Sakurai/pdf.js
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
11,417 Commits 1 Branch 80 Tags
Commit Graph

3460 Commits

Author SHA1 Message Date
Jonas Jenwald
4b15e8566b Remove the Function.prototype.bind polyfill 
This is only relevant for browsers that we don't intend to support with PDF.js version `2.0`.
 2017-11-21 11:00:55 +01:00
Jonas Jenwald
d8cb74d3e3 Remove the btoa/atob polyfills 
This is only relevant for browsers that we don't intend to support with PDF.js version `2.0`.
 2017-11-21 11:00:55 +01:00
Jonas Jenwald
150ac0788f Remove IE9 specific XMLHttpRequest polyfills that utilize VBArray 2017-11-21 11:00:55 +01:00
Jonas Jenwald
935c5c587f Remove the Object.defineProperty polyfill 
This is only relevant for browsers that we don't intend to support with PDF.js version `2.0`.
 2017-11-21 11:00:55 +01:00
Tim van der Meij
25b07812b9
Sanitize the display value for choice widget annotations 2017-11-18 20:37:27 +01:00
Tim van der Meij
9e8cf448b0
Merge pull request #9140 from Snuffleupagus/rm-console-polyfill 
Remove the `console` polyfills
 2017-11-18 15:49:19 +01:00
Tim van der Meij
edaf4b3173
Merge pull request #9037 from Snuffleupagus/refactor-streams-params 
Re-factor how parameters are passed to the network streams
 2017-11-18 15:41:15 +01:00
Tim van der Meij
ae07adf143
Merge pull request #9073 from Snuffleupagus/image-streams-fixes 
Fix the interface of `JpegStream`/`JpxStream`/`Jbig2Stream` to agree with the other `DecodeStream`s
 2017-11-17 23:26:36 +01:00
Tim van der Meij
1d67d9dccd
Merge pull request #9131 from janpe2/svg-empty-paths 
Filling and stroking empty paths in SVG backend
 2017-11-16 22:43:24 +01:00
Jonas Jenwald
42099c564f Remove the console polyfills 
All browsers that we intend to support with PDF.js version 2.0 already supports `console` natively.
 2017-11-16 09:34:51 +01:00
Jonas Jenwald
d5174cd826 Remove the requestAnimationFrame polyfill 
According to https://developer.mozilla.org/en-US/docs/Web/API/window/requestAnimationFrame#Browser_compatibility and https://caniuse.com/#feat=requestanimationframe, the browsers we intend to support with PDF.js version 2.0 should all have native `requestAnimationFrame` support.

Note that the reason for indiscriminately polyfilling `requestAnimationFrame` in iOS, see PR 4961, was apparently because of a bug in iOS 6.
However, according to [Wikipedia](https://en.wikipedia.org/wiki/IOS_version_history#iOS_8): "Support for iOS 8 ended in 2017.", hence the lowest version currently supported is iOS 9.
 2017-11-15 16:08:48 +01:00
Jani Pehkonen
4e8f7070da Filling and stroking empty paths in SVG backend 2017-11-14 18:35:39 +02:00
Jonas Jenwald
745cb73c65 Remove PDFJS.disableRange/PDFJS.disableStream code for now unsupported browsers in src/shared/compatibility.js 
We're currently disabling range requests and streaming for a number of configurations. A couple of those will no longer be supported (with PDF.js version 2.0), hence we ought to be able to clean up the compatibility code slightly.
 2017-11-14 15:28:50 +01:00
Jonas Jenwald
eb3a1f24a3 Remove the PDFJS.disableHistory code from src/shared/compatibility.js 
This compatibility code is only relevant for browsers that will no longer be supported (with PDF.js version 2.0), hence we ought to be able to remove it.
 2017-11-14 15:28:50 +01:00
Tim van der Meij
9686f6652c
Merge pull request #9089 from yurydelendik/rm-chunks 
Extracts OperatorList class and prepares for streaming
 2017-11-13 23:35:40 +01:00
Jonas Jenwald
23699cef1c Re-factor how parameters are passed to the network streams 
*This patch is the result of me starting to look into moving parameters from `PDFJS` into `getDocument` and other API methods.*

