| Issue Number | 3161 |
|---|---|
| Summary | [Media] Automatically calculate length of recording |
| Created | 2010-05-20 15:38:01 |
| Issue Type | Improvement |
| Submitted By | Osei-Poku, William (NIH/NCI) [C] |
| Assigned To | Kline, Bob (NIH/NCI) [C] |
| Status | Closed |
| Resolved | 2010-06-10 10:17:36 |
| Resolution | Fixed |
| Path | /home/bkline/backups/jira/ocecdr/issue.107489 |
BZISSUE::4848
BZDATETIME::2010-05-20 15:38:01
BZCREATOR::William Osei-Poku
BZASSIGNEE::Bob Kline
BZQACONTACT::William Osei-Poku
Is it possible to have the <RunSeconds> element of the media
document
automatically calculate the length of the attached sound file?
(This request was first reported under OCECDR-3157.)
BZDATETIME::2010-05-26 12:08:07
BZCOMMENTOR::Bob Kline
BZCOMMENT::1
Implemented on Mahler; ready for user testing.
I adapted the code I wrote for this from the most widely-used open source package for examining and manipulating information in MP3 files (mp3info). I eliminated one flaw I found. There is one other shortcut taken by the original code that for the moment I have left in place. That code is ignoring differences in the value of the 'padding' flag for the frames in the MP3 file, which could have a slight effect on the accuracy of the number of seconds reported. In theory, the effect could be more than slight, but in actual practice I expect it would typically be negligible. I will rewrite that portion of the code if absolute accuracy is required. I have reported the two anomalies to the current maintainer of the package.
I assume that when you submitted this request you had determined that the anticipated number of times users would have needed to enter this information by hand in the absence of custom code to parse the file will represent enough savings to make it worth while automating the calculation, right?
BZDATETIME::2010-05-26 12:28:05
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::2
(In reply to comment #1)
> Implemented on Mahler; ready for user testing.
>
> I assume that when you submitted this request you had determined
that the
> anticipated number of times users would have needed to enter this
information
> by hand in the absence of custom code to parse the file will
represent enough
> savings to make it worth while automating the calculation,
right?
Yes. The other consideration was how the user who did not record the meeting quickly know how many minutes the recording runs without starting to play the recording? If the user who records the meeting were to also create the document, it would not be an issue to get either the actual length of the recording or an estimate.
BZDATETIME::2010-06-01 09:17:14
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::3
(In reply to comment #1)
> Implemented on Mahler; ready for user testing.
>
Verified on Mahler. Please promote to Bach.
BZDATETIME::2010-06-08 11:28:25
BZCOMMENTOR::Bob Kline
BZCOMMENT::4
Promoted to Bach. I believe I was successful in eliminating the slop in the calculation of the number of runtime seconds.
BZDATETIME::2010-06-10 10:17:20
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::5
(In reply to comment #4)
> Promoted to Bach. I believe I was successful in eliminating the
slop in the
> calculation of the number of runtime seconds.
Verified on Bach. Issues closed. Thank you!
BZDATETIME::2010-06-10 10:17:36
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::6
Now closed.
Elapsed: 0:00:00.001574