Commit Graph

12003 Commits

Author SHA1 Message Date
Chris Young
2c380f818f Fix some memory leakage 2016-01-27 00:01:42 +00:00
Chris Young
780752d276 fix the incorrect copy'n'pasted parameters 2016-01-26 23:55:14 +00:00
Chris Young
49bd536c6e Move OS3 array to list into separate function 2016-01-26 23:49:24 +00:00
Chris Young
f3332a20d7 Don't read the values of gadgets which were not created. 2016-01-26 19:52:39 +00:00
Chris Young
04cbc08c57 Don't use ami_NewMinList as it potentially writes past the MinList structure on OS3.
Instead we use ami_AllocMinList and pointers.
2016-01-26 19:40:59 +00:00
Chris Young
520e81f98d Alloc specifying local charset on OS3 as we can't get this from Locale 2016-01-26 19:37:18 +00:00
Chris Young
9505b8e514 Make most of the prefs GUI display (and hopefully work) under OS3
page.gadget is not a public class in OS3
Some gadgets which will not work have been temporarily #ifdef'd for OS4 only
Some gadgets which are not relevant have been disabled to avoid random values being set
2016-01-26 19:28:07 +00:00
Chris Young
78dec4d515 Show warnings on OS3 with an EasyRequest 2016-01-26 00:10:11 +00:00
Chris Young
2c2dc0f2bc Bring OS3 about requester up on the correct screen 2016-01-25 19:14:04 +00:00
Chris Young
a7ab6c7484 Simple "about" requester for OS3 2016-01-25 19:12:54 +00:00
Chris Young
9879cbbfae Merge branch 'chris/fix-palmap-crash'
Fixes a problem with OS4 freezing up when running NetSurf on a palette mapped screen.
2016-01-25 18:55:30 +00:00
Chris Young
893df2ccf1 Revert "Plot local history background white instead of wholly transparent"
On second thoughts this was correct initially.

This reverts commit 1ea29c6aec.
2016-01-25 18:53:21 +00:00
Chris Young
6eb0bd4579 Remove unneeded pen list 2016-01-25 18:48:49 +00:00
Chris Young
1ea29c6aec Plot local history background white instead of wholly transparent 2016-01-25 18:43:43 +00:00
Chris Young
5c8dccd5f6 When forcing 32-bit render mode, don't create the BitMaps as friends of the screen BitMap.
This fixes a system freeze on OS4
2016-01-25 18:41:50 +00:00
Chris Young
beb6cebf1e conflict 2016-01-25 18:26:22 +00:00
Chris Young
09ac17955b Merge branch 'master' of git://git.netsurf-browser.org/netsurf into chris/fix-palmap-crash
Conflicts:
	amiga/bitmap.c
2016-01-25 18:25:26 +00:00
Vincent Sanders
6847fad653 remove javascript build option from jenkins script 2016-01-25 16:10:42 +00:00
Vincent Sanders
6d63f7959a make declarations match definitions for duk_raw_read_xxx_be 2016-01-25 15:49:31 +00:00
Vincent Sanders
822493af0d fixup some empty logging calls in atari hotlist code 2016-01-25 15:49:31 +00:00
Michael Drake
9e1622a679 Wrap verbose JS logging in debug macro. 2016-01-25 12:20:58 +00:00
Chris Young
8d71c0ae9e track type of native bitmap 2016-01-25 00:29:08 +00:00
Chris Young
c0d031bf73 Make the palette-mapped state relate to the rendering layer rather than globally.
This means we can still render in 32-bit mode internally when the destination isn't the screen.
NB: bitmap_render is currently freezing when the screen is in palette-mapped mode (old bug, pre-dates this change), so for now we skip this routine in that scenario.
2016-01-25 00:01:55 +00:00
Chris Young
452d8ce512 Use itempools for storing our pen locks 2016-01-24 00:00:27 +00:00
Michael Drake
9c57ddf0b2 1 isn't prime. 2016-01-23 17:44:42 +00:00
Chris Young
e38a1f17f2 friend_bitmap should default to false as it causes problems 2016-01-23 15:15:35 +00:00
Chris Young
dc89a2b2cc Move the OS3 statusbar underneath all the scrollbars 2016-01-23 15:11:17 +00:00
Chris Young
97e9086c80 remove missing os3 include 2016-01-23 15:04:45 +00:00
Chris Young
8df392d11d Remove nulling of a function which does actually exist in OS3 gfx.lib 2016-01-23 14:59:22 +00:00
Chris Young
dacdea768c fix argb_to_rgba definition weirdness 2016-01-23 14:58:47 +00:00
Chris Young
a68be9d11a Clear remaining bitmap fields 2016-01-23 14:49:29 +00:00
Chris Young
da9bee4286 Add more appropriate readme for OS3 build 2016-01-23 12:48:24 +00:00
Chris Young
9e5bf6fb68 Don't free the deferred rects using FreeVec on exit as they're itempools 2016-01-23 12:11:33 +00:00
Michael Drake
4b0ebae56a Add a couple of general JS tests.
The primes one prints its execution time:

