DOC: document read_html and to_html

[cpcloud]

No description provided.

[jreback]

FYI after #3693 u won't need to import read_html explicitly anymore

[cpcloud]

ok i will remove

[cpcloud]

i changed a few things around in html.py nothing major, but now read_html attempts to infer row and columns index types if they are not MultiIndexes. rls notes for this? it doesn't try to infer dates

[jreback]

@cpcloud this has nothing to do with your PR< but do you know why when you put an inline function reference its not a link? (what should it be?)

[cpcloud]

i changed it to this because it wasn't linking correctly (still not though since pandas.io.__init__ is empty. must've been looking at a different link when i built the docs...). it should link to the api docs. will fix

[jreback]

its not your link, I see this type of thing elsewhere (e.g. in the Stata docs); if you figure it out we prob need to edit in several places

[cpcloud]

hm this is annoying

[cpcloud]

@jreback got it. u must put the exact module reference here so the above should be :func:~pandas.io.html.read_html``

[cpcloud]

still annoying

[jreback]

you might need to rebase this after #3693

[cpcloud]

done.

[jreback]

Where you show reading with an attribute, I think you are missing something?

It looks the same as the prvious example

http://pandas.pydata.org/pandas-docs/dev/io.html#io-html

[cpcloud]

where r u looking i don't see the attrs anywhere...also a problem

[cpcloud]

oh i c

[jreback]

it's missing -u prob copied the prior example

[cpcloud]

man the io docs take 4ever 2 compile

[cpcloud]

there's probably a very tight correlation btwn amt docs amt of tests and amt of code all pairwise

[jreback]

true!

for 0.11 I had to keep rereading the docs
For all of the indexing chnges
as no tests for language!

[cpcloud]

pr is up