Michael Drake
35eb251244
Fix "error setting certificate verify locations" problem when the Choices file doesn't exist. Now there's a single place for front ends to set options overrides. Fix nsoption_setnull_charp leak.
2012-10-17 14:20:18 +01:00
Michael Drake
d9e7d58106
Fix up ripples from urldb change.
2012-10-11 11:20:02 +01:00
Daniel Silverstone
2363301c94
Show refcnt in lwc iterator at netsurf exit.
2012-07-22 20:24:33 +01:00
Michael Drake
2ad24f39da
Interned string cleanup, phase 2: Create utils/corestrings and use it in css handler.
2012-07-22 15:48:55 +01:00
Daniel Silverstone
8551b9dff0
Move js_finalise(). Fix provided by Ole Loots (mono)
...
svn path=/trunk/netsurf/; revision=14014
2012-06-29 20:37:32 +00:00
Vincent Sanders
4c89c9d5df
improve javascript support
...
svn path=/trunk/netsurf/; revision=13962
2012-06-10 22:17:30 +00:00
Vincent Sanders
67b6f8d4a3
remove unecessary includes
...
svn path=/trunk/netsurf/; revision=13556
2012-03-23 16:56:43 +00:00
Vincent Sanders
1490b52a6b
NetSurf options rework (a=vince r=daniels,jmb)
...
svn path=/trunk/netsurf/; revision=13548
2012-03-22 09:34:34 +00:00
Vincent Sanders
6ebaecccaf
Totaly prefunctry binding to spidermonkey
...
svn path=/trunk/netsurf/; revision=13360
2012-01-01 21:42:38 +00:00
Michael Drake
861f88aa43
Port llcache to nsurl.
...
svn path=/trunk/netsurf/; revision=12904
2011-09-28 11:26:30 +00:00
John Mark Bell
339bed7272
Virtualise content handler finalisation calls. Remove pointless implementations.
...
svn path=/trunk/netsurf/; revision=12797
2011-09-15 22:47:50 +00:00
Vincent Sanders
2d33a8f85a
Make high level cache, low level cache and image cache all be initialised from passed parameters
...
Calculate all cache sizes from single memory cache size option and sanity check
have a single global struct to hold all parameters instead of several individual variables
svn path=/trunk/netsurf/; revision=12784
2011-09-10 00:55:39 +00:00
Vincent Sanders
dec633ca8e
make high level cache take a parameter structure
...
make hlcache have a single global container
svn path=/trunk/netsurf/; revision=12779
2011-09-08 12:04:57 +00:00
Vincent Sanders
69bee6fb71
move bitmap cache initialisation up a level
...
improve stats reporting
svn path=/trunk/netsurf/; revision=12770
2011-09-07 09:15:33 +00:00
John Mark Bell
6807fa854d
Sniff content types where appropriate.
...
We never sniff for CSS, nor for non-page artefacts (e.g. treeview icons)
svn path=/trunk/netsurf/; revision=12707
2011-09-04 06:28:09 +00:00
John Mark Bell
b92ca9cd30
Clean up after content factory
...
svn path=/trunk/netsurf/; revision=12415
2011-05-16 21:16:44 +00:00
John Mark Bell
e71691bae8
Merge branches/jmb/content-factory to trunk
...
svn path=/trunk/netsurf/; revision=12283
2011-05-06 20:40:09 +00:00
Vincent Sanders
74ffc40e92
move logging initialisation
...
svn path=/trunk/netsurf/; revision=12152
2011-04-02 13:05:55 +00:00
Vincent Sanders
fff2520409
commandline option setting
...
svn path=/trunk/netsurf/; revision=12106
2011-03-18 21:58:29 +00:00
Daniel Silverstone
e083b503a0
Ensure we stop the hlcache schedule before we quit the gui
...
svn path=/trunk/netsurf/; revision=12036
2011-03-13 20:20:31 +00:00
Daniel Silverstone
93e9bfe323
Shunt the schedule function definitions to desktop/schedule.h. Shunt the hlcache/llcache to using schedule to get their cleanups run.
