GitLab

Some files like the one from KDE bug 426467 has the same signature
repeated 23 times in the Fields field, just return it one

Fixes https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=25411

    #0 0xf7ef8f19 in [vdso]
    #1 0xf7ccdd08 in gsignal (/lib32/libc.so.6+0x2bd08)
    #2 0xf7ccf206 in abort (/lib32/libc.so.6+0x2d206)
    #3 0xbdb9c2e in grealloc(void*, unsigned int, bool) gdal/poppler/goo/gmem.h:85:5
    #4 0xbdd9e11 in greallocn(void*, int, int, bool, bool) gdal/poppler/goo/gmem.h:171:12
    #5 0xc012373 in SplashPath::addStrokeAdjustHint(int, int, int, int) gdal/poppler/splash/SplashPath.cc:211:35
    #6 0xbfd156f in Splash::makeStrokePath(SplashPath*, double, bool) gdal/poppler/splash/Splash.cc:5987:34
    #7 0xbfcaec2 in Splash::strokeWide(SplashPath*, double) gdal/poppler/splash/Splash.cc:2028:13
    #8 0xbfc8a4d in Splash::stroke(SplashPath*) /src/gdal/poppler/splash/Splash.cc

Based on patch by Even Rouault

Minor gift to new contributors. Makes methods intention more obvious.

These are font faces with reference count 1. Their scaled fonts were already evicted from cairo holdover cache. This should be more functionally equivalent to old behaviour except that eviction is done lazily and not eagerly.

Thankfully coverity caught it before the release :)

As mentioned by @aacid in merge request comments.

Also through general note about helpful title and descriptions.

Source line iteration in Splash::scaleImageYuXuBilinear already tries
to prevent a read behind source image end, as the comment indicates it
and as it's conceptually reasonable. But the check for (currentSrcRow < srcHeight)
is wrong and doesn't do what it claims.

currentSrcRow will only ever increase to srcHeight - 1 after
scaledHeight iterations. Therefore the check always evaluates to
true, and src() is aways called. Intention was to prevented the line
fetch for the last run and leave line2 identical to line1 (the "extra padding").

Nothing bad happened, because SplashOutputDev::imageSrc and alphaImageSrc
gracefully handle the behind-end read. Should be corrected either.

in text selections, by:

 - Ignoring to draw characters with it.
 - Painting the selection's background as transparent.

Fixes issue #157

Based on inital work by Nelson Benitez and changed
to be not tesseract specific by Julian Andres Klode.

Not only one

Fixes KDE bug #425791

With an autotest

and bonus memory leak fixes for existing tests

nullptr means nullptr not null string

While at it move definitions of i and funcs down where used, also remove
the abuse of funcs[0] in one place and just declare a Function for it

WITH_FONTCONFIGURATION_WIN32 is either undefined or 1.

Issue #715

PopplerDocument's creation-date and mod-date properties are 32-bit unix
times, a.k.a. GTime, and will overflow in 2038. Deprecate these
properties and replace with creation-datetime and mod-datetime, which
are GDateTime instead, and add accessor functions.

Fixes #765

GTime was deprecated in glib 2.62 and will overflow in 2038.

The PopplerAttachment struct publicly uses GTime for the 'mtime' and
'ctime' fields. Deprecate these two fields and add accessor functions,
poppler_attachment_get_ctime and poppler_attachment_get_mtime, which
retun GDateTime* instead.

Fixes #765, Related to #715

The spec doesn't say it is, but Adobe Reader acts like it is, so emulate
its bug (or it's a bug in the spec)

Fixes KDE bug #425520