Archives/koreader
2
0
Fork 0
mirror of https://github.com/koreader/koreader synced 2024-11-02 15:40:16 +00:00
Code Issues Projects Releases Wiki Activity
1,280 Commits 1 Branch 262 Tags 936 MiB
fb48b270ef
Commit Graph

1280 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
NiLuJe
b0f7ea5d81 revert the whole 'let's use ld' thing. It's useless, and might wreak 
havoc on EMU builds picking up system libs instead of the bundled ones.
 2012-10-03 12:52:39 -04:00
NiLuJe
9169ac7541 Make sure we're always building luajit statically, so ld doesn't pick a 
shared version. Of course that explained the size difference :D.
 2012-10-03 12:52:18 -04:00
NiLuJe
5a2e4dec99 Fix the luajit clean fix (forgot to quote *_*). 
Let ld do its job when linking, don't pass the static libs as object
files (we end up with a smaller native binary).
 2012-10-03 12:51:27 -04:00
NiLuJe
e29052dbbc Mention open_noshell in the README 2012-10-03 12:50:48 -04:00
NiLuJe
679fce3df4 Make sure we're building for the softfp float ABI, and not in Thumb. 
Might fix #359
 2012-10-03 12:50:17 -04:00
NiLuJe
b5134ee4f2 Honor toplevel settings re: CC/FLAGS when building libdjvu. Define 
_GNU_SOURCE to fix build with CS/MG TC. Fixes #341
 2012-10-03 12:49:50 -04:00
NiLuJe
51345c6f08 Let's do a distclean, actually. Less noise in git status that way. 2012-10-03 12:49:27 -04:00
NiLuJe
9173f0a549 Don't require a cross-tc for cleanthirdparty (still confuzzled about why 
luajit needs CFLAGS for a clean target, but, hey!). Fixes #357
 2012-10-03 12:48:38 -04:00
Dobrica Pavlinušić
40bdee5c1f Merge pull request #361 from houqp/new_ui_code 
merge latest changes from master branch
 2012-10-03 06:36:40 -07:00
Tigran Aivazian
334bedee97 Mention dependency on svn as required for popen_noshell.c 
When we run make fetchthirdparty it does svn co of popen-noshell
 2012-10-03 05:41:33 -04:00
Qingping Hou
7352ff8a65 revert commit 9127d1c9a1 
As NiLuJe points out, that line is not needed, clean
stage of the popen Makefile will take care of it.
 2012-10-03 05:36:59 -04:00
NiLuJe
bccccba794 Reimplement kpdf.sh changes from PR#348 after the PR#352 merge, and fix 
customupdate to actually bundle kpdf.sh in the new location

Conflicts:

	launchpad/kpdf.sh
 2012-10-02 21:40:05 -04:00
Qingping Hou
6c0670740d revert wrong conflict merge for New InfoMessage Concept 2012-10-02 21:21:59 -04:00
Qingping Hou
e82121b72e add HOSTAR for emu mode complie 2012-10-02 21:15:37 -04:00
Qingping Hou
9a1e581f7a add popennslib to cleanthirdparty 2012-10-02 21:14:21 -04:00
NiLuJe
30c3497654 Slight syntax tweak 2012-10-02 21:05:05 -04:00
NiLuJe
6041e5c0eb Forgot to revert that too, useless now :). 2012-10-02 21:04:55 -04:00
NiLuJe
4c5341fb30 Hello, tiny stupid typo! 2012-10-02 21:04:45 -04:00
NiLuJe
9f5cd9d021 Actually, no that's wrong, it's not the parent process, it's a fork. 2012-10-02 21:04:24 -04:00
NiLuJe
2572dd0657 Kill a bit of cruft, explain the reasonning behind the kill/traps, and 
fix a potential buffering issue eating the first slider event
 2012-10-02 21:04:15 -04:00
NuPogodi
0eaf2671fd New InfoMessage Concept 
Part I (still no unireader, crereader, filesearcher & fileinfo)

Conflicts:

	commands.lua
	dialog.lua
	filechooser.lua
	filehistory.lua
	font.lua
	inputbox.lua
	reader.lua
	resources/info-i.png
	screen.lua
	selectmenu.lua
 2012-10-02 21:03:56 -04:00
NiLuJe
4310b68b3b Don't checkout popen_noshell twice (it reverts parts of our patch) 2012-10-02 20:57:06 -04:00
NiLuJe
84e107313c Don't patch popen-noshell twice 2012-10-02 20:56:58 -04:00
NiLuJe
8250d77390 Properly fix the luajit *FLAGS mess 2012-10-02 20:56:46 -04:00
NiLuJe
3672d79fef Restore Makefiles tweaks 2012-10-02 20:56:37 -04:00
NiLuJe
0a14cf7eac Ignore untracked popen_noshell stuff 2012-10-02 20:54:50 -04:00
NiLuJe
10c9b3a58c Fetch popen_noshell during fetchthirdparty, don't bundle it in our repo 2012-10-02 20:53:42 -04:00
NiLuJe
18ed67ba2c Fix a stupid typo 2012-10-02 20:52:59 -04:00
NiLuJe
ead342926c Use popen_noshell as a static lib instead of an object file. Use CHOST 
instead of HOST (and allow it to be set from the env), use $(MAKE)
instead of make to allow using the jobserver properly, and remove the
dash from commands where we do care about the return code (or inhibit
errors the usual way: rm -f instead of -rm)

