Functionality to extract all the names from a font (Thimo Seitz).
Metadata can now be added to PDF and RTF at any time, not only before opening the document. HTML generation will still have the old limitation.
PdfEncodings: the winansi encoding didn't process the char \u00a0.
PdfGraphics2D: addition of hyperlink capability when drawing a string in PDFGraphics2D object. The url is submitted using a RenderingHint. The renderhint is initialized with a HyperlinkOFF value, if a hyperlink should be added setRenderingHint with the url is called during paint of the link. After the link is painted setRenderingHint with null value should be called. The Hyperlink Rendering Hint is present at all times in PDFGraphics2D object; it can be found by iterating all keys and using the toString function.
Fix for invalid array sizes.
Some tiffs have a bogus value in the ROWSPERSTRIP tag.
PdfSmartCopy: PdfSmartCopy also compares the dictionary values up to a nested depth of 10; lifted the restriction to compare templates.
Changes made by Xavier
Code cleanup: adding missing @since tags
directory structure reorganization: core and swing subdirectories in src and test.
Use of x.y.z-SNAPSHOT in Document.java ITEXT_VERSION and in jdiff output
more consistent use of ant property files.
use jdk 1.4 api for javadoc links
jdiff task was using older file in copy without overwrite arg
itext.test.classpath property needs to be setup before calling javac
ant delete target added
new ant targets for rtf parser support: compile.rtf, jar.rtf and test.rtf
javadoc commands use 'maxmemory' argument instead of '-J-Xmx'
maven bundle updates: lower case names to match artifact and javadoc jar added
added "jar" dependency for sign.jar to make sure latest changes are picked up
Changes made by Mark
memory increased for javadoc task
Preparation for the import of the new rtf parser code.
First junit test for the RTF package. Also updated the test.xml so that the test.rtf depends on the jar.rtf.
Fixed a bug with style numbering. The style number is now only set after the inheritance handled, so that the default styles are always added before any user styles.
Changes made by Bruno
Added an extra constructor for List (suggestion by Alexis Pigeon)
Changed (package) private methods/variables into protected or even public on many places in PdfDocument.
Introduction of BadPasswordException
Introduction of the 'numberStyle' concept for Section titles (implemented by Bruno after a question on the list by Friedhelm Hellerhoff).
Introduction of Element.isContent() and Element.isNestable() as suggested by Kris Raney. Indicates if an element contains real content and if it can be nested.
Introduction of the LargeElement interface. Objects implementing this interface (PdfPTable, Table, Section and Chapter) are considered to be 'dangerous' in terms of memory consumption. The LargeElement interface offers a way to add the object to the document (multiple times) before all the content has been added to the large object. Every time the large object is added to the Document, part of the object's content is flushed.
Solved a bug in the old Table class (actually in PdfTable) that occurred with repeating headers in very specific situations.
Added a rewrite of RUPS (Reading / Updating PDF Syntax) to the project. This also involves adding new ANT scripts, a new JNLP and new jars: SUN's PDF Renderer and Dom4j (these jars are not needed for the core iText; only for RUPS).
After a mail communication with Brian Burkhalter (SUN), the license notices for the classes derived from JAI code was updated.
The license notices for TIFFConstants were updated because the list of TIFF constants was originally made by Silicon Graphics.