3754 Commits

Author SHA1 Message Date
Yury Delendik
20a91bcdbf Fixes ignoring of the escaped CR LF 2014-03-20 11:50:12 -05:00
Yury Delendik
878a123e47 Merge pull request #4478 from nnethercote/plain-fnArray
Use a vanilla array for fnArray
2014-03-20 10:30:29 -05:00
Brendan Dahl
10deadd416 Merge pull request #4453 from nnethercote/charToGlyph
Add a cache for glyphs
2014-03-19 16:30:02 -07:00
Yury Delendik
1801fb2c37 Merge pull request #4474 from chriskr/draw-image-subpixel-support
Introduce paintSolidColorImageMask command to handle 1x1 solid image
2014-03-19 14:31:15 -05:00
Yury Delendik
305274cd45 Merge pull request #4481 from bthorben/refactor/pdf_manager.js
refactor src/core/pdf_manager.js: rename pdfModel to pdfDocument
2014-03-19 10:33:42 -05:00
Christian Krebs
fc9edfcf76 Introduce paintSolidColorImageMask command to handle 1x1 solid image masks
See https://github.com/mozilla/pdf.js/issues/4436 for details
2014-03-19 13:27:59 +01:00
Thorben Bochenek
0fa44d7233 refactor src/core/pdf_manager.js: rename pdfModel to pdfDocument
Rename pdfModel to pdfDocument to let the name closer resemble what the
variable actually holds
2014-03-19 09:43:16 +01:00
Yury Delendik
329d0ec30b Merge pull request #4464 from bthorben/issue#4461
Fixes pages without Resources in their dictionary
2014-03-18 13:00:21 -05:00
Thorben Bochenek
8edebc797b Refactor and fix style of core.js 2014-03-18 14:10:44 +01:00
Thorben Bochenek
1ddb019449 Fix Issue #4461: Pages without \Resources in their dictionary fail to load
When no "\Resource" entry could be found return an empty dictionary
2014-03-18 14:10:44 +01:00
Nicholas Nethercote
292b96a0bf Make fnArray always be a plain array.
This makes the code much simpler, and the extra memory use is tiny -- a vanilla
1000 element array is only 4000 bytes larger than a Uint32Array of the same
size.
2014-03-17 17:48:53 -07:00
Nicholas Nethercote
42d175efd4 Fix some indenting within OperatorList. 2014-03-17 17:11:40 -07:00
Brendan Dahl
1802ffffb8 Merge pull request #4447 from nnethercote/object-reduction
Allocate fewer objects
2014-03-17 12:50:23 -07:00
Brendan Dahl
68be273c69 Merge pull request #4470 from yurydelendik/packcmap
CMaps binary packing
2014-03-17 12:27:35 -07:00
Brendan Dahl
ad4be03a08 Merge pull request #4440 from bthorben/refactor/pdf_manager.js
refactor src/core/pdf_manager.js: rename pdfModel to pdfDocument
2014-03-17 10:41:17 -07:00
Jonas Jenwald
5f021b067c Prevent infinite loop in CFFParser_parseHeader 2014-03-17 11:47:14 +01:00
Brendan Dahl
1a6e103c35 Merge pull request #4463 from sukop/patch-1
Fix copy-paste typo
2014-03-14 15:24:16 -07:00
Yury Delendik
69efd9cb96 CMaps binary packing 2014-03-14 16:46:35 -05:00
Tim van der Meij
67c686027c Merge pull request #4465 from Snuffleupagus/src-display-braces
Fix coding style in src/display
2014-03-14 21:11:15 +01:00
Yury Delendik
8266225b73 Merge pull request #4434 from p01/jbig2Optimizations
Jbig2 optimizations
2014-03-14 10:39:41 -05:00
Juraj Sukop
4c2a462a76 Fix copy-paste typo
The Compact Font Format Specification defines the Standard String of SID 237 as "twodotenleader", and not "266 ff".
2014-03-14 14:11:16 +01:00
Jonas Jenwald
a1ee567d60 Fix coding style in src/display/api.js 2014-03-14 13:24:04 +01:00
Jonas Jenwald
bc986a3029 Fix coding style in src/display/canvas.js 2014-03-14 13:11:11 +01:00
Nicholas Nethercote
6c698519cd Avoid allocations in executeCommand(). 2014-03-13 22:15:09 -07:00
Nicholas Nethercote
3759c11f42 Avoid more allocations in bidi(). 2014-03-13 22:15:09 -07:00
Nicholas Nethercote
2e93a0cc98 Avoid allocations in bidi(). 2014-03-13 22:15:09 -07:00
Nicholas Nethercote
6a75e45309 Allocate fewer objects when parsing 2 and 4 byte chunks.
This is achieved by adding getBytes2() and getBytes4() to streams, and by
changing int16() and int32() to take multiple scalar args instead of an array
arg.
2014-03-13 22:15:05 -07:00
Nicholas Nethercote
0866ad5bff Add a cache for glyphs.
This reduces memory consumption for text heavy documents. I tested five
documents and saw hit rates ranging from 97.4% to 99.8% (most of the misses are
due to |width| varying even when |fontChar| matches). On two of those documents
I saw improvements of 40 and 50 MiB.

