2009.02.06 14:56 "Re: [Tiff] assertions, and building with DEBUG/NDEBUG", by Frank Warmerdam

Joris Van Damme (AWare Systems) wrote:

Frank, folks,

This is about http://bugzilla.maptools.org/show_bug.cgi?id=1994. I've currently no bugzilla log-in, and/or I forgot about it, so I'll go ahead and abuse the mailing list.

Joris,

I would appreciate it if you could get a bugzilla account, and be prepared to interact with bugs there. In the meantime I've appended your email to the ticket.

• The thing is, since the compression tag is read in the first stage, and flagged IGNORE, the hacky second and third stages aren't supposed to have to deal with it. Thus, a routine like TIFFFetchNormalTag for example, is never called for the compression tag.

In TIFFFetchNormalTag() we have lots of logic like:

        case TIFF_SETGET_UINT16:
                {
                        uint16 data;
                        assert(fip->field_readcount==1);
                        assert(fip->field_passcount==0);
                        err=TIFFReadDirEntryShort(tif,dp,&data);
                        if (err==TIFFReadDirEntryErrOk)
                        {
                                if (!TIFFSetField(tif,dp->tdir_tag,data))
                                        return(0);
                        }
                }
                break;

Would it be better to change these assertions to a runtime error indicating that this tag is not suitable for processing in TIFFFetchNormalTag(), possibly indicating a corrupt file or a logic error in libtiff?

This is quite a dirty file so I'm pleased to settle with reasonable error reports and recovery.

I presume the "fip" description of the tag cannot be made compatible with TIFFFetchNormalTag for some reason? It does not *seem* like a particularly complicated tag to handle generically. It is currently:

        { TIFFTAG_COMPRESSION, -1, 1, TIFF_SHORT, 0, TIFF_SETGET_UINT16,

TIFF_SETGET_UNDEFINED, FIELD_COMPRESSION, 0, 0, "Compression", NULL },

I a bit lost in aspects of this - particularly the stuff about set_field_type and get_field_type in the TIFFField structure.

Actually, can't we just set the field_readcount to 1 and then this definition would work?

Best regards,
--
---------------------------------------+--------------------------------------

I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam@pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam

and watch the world go round - Rush    | Geospatial Programmer for Rent