>rasterization is the worst of all options and is often insufficient Recent versions of pdftops have a -r option to set the rasterization resolution. Increasing the resolution can create large files, but you can match the resolution to the printer so no detail is lost. The default of 300 is good for viewing on-screen but a little low for printing.
>would you consider adding a "--dont-rasterize" flag to pdftops Another option is writing a comment so pdftops always creates a valid file, but applications can check for the comment and optionally process the pdf through another application. William (on the list but not a maintainer) Date: Thu, 22 Mar 2012 22:58:13 +0100 From: [email protected] To: [email protected] CC: [email protected]; [email protected] Subject: [poppler] Proper Transparency Support in pdftops Dear Poppler Maintainers, Pdftops is a great utility. So great, in fact, that recent versions of CUPS ship with optional compile-time support for using pdftops instead of ghostscript. Gaining the confidence of print software is no small accomplishment, and poppler has proved itself to be a very complete and correct pdf parser. There is one area, however, where poppler unfortunately makes a significant error: rasterize = scan->usesTransparency() || scan->usesPatternImageMask(); The PSOutputDev uses Splash to rasterize PDFs if they're found to use transparency or patterned image masks. While this works well for some use cases, for many other use cases, especially in printing, rasterization is the worst of all options and is often insufficient So I'm wondering -- is including proper support for transparency and patterned image masks on the schedule for the immediate future? If so, wonderful. If not, would you consider adding a "--dont-rasterize" flag to pdftops, so that if PSOutputDev is unable to convert without rasterization, it exits with a known error code? This way, pdftops could be preferred, and ghostscript might function as a fallback, for the case of a scripted conversion utility. Thank you,Jason Donenfeld _______________________________________________ poppler mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/poppler
_______________________________________________ poppler mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/poppler
