fda858ae33
Don't map glyphs to certain problematic General Punctuation Unicode locations (bug 911034)
...
Fixes the remaining missing characters in https://bugzilla.mozilla.org/show_bug.cgi?id=911034 .
For reference, see http://www.unicode.org/charts/PDF/U2000.pdf (and also http://en.wikipedia.org/wiki/General_Punctuation_%28Unicode_block%29 ).
10 years ago
2b1a13ba28
Don't map glyphs to Unicode position 0x0E33, i.e. Thai character SARA AM (bug1046314)
...
*A similar approach as in PR 5705.*
Fixes https://bugzilla.mozilla.org/show_bug.cgi?id=1046314 .
According to https://dxr.mozilla.org/mozilla-central/source/gfx/harfbuzz/src/hb-ot-shape-complex-thai.cc#270-365 , `0x0E33` is treated as a special case (by the font shaping code in Firefox). Hence it seems reasonable to skip it when adjusting the font mapping.
10 years ago
e894a0a4c6
Guess CIDFontType0 subtype based on font file contents (issue 5751)
10 years ago
4a9ff471c4
Correctly detect the presence of the Adobe specific APP14 JPEG marker (bug 1140761)
...
According to the specification, http://partners.adobe.com/public/developer/en/ps/sdk/5116.DCT_Filter.pdf#G3.851943 , the content of the marker segment should begin with `Adobe`, and not `Adobe\x00` as the code currently look for.
Fixes https://bugzilla.mozilla.org/show_bug.cgi?id=1140761 .
Fixes the colour conversion part of issues 4090 and 5623.
10 years ago
97eb215b5f
Add a reduced test-case for issue 5734
10 years ago
82f73bc9f4
Add a |text| test-case for issue 5421
...
Prior to PR 4259, we *incorrectly* ignored `toUnicode` for Type3 fonts. Since we now handle that correctly, this patch adds a `text` test-case to prevent regressions.
10 years ago
f81fc9091a
Correctly detect OpenType font files with CFF data
...
Fixes 5334.
Fixes 215.
Fixes https://bugzilla.mozilla.org/show_bug.cgi?id=1125614 .
According to the specification, http://www.microsoft.com/typography/otspec/otff.htm , OpenType font files with CFF data should have `OTTO` in the header.
10 years ago
0a3341dadc
Don't map glyphs to Unicode "Dotted circle" combining mark (bug 1108301)
...
It seems that `0x25CC` is another bad spot for charCodes.
Fixes https://bugzilla.mozilla.org/show_bug.cgi?id=1108301 .
10 years ago
888cbe0bde
Avoid getting stuck in empty nodes in the Pages tree when calling |Catalog_getPageDict| (issue 5644)
10 years ago
7c8996558a
Fetch parameters if they are a |Ref| in CalGray/CalRGB color spaces (issue 5747)
10 years ago
417800a1b5
Only skip the |!isSymbolicFont| check for TrueType (3, 1) cmap tables if no previous cmap table was found (PR 5703 followup)
...
Fixes https://bugzilla.mozilla.org/show_bug.cgi?id=894572 .
10 years ago
27e3558a41
Fix CCITTStream regression by byte-aligning rows before checking EOL marker
10 years ago
592890a758
Relax the |isSymbolicFont| check for TrueType (3, 1) cmap tables (issue 5701)
10 years ago
d124fc5e08
Test case for issue 5567
10 years ago
01e6565dd4
Try to skip mapping of missing TrueType glyphs
...
Also don't skip mapping of glyphs which are empty, if the corresponding charCode is included in toUnicode.
10 years ago
8174da61fb
Don't skip mapping of glyphs for CIDFontType2 fonts with a CIDToGIDMap
...
Also don't skip mapping of glyphs which are empty, if the corresponding charCode is included in toUnicode.
10 years ago
ad41a2d574
Add support for TrueType (0, 0) cmap tables (issue 5501 and 5574)
10 years ago
1fea44ce14
Ensure that |get fingerprint| doesn't fail if the ID is not an array (issue 5599)
...
According to the [PDF specification](http://www.adobe.com/content/dam/Adobe/en/devnet/acrobat/pdfs/PDF32000_2008.pdf#page=51 ) the `ID` should be an array. To prevent errors if the PDF file violates that part of the spec, check that the `ID` is actually an array.
Fixes 5599.
10 years ago
b835ef425e
CCITTFaxStream: fix regression when EncodedByteAlign is true and EndOfLine is false
10 years ago
75767abb09
Add a reduced test case for issue 4379
...
This should have been part of PR 5136.