  NetSurf without -v: 1300ish
             with -v: 5000ish
      Firefox 38.5.0:  300ish
2016-01-23 10:32:26 +00:00
Chris Young
5b422a3ec9 Fix opening of SSL cert window 2016-01-23 00:17:16 +00:00
Chris Young
7e7ea09000 Allocate generic list objects using itempools
TODO: Allocate the attached structures also using itempools
2016-01-22 18:40:40 +00:00
Vincent Sanders
c2bd86ca96 Fix atari mint duktape build 2016-01-22 13:47:02 +00:00
Chris Young
f51b7780e3 Don't clear memory twice under OS3 2016-01-21 22:39:21 +00:00
Michael Drake
b29bd85274 Disable JavaScript force-on for forthcoming release. 2016-01-21 22:15:14 +00:00
Chris Young
5a58e844e4 Don't clear bitmap structure on allocation 2016-01-21 22:02:39 +00:00
Chris Young
b41685208f Revert "alloc bitmap structures using itempools"
This reverts commit 0cc19b10fc.

This caused memory leakage, as the memory pool is removed before the core has destroyed all the bitmaps.
2016-01-21 20:42:49 +00:00
Chris Young
0cc19b10fc alloc bitmap structures using itempools 2016-01-21 20:30:02 +00:00
Chris Young
7efd9a2c67 Remove unused Amiga async scheduler 2016-01-21 20:13:49 +00:00
Chris Young
7da516067d Use itempools for the scheduler 2016-01-21 19:23:10 +00:00
Michael Drake
d9e92d2032 Add comment on on-demand JS context creation safety
Checking for enable_scripting isn't needed here.
2016-01-21 14:58:38 +00:00
Michael Drake
82b298ac2c Ensure we do on-demand JS context creation only when JS is enabled.
An alternative approach which may be better would be to create the
JavaScript context when the html_content is created, rather than
on demand.

This code checks for the JS context and creates one every time we
add a node to the DOM.  So when JS is on, every doc with a single
node in it has a JS context.  This seems to make on-demand creation
a redundant overhead.
2016-01-21 14:51:17 +00:00
Michael Drake
669448d7b6 Fix place where we're calling a js_* fn without checking for context. 2016-01-21 14:49:57 +00:00
Michael Drake
594012ef52 Ensure constant javascript_enabled for HTML contents.
Now we take the value of the javascript_enabled option when the
content is created.  We then use the content's script_enabled
boolean everywhere else.

This prevents us getting inconsistent values for javascript_enabled
if a user toggles the setting while a page is loading.

It was read frequently during box construction, and also the
parser's script enabled setting could change where we handled
a change of encoding.

Now we only care about the setting of the javascript_enabled
option at time of html_content creation.
2016-01-21 14:11:36 +00:00
Michael Drake
3dcf7d80a1 Pass html_content to box_extract_link. 2016-01-21 14:09:57 +00:00
Michael Drake
ad273a41e8 Simplify place_float_below loop condition. 2016-01-21 09:44:05 +00:00