Activity
From 02/20/2012 to 03/20/2012
03/20/2012
- 04:19 PM Revision 382 (cca): Component info now also printing data ports again
- 03:59 PM Bug #968 (Resolved): Segfault in component info printing, if unconfigured ports
- 03:28 PM Revision 381 (cca): Somehow eliminated calling of state-change hooks. Added again.
- refs #929
03/19/2012
- 01:50 PM Bug #952 (Resolved): Error: Pose has no member setValue
- Applied in changeset r563.
- 11:52 AM Revision 380 (cca): Non-compiling code over the weekend, shame on me!
- I`m sorry, guys.
03/16/2012
- 04:51 PM Revision 379 (cca): Added resetting of CCA components and basic tests
- Needs to be further tested
refs #962 - 04:48 PM Feature #962 (Feedback): Provide onReset hook for components
- onReset hook was already there but reset was not possible from outside.
Reset can now be requested from outside via ... - 04:06 PM Feature #962 (Resolved): Provide onReset hook for components
- It is desirable to reset components to an initial state. In order to conduct the necessary reset steps, the component...
- 03:17 PM Bug #961 (Resolved): Comparing component states does not work
- In order to let a component get to know in which state it is, component states should be comparable, e.g....
- 02:15 PM Support #960 (New): Component exception throwing is catched somewhere
- If I want a component to crash when it is requested to change in a component state that is not supported, I am not ab...
- 02:10 PM Bug #959 (Resolved): Test for component to not be stopped twice
- IMO, the hook 'OnStop' is called *twice*, if the component is requested to change its component state to OFF. Plz che...
03/15/2012
- 02:00 PM Bug #952 (Resolved): Error: Pose has no member setValue
- In order to provide a common interface for DTOs you should ensure that all DTOs are child classes of 'ContainingDoubl...
03/14/2012
- 01:20 PM Feature #929 (Resolved): Provide hooks for changing between the different execution modes
- Applied in changeset r378.
- 01:01 PM Feature #929 (In Progress): Provide hooks for changing between the different execution modes
- 01:18 PM Revision 378 (cca): Hooks for state changes
- fixes #929
03/12/2012
- 04:40 PM Bug #940 (Resolved): Wrong naming of statechange scope
- Applied in changeset r377.
- 04:24 PM Bug #940 (Resolved): Wrong naming of statechange scope
- Statechange input scopes of cca BaseComponents automatically convert the components name to lower cases.
- 04:33 PM Revision 377 (cca): Don`t lower case for scope names
- fixes #940
03/10/2012
- 10:22 PM Revision 376 (cca): arno: removed annoying warning (statechangeconverrter)
- 07:48 PM Revision 375 (cca): * Added BeatProvider to derive from when providing a beat for a CCA system
- * Added tests for BeatProvider
* Started ComponentInfo conversion (for remote sending)
* Cleaning up rsb handlers whe...
03/08/2012
- 10:38 PM Revision 374 (cca): Deriving DTO from rsc printable made trouble
- 10:30 PM Revision 373 (cca)
- 06:20 PM Revision 372 (cca): Fixed two warnings
- 06:03 PM Revision 371 (cca): Extended component info and printing
03/07/2012
- 05:44 PM Enhancement #931 (Rejected): Provide socket transport for remote port configuration
- So far, a REMOTE port configuration uses spread, regardless of which transports are enabled or set in rsb transport c...
- 12:15 PM Feature #929: Provide hooks for changing between the different execution modes
- Hooks installed. To be tested
- 10:57 AM Feature #929 (Resolved): Provide hooks for changing between the different execution modes
- Provide hooks for changing between the different execution modes (execution, online learning, offline learning).
Pro...
03/02/2012
- 07:53 PM Revision 370 (cca): Logging
- 05:48 PM Tasks #917 (Resolved): Review processing strategies for CCA Components
- Fixed through #919
- 04:41 PM Tasks #917 (Resolved): Review processing strategies for CCA Components
- The CCA Base Component creates default input and output ports. This breaks some of the processing strategies, e.g.:
... - 05:47 PM Tasks #919 (Resolved): Statechange and beat port have to be seperated from data ports
- Applied in changeset r367
State and beat ports are now special ports, separated from data ports, with extra handling... - 04:43 PM Tasks #919 (Resolved): Statechange and beat port have to be seperated from data ports
- We probably need to separate between statechange and beat port on the one hand and usual _data ports_ on the other ha...