Based on comments https://github.com/mozilla/pdf.js/issues/4379#issuecomment-51221787 and https://github.com/mozilla/pdf.js/issues/4379#issuecomment-51475463 , it should be OK to include the PDF file in the repo.
10 years ago
d8b905048b
Add fallback for non-embedded "Century Gothic" CIDFontType2 font (issue 4722 and bug 879561)
...
According to practical experiments, falling back to "Helvetica" when we encounter a non-embedded "[Century Gothic](http://en.wikipedia.org/wiki/Century_Gothic )" `CIDFontType2` font seems to work well.
(Also, the section on Wikipedia about "Printer ink usage" *might* provide some anecdotal evidence that Century Gothic is a fairly standard sans-serif font.)
Obviously this patch doesn't make "Century Gothic" fonts render perfectly, as is often the case with non-embedded fonts, but all the text is now legible in the referenced issues.
Fixes 4722.
Fixes https://bugzilla.mozilla.org/show_bug.cgi?id=879561 .
11 years ago
a340927687
Test case for issue 5481
11 years ago
36fce7db7e
Test case for issue 5475
11 years ago
16833159a9
Test case for issue 5549
11 years ago
184880a751
Fix searching for end of inline (EI) images with ASCII85Decode filters (bug 1077808)
...
This patch changes searching for the end of inline image streams to rely on the EOD marker for the filters: ASCII85Decode and ASCIIHexDecode.
11 years ago
3ea81019fe
Add a (linked) test case for bug 1108753 (PR 5276)
...
When submitting PR 5276 there wasn't a good PDF file to include in the test suite. However, with https://bugzilla.mozilla.org/show_bug.cgi?id=1108753 , we now have a better source for a test file, hence this patch.
11 years ago
df1364a993
Add a (linked) test case for bug 1064894
...
This should have been part of PR 5284.
11 years ago
78d0d269dc
Fallback to the |Dummy| Pattern when we fail to parse a Shading Pattern (issue 5509)
11 years ago
5f32f80e34
Add a reduced test case for issue 5470
11 years ago
58cc39f329
Add a reduced test case for issue 5291
11 years ago
bf0d03bccd
Change bug878194 to an |eq| test
11 years ago
88c88d08e1
Add a reduced test case for PR 5286
11 years ago
6337a430b3
Add test case for JPX "Out of packets" error
11 years ago
4ac4d49158
Avoid rendering invisible Type3 fonts (issue 5421)
11 years ago
2d7a34d692
Fixes decoding for CMYK jpegs.
11 years ago
0ffb54b95f
Add a reduced test case for bug 847420
11 years ago
2fc5e6a9ad
Don't map glyphs to unicode non breaking space.
11 years ago
df2a4afd36
Use |toUnicode| when creating the glyph map for standard CIDFontType2 fonts without embedded font file
11 years ago
3c759e296a
Add support for MMType1 fonts with embedded font files
11 years ago
b16c973d9d
Fix Symbol fonts without font file but with Encoding dictionary (issue 5238)
11 years ago
aa8d3d98f8
Fetches params in makeFilter
11 years ago
3ca2c3f4a7
Add ZapfDingbats testcase
11 years ago
be595d0721
Further amend GlyphMapForStandardFonts (issue 5244)
11 years ago
3152085b0d
Remove a duplicate PDF file from the test suite
...
The files issue3115.pdf and issue2337.pdf are identical, the only difference being that the first one is an `eq` test and the second one a `load` test. Hence there is no reason to keep the `load` test, since it's just a subset of the `eq` test.
11 years ago
be998261cc
Fixes Type3 negative font direction
11 years ago
0c5525dc8a
CalRGB: optimize CalRGB calculations
...
Also fix one silly mistake.
11 years ago
1e4a7f981e
implement CalRGB color space
...
Both whitespace and blackspace support are implemented.
11 years ago
8ecbb4da05
Adjust the heuristics to recognize more cases of unknown glyphs for |toUnicode| (issue 5070)
11 years ago
70b934c7a5
Add reduced test-cases for issues 4875 and 4881
11 years ago
26f5b1b2d3
Add a couple of reduced test-cases for PR 4259
...
PR 4259 fixed a large number of font bugs, but none of those where added as test-cases. This was, in my opinion, unfortunate since it increases the risk of regressions in the future when other font bugs are fixed.
This PR simply adds a few more test-cases, to improve our test coverage somewhat.
11 years ago
Jonas Jenwald
Tim van der Meij
fkaelberer
Yury Delendik
Brendan Dahl
Kalervo Kujala