1531 Commits

Author SHA1 Message Date
Yury Delendik
2aea7d7047 Merge pull request from nnethercote/Ref-toString
Optimize Ref_toString().
2014-07-25 10:10:10 -05:00
Nicholas Nethercote
1039791472 Use Array.join to build up strings in readPostScriptTable().
This avoids about 5 MiB of string allocations on one test case.
2014-07-24 16:12:08 -07:00
Nicholas Nethercote
856e1c600b Optimize Ref_toString().
I have a large PDF where this function is called 1.6 million times
during loading. Minimizing the string concatenations reduces the
cumulative allocations done by Firefox within this function from 113 MB
to 48 MB.
2014-07-24 06:49:56 -07:00
Jonas Jenwald
2485f11829 Fix loading of PDF files with invalid or missing Type3 characters (issue 5039) 2014-07-24 15:03:22 +02:00
Nicholas Nethercote
501446ccc4 Optimize common cases in hexToStr().
This avoids the creation of over two million array objects when viewing
http://www.dynacw.co.jp/Portals/3/fontsamplepdf/sample_4942546800828.pdf,
and reduces load time from 76 to 73 ms.
2014-07-22 23:26:03 -07:00
Nicholas Nethercote
c7f02d2c8e Minimize memory usage of font-related arrays.
This patch replaces some vanilla arrays with typed arrays, and avoids
some array copying.

It reduces the peak RSS when viewing
http://www.dynacw.co.jp/Portals/3/fontsamplepdf/sample_4942546800828.pdf
from ~940 MiB to ~750 MiB, and reduces its load time from 83 to 76 ms.
2014-07-22 22:47:45 -07:00
Jonas Jenwald
f13c217b25 Fix another seac regression (issue 4801) 2014-07-22 21:44:13 +02:00
Jonas Jenwald
b950118681 Revert commit fc73e2e (PR 5005) for breaking certain PDF files 2014-07-22 21:17:57 +02:00
Yury Delendik
53320ce734 Merge pull request from Snuffleupagus/issue-5010
Prevent CMapFactory.create from failing by passing the necessary parameters from PartialEvaluator_readToUnicode (issue 5010)
2014-07-22 10:54:35 -05:00
Yury Delendik
584fef90ab Merge pull request from Snuffleupagus/issue-5036
Add |fillRgb| method to LabCS
2014-07-21 09:55:55 -05:00
Jonas Jenwald
9c6316fc15 Merge pull request from fkaelberer/faster_ChunkedStream_getByte
Faster chunkedStream_getByte()
2014-07-18 18:23:49 +02:00
Jonas Jenwald
0237d5036a Merge pull request from nnethercote/share-zero-length-buffers
Improve how DecodeStream handles empty buffers.
2014-07-13 12:13:06 +02:00
Jonas Jenwald
1cb4de2227 Add |fillRgb| method to LabCS 2014-07-10 12:06:19 +02:00
Jonas Jenwald
a7c786775d [CIDFontType2] Map characters missing in toUnicode to the private use area (bug 1028735 and issue 4881) 2014-07-05 00:18:51 +02:00
Nicholas Nethercote
a483c80fc3 Make QueueOptimizer easier to read.
QueueOptimizer is really hard to read. Enough so that it's blocking my
efforts to streamline the representation used for operator lists.

This patch improves its readability in the following ways.

- More descriptive variable names make the sequence checking much clearer,
  as do additional comments.

- The addState() functions now return the index of the first op past the
  sequence, instead of setting context.currentOperation to the last op of
  the sequence.

- The loop in optimize() is clearer.

- The array modification in the fourth addState() function is much clearer
  -- we're just removing trios of ops.

- All four |addState| functions are now more consistent with each other.

I used some debug printfs to find documents where these optimizations are
used and then checked that the number of optimized ops was the same before
and after my changes.
2014-07-03 19:16:31 -07:00
Nicholas Nethercote
db866945b7 Improve how DecodeStream handles empty buffers.
DecodeStream currently initializes its |buffer| field to |null|, which
is reasonable, because lots of DecodeStreams never need to instantiate a
buffer. But this requires various special cases in the code.

This patch change it so DecodeStreamClosure has a single empty
Uint8Array which gets shared between all buffers upon initialization.
This avoids the special cases.

