in the test's case the last '-' character at the end of the paragraph is not encoded correctly.
the '-' is a unicode character, not a regular - sign.
I opened both the input and output files in a browser and they look identical. Even the html source looks identical except for different line wrapping. What exactly is the problem?
I have attached jtidies' output. Have a look at the "incluindo – mas" section. I have tried to shorten the problematic sequence as short as I can. The strange thing is that - when you remove some things from the beginning, everything works nicely.
What version of jtidy did you use and what program/command did you run?
I'm using trunk. I have seen the problem for at least two years but never filed a bug.
Hi, could you try the attached patch please?
It seems to fix the problem but I wonder if it breaks anything.
Btw, it doesn't seem to break any more tests on the java5 branch.
I'll check in two weeks...
I couldn't wait. The problem has gone - thank you! I have re-uploaded all files to create a test case (for the old system and this failure).
Haha, thanks, I'll go ahead and commit then (a bit later).
Well, I committed the fix, but the test case you added doesn't work on trunk - it tries to parse "SUMMARY" (message level) as a number.
Log in to post a comment.
Sign up for the SourceForge newsletter:
You seem to have CSS turned off.
Please don't fill out this field.