quoted law name and statute not rendered

[clkao]

for example: http://lci.ly.gov.tw/LyLCEW/jsp/ldad000.jsp?irKey=&htmlType=agenda&fileName=html/agenda1/01/08/03/01/LCEWA01_080301_00003.htm

"「會計法」第99條之1" should probably still be recognized

[kong0107]

Ah.... that seems to be annonying (but frequently seen in goverment's websites). HTML code of that page contains:

This project use DOM tree and consider each text node (nodeType 3) independently. I haven't thought up a good way to merge or detect texts which cross nodes...

[kong0107]

Hmm, for document pages 立法院議事暨公報管理系統, the following codes work perfectly as pre-processor:

var ps = document.getElementsByTagName("P");
for(var i = 0; i < ps.length; ++i) ps[i].innerText = ps[i].innerText;

But there seems to be different file at that site. The above page @clkao talked about has htmlType=agenda, and there are other page with htmlType=communique in their URL. Pages with later format seem to be OK with this project, but I don't know whether there are different formats.
...well, the biggest problem is that I've no idea about how to handle such contents.

BTW, the page above has a corresponding page http://lci.ly.gov.tw/LyLCEW/html/agenda1/01/08/03/01/LCEWA01_080301_00003.htm with same content but different background.

[kong0107]

Codes

var ps = document.getElementsByTagName("P");
for(var i = 0; i < ps.length; ++i) ps[i].innerText = ps[i].innerText;

spend too much time for large documents. Some even reaches 90 seconds for me (AMD Sempron LE-1200, XPSP3).
Modify the code to ps[i].innerHTML = ps[i].innerText; wouldn't help.

Gotta find some other way...