Conflicts:

	Makefile
 2012-10-02 20:52:49 -04:00
NiLuJe
cb03b33052 Add a small makefile to build popen_noshell as a static lib 2012-10-02 20:49:19 -04:00
NiLuJe
ee7122d6c4 Revert Makefile changes potentially unsuitable for upstream 2012-10-02 20:49:12 -04:00
NiLuJe
7ce8ff58f8 Update the standalone testcase, too. 2012-10-02 20:49:04 -04:00
NiLuJe
1d29b53b84 Kill debug printf 2012-10-02 20:48:57 -04:00
NiLuJe
26e10c6069 Fix input device closing, and fix lipc-wait-event handling (using 
popen-noshell from http://code.google.com/p/popen-noshell/)

Conflicts:

	input.c
 2012-10-02 20:48:43 -04:00
NiLuJe
acae9e618b Makefile tweaks for my TC 2012-10-02 20:43:48 -04:00
Dobrica Pavlinusic
c6ead8690e use git describe to generate version 
Last release had git annotate tag (added with git -a) so this is
now much better alternative to just git sha1
 2012-10-02 20:18:21 -04:00
Tigran Aivazian
47e5065d6a Be minimalistic wrt LDFLAGS 
Passing CFLAGS and LDFLAGS to kpdfview link stage has no effect in our
particular case, so I ripped it out, leaving it only for the crengine.
 2012-10-02 20:18:21 -04:00
Tigran Aivazian
d5fb6c846f Enable LTO for kpdfview and crengine 
Enaling link time optimizations for kpdfview binary and crengine library
causes another slight performance increase --- almost negligible (1-2ms
per page) but still noticeable by precise measurements.
 2012-10-02 20:18:21 -04:00
Tigran Aivazian
15baa52889 ARM-specific optimizations 
Thanks to NiLuJe who pointed out that our generic -march=armv6 can be
replaced (for K2/K3/DX/DXG) with a more specific optimization:
-march=armv6j -mtune=arm1136jf-s -mfpu=vfp. This I have now done and
also passed ARM_CFLAGS value to CXXFLAGS which is then passed to
crengine build. Tested, works fine. The performance improvement is
negligible (a few ms per page, but seems to be consistently better, i.e.
not just plus/minus fluctuations).
 2012-10-02 20:18:21 -04:00
Tigran Aivazian
5dc488e8c8 Build mupdf in release mode, not debug. 
Thanks to NiLuJe who pointed out that we are building mupdf in debug
mode. Switching to "release" build reduced the size of the kpdfview
binary and did not cause any performance degradation (but no noticeable
improvement either --- the page handling times seem to be exactly the
same, i.e. fluctuating a couple of ms in both directions).
 2012-10-02 20:18:21 -04:00
Tigran Aivazian
b7a763cefe Comment out debug printf() in pdf.c 
The printf() about the number of links on a page is very useful, but
only for debugging. As other instances of debugging printf() in djvu.c
and pdf.c are commented out, it seemed consistent to do the same with
this one.
 2012-09-30 03:42:31 +08:00
Tigran Aivazian
69d293a6cc Get rid of the compilation warning. 2012-09-30 03:40:40 +08:00
Dobrica Pavlinusic
0230ccfb59 add LF to end of debug message 2012-09-30 03:38:50 +08:00
Dobrica Pavlinusic
05e0647b9c getPageLinks implementation for mupdf #72 
This is rough first draft, and provides just dump of all links
available on pdf page. Binding to Shift+L is temporary and just for
debugging!

Conflicts:
	pdfreader.lua
	unireader.lua
 2012-09-30 03:37:37 +08:00
Dobrica Pavlinusic
db7e3770ad fix indenting to tabs 2012-09-30 03:36:58 +08:00
Tigran Aivazian
2ac7c7d4cf Optimize compilation of cre.cpp 
1. Remove unused "-lsdtc++" from the compilation stage as no linking is
done then.
2. Add our standard CFLAGS to the compilation of cre.cpp which brings in
-O3 -march=armv6 which enables optimizations.
It seemed strange that we compile a cpp file with gcc (as opposed to
g++), but I left it as is for now.
 2012-09-30 03:30:48 +08:00
Tigran Aivazian
c5565146f9 Typo: --disable-largefiles -> --disable-largefile 
As a result of a simple typo we were continuing to include LFS
support unnecessarily...
 2012-09-30 03:30:00 +08:00
Qingping Hou
8014773804 update gitignore 2012-09-29 01:10:19 +08:00
Dobrica Pavlinušić
fe7292f55c Merge pull request #330 from houqp/new_ui_code 
merge latest patches from master
 2012-09-27 00:42:29 -07:00
Dobrica Pavlinusic
66efe66f61 invertRect should handle negative x or y values 
two-column mode combined with showing links in pdf files triggers this bug
 2012-09-27 12:42:26 +08:00