Re: [Libcdio-devel] CD-TEXT documentation

[Rocky Bernstein]

On Sat, Feb 4, 2012 at 6:27 AM, Thomas Schmitt <address@hidden> wrote:

> Hi,
>
> now that i have released libburn and cdrskin with CD-TEXT features,
> i want to contribute to the libcdio docs about that topic.
>

Many thanks! Independent of libcdio, public CD Text documentation at an
implementation level is much needed.


> Are there files to read and update, other than these ?
>  doc/glossary.texi item "CD Text"
>  doc/libcdio.texi node "CD Text"
>

No this is all am aware of.

>
> ---------------------------------------------------------------------------
> doc/glossary.texi:
>
> Consider to add a pointer which refers to libburn's doc/cdtext.txt.
>  http://libburnia-project.org/browser/libburn/trunk/doc/cdtext.txt


It is a little weird to have a documentation link to a file inside the trac
svn project.

or
>
> http://libburnia-project.org/browser/libburn/trunk/doc/cdtext.txt?format=txt
>

When I tried this from windows it offered to open notepad and then the was
not very readable because of the lack of  carriage returns. I realize this
is highly anomalous of me, since I prefer not to use MS Windows and even
then should have opened with Emacs which rather than using  MS Windows'
suggested default. But it does suggest someone else may have this problem.


> (The latter triggers on my elderly browser download rather than viewing.)
>

>
> ---------------------------------------------------------------------------
> doc/libcdio.texi:
>
> Some of its aspects go beyond my own CD knowledge.
> Nevertheless, i have to question some statements:
>
> > The first place the information can be recorded is in the R-W sub
> > codes in the lead in area of the CD giving a data capacity of about
> > 5,000 ASCII characters (or 2,500 Kanji or Unicode characters).
>
> How was the payload of 5,000 bytes computed ?
>

Probably from  http://web.ncf.ca/aa571/cdtext.htm


> The highest lead-in start for recordable CD media in my list is
> 97m 49s 00f which means LBA -9975. Lead-in reaches up to LBA -150.
> Each sector can take 4 text packs of 12 bytes payload = 471,600 bytes.
>
> MMC command READ TOC/PMA/ATIP could retrieve 65,534 bytes from lead-in.
>
> A CD-TEXT block can have only 253 payload packs of 12 bytes each.
> There can be 8 blocks at most. 253 * 12 * 8 = 24,288 bytes.
>
> A single block (i.e. one language) can have 253 * 12 = 3036 bytes.
>
>
> > The second place the information can be recorded is in the R-W sub
> > codes in the program area of the CD giving a data capacity of roughly
> > 31MB. This information is stored in a format that follows the
> > Interactive Text Transmission System (ITTS) which is the same data
> > transmission standard used by such things as Digital Audio
> > Broadcasting (DAB), and virtually the same as the data standard for
> > the MiniDisc. Traditionally the R-W sub codes have been used for text
> > and graphics in applications such as CD+G
>
> Shouldn't the statement about the format be separated in an own paragraph ?
>

There's some leeway in how writing works, but I do prefer Franz Kafka
simpleness to  William Faulkner or James Joyce. So I've split this now into
two paragraphs.

I wasn't totally sure where you were suggesting the split. Since you too
have commit rights feel free to just make the change
or post a patch (which git can help with).

"This information is stored in a format [...] the MiniDisc."
>
> Afaik, the format is the same in lead-in and in program area.
>
> Is the statement about CD+G particularly about program area or does it
> apply to lead-in too ?
>

Dunno. CD+G is another mystery area.


>
>
> > In the case of ATAPI drives, the
> > SFF8020 spec covers the reading of the RW subcodes.
>
> This is covered by MMC meanwhile.
> Reading from lead-in is done by command 43h READP TOC/PMA/ATIP, Format
> 0101b.
> Reading from program area is supposed to be done by BEh READ CD.
> (I did not explore the latter yet.)
>
> Are any non-MMC drives still worth to be mentioned ?
>

Dunno, but it's better to err on conservatism here.


>
>
> How much degree of detail from
>  http://libburnia-project.org/browser/libburn/trunk/doc/cdtext.txt
> would be appropriate for doc/libcdio.texi resp. other libcdio
> documentation ?
>
> You are of course invited to simply copy the whole thing. But near the
> end it contains some libburn specifics.
> (Proff-reading would be appreciated, too.)
>

That should be "Proof reading ..."  :-)

I'll make a pass over the document, convert it to texinfo format and add it
to libcdio's documentation area adding links back to libburn. Then feel
free to copy this back to libburn.

Thanks again for putting together more detailed CD Text information and
making that available to everyone.


>
> Have a nice day :)
>
> Thomas
>
>
>