Age | Commit message (Collapse) | Author | |
---|---|---|---|
2016-01-20 | epub: Adjust text-align based on paragraph directionality. | Tor Andersson | |
In RTL context, text-align: left should be right, and vice versa. Since we currently ignore the dir="xxx" HTML attributes and CSS properties, always pick up the directionality from the text as if dir="auto". | |||
2016-01-20 | Various formatting cleanups. | Tor Andersson | |
2016-01-20 | Tidy bidirectional source. | Robin Watts | |
Make the import follow mupdf style (better, if not perfect). Use ucdn where possible to avoid duplicating tables. Shrink the types, make them explicit (e.g. use fz_bidi_level rather than int) and make tables const. Use 32-bit integers for text. | |||
2016-01-18 | Add support for bidirectional mirror characters | Robin Watts | |
2016-01-18 | Save some space in fz_html_flow with a union. | Robin Watts | |
2016-01-18 | Implement R2L characterwise plotting in the html layout. | Robin Watts | |
2016-01-18 | Ensure layout honours text directional (block level). | Robin Watts | |
2016-01-18 | Process HTML text for directionality. | Robin Watts | |
After we parse html text from an ebook run it through the unicode bidirectional algorithm to determine the directionality of each fragment. This may require splitting of fragments. Currently we don't do anything with this information. | |||
2016-01-18 | First import of bidi code. | Robin Watts | |
2016-01-18 | Move fz_html_flow to use a bitfield. | Robin Watts | |
Saves a bit of memory, and prepares for the addition of later fields (such as direction). | |||
2016-01-18 | Add consts to fz_keep/drop_path. | Robin Watts | |
Forgot these in my previous pass. | |||
2016-01-18 | Simplify try/catch macros. | Tor Andersson | |
Use a pointer to the top error stack slot instead of access via array and index. Return the stack slot from fz_push_try. | |||
2016-01-15 | pdf: Consume entire token before lexing numbers. | Tor Andersson | |
"0.00-70" should be parsed as one token, not two tokens as we did. | |||
2016-01-13 | Add lots of consts. | Robin Watts | |
In general, we should use 'const fz_blah' in device calls whenever the callee should not alter the fz_blah. Push this through. This shows up various places where we fz_keep and fz_drop these const things. I've updated the fz_keep and fz_drops with appropriate casts to remove the consts. We may need to do the union dance to avoid the consts for some compilers, but will only do that if required. I think this is nicer overall, even allowing for the const<->no const problems. | |||
2016-01-08 | epub: Fix memory leaks. | Tor Andersson | |
2016-01-08 | epub: Use pool allocator for fz_html box tree. | Tor Andersson | |
2016-01-08 | Add pool allocator. | Tor Andersson | |
2016-01-08 | Use a binned cache for font encoding lookups. | Tor Andersson | |
It's slower, but will work for CJK fonts as well. | |||
2016-01-08 | Optimize font advance and encoding caches. | Tor Andersson | |
2016-01-08 | pdf: Add function to look up the page for a named destination. | Tor Andersson | |
2016-01-08 | pdf: Fix pdf_annot memory leak. | Tor Andersson | |
2016-01-08 | Fix conditional jump or move depends on uninitialised value. | Tor Andersson | |
stm.avail_out is only set if !only_metadata. Reverse the order of the if tests to silence valgrind. | |||
2016-01-08 | Tweak lex_number to avoid (or minimise) underflow | Robin Watts | |
Keeps operations in the int domain as long as possible, and only resorts to floats if required. | |||
2016-01-06 | Squash some warnings | Robin Watts | |
2016-01-06 | epub: Cache font encoding lookups. | Tor Andersson | |
Add a caching table for the lower planes of unicode, which cover the latin, greek, cyrillic, hebrew and arabic scripts. | |||
2016-01-06 | epub: Only look for CSS in html head node. | Tor Andersson | |
2016-01-06 | epub: Speed up css application by sorting the matched property list. | Tor Andersson | |
2016-01-06 | js: Fix error handling bugs and one scoping bug. | Tor Andersson | |
js_pcall was leaving crap on the stack in case of errors. AFSimple_Calculate needs access to the doc object, but cannot use the usual 'this' binding since it's a function. Add a global reference to the doc object MuPDF_Doc. | |||
2016-01-06 | epub: Optimize CSS selector matching. | Tor Andersson | |
Remove strcpy and strtok calls when matching class conditions. | |||
2016-01-05 | Separate pdf_drop_annots (that drops lists) and fz_drop_annot. | Tor Andersson | |
2016-01-05 | Add pixmap struct accessors. | Tor Andersson | |
2016-01-05 | Stylistic naming cleanups. | Tor Andersson | |
2016-01-05 | Clean up trailing whitespace. | Tor Andersson | |
2016-01-05 | Change fz_ftoa to fz_grisu to remove one extra layer of function calls. | Tor Andersson | |
2016-01-05 | Speed up fz_ftoa. | Simon Reinhardt | |
During pdf_save_document the main performance bottleneck is the formatting of floats to decimal ASCII representations in fz_ftoa. Fix this by using the Grisu2 algorithm, the fastest known algorithm for accurate printing of IEEE floating point numbers while minimizing the number of produced decimal digits. This requires no libc support, only integer arithmetic. | |||
2016-01-05 | epub: Speed up CSS style application by using faster strtod. | Tor Andersson | |
CSS doesn't need precision, and doesn't support exponential notation. Use a simpler string to float conversion algorithm, which is magnitudes of order faster. | |||
2016-01-05 | epub: Speed up font layout by caching freetype advance widths. | Tor Andersson | |
2016-01-05 | Clarify snprintf length when printing to fz_buffer and fz_output. | Tor Andersson | |
The +1's for zero terminating bytes are not needed: printf to a fz_buffer or fz_output does not write a zero terminator. The extra code to add space for a zero terminator when calling snprintf internally are merely confusing. | |||
2016-01-05 | Remove fz_page argument from fz_annot function calls. | Tor Andersson | |
2015-12-28 | Drop 'jsimp' abstraction and use mujs directly. | Tor Andersson | |
2015-12-28 | Rename fz_image_get_pixmap to fz_get_pixmap_from_image. | Tor Andersson | |
2015-12-18 | Rename fz_image_get_sanitised_res to fz_image_resolution. | Tor Andersson | |
2015-12-18 | Remove fz_save_document and use pdf_save_document directly instead. | Tor Andersson | |
In preparation of adding pdf_write_document that writes a document to a fz_output stream. | |||
2015-12-18 | Rename fz_halftone_pixmap to fz_new_bitmap_from_pixmap. | Tor Andersson | |
2015-12-15 | Rename fz_buffer_cat to fz_append_buffer. | Tor Andersson | |
2015-12-15 | Rename fz_output_x to fz_write_pixmap_as_x etc. | Tor Andersson | |
2015-12-15 | Rename fz_write_x to fz_save_pixmap_as_x or fz_save_bitmap_as_x. | Tor Andersson | |
Separate naming of functions that save complete files to disk from functions that write data to streams. | |||
2015-12-14 | Add fz_clone_path | Robin Watts | |
Will be required for JNI bindings. | |||
2015-12-14 | Add fz_clone_text | Robin Watts | |
Will be required for JNI bindings. | |||
2015-12-14 | Add 'const' to the context lock and alloc structure pointers. | Robin Watts | |
This highlights the fact that we never alter the internals, and allows callers to pass in static const pointers. |