| Age | Commit message (Collapse) | Author |
|---|
|
Do not cache PDF fonts, since they have custom metrics and encodings that
cannot be shared. This is primarily to speed up font loading for EPUB, PDF
appearance synthesis, and any other uses of default metrics and unicode
encoded fonts.
|
|
|
|
|
|
It is safe to save a document with unsaved signatures to an fz_output,
provided it supports fz_stream_from_output.
|
|
Looks like x64 ones didn't 'stick' last time I edited them.
|
|
When updating from VS2005 to VS2017, the .rules mechanism is
converted to a .props based scheme. This is not entirely
equivalent, and so causes problems with our "build both
32 and 64bit resources" scheme.
Instead, we amend the flag handling in bin2coff so that it
will accept "Win32" and "x64" as flags. These can then be driven
easily from the project files, giving us a "more standard"
way of working.
Also, remove some stray fonts that had crept into the resources
build.
|
|
|
|
Having libluratech set to not build in non-commercial builds is
not enough to make stuff work, as the dependencies will still
try to link it in.
Instead, we now have a Release configuration for libluratech
with every file disabled, and the non-commercial builds depend
on that.
Also, tweak bin2coff build to put it's buildlog in the usual
place.
|
|
Everything builds in a batch build for me now, except for the OpenSSL
stuff, which fails due to me not having it available.
|
|
|
|
This is the windows equivalent of objcopy or ld -r binary.
We need different bin2coff font object files for 32 and 64 bit builds,
so put the font object files in AdditionalDependencies.
Revamp Windows builds to use bin2coff.
Fix bin2coff to not use '-' in symbol names.
Add a new project file to make bin2coff.
Add a .rules file to tell MSVC how to call the built version of bin2coff.
Update libresources build to use this rules file.
Update noto.c to correctly find the symbols we make.
|
|
|
|
|
|
|
|
Example:
mutool show $PDF pages/*/Resources/Fonts/*/Subtype
|
|
|
|
|
|
|
|
|
|
|
|
fz_xml_tag returns NULL for non-tag (i.e. text) elements.
Use the fz_xml_is_tag function instead of comparing strings.
|
|
|
|
|
|
|
|
|
|
|
|
Also don't bother adding an indirect object for the top resource dict.
|
|
|
|
|
|
|
|
The Noto symbol font has been split into two. The second symbol font only
contains a subset of the emojis, so reinstate the real emoji font and add
both symbol fonts to the fallback chain.
|
|
Previously if a glyphs units per em was 0 we'd try to divide
by zero, now mupdf assumes a default value and continues.
|
|
Previously if both cleaning and sanitizing content streams the
pages' resource dictionaries would retain the actually used
resources. If the content streams were only cleaned and not
sanitized the page's resource dictionaries were incorrectly
emptied. All resources, whether used or not, ought to be
retained, as is the case after this commit.
|
|
Suppose an application that uses MuPDF implements its own
allocator that limits the amount of memory that may be allocated.
When that condition occurs fz_malloc() will throw an exception,
this exception would then be thrown right back to MuPDF without
allowing for libjpeg to free any of the memory it has allocated.
After this commit fz_malloc_no_throw() is called, which under the
same conditions would simply return NULL. allowing for libjpeg to
free what ever allocations it has done, and return to MuPDF with
an error just like before.
|
|
Add some missing convenience data objects: font, highlight, position, etc.
Add color.convert and color.equal functions.
|
|
|
|
|
|
A 'null' pdf_obj is also NULL.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
PNaCl was deprecated in 2015.
WebOS hasn't seen use since 2012.
The beagleboard and tizen builds are no longer used.
The mingw and ios sections should suffice as examples.
|
|
|
|
|
|
|
|
|
