> I apologize and try to be more explicit in the future. No pb. In an ideal world, the regression would have been catched by the automated regression tests, but as it only showed in Acrobat and not with poppler rendering, this is not really doable.
> I repeated the test with r25261 and layer names with spaces are > OK and model tree works also for the labeled layers. Model tree also shows > non-ASCII characters in layer names now which is nice. Great. Note that for texts in LABELs, there's a limitation to the ASCII character set. (Font and text management in PDF is really intimidating and would probably require some helper library) > I have tested almost all the options of the PDF driver, most important > exception being EXTRA_STREAM which seems to be something that a > regular power user can't test. Everything looks good to me now. Yes EXTRA_STREAM is admitedly not aimed at novice end-users. It is mostly something that an application developer could use to add some custom PDF content (title, legend, etc..). However if you want to play with it, you can have a look at the pdf_custom_layout() (around line 1275) of http://trac.osgeo.org/gdal/browser/trunk/autotest/gdrivers/pdf.py that adds some ungeoreferenced text to the page. With a bit of curiosity, you could make some sense of it by looking at "Appendix A - Operator Summary" of http://www.adobe.com/devnet/acrobat/pdfs/pdf_reference_1-7.pdf and following the links for the operators of interest. Even _______________________________________________ gdal-dev mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/gdal-dev
