v2.3.1Minor parsing bugfixes, gz decode fallback

What's Changed

• Add fallback gz decode mechanism when gz headers are missing by @PrinsFrank in #200
• Properly retrieve content from uncompressed objects by @PrinsFrank in #201
• Accept whitespace around array values by @PrinsFrank in #202
• Support widths array spread over multiple lines by @PrinsFrank in #204
• Fix exception message length on decode issue by @PrinsFrank in #205

Full Changelog: v2.3.0...v2.3.1

v2.3.0 More extensive image support: JBIG2, TIFF, CMYK, several other formats and bugfixes

What's Changed

• DecodeParams can be array, Pattern can be string by @PrinsFrank in #159
• Add support for array filters in images by @PrinsFrank in #160
• Accept space after byte offset last crossReference source by @PrinsFrank in #161
• Accept space after start crossReference source by @PrinsFrank in #162
• Accept reference value for MediaBox and CropBox by @PrinsFrank in #163
• Scale and font size can both be negative by @PrinsFrank in #165
• Increase logging for incorrect matrix transformation by @PrinsFrank in #166
• Add support for padding and extra whitespace in reference array values by @PrinsFrank in #167
• Add support for TIFF images (CCITTFaxDecode) by @PrinsFrank in #169
• DecodeParams cannot be plain array by @PrinsFrank in #170
• Accept subdictionary for decodeParams by @PrinsFrank in #171
• Dictionary arrays can contain null by @PrinsFrank in #173
• Properly support array filter types when retrieving image content by @PrinsFrank in #174
• Properly handle newlines in CIDFonts widths by @PrinsFrank in #175
• Add support for JBIG2 images by @PrinsFrank in #176
• Add FillSignData as valid TypeNameValue by @PrinsFrank in #182
• Accept scale values that are floats besides ints by @PrinsFrank in #184
• Recover from invalid byte offset last cross reference section by looking for stream/table markers at the end of the document instead by @PrinsFrank in #186
• Add support for PNG predictor algorithm None by @PrinsFrank in #187
• Add support for CMYK rasterized images by @PrinsFrank in #189
• Handle spaces between components of matrix transformation in graphics state operator by @PrinsFrank in #192
• CONTENTS can be a ReferenceValueArray, not a simple ArrayValue by @PrinsFrank in #193
• Don't parse characters in resource names as operators by @PrinsFrank in #195
• Array values can be multiple resource names not seperated by space by @PrinsFrank in #196

Full Changelog: v2.2.1...v2.3.0

v2.2.1 Several small parsing related bugfixes, add JPEG2000 support

What's Changed

• Simplify working with underlying positioned text elements by moving retrieval of fonts to PositionedTextElement and presorting positionedTextElements by @PrinsFrank in #136
• Properly handle newline between dictionary key and value by @PrinsFrank in #141
• Use the length from a crossReferenceStream when it is available by @PrinsFrank in #143
• Don't try to decode values with JPX_DECODE filter by @PrinsFrank in #145

Full Changelog: v2.2.0...v2.2.1

v2.2.0 Add support for extraction of Rasterized Images

What's Changed

• Delete stale workflow to prevent issues from being marked as stale by @PrinsFrank in #130
• Implement rasterized image extraction by @PrinsFrank in #132
• Add support for other colorspaces that are not DeviceColorSpaces but are stored as luts in objects by @PrinsFrank in #134

Full Changelog: v2.1.1...v2.2.0

v2.1.1 Fixes bug when parsing PDFs where trailer marker is not followed by newline

What's Changed

• Don't assume newline after trailer marker, by @PrinsFrank in #128

Full Changelog: v2.1.0...v2.1.1

v2.1.0 Official PDF2.0 support, classes now non final to allow extension and mocking

What's Changed

• Make classes non-final to open them for extension and mocking by @PrinsFrank in #124
• Add official support for pdf 2.0 by @PrinsFrank in #125

