mythtranscode --infile fails on non default-format filenames

[ylee@…]

The

--infile xyz.nuv

flag tells mythtranscode to do its best to figure out the channel and starttime of the recording from the filename itself; non default-format filenames (i.e., those not in the traditional

1160_20070131202900.mpg (or .nuv)

style) prevent mythtranscode from doing so (and which it duly complains about). This is all well and good, except that mythtranscode still proceeds anyway, confusing the user. In other words, the resulting mythtranscode run is utterly useless because the utility doesn't know what seektable to update in the first place. The only way around this is 1) first rename the recording back to the default format or 2) instead of using the more-convenient --infile, manually specify the data mythtranscode needs with -c and -s.

I suggest that mythtranscode either a) abort immediately if it is unable to match the recording with the relevant database entries, instead of proceeding and misleading the user, or preferably b) use the filename from --infile (which, of course, is also already in the database) to identify the proper database entries.

[cpinkham]

It has been on my TODO list for a while to convert mythtranscode and the mythtv test binary to use the Programinfo::GetProgramFromBasename?() that I added a while back to see if the filename they are given is for a recording in the database.

Moving ticket to my queue.

[cpinkham]

(In [12786]) Convert mythtranscode and the mythtv test binary to use ProgramInfo::GetProgramFromBasename?() on the filename they are given when looking to see if the file is a recording currently in the database. This is much better than parsing the chanid_starttime format that mythtranscode was previously doing.

Closes #3032.