SetaPDF-Extractor

Rev. 1026 to 1049

SetaPDF-Extractor Component

Feature

• Implemented handling of invisible glyphs above visible glyphs. They are ignored now and will not trigger a word break in both Word and ExactPlain strategies.
• Added setCleanStreamCallback() and getCleanStreamCallback() in the abstract strategy class.
• Added SetaPDF_Extractor_ContentStreamCleaner class and implemented it as a default callback in all strategies. It allows you to clean up a content stream before it is parsed and will bring a performance boost for documents with e.g. very much vector operations.

Bugfix

• Pass $_ignoreSpaceCharacter property in SetaPDF_Extractor_Strategy_ExactPlain to sub-instance.

Tweak

• Optimized code in sorting functions.

SetaPDF-Core Component

Feature

• Added setTabOrder() and getTabOrder() in SetaPDF_Core_Document_Page_Annotations class.
• Added method to normalize line breaks: SetaPDF_Core_Text::normalizeLineBreaks().
• Added SetaPDF_Core_Document::getSaveMethod() and renamed $update parameter in save() method to $saveMethod (meaning is the same as before).
• Added support for PDF documents with invalid data before their PDF file header.

Bugfix

• Fixed a bug which occurs if AES encryption is used while the Length value of a stream is an indirect object which is already written because of the SetaPDF_Core_Document::SAVE_METHOD_REWRITE_ALL save method.
• Prevent "String offset cast occured" notice in abstract reader class.
• Fixed SetaPDF_Core_Document_Destination::findByName() when name was not found.

Tweak

• Normalizes line breaks in SetaPDF_Core_Text_Block class.
• Check for null bytes in file names before passing them to PHP functions to prevent warnings or notices.
• Added support for faulty objects with a generation number higher than zero in object streams.
• Allow date strings without "D:" prefix.

Demo

• Updated get image sizes demo to handle rotated pages correctly.