Markdown in Python: stripToLevelTags fails when input only contains link references.

Anonymous users must enter captcha below.

Ticket Information

Ticket Title

The following issue:

File "/base/data/home/apps/tapingya/2.333952260742068510/markdown/__init__.py", line 400, in convert
    start = output.index('<%s>'%DOC_TAG)+len(DOC_TAG)+2
    ValueError: substring not found

Was produced when formatting this text:

[Data Binding presentation]:http://link.brightcove.com/services/player/bcpid1733261879?bclid=1729365228&bctid=1741212660

I fixed it by using try, except: around the md.convert(text) call, but thought I should bring it to your attention.

Barry

###Comments

__Waylan on 6/5/09__

Thanks for the report Barry. First of all, the simplest input that produces the error is this:

[foo]:bar

As it turns out, markdown sees this as a reference link and removes it from the document. However, the document is now empty, but markdown has already checked for an empty document - so the empty document shortcut is missed. Of course, when trying to strip the top level tags if finds an empty div: `<?xml version='1.0' encoding='utf8'?><div />` instead of: `<?xml version='1.0' encoding='utf8'?><div>{{ content here }}</div>`. For obvious reasons, it chokes on that.

Fixed in [7bfdc83](http://gitorious.org/python-markdown/mainline/commit/7bfdc831d2ad4fc0350ec4f020a4a79ff8d2845f).

Reported by

Assigned to

Status

Resolution

Don't put anything here

Advanced Fields

Priority

Resolution Explanation

Component

Don't put anything here

About This Edit

Minor Edit

Edit Summary

Don't put anything here