[ Sort of re-try here. More justified, I hope. ] I have used the term "scrapbook copy" but really it is at the moment the only copy which works for math texts and for textual images, e.g., scanned papers in PDF available from digital libraries such as of IEEE. As it is only copy which really works, it should be in any reader.
I have used different term because the copy in PDF readers is the copy defined in PDF standard: totally different operation working on the source code of PDF and restricted by the copy-protection flag of the source code. That copy seems to be restricted to text only and works poorly because digital libraries have PDFs which use characters which does not match, e.g., with Emacs. We users need a working and intuitive copy, not a half-complete engineer-oriented copy of the source code. Scrapbook copy is such a tool and has been proven to work for many decades, prior computers were invented. I don't know why engineers have failed to implement a working copy by using the source code, but I have waited the fix for as long as Acroread has been available. It is highly unlikely they will fix it for the years to come. The problem requires an alternative solution sooner. The scrapbook copy can be equipped with OCR tool if user wishes copy the text as text, otherwise it works on renderings in non-restricted manner. Because OCR technology is not perfect, the source code copy and the scrapbook copy are disjoint copy operations, and may live together if so is wished. OK. Do you wish to have the scrapbook copy tool in Evince? Technically, the scrapbook copy is done by rendering to offscreen pixmap with wanted resolution and copying that. Juhana _______________________________________________ Evince-list mailing list [email protected] http://mail.gnome.org/mailman/listinfo/evince-list
