[xwiki-devs] [GSoC][OfficeConverter]ppt conversion approach

Hi, When converting a ppt or odp document, the result of the conversion is some html files contain text and html files contain jpg images. A text html and a image html snapshot for per slide. And have the navigation links in per html, some thing like "First page" "Back" "Continue" "Last page" "Text"/"graph". For instance, the ppt with 3 slides will turn out: text1.html text2.html text3.html img1.html img2.html img3.html img1.jpg img2.jpg img3.jpg The img1.html is just the html file contain the img1.jpg. The return of the conversion only is the first page html, img1.html, which is not enough for the conversion result. The text1.html is the text version of the slide which doesn't contain any images in the source document. The img1.html is the snapshot of the original ppt slide which can display everything in the source document. There are three way to handle the ppt conversion. 1. assemble the text version of slide "text1.html, text2.html ..." to a final result. gmail's "view as html" of the attachment go this way. But this can lose the images of original document. 2. compress all the output file to a zip file and attach the zip file to the xwiki page. Then view the zip file content, like http://code.xwiki.org/xwiki/bin/view/Snippets/ViewOfficeDocumentSnippet does. 3. assemble the image version of slide to a final result. BTW, I'm working on the "view as html" of office document in xwiki page attachment. I have two way to display the result html. 1. display the result html in a pure normal html page. This way is hard to display the image. Maybe "view as html" can ignore the image of the documents, as the office import feature can handle the image well. 2. put the result html page in a xwiki page content and display the xwiki page. This way can address the image as page attachment easily. WDYT? Please help me to make those two decisions. Thanks. Wang Ning