[Poppler-bugs] [Bug 90795] [patch] Allow configuring SPLASH_CMYK support with --enable-cmyk as in xpdf

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Mon Jul 6 08:24:47 PDT 2015


https://bugs.freedesktop.org/show_bug.cgi?id=90795

--- Comment #10 from William Bader <williambader at hotmail.com> ---
I was thinking of something similar -- make a new OutputDev flag like
enableSplashCMYK to protect all of the SPLASH_CMYK code that is not obviously
protected inside a "case" or "if" for splashModeCMYK8 or splashModeDeviceN8,
although gcc might complain about possibly uninitialized variables.

poppler/PSOutputDev.cc could set the flag when
  (level == psLevel1Sep || level == psLevel2Sep ||
   level == psLevel3Sep || globalParams->getOverprintPreview())

My main goal is to have pdftops -level1sep run as fast as possible because I
periodically have to push batches of tens to thousands of PDFs through it, and
I expect that I will always need to build poppler specially for pdftops with
options like -DSPLASH_CMYK=1 -Ofast -DNDEBUG --enable-static --disable-shared
that would not be acceptable for someone trying to build libpoppler.so. That is
why I proposed making a configure option for a release build.
https://bugs.freedesktop.org/show_bug.cgi?id=90796

I would be happy enough to see a configure option for SPLASH_CMYK that explains
why you would want to enable it.  I think that most distributions build pdftops
SPLASH_CMYK disabled, and someone who does not know the poppler internals would
have a hard time connecting invalid output from pdftops with the currently
undocumented SPLASH_CMYK flag that they must manually add to CXXFLAGS.

pdftops is not the only affected utility. pdftoppm currently has a bug when
SPLASH_CMYK is enabled. https://bugs.freedesktop.org/show_bug.cgi?id=90570

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/poppler-bugs/attachments/20150706/5e899650/attachment-0001.html>


More information about the Poppler-bugs mailing list