When familiarizing myself with the code, the signatures of the various network streams seemed to be unnecessarily cumbersome since `disableRange` is currently handled separately from other parameters.
I'm assuming that the explanation for this is probably "for historical reasons", as is often the case. Hence I'd like to clean this up *before* we start the larger, and more invasive, `PDFJS` parameter re-factoring.
 2017-11-11 11:23:29 +01:00
Jonas Jenwald
de5297b9ea Fix the interface of JpegStream/JpxStream/Jbig2Stream to agree with the other DecodeStreams 
The interface of all of the "image" streams look kind of weird, and I'm actually a bit surprised that there hasn't been any errors because of it.
For example: None of them actually implement `readBlock` methods, and it seems more luck that anything else that we're not calling `getBytes()` (without providing a length) for those streams, since that would trigger a code-path in `getBytes` that assumes `readBlock` to exist.

To address this long-standing issue, the `ensureBuffer` methods are thus renamed to `readBlock`. Furthermore, the new `ensureBuffer` methods are now no-ops.
Finally, this patch also replaces `var` with `let` in a number of places.
 2017-11-11 11:22:16 +01:00
Jonas Jenwald
36593d6bbc Move JpegStream and JpxStream to their own files 2017-11-11 11:22:16 +01:00
Yury Delendik
5fa56f6a9d For backwards compatibility: use addOp amount instead of queue size. 2017-11-09 18:46:48 -06:00
Yury Delendik
877c2d7743 Changing QueueOptimizer to be more iterative. 2017-11-09 18:46:48 -06:00
Max Schaefer
3ae37d1b06 Remove a few useless assignments. 2017-11-03 11:36:48 +00:00
Max Schaefer
bc8f673522 Remove spurious arguments to NullStream constructor. 2017-11-03 10:14:32 +00:00
Max Schaefer
3ab1a9922a Rearrange a few declarations so that they precede their uses. 2017-11-03 10:14:32 +00:00
Jonas Jenwald
2dbd3f2603 [api-major] Remove the TypedArray polyfills 2017-11-01 10:31:28 +01:00
Brendan Dahl
b46443f0c1
Merge pull request #9077 from yurydelendik/v2 
Version 2.0 merge
 2017-10-31 14:24:20 -07:00
Jonas Jenwald
83e8398ff2 For non-embedded fonts, map softhyphen (0x00AD) to regular hyphen (0x002D) (issue 9084) 
In the PDF file, the `ToUnicode` data first maps the hyphen correctly, and then *overwrites* it to point to the softhyphen instead. That one cannot be rendered in browsers, and an empty space thus appear instead.

Fixes 9084.
 2017-10-31 13:26:04 +01:00
Jonas Jenwald
92fcfce685
Merge pull request #9082 from brendandahl/issue7562 
Overwrite glyphs contour count if it's less than -1.
 2017-10-30 20:44:01 +01:00
Yury Delendik
85f544f55a Moves OperatorList and QueueOptimizer into separate file. 2017-10-30 13:29:58 -05:00
Brendan Dahl
17037b5e51 Overwrite glyphs contour count if it's less than -1. 
The test pdf has a contour count of -70, but OTS doesn't
like values less than -1.

Fixes issue #7562.
 2017-10-30 09:16:51 -07:00
Yury Delendik
b4e25fb2e8 Merge remote-tracking branch 'mozilla/version-2.0' into v2 2017-10-27 14:01:45 -05:00
Jonas Jenwald
5e627810e4 Use stringToBytes in more places 
Rather than having (basically) verbatim copies of `stringToBytes` in a few places, we can simply use the helper function directly instead.
 2017-10-26 11:01:13 +02:00
Jonas Jenwald
e94a0fd4e7 Extract the actual decoding in CCITTFaxStream into a new CCITTFaxDecoder "class", which the new CCITTFaxStream depends on 2017-10-24 16:03:08 +02:00
Jonas Jenwald
bb35095083 Move CCITTFaxStream and Jbig2Stream, from src/core/stream.js, to separate files 2017-10-24 12:00:40 +02:00
Jonas Jenwald
d71a576b30 Merge pull request #9045 from brendandahl/sani-name 
Sanitize name index in compile phase of CFF.
 2017-10-24 11:48:03 +02:00
Brendan Dahl
6b12612a52 Sanitize name index in compile phase of CFF. 
Fixes #8960
 2017-10-23 17:13:49 -07:00
Tim van der Meij
a0ec980e63 Merge pull request #9055 from Snuffleupagus/core-js-Number 
Replace `Number` polyfills with the ones from core-js
 2017-10-21 16:15:11 +02:00
Jonas Jenwald
dbbb763eaf Replace Number polyfills with the ones from core-js 
Since we're already using core-js elsewhere in `compatibility.js`, we can reduce the amount of code we need to maintain ourselves.

