antisamy ignoring anything rule? about antisamy

[clqian]

hi @davewichers @spassarop yes the changes work for me. thank you so much!

[davewichers]

@spassarop - Can you answer this?

[spassarop]

Hi @clqian, DOM or SAX, any of them have the same issue with an LSEP character because the compiled Pattern for the regexp does not have the Pattern.DOTALL flag. Then, although you may want to match any character, it does not consider line separators.

If every compiled regexp has that flag on Policy parsing, it works. @davewichers do you think that would be ok? Change should be on 5 occurrences only in Policy class.

[davewichers]

I'm not qualified to comment on that. @nahsra - Arshan? What do you think?

[jdr023]

any updates on this?

[nahsra]

I agree with @clqian -- I think the character set being called literally "anything" should suggest that LSEP be allowed characters. We can't say for certain about how allowing LSEP would affect markup parsing in the browsers because they're essentially black boxes with unpredictable behavior at times, but I am comfortable with the change overall.

[spassarop]

Something I can add is that the LSEP char may not be included in the output. At least I tested it as a unicode char and I could not make it appear after the sacan :(

That change can be done tomorrow I think. It's really simple.

[davewichers]

Hey @clqia, @spassarop implemented a change to address this issue and I just merged it into the 1.6.0 branch. Can you review the change and test. And let us know if you think this fixes this issue properly? Can if you could do that tomorrow that would be great, as we are trying to get the 1.6.0 release out in the next few days. Thx

[davewichers]

OK. This issue will then be fixed in the 1.6.0 release, which should be coming out soon.

[davewichers]

This was addressed in commit 3f446c5 and AntiSamy v1.6.0 was released today with this included.