I have been puzzled by the issue.When I use pdftohtml-0.39-win32 to convert pdf to xml (command:pdftohtml.exe -xml jstl-1_0-fr-spec.pdf test)
then sometimes it can be converted perfectly,but mosttimes it is done partly and miss some information like:
<text top="143" left="168" width="23" height="10" font="23">key</text>
I try another pdf files, the issue remains sometimes.
Anybody has the solution,contact me:
Thank you very much :)