DecodeStream.prototype.ensureBuffer() is really hot, and this removes a
test from the fast path. For one 226 page scanned document this sped up
rendering by about 2%.
2014-07-02 18:53:21 -07:00
fkaelberer
fc73e2e173 use getBytes() instead of looping over getByte() 2014-06-27 09:09:54 +02:00
Jonas Jenwald
c5f4051a75 A few small optimizations of adjustMapping
Replace a couple of |in| checks with comparisons against undefined.
2014-06-27 00:59:42 +02:00
Jonas Jenwald
c121def806 A few small optimizations for CIDFontType2 fonts
Cache a constant length and replace one usage of |in| with a comparison against undefined.
2014-06-27 00:52:54 +02:00
Jonas Jenwald
04975acceb Prevent CMapFactory.create from failing by passing the necessary parameters from PartialEvaluator_readToUnicode (issue 5010) 2014-06-27 00:46:16 +02:00
fkaelberer
9a41659ae7 Faster chunkedStream_getByte() 2014-06-26 22:34:00 +02:00
Yury Delendik
6d5a04149b Merge pull request from pramodhkp/rectelmnt
Combine re element into constructPath
2014-06-24 09:27:21 -05:00
pramodhkp
8407d28c9e Combine re element into constructPath 2014-06-25 00:27:42 +05:30
Yury Delendik
10db93be29 Merge pull request from Snuffleupagus/bug-1027533
Additional heuristics to recognize unknown glyphs for toUnicode (bug 1027533)
2014-06-23 21:56:13 -05:00
Yury Delendik
bb7e7d33c5 Merge pull request from CodingFabian/restructure-evaluator-read
Restructured EvaluatorPreprocessor_read to be more natural.
2014-06-23 21:50:14 -05:00
Yury Delendik
c28839b2f3 Merge pull request from Snuffleupagus/issue-4934
Don't blindly trust toUnicode when building toFontChar for non-standard fonts without a font file (issue 4934)
2014-06-23 21:49:24 -05:00
Fabian Lange
60f67c3961 Restructured EvaluatorPreprocessor_read to be more natural. 2014-06-23 23:35:25 +02:00
Yury Delendik
3ad58db7e8 Merge pull request from nnethercote/use-null-for-zero-args
Use null instead of [] for ops with no args.
2014-06-23 15:38:48 -05:00
Nicholas Nethercote
081866a184 Use null instead of [] for ops with no args.
This reduces peak RSS on one test file from ~600 to ~560 MiB.
2014-06-22 16:03:48 -07:00
Jonas Jenwald
b19bb74813 Additional heuristics to recognize unknown glyphs for toUnicode (bug 1027533) 2014-06-20 09:57:16 +02:00
Yury Delendik
84157e039d Merge pull request from nnethercote/better-ref-keys
Factor out repeated Ref key string generation code.
2014-06-19 21:00:09 -05:00
Nicholas Nethercote
1ad3ffbc7b Factor out repeated Ref key string generation code.
In src/core/obj.js, we convert a Ref to a string to index into a table like
this: 'R1.0'.  This conversion is repeated numerous times.

This patch factors out the conversion into a new function.
Ref.prototype.toString().
2014-06-19 18:22:39 -07:00
Yury Delendik
c0a6b0f308 Merge pull request from yurydelendik/rm-suppressEncryption
Removes error catch from fetchUncompressed()
2014-06-18 22:03:09 -05:00
Yury Delendik
b557b87fc9 Merge pull request from nnethercote/preprocessor-read
Avoid allocating return object in EvaluatorPreprocessor_read().
2014-06-18 22:00:31 -05:00
Nicholas Nethercote
17170af3c7 Avoid allocating return object in EvaluatorPreprocessor_read().
This function can be called 100s of 1000s or even millions of times, and the
allocated return object accounts for 10% of all GC thing allocations for some
documents. It's easy to avoid, which reduces stress on the garbage collector,
and this patch does that.
2014-06-18 16:41:29 -07:00
Yury Delendik
623fa29300 Removes error catch from fetchUncompressed() 2014-06-18 18:30:27 -05:00
Yury Delendik
fbdab2c7c5 Not ignoring MissingDataException exception. 2014-06-18 18:24:54 -05:00
Yury Delendik
cf4bc42e33 Merge pull request from nnethercote/glyphBuf
Build up textChunk.str more efficiently.
2014-06-18 17:51:07 -05:00
Yury Delendik
88fd1aa78b Removes PDFJS.Annotation 2014-06-18 16:58:11 -05:00
Jonas Jenwald
2282c98500 Merge pull request from yurydelendik/annotations
Splits shared/annotation.js into core/ and display/
2014-06-18 17:01:38 +02:00
Nicholas Nethercote
bce7601480 Build up textChunk.str more efficiently.
PartialEvaluator.getTextContent() builds up textChunk strings 1 char at a time,
creating many 100s of 1000s of intermediate strings along the way. This patch
make it instead push chars to an array and then join them at the end, as we
have done in numerous other places.
2014-06-18 07:48:22 -07:00
Nicholas Nethercote
4428cebdbc Add ChunkedStream.ensureByte().
This new function is much faster than ensureRange(pos, pos+1), which is a very
common case.

This speeds up the rendering of some test cases (including the Tracemonkey
paper) by 4--5%.
2014-06-17 21:33:48 -07:00
Yury Delendik
bdeca30fbf Splits shared/annotation.js into core/ and display/ 2014-06-17 17:43:33 -05:00
Yury Delendik
bad24bf707 Merge pull request from fkaelberer/fasterJPEGtransform
Faster JPEG transform
2014-06-17 09:03:23 -05:00
Yury Delendik
5a2e511cbd Merge pull request from timvandermeij/rename-concatenate
Renames concatenateToArray to appendToArray
2014-06-17 08:21:47 -05:00
Jonas Jenwald
ab67e1c272 Let Parser_makeFilter return NullStream when an invalid stream is encountered (issue 3417) 2014-06-17 12:03:34 +02:00
fkaelberer
f9cde5d93e faster JPEG transform 2014-06-17 10:09:17 +02:00
Jonas Jenwald
22cfcbcf8a Merge pull request from yurydelendik/telemetry
Collect More Telemetry Data
2014-06-17 00:36:58 +02:00
Yury Delendik
0cd28ebfa3 Telemetry for used stream and font types 2014-06-16 16:41:04 -05:00
Tim van der Meij
9c072a5d4b Renames concatenateToArray to appendToArray 2014-06-16 22:10:10 +02:00