and
compression attributes in Create-Job requests (STR #4086)
- The libusb-based USB backend did not work on non-Linux platforms
(STR #4088)
- httpReconnect() did not reset the read/write buffers (STR #4065)
- Compiling without threading support failed (STR #4060)
- Fixed compile problem with old versions of OpenSSL (STR #4036)
- The network backends did not check SNMP supply levels regularly
(STR #4040)
- The online help always included the "help on help" text (STR #4042)
- Fixed a SSL handshake issue on OS X (STR #4045)
- The scheduler could crash if a PPD file contained an invalid paper
size (STR #4049)
- The CUPS polling daemon did not reinitialize its connection to the
remote server on errors in all cases (STR #4031)
- PostScript auto-configuration was slow and unreliable with some
printers (STR #4028)
- Missing localizations caused empty output (STR #4033)
- The cups-driverd program could temporarily "forget" a PPD file if it
was updated in place.
- The dnssd backend now prefers IPPS over IPP.
- The USB backend now uses and requires LIBUSB 1.0 or later (STR #3477)
- The LIBUSB-based USB backend now supports the back-channel (STR #2890)
- Changed how timeouts are implemented in the LPD backend (STR #4013)
- Added more supported color names for SNMP supplies (STR #3981)
- The default InputSlot setting was never used (STR #3957)
- POSIX ACLs are now set properly on certificate files (STR #3970)
- Supplies with commas in their names were not reported correctly
(STR #4020)
- The cupsGetPPD3() function created a temporary file when one was not
needed (STR #4018)
- The scheduler now ensures that job notifications contain a value for
the notify-printer-uri attribute (STR #4014)
- The lp and lpr commands did not cancel jobs queued from stdin on an
error (STR #4015)
- Fixed the IPP backend's handling of HTTP/1.0 compatibility (STR #3988)
- The IPP backend did not always setup username/password authentication
for printers (STR #3985)
- The IPP backend no longer re-queues print jobs that are too large for
the printer/server (STR #3977)
- The RPM spec file did not work (STR #4021, STR #4057)
- Encryption did not work when the server name ended with "."
(STR #4011)
- The multi-purpose tray is now mapped to the IPP "by-pass-tray"
(STR #4009)
- The correct media size was not always passed to IPP printers
(STR #4001)
- Finishing options were not passed to IPP printers (STR #3995)
- Fixed iCloud-based Back to My Mac printing (STR #3996)
|
|
v1.2.0 (28th AUgust 2012)
- Feature: correctly extract text using surrogate pairs and ligatures
(thanks Nathaniel Madura)
- Speed optimisation: cache tokenised Form XObjects to avoid re-parsing them
- Feature: support opening documents with some junk bytes prepended to file
(thanks Paul Gallagher)
- Acrobat does this, so it seemed reasonable to add support
|
|
|
|
(additionaly, reset PKGREVISION of qt4-* sub packages from base qt4 update)
|
|
to avoid blocked bulk builds.
|
|
|
|
This is in modular form, with the necessary dependencies noted.
|
|
|
|
|
|
changes:
-Improvements regarding embedded file handling
-bugfixes
-cleanup
pkgsrc note: shlib major changes, revbumps needed. also switch to
lcms2 for the "cms" option
|
|
|
|
|
|
crashes or possible code injection (CVE-2012-4405)
bump PKGREV
|
|
|
|
Upstream changes:
2012-08-29 Andrew Ford <a.ford@ford-mason.co.uk>
* version 0.06
* added script "latex-encode" as a simple filter that uses LaTeX::Encode
* added a couple more encodings
* reverted to 'base' from 'parent' for setting up inheritance
* updated author tests
|
|
This is a network support improvement release!
=== Changes
==== Ruby/GLib2
* Improvements
* windows: Bundled glib-networking.
==== Ruby/GIO2
* Improvements
* Added Gio::TlsCertificate.
* Added Gio::TlsCertificateFlags.
==== Ruby/GTK2
* Improvements
* Added Gtk.show_uri.
* Fixes
* Gtk::TreeStore#reorder:
Fixed wrong the number of reorder values check.
[SF.net#3561748] [Reported by Nounou]
|
|
Upstream changes:
2012-08-28 Andrew Ford <a.ford@ford-mason.co.uk>
* version 0.05
* added more symbols
* build-character-table now builds a %provided_by map that gives the LaTeX package
that provides the LaTeX command used in the encoding
* removed unimplemented "use_textcomp" option
* added "packages" option to latex_encode() to pass a reference to a hash that is
updated with the names of optional LaTeX packages (styles) that implement the
commands included in the encoded string
2012-08-27 Andrew Ford <a.ford@ford-mason.co.uk>
* version 0.04
* added requirement for Perl 5.8.1 or later due to deficiencies in
Unicode handling of earlier Perl versions.