...
svn path=/trunk/netsurf/; revision=12029
2011-03-13 18:26:46 +00:00
John Mark Bell
56413fd61c
Lose nasty code
...
svn path=/trunk/netsurf/; revision=11984
2011-03-12 19:14:23 +00:00
John Mark Bell
ceadbafcbc
Clean up blank_import on exit
...
svn path=/trunk/netsurf/; revision=11818
2011-02-26 01:19:11 +00:00
Vincent Sanders
fbb1139214
add default system colour handlers to each frontend
...
svn path=/trunk/netsurf/; revision=11530
2011-01-30 13:40:47 +00:00
Daniel Silverstone
3e7bf7cfb5
Ensure we correctly release all icon resources associated with trees, SSL certs, search providers, etc.
...
svn path=/trunk/netsurf/; revision=11421
2011-01-20 13:51:41 +00:00
John Mark Bell
43ce8206b7
Whee. Another runtime that doesn't support stdout = stderr.
...
We really should stop doing this.
svn path=/trunk/netsurf/; revision=11148
2010-12-29 21:25:47 +00:00
John Mark Bell
c96f682d9d
Fix bug #3138151 : shut down caches before killing off fetch layer
...
svn path=/trunk/netsurf/; revision=11069
2010-12-16 00:03:57 +00:00
Michael Drake
5eaeaed909
Remove Hubbub and Wapcaplet initialisation and finalisation.
...
svn path=/trunk/netsurf/; revision=10980
2010-12-04 20:14:33 +00:00
Vincent Sanders
bda534e12a
Clean up how GTK frontend finds resources
...
svn path=/trunk/netsurf/; revision=10668
2010-07-28 15:22:44 +00:00
John Mark Bell
000e6ad3de
It turns out that realloc(ptr, 0) --> free(ptr) is not actually required by the C standard (whereas realloc(NULL, size) --> malloc(size) is).
...
Therefore, explicitly model the behaviour expected by our libraries (that realloc of 0 size is equivalent to free).
svn path=/trunk/netsurf/; revision=10524
2010-04-30 07:00:58 +00:00
John Mark Bell
2276afb9f6
Log any remaining interned strings (there should be none)
...
svn path=/trunk/netsurf/; revision=10502
2010-04-27 22:14:52 +00:00
John Mark Bell
0de2013f25
Clean up caches on exit
...
svn path=/trunk/netsurf/; revision=10494
2010-04-27 06:48:09 +00:00
John Mark Bell
ad8d0b3350
Handle llcache queries (GTK only for now)
...
svn path=/trunk/netsurf/; revision=10403
2010-04-14 23:03:49 +00:00
John Mark Bell
9e8849c04c
Take the easy way out wrt loading messages files: leave it to the RISC OS frontend to do this.
...
svn path=/trunk/netsurf/; revision=10374
2010-04-11 22:40:41 +00:00
John Mark Bell
71de3618f1
Implement hlcache_poll(), which drives the low-level cache event loop, and attempts to clean the high-level cache.
...
Call this, instead of llcache_poll().
svn path=/trunk/netsurf/; revision=10371
2010-04-11 20:21:13 +00:00
Chris Young
103998165a
Revert previous change
...
svn path=/trunk/netsurf/; revision=10215
2010-03-30 18:20:48 +00:00
Chris Young
9b9c26b569
Avoid crash if pargc is NULL. This might happen, for example, when launching NetSurf
...
from Workbench under AmigaOS :)
svn path=/trunk/netsurf/; revision=10214
2010-03-30 18:14:40 +00:00
Vincent Sanders
e00fb7bd2d
second pass at startup refactor removing the gui_init callback
...
svn path=/trunk/netsurf/; revision=10205
2010-03-29 20:03:44 +00:00
Vincent Sanders
7e67527267
first part of initialisation refactor move gui_init2 into frontends
...
svn path=/trunk/netsurf/; revision=10202
2010-03-29 16:59:30 +00:00
Daniel Silverstone
cb7412181f
Replace fetch_poll with llcache_poll throughout. Allow CONTENT_OTHER to complete properly.
