[DVDnav-discuss] libdvdread patches from my tree

[Erik Hovland]

>> --- a/src/dvd_udf.c
>> +++ b/src/dvd_udf.c
>> @@ -611,7 +611,9 @@ static int UDFScanDir( dvd_reader_t *device, struct AD Dir, char *FileName,
>>              memcpy(FileICB, &tmpICB, sizeof(tmpICB));
>>              found = 1;
>>            }
>> -          UDFMapICB(device, tmpICB, &tmpFiletype, partition, &tmpFile);
>> +          if(!UDFMapICB(device, tmpICB, &tmpFiletype, partition, &tmpFile))
>> +         return 0;
>> +
>
> Mix of tabs and spaces, should be all spaces.

Thanks. Fixed in my tree.

>> From e6a378b53b7d8d961dfe87834eb6aed9035a1bb3 Mon Sep 17 00:00:00 2001
>> From: Erik Hovland <erik at hovland.org>
>> Date: Mon, 26 Sep 2011 10:08:12 -0700
>> Subject: [PATCH 2/8] Use a macro instead of a magic constant
>>
>> The levels value is in a macro in the very next line of
>> ifo_types.h. The static array declared above it should
>> use it.
>> ---
>>  src/dvdread/ifo_types.h |    2 +-
>>  1 files changed, 1 insertions(+), 1 deletions(-)
>>
>> diff --git a/src/dvdread/ifo_types.h b/src/dvdread/ifo_types.h
>> index 7db7d34..6e5f4a1 100644
>> --- a/src/dvdread/ifo_types.h
>> +++ b/src/dvdread/ifo_types.h
>> @@ -475,7 +475,7 @@ typedef struct {
>>   * Parental Management Information Unit Table.
>>   * Level 1 (US: G), ..., 7 (US: NC-17), 8
>>   */
>> -#define PTL_MAIT_NUM_LEVEL 8
>> +#define PTL_MAIT_NUM_LEVEL 8U
>>  typedef uint16_t pf_level_t[PTL_MAIT_NUM_LEVEL];
>
> I wonder how the log message and the change match - they don't AFAICT.

This is a vestige of a merge between my tree and another tree after the
svn commit on Jan 30. This commit is reverted in my tree.

>> From e170cbc56844b03db6bed41fc7013265be01c1f0 Mon Sep 17 00:00:00 2001
>> From: Erik Hovland <erik at hovland.org>
>> Date: Mon, 10 Oct 2011 15:08:24 -0700
>> Subject: [PATCH 5/8] Use NULL instead of 0, this is C kids
>>
>> --- a/src/ifo_read.c
>> +++ b/src/ifo_read.c
>> @@ -475,7 +475,7 @@ ifo_handle_t *ifoOpenVTSI(dvd_reader_t *dvd, int title) {
>>
>> -  if(ifoRead_VTS(ifofile) && ifofile->vtsi_mat)
>> +  if(ifoRead_VTS(ifofile) && ifofile->vtsi_mat != NULL)
>>      return ifofile;
>
> This makes no sense except bloating the code IMO.

Not a problem to remove.

>> From 5a6374c00f173101b1914f5862f91c4704dfbece Mon Sep 17 00:00:00 2001
>> From: Erik Hovland <erik at hovland.org>
>> Date: Mon, 10 Oct 2011 15:49:49 -0700
>> Subject: [PATCH 6/8] Update the TODO file
>>
>> ---
>>  TODO |    9 +++++----
>>  1 files changed, 5 insertions(+), 4 deletions(-)
>>
>> diff --git a/TODO b/TODO
>> index 3fe2eed..870ffe6 100644
>> --- a/TODO
>> +++ b/TODO
>> @@ -8,5 +10,4 @@
>>     3 - fully respect user prohibitions
>>  * cleanup public API and fix libtool versioning
>>  * Update decoder.c with some of the more rare commands. Update already done to vmcmd.c
>> -* Replace the auto* build system with a custom and clean one based on ffmpeg's (partly done)
>> +* Replace the auto* build system with a custom and clean one based on ffmpeg's (mostly done)
>
> How far along is this?  I never checked how this was done and how much
> code was actually copied over.  Are you seriously interested in redoing
> the build system libav style or do you just want to fix up the auto*
> implementation you already have?

I doubt anyone is willing to take this any farther. Which is why I
changed the TODO. Since no one wants to really take it any farther it is
done as much as anyone ever really needed.

Thanks for the review. I appreciate it.

E

-- 
Erik Hovland
erik at hovland.org
http://hovland.org/