The patch also introduces the Glyph constructor, and renames the |unicodeChars|
local variable as |unicode| for consistency with the corresponding Glyph
property.
2014-03-13 19:51:48 -07:00
p01
bd9419f1ef Micro optimization of ArithmeticDecoder_readBit 2014-03-13 17:38:12 +01:00
p01
2b6cb317db Optmized decodeRefinement, moving the prediction error outside the inner loop
There is no need to slow down the inner loop with a test for ltp as it can only
change if prediction is true in which case it only changes in the outer loop.
2014-03-13 17:38:12 +01:00
p01
840d9d40b6 Reuse the row Uint8Array in JBIG2's decodeBitmap 2014-03-13 17:38:11 +01:00
p01
b2b3836683 Process only once the offset and starting mask JBIG2's drawBitmap 2014-03-13 17:38:11 +01:00
p01
0bca56cbb2 Use changingTemplateX, Y and Bit in JBIG2's decodeBitmap 2014-03-13 17:38:01 +01:00
p01
e2a4328ff2 Optimized bounding box tests in JBIG2's decodeBitmap 2014-03-13 17:38:00 +01:00
p01
2a9fe573c0 Nest JBIG2 prediction if statements together
ltp can only become truthy if predition is true.
2014-03-13 17:37:59 +01:00
p01
02a5f9435a More bit friendly decodeIAID and decodeInteger 2014-03-13 17:37:55 +01:00
p01
e862c64bb8 JBIG2 20% faster SimpleSegmentVisitor_drawBitmap
Moved the switch(combinationOperator) outside the width and height loops
Check bitmap[i][j] before applying the switch(combinationOperator)
2014-03-13 17:37:53 +01:00
Yury Delendik
1aaeec8020 Heuristic to optimize chunking 2014-03-13 11:01:34 -05:00
Yury Delendik
257898b359 Caching inlined mask images 2014-03-13 11:01:34 -05:00
Yury Delendik
f48f57e30a Caches last parsed resource image, recornizes image repeats 2014-03-13 11:00:10 -05:00
Yury Delendik
2c23be369b Optimizing pdf when text printed char-by-char 2014-03-13 11:00:10 -05:00
Yury Delendik
dbb3d17d8f Refactors optimization list 2014-03-13 11:00:10 -05:00
Yury Delendik
1c0e1cc591 Merge pull request #4444 from nnethercote/min-length
Estimate the size of decoded streams in advance.
2014-03-13 10:58:31 -05:00
Yury Delendik
2778d75d2c Merge pull request #4437 from yurydelendik/issue4398
Fixes corrupted by Aspose fonts
2014-03-13 10:44:02 -05:00
Yury Delendik
03bedaa316 Merge branch 'master' of github.com:mozilla/pdf.js into issue-4370 2014-03-13 10:38:11 -05:00
Jonas Jenwald
b0824e1626 Prevent circular reference when checking for blend modes (issue 4370) 2014-03-13 10:36:49 -05:00
Yury Delendik
7963f22545 Merge pull request #4259 from brendandahl/built-in-cmaps-squash
Adds built in CMaps and unifies the glyph mapping.
2014-03-13 10:27:48 -05:00
Yury Delendik
31ea4e0f4d Merge pull request #4420 from prometheansacrifice/RemovingCheckSetPresence
Removing set-presence-in-ImageData check from canvas.js
2014-03-13 08:31:53 -05:00
Yury Delendik
5c20d17c87 Merge pull request #4431 from fkaelberer/ExtractArithmeticDecoder
Extract arithmetic decoder to own file
2014-03-13 08:19:11 -05:00
Nicholas Nethercote
b3024db677 Estimate the size of decoded streams in advance.
When decoding a stream, the decode buffer is often grown multiple times, its
byte size increasing like so: 512, 1024, 2048, etc. This patch estimates the
minimum size in advance (using the length of the encoded stream), often
allowing the smaller sizes to be skipped. It also renames numerous |length|
variables as |maybeLength| to make it clear that they can be |null|.

I measured this change on eight documents. This change reduces the cumulative
size of decode buffer allocations by 0--32%, with 10--20% being typical. This
reduces peak RSS by 10 or 20 MiB for several of them.
2014-03-13 02:06:58 -07:00