[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Libcdio-devel] CD-Text patches
From: |
Rocky Bernstein |
Subject: |
Re: [Libcdio-devel] CD-Text patches |
Date: |
Fri, 15 Apr 2016 22:31:09 -0400 |
Looked over the code in a little more detail. The code looks fine to me.
Also "make distcheck" worked as well.
I haven't tried with a CD with CD-Text yet.
On Thu, Apr 14, 2016 at 3:34 AM, Rocky Bernstein <address@hidden> wrote:
> As always, many thanks. I'll see if I have time to look at this over the
> weekend.
>
> On Thu, Apr 14, 2016 at 2:49 AM, Leon Merten Lohse <address@hidden>
> wrote:
>
>> As of 9 hours ago, I pushed the branch cdtext-testing into libcdio git.
>>
>> It contains the patch I proposed earlier and 4 additional commits.
>>
>> 1) Interpretation of TAB characters, as discussed earlier on the list
>>
>> 2) Avoid a buffer overflow in the CD-Text parser, which could occur
>> when single text fields are longer than 254 bytes - the spec allows
>> for ~160
>>
>> 3) Store the min and max track numbers in the cdtext struct and add
>> public api functions to access them.
>> Also export the function to parse cdtext from the binary format. The
>> latter is useless without the track numbers.
>>
>> 4) Add a new example to demonstrate the new api functions.
>>
>> My idea behind 3) and 4) is that a standalone CD-Text parser might be
>> useful to some people and access to the parser functions allows
>> implementing a test case that only tests this function and does not
>> depend on the cue sheet parser.
>>
>> Item 3) includes this diff:
>> --- a/lib/driver/cdtext_private.h
>> +++ b/lib/driver/cdtext_private.h
>> @@ -116,6 +116,8 @@ struct cdtext_block_s {
>> cdtext_genre_t genre_code; /**< genre code of the disc */
>> cdtext_lang_t language_code;/**< language of this block */
>> bool copyright; /**< CD-TEXT copyright */
>> + track_t first_track; /**< first track number */
>> + track_t last_track; /**< last track number */
>> };
>>
>> This should not change the ABI though, because cdtext_block_s is not
>> exported.
>>
>> The test cases are also in the works and only need a little
>> more polishing.
>>
>> Best regards
>> Leon
>>
>>
>>
>