e737b9ef1b
Lumiera renaming -- source code
2008-03-10 06:09:44 +01:00
0baf15ca94
Cinelerra-3 renamed to Lumiera -- copyright and documentation
2008-03-10 04:25:03 +01:00
1dd9989516
some tweeks to the (scons) build system, using precompiled header
2008-01-27 02:39:13 +01:00
8fe1a901ba
use typelists to declare to visit a collection of types
2008-01-05 18:54:10 +01:00
bf93301211
add an type-list facility inspired by Loki
2008-01-05 17:23:38 +01:00
9aa3cc11e8
new visitor implementation now running.
...
Veryfied the tests, cleaned up. TODO: concurrency, error checks and convienience shortcuts.
2008-01-05 14:26:28 +01:00
cb602663d0
WIP ... obviousely this will segfault
...
And the lession is: class static variables and dynamic linking are dangerous
2008-01-05 12:16:07 +01:00
0bb8051fc5
WIP adjusted BuilderTool (specialisation) and tests....
...
compiles ok, but segfaults
2008-01-05 01:38:32 +01:00
2bd931b6da
WIP replaced former implementation with my new visitor concept.
...
The former implementation was more closely based on Loki's Acyclic Visitor. Swapped in the working new
concept and adapted the policies and tests. Doesn't compile yet
2008-01-04 15:10:18 +01:00
809301a60e
proof-of-concept visitor implementation now working
2008-01-04 01:44:40 +01:00
527ce96f83
WIP draft of a visitor implementation promising to deliver the necessary features
2007-12-24 06:12:29 +01:00
46b200809e
WIP desperately trying to work out a really usable visitor implementation
2007-12-22 08:45:09 +01:00
89c9d2f3df
covered the more esoteric cases for visiting tools and made BuilderTool work
2007-11-29 07:07:14 +01:00
78a4bf8e6b
introcuded a catch-all hook when applying visiting tools to a hierarchy
2007-11-28 04:19:21 +01:00
6d4133cefe
WIP generalizing my builder tools to use the acyclic visitor.
...
inspired by Loki; ichthyo intends to use this pattern for typesafe visiting within the EDL too...
2007-11-27 06:40:51 +01:00
6bdf0e2cdc
put my name in the copyright claim of those files I have written in the last months.
...
(search & replace). Please feel free to add your name to any files to which you added any contributions.
ALL SOURCECODE IS LICENCED UNDER THE TERMS AND CONDITIONS OF THE GPL (2 and above)
2007-11-27 03:19:35 +01:00
940d6de201
Asset/Clip now using the dependency relation.
...
DependantAssets_test now complete, finally...
2007-11-26 04:27:27 +01:00
855d0706d9
implemented enabling/disabling of dependant assets,
...
WIP: passes test, but code still needs some cleanup...
2007-11-24 04:18:54 +01:00
393f0944c1
DependantAssets_test now (partially) running
2007-11-23 04:37:50 +01:00
95af2d50cc
worked out some important details of deleting and unlinking assets
2007-11-22 06:26:55 +01:00
66daf409d4
specify (as test) how asset dependency should work
2007-11-21 06:08:01 +01:00
49459b4bf7
Fix inheritance of asset::Unknown, fix broken test, enable MakeClip_test
2007-11-20 05:20:01 +01:00
b07d8b9687
passes the compiler again.
...
(using direct references instead of smart-ptrs for the 'backlink' from Clip-MO to asset::Media)
2007-11-14 10:10:52 +01:00
5c2fa504e7
WIP update test code to reflect the new placement properties.
...
still have to convince the compiler that i am not a terrrorist....
2007-11-10 23:09:15 +01:00
08f10c2a2e
WIP trying to get the Placement stuff right,
...
but seemingly still having some design mismatch...
2007-10-13 05:13:28 +02:00
fc1c39dc44
WIP: sketch handling placements and mobjects (as test)
2007-10-12 04:21:36 +02:00
fadd31282f
implement access to current session.
...
add lots of UNIMPLEMENTED stubs and finally get the sourcebase through the compiler....
2007-09-27 23:26:54 +02:00
82e9519e73
WIP specify some details of the session interface and the relation of session and EDL
2007-09-27 04:45:06 +02:00
3fdd16ff48
draft adding clips and simple editing operations
...
WIP: lots of tests, no implementation, doesn't compile...
2007-09-25 23:39:46 +02:00
38b47b7f93
proposal for the MediaAccessFacade (interface to bakend), incl. Mock test
2007-09-24 16:20:41 +02:00
1d462845dd
oops... class TargetObject defined twice
2007-09-24 02:35:12 +02:00
eef591f873
WIP injecting a test-mock instead of a singleton
2007-09-23 16:50:05 +02:00
5d54cfc89e
planning next implementation steps
2007-09-20 05:36:08 +02:00
dc97acde5e
added test covering Asset category tuple
2007-09-19 05:05:25 +02:00
7ef8372129
equality and ordering of assets implemented
2007-09-19 03:55:45 +02:00
14d1c5a34b
basic AssetManager works now (creating, registering, querying),
...
including memory management. Ordering is still WIP.
2007-09-18 05:16:56 +02:00
3927f7d5d8
WIP: ordering of Assets, dump AssetManager, bugfix (partially...)
2007-09-17 05:47:22 +02:00
3a416f9e41
WIP test coverage, debugging...
...
still have to fix a segfault :-)
2007-09-16 17:17:54 +02:00
c746629312
util for sanitizing a string, making it usable as identifier
2007-09-14 19:18:11 +02:00
814f6fc734
CreateAsset_test now passes the compiler, after quite some tinkering....
2007-09-12 06:53:12 +02:00
26374ce409
WIP: fleshed out a good deal of the AssetManager foundation.
...
compiles ok, singleton test pass (most of asset creation just commented out...)
2007-09-10 06:45:36 +02:00
79c88c2c87
outlined Asset and AssetManager interfaces...
...
urrgh. this was rather tough to get right. Main problem is avoiding switch-on-type.
2007-09-07 23:24:13 +02:00
45ed4b4264
extracted Singleton pattern into a separate Template.
...
I prefer this approach, so I can ignore threadsafety for the moment and add it later just by switching policy
2007-09-06 22:13:26 +02:00
737da9c868
continued wishful thinking in Test form
2007-09-04 05:09:08 +02:00
1cadae2d75
TDBS (test driven brain storming) :-)
2007-09-03 02:33:47 +02:00
c0dcccebd9
Factory impl now complete (hopfully).
...
Added Tests/Examples for the more advanced use cases
* using custom allocation (e.g. C-style malloc)
* using placement new
* creating Objects via private ctor
* example for the 'PImpl' pattern
2007-08-31 20:32:26 +02:00
b1ea774ef1
removed all custom allocator / deleter stuff (it's the wrong aproach)
2007-08-31 16:53:23 +02:00
b11ac46b76
added possibility to pass a custom deleter function down to shared_ptr.
...
Note: actually all of this seems against the spirit of C++ (and OO in general). This is a do-it-yourself aproach
which adds far to much complexity (accidental, not fundamental complexity). So I checked it in for the record,
but will abandon this aproach and rely on overloading of operator new / delete if necessary (and when we really need it)
2007-08-31 16:25:04 +02:00
2b0ea4650b
reworked factory to get better separation of concerns. But custom allocation can't be done as I intended
2007-08-31 14:57:49 +02:00
307945b629
wrote test: object smart pointer creation factory
2007-08-29 05:03:21 +02:00
