Hi, I'm not sure if I should post this here, or simply file a bug report - this is my first contact with the odftoolkit community.
Procedure: Using LibreOffice: 1. Create a spreadsheet with a cell containing text, and set the font to one not normally used - eg, Comic Sans. Save the document to a file. I'll refer to this as the source document. Using OdfToolkit-0.5-incubating: 2. Load the source SpreadsheetDocument from file and get the table from it. I'll refer to this as the source table. 3. Create a new SpreadsheetDocument instance. I'll refer to this as the target document. 4. Call appendSheet( sourceTable, "MyName" ) on the target document. 5. Save the target document to file. Tests: 1. Load the target document using LibreOffice. The cell does not have the correct font, instead it's the default font (eg, Arial). Analysis: Comparing the content.xml from the source and target documents, the source document contains a <style:font-face> element (within the <office:font-face-decls> element) corresponding to the font used. The target document does not. Conclusion: It appears that appendSheet does copy most referenced style-related metadata, but may not copy or create referenced <style:font-face> elements from the <office:font-face-decls> element. Additional Comments: I'm using a version of the toolkit cross-compiled to the .NET framework using IKVM, with Xerces 2.11 and Xalan 2.7.1. This is why I've not provided a java code example. Apart from the the issue above, everything I've used so far works well, so I have no reason to suspect that the .NET conversion is to blame. Many thanks, Tom
