summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
...
| * update amiga frontend to new set_scroll APIVincent Sanders2017-04-261-3/+18
| |
| * simplify the browser window operations by removing scroll APIVincent Sanders2017-04-264-110/+105
| | | | | | | | | | The browser window scrollingAPI was duplicated in window operation table, this simplifies it to a single set_scroll API.
* | HTML layout: Extend :after pseudo element handling to cover display:table.Michael Drake2017-04-261-2/+5
| | | | | | | | | | | | This is used in a common clearfix variant. http://nicolasgallagher.com/micro-clearfix-hack/
* | Box tree dump: If box has element, include element name in dump.Michael Drake2017-04-261-1/+8
| |
* | GTK: Optimise tiled bitmap plotting, and scaled bitmap rendering.Michael Drake2017-04-261-149/+79
|/ | | | | | We now let Cairo deal with the tiling. Also we don't keep cached scaled bitmaps any more. The speedup for small bitmap tile sizes is huge.
* update framebuffer frontend to remove reformat APIVincent Sanders2017-04-261-14/+13
|
* update beos frontend to remove reformat APIVincent Sanders2017-04-261-35/+25
|
* update atari frontend to remove reformat APIVincent Sanders2017-04-261-19/+10
|
* update monkey frontend to remove reformat APIVincent Sanders2017-04-261-11/+12
|
* update riscos frontend to remove reformat APIVincent Sanders2017-04-261-19/+10
|
* update amiga frontend to remove reformat APIVincent Sanders2017-04-261-29/+20
|
* update windows frontend to remove reformatVincent Sanders2017-04-261-21/+6
|
* update cocoa frontend with reformat API changeVincent Sanders2017-04-263-22/+13
|
* update GTK frontend for reformat API changeVincent Sanders2017-04-261-23/+20
|
* remove reformat from browser window operation tableVincent Sanders2017-04-265-41/+79
| | | | | | the reformat callback was completely unecessary and implementations appeared potentialy buggy. This rationalises the API and reduces the number of operations a frontend must provide.
* Core hotlist: Ensure any hotlist save callback is removed on hotlist_fini.Michael Drake2017-04-251-0/+4
|
* Hotlist: Save hotlist modifications.Michael Drake2017-04-251-71/+110
| | | | | | When URLs are added, or address entries are edited or deleted, a hotlist save is scheduled to happen after 10 seconds, if there isn't already a hotlist save scheduled.
* Windows: Update for core hotlist API change.Michael Drake2017-04-252-2/+3
|
* RISC OS: Update for core hotlist API change.Michael Drake2017-04-252-2/+3
|
* GTK: Update for core hotlist API change.Michael Drake2017-04-251-2/+3
|
* Cocoa: Update for core hotlist API change.Michael Drake2017-04-251-2/+2
|
* Atari: Update for core hotlist API change.Michael Drake2017-04-251-2/+2
|
* Amiga: Update for core hotlist API change.Michael Drake2017-04-251-2/+3
|
* Core hotlist API: Take save path at init, rather than fini.Michael Drake2017-04-252-8/+24
|
* Merge branch 'vince/invalidate-api'Vincent Sanders2017-04-2313-233/+288
|\
| * Update beos frontend to use invalidate window area APIVincent Sanders2017-04-231-26/+28
| |
| * Update framebuffer frontend to use invalidate window area APIVincent Sanders2017-04-231-14/+25
| |
| * Update monkey frontend to use invalidate window area APIVincent Sanders2017-04-231-14/+22
| |
| * Update windows frontend to use invalidate window area APIVincent Sanders2017-04-231-36/+27
| |
| * update gtk frontend with invalidate window API changeVincent Sanders2017-04-231-15/+25
| |
| * update atari frontend for invalidate window API changeVincent Sanders2017-04-231-23/+29
| |
| * update amiga frontend invalidate window API changeVincent Sanders2017-04-191-35/+48
| |
| * update riscos frontend to invalidate window APIVincent Sanders2017-04-193-49/+58
| |
| * replace redraw and update methods with invalidate in window table APIVincent Sanders2017-04-193-21/+26
| |
* | Duktape: Make declarations match definitions for fastint-enabled functions.Michael Drake2017-04-211-2/+2
| |
* | Duktape: Enable fastints.Michael Drake2017-04-201-0/+1
|/ | | | Suggested by svaarala.
* Adjust the requested class versions to be the minimum required.Chris Young2017-04-166-26/+50
| | | | | | | Some features of later versions are used but are either restricted to OS4 only, or runtime version checks select the appropriate routines. Where the lowest required version cannot be determined, the version provided in the ClassAct 3.3 archive is requested. Bugs may mean these older classes do not work as expected. In addition, restrict the opening of listbrowser.gadget and the custom Stringview class to OS4 only.
* Duktape: Make declaration match definition for duk_refzero_check_fast()Michael Drake2017-04-161-1/+1
|
* Duktape: Make declarations match definitions for duk_raw_read_xxx_beMichael Drake2017-04-161-6/+6
| | | | Restore 6d63f7959af64a45b0643d0610fcbdb0c07bfbc4 for 2.1.0 import.
* Duktape: Prevent clang static analysis.Michael Drake2017-04-161-0/+3
| | | | Restore 336326af3aab93f31474fa6de28782457ae4a1c0 for 2.1.0 import.
* Duktape: Update to 2.1.0 release.Michael Drake2017-04-163-5539/+7089
|
* GTK: Slight optimisation to nsfont_split().Michael Drake2017-04-111-1/+1
| | | | | When we aren't modifying the line, pango_layout_get_line_readonly() is a faster alternative to pango_layout_get_line().
* GTK: Font rendering: Use same pango layout for painting as for measuring.Michael Drake2017-04-111-7/+4
| | | | | | | | | | | | | | | With this change we are consistent about how the pango layout we use is created. Now it always comes from a pango_layout_new() call on a pango context that comes from gdk_pango_context_get(). Previously the pango layout used for painting came from a call to pango_cairo_create_layout(), which required a global called "current_cr" (a cairo drawing context), which is only valid during redraw (painting). Since it was only valid during painting, this source could not be used for pango layout creation for the measuring code.
* CSS: Handle presentational hint for HR's "width" attribute.Michael Drake2017-04-101-0/+1
|
* Presentational hints: Ensure length is initialised for auto margins.Michael Drake2017-04-061-0/+2
| | | | | | | | | | | Although the length is unused when margin is auto, having uninitialised values present in a computed style means that the style hashing may give the same style different hashes depending on the unintitialised info. The effect of this would have been be to reduce the chance of computed style sharing. It would have had no effect on page rendering.
* Support coverage flushing on assert()Daniel Silverstone2017-03-312-4/+68
| | | | | | | | | | | | When assert() is called, which is not uncommon in utility code within NetSurf, we lose coverage data for anything done before the assert() in the test. This commit corrects that oversight but is at least slightly GCC specific and may need tweaks for non-Linux platforms. By default, 'make coverage' will enable assert coverage, and it can be disabled with 'make coverage NOASSERTCOVERAGE=yes' if necessary. Signed-off-by: Daniel Silverstone <dsilvers@netsurf-browser.org>
* fix doc comment spelling mistakeVincent Sanders2017-03-281-1/+1
|
* nsurl: Remove redundant code path.Michael Drake2017-03-271-8/+9
|
* Tests: Increase nsurl coverage.Michael Drake2017-03-271-0/+11
|
* Tests: Add nsurl insensitivity test for HTTPS scheme.Michael Drake2017-03-271-0/+1
|