Full Changelog: v2.0.0...v2.1.0

v2.0.0 Multiline/positional text extraction, image extraction, many bugfixes

What's Changed

• More strictly parse dictionary arrays by @PrinsFrank in #53
• Add new PDF2.0 tabs nameValues by @PrinsFrank in #54
• Fix issue where nested arrays in dictionaries were closed too early by @PrinsFrank in #56
• Fix comment state in dictionary parsing by @PrinsFrank in #58
• Fix nesting issues in text operator parsing by @PrinsFrank in #60
• Correctly parse dictionary array values where items are seperated over several lines by @PrinsFrank in #62
• Add missing public key security handlers to FilterNameValue by @PrinsFrank in #65
• Add support for multiple codespaceranges by @PrinsFrank in #67
• Remove extra decoding in uncompressed object by @PrinsFrank in #68
• Allow dashes in font names by @PrinsFrank in #70
• Fix issues where font operator contains font names by @PrinsFrank in #72
• Rename TextObjects and collection to contentStream to better reflect expected content by @PrinsFrank in #73
• Tc operator sets char space not char size by @PrinsFrank in #74
• Keep track of content stream commands outside text objects by @PrinsFrank in #75
• Remove dependency to internal phpunit method and ignore non-testing tests by @PrinsFrank in #76
• Organize content stream classes by @PrinsFrank in #77
• Organize text operators by @PrinsFrank in #78
• Parse textObjects to intermediate PositionedTextElement by @PrinsFrank in #79
• Implement remaining content stream operators by @PrinsFrank in #81
• Fix false positives on content stream commands by @PrinsFrank in #82
• Implement a transformation state stack to fix text lines appearing out of order by @PrinsFrank in #83
• Fix a typo in CONTRIBUTING by @szepeviktor in #84
• Fix content stream unit test after matrix multiplication fix resulting in correct x and y offsets by @PrinsFrank in #85
• Add test for compressed byte offsets by @PrinsFrank in #87
• Parse CIDFontWidths in dictionary values by @PrinsFrank in #89
• Implement space insertion based on text width by @PrinsFrank in #90
• Update samples dependency and increase space insertion threshold by @PrinsFrank in #91
• Don't try to parse EMCs and other content stream content as dictionaries by @PrinsFrank in #92
• Don't continue matching operators when in dictionary key or escaped string by @PrinsFrank in #94
• Implement encryption detection by @PrinsFrank in #97
• Use object length instead of searching for end marker when set in dictionary by @PrinsFrank in #95
• Add missing tests for inMemoryStream by @PrinsFrank in #98
• Implement missing features for literal string escape sequences by @PrinsFrank in #104
• Implement toUnicodeCMap unicode mappings for string literals by @PrinsFrank in #106
• Fix issue where string literals ended up as part of multibyte character groups by @PrinsFrank in #110
• Fix issue when text state is set outside of text object by @PrinsFrank in #112
• When in a string literal, don't keep track of other nesting levels as that is not possible, fixes several array delimiter issues in string literals by @PrinsFrank in #113
• Add missing resource TypeNameValue by @PrinsFrank in #114
• Fix: mb_convert_encoding can output false by @PrinsFrank in #115
• Implement image extraction by @PrinsFrank in #118

New Contributors

• @szepeviktor made their first contribution in #84

Full Changelog: v1.1.0...v2.0.0

v2.0.0 Alpha 5

What's Changed

• Implement image extraction by @PrinsFrank in #118

Full Changelog: v2.0.0-alpha.4...v2.0.0-alpha.5

v2.0.0 Alpha 4

What's Changed

• Fix issue where string literals ended up as part of multibyte character groups by @PrinsFrank in #110
• Fix issue when text state is set outside of text object by @PrinsFrank in #112
• When in a string literal, don't keep track of other nesting levels as that is not possible, fixes several array delimiter issues in string literals by @PrinsFrank in #113
• Add missing resource TypeNameValue by @PrinsFrank in #114
• Fix: mb_convert_encoding can output false by @PrinsFrank in #115

Full Changelog: v2.0.0-alpha.3...v2.0.0-alpha.4

v2.0.0 Alpha 3

What's Changed

• Implement toUnicodeCMap unicode mappings for string literals by @PrinsFrank in #106

Full Changelog: v2.0.0-alpha.2...v2.0.0-alpha.3