Quote:
Originally Posted by davidfor
I'll have a play and see what I can see.
|
Hm. I've tinkered a bit more; see if this build does the trick for you. Also, stupid question: are the samples you attached the originals, or had they already been through unpretty? (I swear, even before I started tinkering, they unprettied just fine for me...)
EDIT: It looks like the difference in results is due to a different protocol; I just set up an "unpretty, and nothing else" test, and now I see the results you're getting. I'm on the case.
EDIT 2: Aha! Seems your HTML files were using CRLF EOL markers, and an earlier step in my previous process was changing those to plain LF markers. I've added an explicit step to unpretty which replaces CR(LF?) with LF, which appears to fix the issue.
The attached version has been updated with this fix, as of 9:20pm EDT.