If it still builds (I didn't bother trying), http://mxr.mozilla.org/mozilla-central/source/parser/htmlparser/tests/htmlgen/htmlgen.cpp would build you a utility to create HTML test files replacing the macros you put in your template with things from "the set of known HTML tags" including such popular delights as <ILAYER> and <SERVER>. We have impressionable young people looking at the codebase (I only saw this because of Alex wondering whether he could get a security bounty for an obvious buffer overflow in it), we shouldn't be exposing them to <SOUND> or <PLAINTEXT> or <MENU>. Just removing it does leave a dangling mention of it in http://mxr.mozilla.org/mozilla-central/source/parser/htmlparser/tests/html/residual.html but I think it's safe to say that nobody will find that any more distressing than the way that half the linked testcases aren't actually where it thinks they are, and when I've got more time to look at dozens and dozens of dead files, I'll sweep out all that trash too.
Attachment #499625 - Flags: review?(hsivonen)
For anyone arriving here after following the 2011-01-02 nightly's pushlog entry: bbb7cd978802: "Don't run StopPluginInstance multiple times on the same instance. Also, let StopPluginInstance do the call to instance->Stop()" ...that was actually for bug 621250 and was mislabelled.
Comment on attachment 499625 [details] [diff] [review] rm AFAICT, the new parser isn't being tested with this code.
Attachment #499625 - Flags: review?(hsivonen) → review+
Yup, nothing's being tested by it, it's a NPOTB tool to generate manual testcases, which could conceivably have been used to create some of the NPOTB and unused testcases in http://mxr.mozilla.org/mozilla-central/source/parser/htmlparser/tests/html/ though none of them really show any sign of having been created by it.
Whiteboard: [needs review] → [needs push]
Status: NEW → RESOLVED
Closed: 9 years ago
Resolution: --- → FIXED
Whiteboard: [needs push]
Target Milestone: --- → mozilla2.0b10
You need to log in before you can comment on or make changes to this bug.