...
svn path=/trunk/netsurf/; revision=10187
2010-03-28 16:00:54 +00:00
Daniel Silverstone
05a4eceae2
Remove fetchcache.[ch]
...
svn path=/trunk/netsurf/; revision=10183
2010-03-28 15:34:07 +00:00
Daniel Silverstone
270ef59a98
Merge jmb/new-cache; r=dsilvers,rs=vince
...
svn path=/trunk/netsurf/; revision=10180
2010-03-28 12:56:39 +00:00
Vincent Sanders
55162445db
Add Windows frontend
...
svn path=/trunk/netsurf/; revision=9940
2010-01-31 00:36:02 +00:00
Vincent Sanders
e01d93f344
Refactor main into the frontends
...
remove unused embedded global
svn path=/trunk/netsurf/; revision=9851
2010-01-20 20:32:30 +00:00
Rob Kendrick
c72b3cf71a
OpenBSD 4.4 compilation fixes. Does not deal with lack of -Wextra or -Wstrict-aliasing for lpu or hubbub.
...
svn path=/trunk/netsurf/; revision=6673
2009-03-02 19:26:11 +00:00
François Revel
e4407cd7e9
Had to split main() into parts to have netsurf_init() called from one thread, and the loop in another to avoid a deadlock in replicants... hope nobody minds.
...
svn path=/trunk/netsurf/; revision=5586
2008-10-16 20:49:12 +00:00
François Revel
6330f8e475
Have netsurf_main() do the real work and main() call it. This allows ports to call it from elsewhere to embed NetSurf in another application.
...
The netsurf_embedded bool remembers it for later (like, we shouldn't probably save the hotlist and preferences in that case).
svn path=/trunk/netsurf/; revision=5579
2008-10-16 11:10:43 +00:00
John Mark Bell
6c49f1fcd5
r4153 was mostly right. It does help, however, if you actually force the locale to "C" in the first place. The issues listed in the commit log for r4153 still apply. This should fix the core's parsing of floats in locales that use something other than '.' as the decimal separator.
...
svn path=/trunk/netsurf/; revision=5568
2008-10-13 23:32:21 +00:00
Rob Kendrick
e7c5e16b9b
Performance improvements: rather than calling content_clean() every poll, we now call it no more frequently than once every 5 seconds. Additionally, we cache the result of talloc_total_size() in content_clean() rather than calculating it twice. On large documents, this function took 25% of CPU time. This makes the fetching/rendering/scrolling/redrawing of large documents over twice as fast.
...
svn path=/trunk/netsurf/; revision=4527
2008-07-07 14:05:29 +00:00
John Mark Bell
23fb72ea6b
The core code has always assumed a locale of "C".
...
Do not change the locale globally, else things will break in weird and
wonderful ways.
Introduce utils/locale.[ch], which provide locale-specific wrappers for various
functions (currently just the <ctype.h> ones).
Fix up the few places I can see that actually require that the underlying
locale is paid attention to.
Some notes:
1) The GTK frontend code has not been touched. It is possible that reading of
numeric values (e.g. from the preferences dialogue) may break with this
change, particularly in locales that use something other than '.' as their
decimal separator.
2) The search code is left unchanged (i.e. assuming a locale of "C").
This may break case insensitive matching of non-ASCII characters.
I doubt that ever actually worked, anyway. In future, it should use
Unicode case conversion to achieve the same effect.
3) The text input handling in the core makes use of isspace() to detect
word boundaries. This is fine for western languages (even in the C locale,
which it's currently assuming). It will, however, break for CJK et. al.
(this has always been the case, rather than being a new issue)
4) text-transform uses locale-specific variants of to{lower,upper}. In future
this should probably be performing Unicode case conversion. This is the
only part of the core code that makes use of locale information.
In future, if you require locale-specific behaviour, do the following:
setlocale(LC_<whatever>, "");
<your operation(s) here>
setlocale(LC_<whatever>, "C");
The first setlocale will change the current locale to the native environment.
The second setlocale will reset the current locale to "C".
Any value other than "" or "C" is probably a bug, unless there's a really
good reason for it.
In the long term, it is expected that all locale-dependent code will reside in
platform frontends -- the core being wholly locale agnostic (though assuming
"C" for things like decimal separators).
svn path=/trunk/netsurf/; revision=4153
2008-05-13 14:37:44 +00:00