Hi Ginnie,
I have no further comments on your responses except 1.
So, I removed all the other comments.
On 07/23/10 01:35 PM, Ginnie Wray wrote:
Hi Karen -
Thanks for the review. See inline.
ginnie
On 07/22/10 17:48, Karen Tung wrote:
- line 182: Looks like tmp_progress_holder variable is only used in
the emit()
function. Is there a reason why it needs to be defined here.
If we have more than one progresshandler, the record.progress value
needs to be preserved. In the emit method for the progress handler (lns
224 - 246), the weighted progress is calculated. Then the
record.progress variable is set equal to that and sent to be published
to the logging output. If there is more than one ProgressHandler, the
whole cycle will happen again with the same record. If I don't put the
original value back in, it doesn't get calculated correctly for
subsequent progress handlers (ln 242). It looks like this instead:
ProgressHandler 1:
PROGRESS REPORT: progress percent:100 this is a progress message
percent 100
PROGRESS REPORT: progress percent:50 this is a progress message for
installlogger 50
Progresshandler 2:
PROGRESS REPORT: progress percent:0.01 this is a progress message
percent 100
PROGRESS REPORT: progress percent:0.005 this is a progress message for
installlogger 50
Let me know if you think there is a cleaner way to do this.
I understand why you need to save the value for "progress" in the
record, and restore
them. My question is why the "tmp_progress_holder" variable need to be
an "class" wide
variable. Why not just define it and use it in the emit() function itself?
Another approach to solve your problem above that does not involve
saving and restoring
the value for progress in the record is to create a copy of the
"record", and then,
modify the copy so the original record is un-modified and ready for
other progress handlers.
Thanks,
--Karen
_______________________________________________
caiman-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
