| Age | Commit message (Collapse) | Author |
|---|
|
min is a windows only define.
|
|
All of the group are updated which can make them act like radio buttons
in the case that all have different "on" values
|
|
Some fields have separate widget annotations. Some have the widget rolled
in. This commit correctly identifies the field when setting the field value,
thus fixing bug 693145
|
|
|
|
Currently pdf_lexbufs use a static scratch buffer for parsing. In
the main case this is 64K in size, but in other cases it can be
just 256 bytes; this causes problems when parsing long strings.
Even the 64K limit is an implementation limit of Acrobat, not an
architectural limit of PDF.
Change here to allow dynamic buffers. This means a slightly more
complex setup and destruction for each buffer, but more importantly
requires correct cleanup on errors. To avoid having to insert
lots more try/catch clauses this commit includes various changes to
the code so we reuse pdf_lexbufs where possible. This keeps the
speed up.
|
|
|
|
The About box was resized on master, and this change was lost
during an earlier merge.
|
|
Avoid overflowing fz_buffer_printf by using a dedicated string
output function; this also solves potential problems with us not
escaping chars in strings correctly.
Unfortunately this doesn't completely cure the bug as we run
straight into problems in the lexer.
|
|
Mudraw is updated so when the -j flag is used it guesses the length
of the text it should output for each field. It also outputs a
"Lorem Ipsum" string, which should hopefully test the layout
facilities a bit better.
Mujstest is updated so it can read the escaped chars from that
string.
|
|
This fixes bug 693126
|
|
|
|
|
|
This fixes the problem with the test file in bug 693125, although it
doesn't make radio buttons act correctly. The file, rather than using
radio buttons, was actually using check boxes with a reset-form action
to achieve the exclusivity.
|
|
We now create pdf_annot objects for PDF annotations even if they have no
appearance stream (i.e. even if invisible). That is necessary because
even invisible annotations can be targets of user interaction.
This is at least a partial fix for bug 693131
|
|
Previously, we've been working with a prebuild v8 lib, only available
in a single (debug) configuration.
Here, we move to reading the include files from
thirdparty/v8-3.9/include and getting the built libs from
thirdparty/v8-3.9/build/{Debug,Release}/lib/ as appropriate.
|
|
If the v8 lib is present in thirdparty, add the above binaries
to the 'all' target.
|
|
Mostly char * <-> unsigned char * conversions.
|
|
"verbose" seems to be a reserved word.
Various unused vars, and warnings that didn't show up on windows.
|
|
|
|
Remove the shim indirection layer for fz_document. A little less type
safe, but a lot less boiler plate.
|
|
|
|
|
|
|
|
Rather than having a dedicated call to enumerate the rectangles for the
annotations on a page, add an interface for enumerating annotations
with accessor functions. Currently the only accessor function is
the one to get the annotation rectangle.
Use this new scheme in place of fz_bound_annots within mudraw.
Also use this scheme to set the caret cursor in the viewer when over
a data field.
|
|
Add/Remove a few consts to pacify MSVC.
|
|
Refactor the text-widget updating code to use the later-written, button-updating
technique, which accounts for rotation. Also now delays generation of
appearance streams until required for rendering.
|
|
|
|
This is initialised to rgb or bgr according to whether _WIN32 is set
as has always been the case. This allows apps that want to override
it (such as mujstest) to do so though.
|
|
When clicking on a form field, especially with breakpoints in play,
it is possible for the up/down click logic to get confused.
Improve it by using a flag to avoid processing a down click twice.
|
|
When encountering top bit set chars in script files, behave well.
|
|
The -v flag tells mujstest to echo the lines read from the script file
as it runs (useful to see at what point things crash).
The -p flag gives a prefix to apply to the paths from in the script
file.
|
|
Currently we were attempting to load the javascript for a document
immediately on opening it. Here we delay it until 1) the encryption
for a document has been loaded, and 2) any repair required to a
document has been done.
This solves various problems, which were leading (indirectly) to bug
693128.
|
|
Fix one pdf_js function that didn't protect agains doc->js == NULL.
Also change exception handling in pdf_new_js so that it doesn't
abort just because of a single failed stream load.
|
|
|
|
We only seem to have a debug build of the v8 build to work with at
the moment though...
|
|
We add a new fz_bound_annots function (and associated pdf_bound_annots
function) that calls a given callback with the page rectangle of the
annotations on a given page. This is marked as being a 'temporary'
function, so we can remove it/change it in future if required. It
seems likely that we'll want to have some sort of 'iterate over
annotations' function eventually, and this does the job for now.
Add a -j flag to mudraw that outputs a simple mujstest script.
For each page with annotations, the script jumps to that page, then
for each annotation on the page, it sets some text to be entered,
and clicks the annotation. In the case of text fields, this will cause
the text to be entered into that text field; in the case of buttons
it will execute the button.
At the end of each page with annotations, the script is told to
snapshot the page.
These test scripts are not designed to be full tests, but they do
at least provide an easy way for us to generate scripts where
every field in our test suite is interacted with.
|
|
Simple command line tool made from cutting all the windows specifics
out of win_main.c and adding a simple script handler in.
Read lines from the script, and feed those events to pdfapp. Screenshot
pages as required.
|
|
This is important, otherwise we can get into an unexpected state and
subsequent mouse moves appear as pans.
|
|
Testing tests_private/v1.3/09+20anim+20fx-fo-dx-adr-mx.pdf with
mujstest shows an infinite loop due to an initialisation that
uses FLT_MAX rather than INT_MAX.
Even using INT_MAX still shows problems until a cast is added.
|
|
These functions currently call pdf_array_put, but this fails to
extend the array. Change to use pdf_array_push instead.
|
|
|
|
|
|
Conflicts:
fitz/fitz-internal.h
fitz/stm_buffer.c
pdf/mupdf-internal.h
|
|
|
|
|
|
Harmless, since the context wasn't used, but confusing.
|
|
|
|
|
|
Make mudraw pass a cookie in to the rendering procedures. If any errors
are reported for any page, remember this, and set the return code to 1
on exit.
|
|
|