https://github.com/zloirock/core-js#ecmascript-6-number
 2017-10-21 12:58:53 +02:00
Jonas Jenwald
8684aa0ee5 Replace our Promise polyfill with the one from core-js 
Since we're already using core-js elsewhere in `compatibility.js`, we can reduce the amount of code we need to maintain ourselves.

https://github.com/zloirock/core-js#ecmascript-6-promise
 2017-10-21 12:51:14 +02:00
Tim van der Meij
10fd590c09 Merge pull request #9032 from Snuffleupagus/nativeImageDecoderSupport-2.0 
Simplify the check, and remove the warning, for the `nativeImageDecoderSupport` API parameter
 2017-10-20 21:30:47 +02:00
Brendan Dahl
fcc9943d04 Use charstring as plain text when lengthIV is -1. 
Fixes #7769
 2017-10-18 14:19:59 -07:00
Tim van der Meij
17cc94db4e Merge pull request #9034 from Snuffleupagus/javascript-null 
[api-major] Change `getJavaScript` to return `null`, rather than an empty Array, when no JavaScript exists
 2017-10-17 21:58:45 +02:00
Tim van der Meij
7d7edd9cc6
[api-major] Remove the PDFJS_NEXT option 
Nothing uses this option anymore, so setting it is a no-op now. We can
safely remove it.

Use `SKIP_BABEL` (instead of `PDFJS_NEXT`) now if you want to skip Babel
translation for a build.
 2017-10-16 23:16:51 +02:00
Jonas Jenwald
b5794bb26b Simplify the check, and remove the warning, for the nativeImageDecoderSupport API parameter 
As discussed in PR 8982.
 2017-10-16 09:11:39 +02:00
Jonas Jenwald
bcb29063c1 Polyfill Object.values and Array.prototype.includes using core-js 
See https://github.com/zloirock/core-js#stage-4-proposals.
 2017-10-16 09:11:39 +02:00
Jonas Jenwald
1cd1582cb9 [api-major] Change getJavaScript to return null, rather than an empty Array, when no JavaScript exists 
Other API methods already return `null`, rather than empty Arrays/Objects, hence it makes sense to change `getJavaScript` to be consistent.
 2017-10-15 22:17:14 +02:00
Tim van der Meij
815bc53a16 Merge pull request #9029 from Snuffleupagus/eslint--report-unused-disable-directives 
Enable the `--report-unused-disable-directives` ESLint command line option
 2017-10-15 15:12:20 +02:00
Jonas Jenwald
04b46831c1 Enable the --report-unused-disable-directives ESLint command line option 
This option was added in [version `4.8.0` of ESLint](https://github.com/eslint/eslint/releases/tag/v4.8.0), which is already listed as the minimum version in our `package.json` file; please refer to https://eslint.org/docs/user-guide/command-line-interface#--report-unused-disable-directives for additional details.

Despite the caveat listed in the link above, I still think that using this option makes sense since it will help ensure that no longer necessary disable statements are removed.
 2017-10-15 13:45:12 +02:00
Jonas Jenwald
0f90d5130c [api-major] Remove all remaining deprecated functions/methods 2017-10-15 13:27:10 +02:00
Jonas Jenwald
2f32131601 Replace our WeakMap polyfill with the one from core-js 
Since we're already using core-js elsewhere in `compatibility.js`, we can reduce the amount of code we need to maintain ourselves.

https://github.com/zloirock/core-js#weakmap
 2017-10-15 10:22:06 +02:00
Diego Casorran
11b1daa72d Mispelled isEvalSupported property at FontFaceObject() creation. 2017-10-07 20:05:21 +02:00