- 05:45 PM Revision 369 (cca): More component lifecycle tests
- 05:38 PM Revision 368 (cca): Fixed component state change to 'off'
- Extended component tests
- 05:23 PM Revision 367 (cca): Refactored components default ports to be special ports
- 02:54 PM Revision 366 (cca): Minor bugfix
- 02:48 PM Revision 365 (cca): Component now has helper method to publish it`s own state
- Component automatically publishes new state if state transition successful
- 02:32 PM Revision 364 (cca): Fixed port numbering and component tests
- refs #913
- 02:18 PM Revision 363 (cca): Minor bugfix
- 02:10 PM Tasks #913 (Resolved): Default configuration for component info and component state ports
- Applied in changeset r362.
- 11:57 AM Tasks #913 (Resolved): Default configuration for component info and component state ports
- Default configuration for component info and component state ports.
* Component state should be published on default... - 02:09 PM Revision 362 (cca): Defaut configuration for components default ports
- fixes #913
- 11:50 AM Revision 361 (cca): Bugfix in output port configuration of components
- 11:38 AM Tasks #912 (Resolved): Default configuration for components
- Provide solutions for default configuration inside components as much as possible. The goal would be that one can alw...
03/01/2012
- 04:26 PM Revision 360 (cca): Extended Component
- * Now has a component state
* Extended lifecycle
* Processing now separated into three different states:
** Execution... - 04:24 PM Revision 359 (cca): Component state naming
- 04:24 PM Revision 358 (cca): Component state naming
- 03:26 PM Revision 357 (cca): Fixed for changes in RSB (namespace change of AnnotatedData)
- 02:50 PM Revision 356 (cca): Fixed and tested ComponentState Converter
- 02:35 PM Revision 355 (cca): ComponentState Converter (untested)
- 02:34 PM Revision 354 (cca): ComponentState Converter (untested)
- 02:34 PM Revision 353 (cca): Rough draft of ComponentState DTO
- * ComponentState Converter
02/28/2012
- 05:08 PM Revision 352 (cca): * Some restructuring
- * Introduced Component (extending CCANode)
02/27/2012
- 03:52 PM Revision 351 (cca): Updated test environment
- 03:28 PM Revision 350 (cca): Fixed spread test config
02/25/2012
- 04:12 PM Feature #898 (Resolved): Implement reset mechanism in component lifecycle
- Applied in changeset r349.
- 02:57 PM Feature #898 (Resolved): Implement reset mechanism in component lifecycle
- Implement reset mechanism in component lifecycle, required in a downstream project for resource node components.
- 04:11 PM Revision 349 (cca): Empty all input ports (buffers) on reset
- fixes #898
- 04:00 PM Feature #899 (Resolved): Purge buffers of input ports
- Applied in changeset r348.
- 03:34 PM Feature #899 (Resolved): Purge buffers of input ports
- For resetting logic in components we need to be able to empty the buffers of the input ports to get the components in...
- 03:57 PM Revision 348 (cca): Implemented and tested purgin (emptying) of buffers
- fixes #899
- 03:37 PM Revision 347 (cca): Implemented and tested reset state transitions in components
- * Needs purging of the component`s input buffers in order to be complete reset
#refs 898
#refs 899 - 02:26 PM Revision 346 (cca): amarsi-fixes #219
- 01:24 PM Revision 345 (cca): Missed to include nemobench include dirs
02/24/2012
- 05:56 PM Revision 344 (cca): Synchronized access of single latest item through mutex
- 05:55 PM Revision 343 (cca): Synchronized access through mutex
02/23/2012
- 06:01 PM Revision 342 (cca): * More benchmarking
- * Minor performance tweak
- 05:26 PM Revision 341 (cca): Marked tryGet method in buffer deprecated
- 05:16 PM Revision 340 (cca): Benchmarking buffers
- 05:07 PM Revision 339 (cca): Buffers now only provide one get method (tryGet now deprecated)
- * Waiting until new item arrives breaks with CCA managed processing
- 03:52 PM Revision 338 (cca): * Static declarations for degtorad and radtodeg mappings
- * More constness where allowed
02/22/2012
- 06:59 PM Revision 337 (cca): Performance improvement on DTO splitting based on benchmarks
- 06:56 PM Revision 336 (cca): Changed semantics of Constructor with one numeric value
- * Formerly: interpreted as dimensionality
* Is now: Interpreted as desired initial value of a one-dimensional DTO - 06:39 PM Revision 335 (cca): Some benchmarks for ContainingDoubles DTOs
- 01:50 PM Revision 334 (cca): Introduced Benchmarking
- * First small benchmark example for DTO splitting
02/20/2012
Also available in: Atom