It turns out it is really not a good idea to render _anything_ on the
first run of performIn/performOut in classes derived from
SlideChangeBase. That is because the functions are called separetely for
every view, so, if there is more than one view, like when the presenter
console is active, it will be done only for the first of them. This
commit introduces another function, prepareForRun, that can be used
reliably for that purpose.
Change-Id: I2c690fb145ccc05a680d4c2991c73900336e6f42
2007/06/15 20:38:59 thb 1.7.12.8: #i78433# Posting myself an event, to delay-prefetch incoming slide bitmap (follow-up issue for the root cause fix: i78544
2007/05/10 20:32:41 thb 1.7.12.7: #i37778# Reworked LayerManager::updateShapeLayers - now much better aligned with general shape update, and avoids superfluous rendering; made unit tests work again; passing down slide background repaint status to LayerManager; relaxed preconditions for Shape::getUpdateArea() - no longer requires views to work; now catching singular view matrix and using sensible default at API border
2007/05/02 21:37:24 thb 1.7.12.6: #i37778# Changed calls to cppcanvas::Canvas::getTransformation() to ViewLayer::getTransformation() where appropriate (because that's more direct); added initial Slide rendering back in (this time optional)
2007/04/30 07:26:34 thb 1.7.12.5: #i37778# Rendering slide bitmap at the end of a slide transition, instead of during Slide::show(), to avoid temporary display of previous slide after transition sprite vanished.
2007/01/31 14:30:33 thb 1.7.12.4: #i37778# removed View::isContentDestroyed() and mbContentValid distinction on View::clear() - clear() now always clears view the hard way; added explicit screen update to CombTransition, which bypasses SlideChangeBase functionality
2007/01/31 12:29:06 thb 1.7.12.3: #i37778# Slide animations now notify screen updates themselves
2007/01/31 11:25:17 thb 1.7.12.2: #i37778# Added prefetch to Animation interface (to facilitate prefetching - nice for slide transitions, which otherwise lag noticeably while generating the slide bitmap); brought tests up to par, re-enabling unit tests and demo show
2007/01/29 14:02:17 thb 1.7.12.1: Issue number: #i37778#
Larger slideshow refactoring. Wrote design and coding style manifest,
and adapted the code to actually conform to this. In detail:
- cleaned up ownership/disposable/weak_ptr story. removed hacks and
explicit Disposable implementations, where workaround were available
- removed object mutices, where superfluous
- reworked EventMultiplexer (using templatized listener class now), added
more events. EventMultiplexer now serves as a true blackboard
- reworked directory structure: disjunct parts are now physically separated
into directories, instantiation happens via factories & abstract interfaces
- added CursorManager, to make setting mouse cursor less hackish
- reworked DrawShape, to implement SeparateListener pattern
- reworked IntrinsicAnimationActivity, to avoid cyclic references
- modified hyperlink & shape cursor handling to communicate via
EventMultiplexer
- renamed & cleaned up files (presentation.cxx now named slideshowimpl.cxx,
etc.)
- added first version of the z-order fix to layer/layermanager
- cleaned up include guards and include syntax
2006/10/20 14:55:55 thb 1.4.16.9: #i10000# Fixed post-merge build errors (mostly warnings)
2006/10/18 19:59:16 thb 1.4.16.8: RESYNC: (1.4-1.6); FILE MERGED
2006/09/18 15:26:50 thb 1.4.16.7: #i10000# Fixed post-MERGE breakages
2006/04/24 13:25:33 thb 1.4.16.6: #i53194# Unified include statements (local headers always have double quotes; external headers angle brackets); reverted EventMultiplexer pause events to shared_ptr; removed EventMultiplexer::removeViewHandler(), since the handler is held weakly, anyway.
2006/04/12 20:40:08 thb 1.4.16.5: #i37778# Replaced all shared_ptr.get() != NULL places with the more elegant automatic-conversion-to-bool version (at least where the compiler tolerated that)
2006/04/03 16:19:01 thb 1.4.16.4: #i37778# Now passing down ComponentContext to all interested parties; building a second, all-exports version of the slideshow component (to facilitate unit testing also for internal classes) - this made necessary renaming ImportFailedException to ShapeLoadFailedException (because of silly i63703); applied relevant parts of #i63770# (const-correctness); reworked view handling in such a way that views are now kept in one central repository (and are not duplicated across all interested objects); moved code from namespace presentation to namespace slideshow
2006/03/24 18:23:26 thb 1.4.16.3: #i37778# Moved whole slideshow engine from namespace presentation (which conflicts with one of the UNO subnamespaces) to slideshow
2006/03/08 18:08:07 thb 1.4.16.2: #i49357# Removed cruft from gdimtftools; moved slideshow component to ServiceDecl; refactored multiple places of filling a rectangle in XCanvas into a tools method; now clearing the shape area to grey50 for applets/plugins (to at least show _something_ during slide transitions)
2006/03/06 22:14:32 thb 1.4.16.1: #i53194# #i55294# #i59324# Overhauled IntrinsicAnimationActivity; fixes GIF animation import; corrected rehearse timings sprite size; several cosmetic changes (removed external header guards); prepared scene for sprite prio
2005/02/15 23:51:00 thb 1.1.2.5: #i42440# Moved sprite-growing code to slideshow, some minor issues with size calculations for empty slides (when a 'leaving' slide is needed, but there's no previous one, an empty bitmap is generated)
2005/02/15 12:45:09 thb 1.1.2.4: #i42440# Moved sprite-growing code to slideshow, we already handle the case there for update areas of antialiased shapes; Improved/reformatted documentation; (Hopefully) fixed the problem that sometimes, the first effect preview did not work. The reason there was presumably the clearing of the EventQueue before every slide, and the fact that EventMultiplexer only generated new tick events when it thought ticking was switched off.
2005/02/07 01:05:37 thb 1.1.2.3: #i38960# Slight adaptions for new timer handling: event and activities queue now share a timer, such that e.g. a hold on this timer affects the other queue, too; slight reworks to slidechangebase, now creating the slide bitmap(s) on start(), too.
2005/01/30 15:54:43 dbo 1.1.2.2: #i39787# correct bitmap dimensions
Issue number:
Submitted by:
Reviewed by:
2005/01/26 11:18:32 dbo 1.1.2.1: #i39787# slide transition revision: recalc bitmaps/sprites in case of view changes
Issue number:
Submitted by:
Reviewed by:
