| Age | Commit message (Collapse) | Author |
|---|
|
Correctly transformed target coordinates for PDF.
Target coordinates for EPUB and HTML.
|
|
|
|
Just some typecasting required.
|
|
Add API to:
* allow enumeration of layer configs (OCCDs) within PDF files.
* allow selection of layer configs.
* allow enumeration of the "UI" (or "Human readable") form of layer
configs.
* allow selection/toggling of entries in the UI.
|
|
build.
|
|
|
|
|
|
|
|
In particular ensure that curl builds properly for 32 and 64bit
modes.
|
|
No point in building libfonts in anything other than release
mode. It's just data. This avoids needless rebuilds in batch
builds.
|
|
No point in building the 'generated' tools in anything other
than debug mode. This avoids repeatedly building the same thing
in batch builds.
|
|
This will only be built in "DebugCommercial" "ReleaseCommercial" and
"MementoCommercial" configuration builds.
A note on the way MSVS solutions work; libluratech is listed as
a dependency of the executables - this applies to ALL configurations.
In order to stop normal builds failing when we have no source, we
set each non "Commercial" configuration to not build libluratech.
The linker still looks for the output from the libluratech stage
to link in though, so for all non "Commercial" configurations, we
have the libluratech stage set to produce output named " ". This
is copied into the linker command line, and all is happy.
The sole downside to this is that if we try to build libluratech
in "Debug" (say) rather than "DebugCommercial", the build will
fail at the library making step. This will never happen unless
we manually try to build that project in that configuration though.
|
|
|
|
in 1dec53c.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
All link destinations should be URIs, and a document specific function
can be called to resolve them to actual page numbers.
Outlines have cached page numbers as well as string URIs.
|
|
|
|
|
|
|
|
|
|
In addition, make all callers passing 0 as a point pass NULL
instead.
|
|
As we skip through pages very quickly, it is apparently possible
to trigger a SEGV. Alex Talis has given a clear description of
the problem on the bug, and proposed this solution.
Essentially this tweaks our CancellableAsyncTask class to ensure
that we do not destroy the cookie before it has finished being
accessed.
|
|
|
|
|
|
As fz_drop_*()/fz_free() all must handle NULL.
|
|
|
|
|
|
|
|
Loading the outline automatically triggers a layout.
|
|
Code MUST pass a non-null context to all functions.
Checking ctx for null and failing silently is no more useful
than segfaulting.
fz_keep_imp and fz_drop_imp handle NULL pointers safely, so
the NULL checks for this can also be dropped at the same time.
|
|
Increase zoom limit to 64 times.
|
|
Only Windows and Android have form filling text input implemented
properly so far. Neither of those build using the Makefile. We
therefore disable text input in Makefile built viewers.
|
|
|
|
|
|
New Devanagari serif font, several other updated fonts.
|
|
|
|
|
|
Use a flag in the pdf_annot struct instead.
Don't pass pdf_document to annotation edit functions.
|
|
The implementation does not need to be in the public API.
|
|
When updating openjpeg, I forgot to update the Android
JNI makefiles with a newly created file.
Thanks to szukw000 for reporting this!
|
|
Move the definition of fz_font to be in a private header file
rather than in the public API. Add accessors for specific
parts of the structure and use them as appropriate.
The font flags, and the harfbuzz records remain public.
This means that only 3 files now need access to the font
implementation (font.c, pdf-font.c and pdf-type3.c). This
may be able to be improved further in future.
|
|
Part of the change in this code is to require opj_malloc
and co. Sadly, opj_malloc and co do not take a context field,
so a we need to wrap calls to openjpeg with a lock.
I am reusing the freetype lock here for simplicity.
|
|
|
