summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* change the persistant data store to owning the allocationsVincent Sanders2014-11-224-137/+263
|
* Ensure we're not running events ahead of time.Chris Young2014-11-221-2/+9
|
* Process scheduled event changes before incoming timer events.Chris Young2014-11-221-10/+11
| | | | This stops things being run that have just been deleted. However, may cause the next event to run early because we don't check the time.
* Re-enable asynchronous backing store writesChris Young2014-11-225-7/+124
|\
| * Fix llcache ops tableChris Young2014-11-221-0/+1
| |
| * Revert "Revert "Write out to the backing store asynchronously.""Chris Young2014-11-225-7/+123
|/ | | | This reverts commit 75623179aa7a0259477ef93dcd2a3562c4884c74.
* fix warningChris Young2014-11-221-0/+1
|
* Reduce stack to 2KChris Young2014-11-221-1/+1
|
* Reply to the timer before running the scheduled task.Chris Young2014-11-221-1/+4
|
* New asynchronous schedulerChris Young2014-11-223-66/+286
|\ | | | | | | This ensures that if other processes other than the main NetSurf process try to create scheduled tasks, they are always run on the main process.
| * New schedulerChris Young2014-11-223-57/+186
| |
| * Initial steps for a scheduler processChris Young2014-11-223-38/+129
|/
* Revert "Should be a while loop in case of multiple simultaneous event triggers."Chris Young2014-11-221-2/+2
| | | | | | This reverts commit 9dcf001a0bf6d09864783125390e20c8eb53c12f. For ehatever reason this is entirely broken, but I'm in the process of rewriting so just revert for now to get the working builds again.
* Fix brackets.Michael Drake2014-11-221-1/+1
|
* Shouldn't need browser_private.h now.Michael Drake2014-11-221-1/+1
|
* Use API to find if content is selectable, instead of guessing.Michael Drake2014-11-221-1/+1
|
* Simplify ami_history_open API and use correct call to get history object.Michael Drake2014-11-224-7/+11
|
* Use API to get URL.Michael Drake2014-11-221-3/+2
|
* Use browser_window_get_content where we need to.Michael Drake2014-11-221-5/+5
|
* Should be a while loop in case of multiple simultaneous event triggers.Chris Young2014-11-211-2/+2
|
* Should be free of reliance on browser window internals here now.Michael Drake2014-11-211-1/+1
|
* Use browser_window_get_content() to get the current content.Michael Drake2014-11-211-1/+1
| | | | Although I dislike this API, and it may go at some point.
* Use API for getting browser window history.Michael Drake2014-11-211-5/+3
|
* Use browser_window_get_url() instead of getting URL from current content.Michael Drake2014-11-211-5/+5
|
* Use browser_window_can_select(), instead of assuming based on content type.Michael Drake2014-11-211-2/+1
|
* fix missing semicolonVincent Sanders2014-11-211-1/+1
|
* Shouldn't need browser_private.h now.Michael Drake2014-11-211-1/+1
|
* Use browser_window_stop_available()Michael Drake2014-11-211-1/+1
| | | | | Looking at bw->loading_content was bad and wrong in the cases of frames, and object fetches.
* Use API to get content for view source.Michael Drake2014-11-211-1/+1
|
* Use API to get current scale.Michael Drake2014-11-211-3/+3
|
* remove atari need to use browser internal data structuresVincent Sanders2014-11-212-16/+10
|
* Remove usage of browser private interfacesVincent Sanders2014-11-213-12/+15
| | | | | | | | The cocoa frontend was directly acessing browser internals instead of using the API. In the case of gui.m there was a check for the browser window not being root (browser->parent != NULL) . As gui windows can only ever be associated with the root brower window (i.e. its parent will always be NULL) this was completely redundant.
* Use public bw header.Michael Drake2014-11-211-1/+1
|
* Should be able to switch to public bw header now.Michael Drake2014-11-211-1/+1
|
* Use browser_window_get_url, instead of fishing content out of bw and getting ↵Michael Drake2014-11-211-1/+1
| | | | URL from that.
* Use public bw header.Michael Drake2014-11-211-1/+1
|
* Don't appear to need browser_private.h here.Michael Drake2014-11-211-1/+0
|
* update entry points to backing store ready for allowing differing object ↵Vincent Sanders2014-11-204-15/+92
| | | | lifetimes
* Revert "Write out to the backing store asynchronously."Chris Young2014-11-195-123/+7
| | | | | | This reverts commit 1ddf8215cfde443a48043274cef17e45c68f0c91. Appears to be unsafe to run this in a new process.
* Limit creation of store path directories.Vincent Sanders2014-11-191-9/+10
| | | | | Only create all elements of a path in the store when it is being opened for creation instead of causing reads to create directories.
* Improve backing store control data serialisationVincent Sanders2014-11-191-60/+90
| | | | | | | By scheduling the control data to be maintained (entries index written and headers updated) once activity occurs to update these control structures rather than a single serialisation at browser exit the data is more likely to be up to date and not lost on a crash.
* make backing store use rename to atomicly replace entries fileVincent Sanders2014-11-191-13/+38
|
* Write out to the backing store asynchronously.Chris Young2014-11-185-7/+123
| | | | | | | | | | | | Despite the fact AmigaOS filesystems run as separate processes, DOS filesystem writes occur synchronously waiting for the disk write to finish before returning. Here we have effectively taken the core's fs backing store and extended it slightly so the store operation is spun off into a new process, running at a lower task priority than NetSurf itself. This stops the UI pauses that occur shortly after loading pages as NetSurf commits the cache to disk. Somewhat experimental, pending more extensive user testing.
* Fix my spelling of teh everywhereVincent Sanders2014-11-1814-15/+15
|
* Remove no longer used poll option from schedule runnerChris Young2014-11-173-16/+5
|
* Improve data scheme fetcher allocationVincent Sanders2014-11-171-10/+8
| | | | | | The data scheme fetcher was over allocating the space for decoded base64 encoded urls and not using the base64 API that allocated the correct size storage.
* add libnsutils to default conveniance library setupVincent Sanders2014-11-171-7/+13
|
* update commentsChris Young2014-11-162-3/+6
|
* Allocate screen title when first needed and keep it private to amiga/gui.cChris Young2014-11-1611-19/+33
|
* Make win_destroyed variable private to amiga/gui.cChris Young2014-11-163-18/+16
|