[mythtv-commits] Ticket #646: Update myth.rebuilddatabase.pl to support the basename, progstart, and progend columns

[MythTV]

#646: Update myth.rebuilddatabase.pl to support the basename, progstart, and
progend columns
----------------------------------------------+-----------------------------
 Reporter:  sphery <mtdean at thirdcontact.com>  |       Owner:  ijr
     Type:  patch                             |      Status:  new
 Priority:  trivial                           |   Milestone:     
Component:  mythtv                            |     Version:     
 Severity:  low                               |  
----------------------------------------------+-----------------------------
The attached patch updates myth.rebuilddatabase.pl to support the
 basename, progstart, and progend columns that are now a part of the
 database schema.  Because of the infrequency with which the
 myth.rebuilddatabase.pl script is used, I didn't feel it warranted the
 effort required to redesign the script to work more appropriately with the
 new schema; however, I discussed an approach I felt would be ideal for use
 with the new schema in the thread starting at the post http://www
 .gossamer-threads.com/lists/mythtv/dev/155680#155680 (the approach is
 heavily based off Chris's mythrename.pl script).

 Instead, this patch simply adds the new columns into the insert
 statements.  Users must name the files according to the old file naming
 scheme (i.e. chanid_progstart_progend.nuv, where the progstart and progend
 must be of the format YYYYMMDDHHiiSS) for the script's automatic program
 information lookup to work (otherwise, the user will get prompted for all
 information).  Regardless of the format of the name used, the filename
 extension *must* be ".nuv"--even if it's an MPEG file--for the script to
 even find the file.  If the user does not like the filename (format of the
 name or extension), he/she may change it after import with mythrename.pl.

 Instead of forcing the user to specify a real starttime and endtime, the
 script simply uses the progstart and progend (which it calls
 starttime/endtime) as the starttime and endtime, respectively.  This does
 mean that the length of the recording is not necessarily shown correctly,
 but the extra effort required of the user to specify the actual start and
 end times of the recording does not seem worth the benefit of doing so.
 Users wanting actual starttime and endtime information may name their
 files chanid_starttime_endtime.nuv and type in the appropriate title,
 subtitle, and description when prompted (since the program information
 lookup will not succeed).

 In the event that a user has a software encoder card (i.e. still uses the
 ".nuv" file extension with filenames that don't contain the endtime), all
 files that do not have the endtime (i.e. chanid_starttime.nuv) will result
 in a prompt, "Non-nuv file <filename> found.  Do you want to import?
 (y/n)" and the user will need to answer, "n".  Note that this will also
 apply to any recording--even from a hardware encoder using MPEG streams--
 created on a machine with SVN revs [7240] (inclusive) through [7246]
 (exclusive) (from the time the endtime was removed from the filename until
 the file extensions were changed to use proper extensions).  On a system
 with hundreds of recordings using the ".nuv" extension without endtime,
 this will get very tedious.  This is why, IMHO, the filename scanning code
 no longer makes sense, and why I recommended a new script in the posts at
 the link above.

 Although this patch is not the best way to solve the problem, it is
 definitely the quickest solution, and it works (with qualifications :).

-- 
Ticket URL: <http://svn.mythtv.org/trac/ticket/646>
MythTV <http://www.mythtv.org/>
MythTV