Wednesday, August 10 2005
From LDTP
Aug 10 16:30:59 nags hey guys meeting time :)
Aug 10 16:31:43 prem nags: I am in :)
Aug 10 16:31:43 jess yes thanks prav33n
Aug 10 16:31:45 jess prem
Aug 10 16:31:51 prem jess: fine :)
Aug 10 16:32:52 shaheed i am in
Aug 10 16:33:10 nags kaveri: ?
Aug 10 16:33:16 nags Raghu: ?
Aug 10 16:33:17 nags Casanova: ?
Aug 10 16:33:39 Casanova hi nags :)
Aug 10 16:33:45 nags Casanova: Hi
Aug 10 16:34:14 nags Updates from my side
Aug 10 16:34:37 nags Reviewed new logic proposed by prem
Aug 10 16:34:40 nags things look fine
Aug 10 16:34:48 nags I have asked him to work on that
Aug 10 16:35:02 * prem prem goes yeepppe
Aug 10 16:35:05 nags Updated two component function
Aug 10 16:35:18 nags and nothing much last week
Aug 10 16:35:25 nags hey forgot onething
Aug 10 16:35:37 nags now discussing the recording module with varadhan
Aug 10 16:35:51 nags soon, we will try to rectify the major issues that we are facing
Aug 10 16:36:21 nags For August 17th release, from my side only those two component functions
Aug 10 16:36:28 nags guys from your side ?
Aug 10 16:36:32 nags prem: ?
Aug 10 16:36:33 nags mnagashree: ?
Aug 10 16:36:38 nags Casanova: ?
Aug 10 16:36:43 nags shaheed: ?
Aug 10 16:36:46 nags varadhan: ?
Aug 10 16:36:54 Casanova nags: i will get the deb ready by this weekend
Aug 10 16:37:04 nags Casanova: cool :)
Aug 10 16:37:16 nags Casanova: any updates from tuxjr ?
Aug 10 16:37:23 prem I have done three new component functions -- those that i have not chosen from the TODO ;)
Aug 10 16:37:32 Casanova nags: and will start with gaim after that... need a lot of help from you starting off :D
Aug 10 16:37:42 mnagashree file-roller verifications that were pending should be done my next release thats on 17th
Aug 10 16:37:42 nags Casanova: sure :)
Aug 10 16:37:49 Casanova nags: no i think his exams are going on.. he hasnt logged on for a while
Aug 10 16:37:56 nags mnagashree: cool :)
Aug 10 16:38:03 nags Casanova: okay
Aug 10 16:38:15 prem Currently working on fixing the broken scripts in Evolution mailer and calendar, for making it ready for the Aug 17 release
Aug 10 16:38:28 nags prem: thats nice
Aug 10 16:38:42 nags Raghu: about your gedit test-scripts ?
Aug 10 16:38:51 prem Also working on implementing the new appmap generation and searching logic ...
Aug 10 16:38:52 nags Raghu: also biswa's eog scripts ?
Aug 10 16:39:02 nags kaveri: any updates from your side ?
Aug 10 16:40:26 nags shaheed, bandini: updates gnome-terminal test scripts ?
Aug 10 16:41:10 nags it seems wipro guys are bit busy
Aug 10 16:41:19 shaheed nags: not yet started, but will make it ready by the date we have planned ;)
Aug 10 16:41:28 nags shaheed: cool
Aug 10 16:41:47 nags prem: about your new pyldtp algorithm ?
Aug 10 16:42:43 prem I have sent a mail to the mailing list, describing the core logic used and its difference from the current approach
Aug 10 16:43:00 prem Sorry for the late mail ;)
Aug 10 16:43:37 nags prem: a small short note here ?
Aug 10 16:43:52 prem Sure ... For those who haven't checked their mail let me give brief note on that
Aug 10 16:44:18 nags prem: also me, who haven't checked the mail yet :D
Aug 10 16:44:45 nags what all modules this will affect ?
Aug 10 16:45:08 prem In the current approach for getting the accessible handle of an object we browse through all the objects in the entire dialog till we reach the desired one
Aug 10 16:45:52 prem This will obviously cost lot of CPU and memory
Aug 10 16:46:28 nags prem: okay, so this new approach is going to increase the execution performance. Thats a next mile stone for LDTP
Aug 10 16:47:03 shaheed prem: curious to know, in evolution for few objects the index changes after doing some operations, how you are going to address this issue with your new logic
Aug 10 16:47:04 prem In the new model we can store the child_index with respect to the immediate parent ... thus at runtime with few calls to Accessible_ChildAtIndex
Aug 10 16:47:17 prem we can get the required object
Aug 10 16:47:37 nags where we are going to store the child_index ?
Aug 10 16:48:05 prem shaheed: This new approach will help in readjusting the appmap at runtime ( thought it is not planned for the first step )
Aug 10 16:48:34 prem nags: We can store it in the attributes hashtable corresponding to evey object
Aug 10 16:49:18 prem I have planned to deliver the primary appmap generation and searching algorithm for the sept first week release
Aug 10 16:50:00 prem Other features like dynamic appmap for a particular object, a small GUI for viewing the generated appmap (like we do using at-poke)
Aug 10 16:50:09 prem can be planned for future releases
Aug 10 16:50:39 nags prem: you mean to say at run time when running pyldtp ?
Aug 10 16:50:50 prem Guys , if you have queries and suggestions please let me know ...
Aug 10 16:51:10 nags prem: sure :)
Aug 10 16:51:31 prem nags: Do you mean the GUI feature
Aug 10 16:51:54 nags prem: we will take that offline ;)
Aug 10 16:52:07 prem nags: fine
Aug 10 16:53:53 nags guys anything from your side ?
Aug 10 16:54:22 prem nags: I am done ...
Aug 10 16:54:26 nags if not we can conclude this meeting :)
Aug 10 16:55:08 nags Thanks guys for your time, see you all by next week with the new release !!!
