Re: [Libcdio-devel] How tolerant to be towards CD-TEXT character set mislabeling ?

[Serge Pouliquen]

Hi,

> Serge, may i impose another round of tests on you ?
I made the new test run.

> You have the best ill CDs ever. :))
lucky me!

diff --git a/lib/driver/cdtext.c b/lib/driver/cdtext.c
index 868cef5..0c1368e 100644
--- a/lib/driver/cdtext.c
+++ b/lib/driver/cdtext.c
@@ -717,7 +717,7 @@ cdtext_data_init(cdtext_t *p_cdtext, uint8_t *wdata,
size_t i_data)
             charset = (char *) "ISO-8859-1";
             break;
           case CDTEXT_CHARCODE_ASCII:
-            charset = (char *) "ASCII";
+            charset = (char *) "ISO-8859-1";
             break;
           case CDTEXT_CHARCODE_SHIFT_JIS:
             charset = (char *) "SHIFT_JIS";


cd-info output results are:

bulle_bob_plage (declaring iso-8859-1 using unknown) : no iconv error
  TITLE: Tiens voil la pelle
  TITLE: Douce et salŽe
  TITLE: Il va o notre chateau
  some characters are not displayed

I promise you, it's a real cd. editor link :
https://didier-jeunesse.com/collections/polichinelle/bulle-et-bob-la-plage-9782278061884
looks OK to me as encoding issue is not on software

> But i could not find any character set where a-accent-grave is 0x88.
an other option is human mistake

starmania (declaring ascii using iso-8859-1) : OK (no error + everything
is displayed)

other cd with ascii only characters : OK


enhancement suggestion
In CD Analysis Report, do you think of any interest to add/display the
encoding used.
If I was able to read that my first cd was declaring ascii encoding, and
not displaying line with accent, I may not have filled a bug report...


Leon wrote:
> How about introducing a second step that, in case inconv
ISO-8559-1->UTF-8 fails, simply ignores the illegal characters?

If you want me to do another test run, I will do. But I let you do the
code which may be accepted on project.
I think I can enter the second step with the cd declaring ascii using
iso-8859-1, if using ascii for the first step.

Regards,
Serge