Hi folks
I'm beginning to think I've misunderstood something rather severely,
since even after applying basic close order fixups I still can't handle
the PDF reference when enforcing (otherwise) strict open/close nesting.
For example, in the PDF Reference 1.5 (v5), page 167, the following code
appears:
1366 Keyword: q
1367 Variant: 94.500000
1368 Variant: 655
1369 Variant: 424
1370 Variant: -112
1371 Keyword: re
1372 Keyword: W
1373 Keyword: n
1374 Variant: /Cs8
1375 Keyword: cs
1376 Variant: 0
1377 Keyword: scn
1378 Variant: /GS3
1379 Keyword: gs
1380 Variant: 96.667000
1381 Variant: 654
1382 Variant: 421
1383 Variant: -111
1384 Keyword: re
1385 Keyword: f
1386 Variant: /EmbeddedDocument
1387 Variant: /MC8
1388 Keyword: BDC
1389 Keyword: Q
1390 Keyword: q
1391 Variant: 96.667000
1392 Variant: 653.570000
1393 Variant: 420.500000
1394 Variant: -110.570000
1395 Keyword: re
1396 Keyword: W
If PDF content streams *do* obey strict XML-style nesting rules, as they
seem to *most* of the time, the above sequence would be illegal, since
the nesting would be wrong:
q
BDC
Q %% illegal here without preceeding EMC?
q
I dont't think I can just swap the order of operators like I can when I
see a "Q EMC" and I expect an "EMC Q" or vice versa. It'd make more
sense if the BDC and its arguments followed, rather than preceeded, the
Q operator, and I assume that's the intent, but is it really safe to
assume that's what the writing application (Acrobat Distiller, in this
case) intended?
The PDF reference does suggest at several points that proper nesting
must be maintained in certain pairs of operator types, but I couldn't
find anywhere that it made a strong statement about nesting in general.
--
Craig Ringer
-------------------------------------------------------------------------
SF.Net email is sponsored by: The Future of Linux Business White Paper
from Novell. From the desktop to the data center, Linux is going
mainstream. Let it simplify your IT future.
http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4
_______________________________________________
Podofo-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/podofo-users
