[xwiki-devs] [GSoC][officeimporter]option to convert document to xhtml or xwiki syntax 2.0
Hi Vincent, AFAIK, the conversion from xhtml to xwiki syntax is hard because many things about how to render the html elements in xwiki syntax are still not sure, like table, img, {style} macro. Furthermore, the xhtmlparser is buggy, like <img> <link> list. I will try my best to make the conversion work, but to make a option, I want develop a feature that just convert office doument to xhtml. So the user can just use the convert2html to get the xhtml and fill xhtml into xwiki page with xwiki syntax 1.0. After we finish the new rendering and the xhtmlparser, the use can change to convert2xwikisyntax. The convert2html is easy but some thing to handle * clean the html (htmlcleaner) * remove the <style> in head (htmlcleaner) * remove empty link (htmlcleaner) * remove <p> in <li> (jdom) * replace <img> with {img} (jdom) * replace link <a> to [] (jdom) * handle the ppt and odp to the single page All these step is easy to implement. The convert2html may be not perfect, but it can work for most conversion requirement for now. As I am runnig for deadline, I think a workable feature is useful. However, I will work on the finial gold "convert to xwiki syntax2.0" continually. WDYT? -- Thanks Wang Ning
On Aug 19, 2008, at 6:47 PM, Wang Ning wrote:
Hi Vincent,
AFAIK, the conversion from xhtml to xwiki syntax is hard because many things about how to render the html elements in xwiki syntax are still not sure, like table, img, {style} macro. Furthermore, the xhtmlparser is buggy, like <img> <link> list. I will try my best to make the conversion work, but to make a option, I want develop a feature that just convert office doument to xhtml. So the user can just use the convert2html to get the xhtml and fill xhtml into xwiki page with xwiki syntax 1.0. After we finish the new rendering and the xhtmlparser, the use can change to convert2xwikisyntax. The convert2html is easy but some thing to handle * clean the html (htmlcleaner) * remove the <style> in head (htmlcleaner) * remove empty link (htmlcleaner) * remove <p> in <li> (jdom) * replace <img> with {img} (jdom) * replace link <a> to [] (jdom) * handle the ppt and odp to the single page All these step is easy to implement. The convert2html may be not perfect, but it can work for most conversion requirement for now. As I am runnig for deadline, I think a workable feature is useful. However, I will work on the finial gold "convert to xwiki syntax2.0" continually.
WDYT?
I agree. Both conversion to XHTML and to wiki syntax should be pursued. Thanks -Vincent
On Aug 19, 2008, at 6:51 PM, Vincent Massol wrote:
On Aug 19, 2008, at 6:47 PM, Wang Ning wrote:
Hi Vincent,
AFAIK, the conversion from xhtml to xwiki syntax is hard because many things about how to render the html elements in xwiki syntax are still not sure, like table, img, {style} macro. Furthermore, the xhtmlparser is buggy, like <img> <link> list. I will try my best to make the conversion work, but to make a option, I want develop a feature that just convert office doument to xhtml. So the user can just use the convert2html to get the xhtml and fill xhtml into xwiki page with xwiki syntax 1.0. After we finish the new rendering and the xhtmlparser, the use can change to convert2xwikisyntax. The convert2html is easy but some thing to handle * clean the html (htmlcleaner) * remove the <style> in head (htmlcleaner) * remove empty link (htmlcleaner) * remove <p> in <li> (jdom) * replace <img> with {img} (jdom) * replace link <a> to [] (jdom) * handle the ppt and odp to the single page All these step is easy to implement. The convert2html may be not perfect, but it can work for most conversion requirement for now. As I am runnig for deadline, I think a workable feature is useful. However, I will work on the finial gold "convert to xwiki syntax2.0" continually.
WDYT?
I agree. Both conversion to XHTML and to wiki syntax should be pursued.
One feature we need is some {{office}} macro that will render the document specified in the document's view. This feature could be implemented using the XHTML generation for now. The macro will simply have to generate a {{xhtml}} macro block element containing the generated XHTML to work. Then later on when the wiki conversion is working we could simply remove the {{xhtml}} macro block. Thanks -Vincent
Hi Vincent & all devs, On Wed, Aug 20, 2008 at 12:59 AM, Vincent Massol <[email protected]> wrote:
On Aug 19, 2008, at 6:51 PM, Vincent Massol wrote:
On Aug 19, 2008, at 6:47 PM, Wang Ning wrote:
Hi Vincent,
AFAIK, the conversion from xhtml to xwiki syntax is hard because many things about how to render the html elements in xwiki syntax are still not sure, like table, img, {style} macro. Furthermore, the xhtmlparser is buggy, like <img> <link> list. I will try my best to make the conversion work, but to make a option, I want develop a feature that just convert office doument to xhtml. So the user can just use the convert2html to get the xhtml and fill xhtml into xwiki page with xwiki syntax 1.0. After we finish the new rendering and the xhtmlparser, the use can change to convert2xwikisyntax. The convert2html is easy but some thing to handle * clean the html (htmlcleaner) * remove the <style> in head (htmlcleaner) * remove empty link (htmlcleaner) * remove <p> in <li> (jdom) * replace <img> with {img} (jdom) * replace link <a> to [] (jdom) * handle the ppt and odp to the single page All these step is easy to implement. The convert2html may be not perfect, but it can work for most conversion requirement for now. As I am runnig for deadline, I think a workable feature is useful. However, I will work on the finial gold "convert to xwiki syntax2.0" continually.
WDYT?
I agree. Both conversion to XHTML and to wiki syntax should be pursued.
One feature we need is some {{office}} macro that will render the document specified in the document's view.
The {{office}} is for xwiki syntax 2.0. So I will still implement a convert2html for xwiki syntax 1.0 which can be used by the all the xwiki user.
This feature could be implemented using the XHTML generation for now. The macro will simply have to generate a {{xhtml}} macro block element containing the generated XHTML to work. Upload a office document to the xwiki page, and add {{office}}attachment_name{{office}},then save&view. But there is a question: * how to handle the images in the office document. XHTML can't refer to the images if the images are not attached to the xwiki page. where to put the generation images? I can't come to a good idea. please help me.
Thanks. Wang Ning
Then later on when the wiki conversion is working we could simply remove the {{xhtml}} macro block.
Thanks -Vincent
_______________________________________________ devs mailing list [email protected] http://lists.xwiki.org/mailman/listinfo/devs
participants (2)
-
Vincent Massol -
Wang Ning