|Atomik Xport SE: Reference > Chapter 12 Extracting Tables||<< >>|
The next step is to understand the relationship of the Table element declaration to the parts of a table as they exist on the QuarkXPress page. You will see two diagrams below that illustrate this based on a table of stock prices. The two illustrations represent the top and bottom portions of the total page shown at the start if the start of the chapter.
Rules for tables are created in exactly the same way as for other elements in your DTD. These rules can be setup through direct association of style sheets to Table elements or through direct drag and drop of representative styling from the QuarkXPress page (see Chapter 5 - Working With Rulesets for further details).
When making rules you should map the appropriate style sheets for each row to the appropriate row element in the Atomik Xport DTD, (i.e. ‘TableDataRow’, ‘TableHeadRow’), or to the appropriate section element in the CALS DTD (i.e. thead, tbody).
Note: You should then create an additional mapping from your rows’ style sheets, this time to the cell element. The cell element in Atomik Xport is TableData. In CALS it is data, while in XHTML the td and th elements represent table head cand body cells respectively.
Creating mappings to both the cell and row elements removes ambiguity from the automated extraction.
Table and elements and how they relate to the Table element declaration.
To successfully extract a tabbed table, it has to be in a single text box. A table that is spread across multiple boxes will not be extracted as a single table (although the table may be reconstructed from the individual tables after the extraction process is complete using custom XSLT).
Please also be aware that the Table element cannot be used as the root element in your DTD.
In order to extract the correct table geometry for tabbed tables you will need to ensure that you have the same tab stop positions in every paragraph of your tabbed data.