* updated encoding table to add Turkish characters (RT #55526) and to not include
spaces after accented characters (RT #49357)
* switch from 'base' to 'parent' for setting up inheritance
* added add_latex_encodings() and remove_latex_encodings() functions to update the
encoding table.
* allow 'add' and 'remove' specifiers to be provided on the 'use' statement
|
|
Bump PKGREVISION.
|
|
|
|
--gc-sections on netbsd because --gc-sections doesn't work with the
netbsd-4 and netbsd-5 linker. See PR 46698 and PR 40401.
Bump PKGREVISION because this changes the compiled package on netbsd-6.
|
|
|
|
changes:
The command line tools have been combined into one tool that does all:
mubusy. Where previously you would have called "mupdfclean blah blah
blah", now call "mubusy clean blah blah blah" instead.
Support for UserUnits
bugfixes and optimisations
|
|
This fixes PR 46787.
|
|
avoid to specify nasty C string macro in the command line.
|
|
Changes:
- `c' config file include directive.
|
|
Remove a dependency to poppler-0.16 in the poppler-qt.pc file so
that pkg-config --cflags "poppler-qt >= 0.3.1" returns valid information.
Default poppler version is now 0.20.x. Since this package explicitly uses
0.16.x, @PC_REQUIRES_PRIVATE@ introduces a dependency to poppler-0.16.x in
the .pc file; pkg-config will then print messages such as "Package
'poppler-qt' requires 'poppler = 0.16.7' but version of poppler is 0.20.2"
instead of the required information. Removing the Requires.private line
entirely fixes the issue.
After applying this patch and rebuilding poppler-qt, graphics/kdegraphics3
and misc/koffice now build on DragonFly. Previously they failed with
poppler-related errors.
Patch provided by F.Tigeot.
|
|
|
|
|
|
fixes build w/ recent poppler
|
|
|
|
|
|
changes: minor fixes, cleanup
pkgsrc note: shlib major changed (for no obvious reason), need revbumps
|
|
bump PKGREV
|
|
libspectre 0.2.7 (08 August 2012)
=================================
This is another bugfix only release in the libspectre's 0.2 series.
This release fixes two crashes than can happen when %%Pages: or
%%Page: comments are missing in the document or when %%Pages: comment
is present but it's invalid.
|
|
Changelog:
2.1.1
- Tiny bug fix to make sure that the help file is in the right language.
2.1.0
- Dragging and and dropping files from Windows Explorer, Thunar, etc.,
now works (at least for me, on Windows 7 and Xubuntu).
- Updated Czech translation from Pavel Fric.
- French translation from Pierre-Alain Bandinelli.
- German translation (and many suggestions) from Rainer Krachten.
- Translations should "just work"; i.e., if you use a French locale then
the user interface and help text should appear in French. However, you
can force DiffPDF to use any language it supports by using the
--language command line option (e.g., --language=fr). If an
unsupported locale or --language option is used, DiffPDF will fall
back to English.
- Now support -h as a synonym for --help.
- Initial path defaults to home directory instead of DiffPDF's
installation directory.
- Improved the Options dialog's layout.
- A subtle bugfix to the --words command line option.
- Fixed a crash: clicking to set a margin when there're no PDFs loaded
now safely does nothing.
2.0.0
- Can now have comparisons exclude text that's outside user-specified
margins. This feature was sponsored by a company that prefers to
remain anonymous.
- Save As can now save images as well as PDFs.
- Minor bug fixes and improvements.
- Command line help (--help) will no longer work on Windows. This is to
avoid a spurious console window appearing. I've put the output in the
online help though.
- Improved dock window handling so that docks can now be stacked on top
of each other (useful for small screens).
|
|
The package assumes that if the dc program doesn't support -V, it must
not support -e either. This is a bad assumption; DragonFly supports
-e but doesn't support -V. Rather than try to fix a bad conftest, just
remove it completely. There's a previous test for dc presence. If dc
doesn't support -e, let it break during the build rather than before it.
It doesn't take long to build.
|
|
WRKSRC should be the toplevel, while CONFIGURE_DIRS can be used
for sub-hierarchies. Make it so.
|
|
|
|
|