Started looking into this one and I think I uncovered a somewhat related
bug. If a child element's tail is assigned, then neither the text nor
tail show up in the output.
For example:
import markdown
from markdown.inlinepatterns import Pattern
class MyPattern(Pattern):
def handleMatch(self, m):
el = markdown.etree.Element('p')
el.text = markdown.AtomicString('an *atomic*')
c1 = markdown.etree.SubElement(el, 'span')
c1.text = markdown.AtomicString('*string*')
c1.tail = markdown.AtomicString('`not code`')
return el
class MyExtension(markdown.Extension):
def extendMarkdown(self, md, md_globals):
patt = MyPattern(r'::(.*?)::')
md.inlinePatterns.insert(0, 'foobar', patt)
ext = MyExtension(None)
md = markdown.Markdown(extensions=[ext], output_format="xhtml1")
html = md.convert('here is ::mypattern::')
print html
The resulting output is:
<p>here is <p>an *atomic* </p>
</p>
As opposed to:
<p>here is <p>an *atomic*<span>*string*</span>`not code`</p>
</p>
This is probably out in the weeds as far as usage goes. Also, I'm
surprised that this doesn't cause normal usage issues- surely the tree
parser receives similar input in normal use situations(ie when parsing a
file of Markdown text). An element with a child with text and tail
can't be that unusual.
Anyway, I'll be investigating it a bit more.
Regards-
Gerry LaMontagne
|
