BUG: read_html with a single column table #9178

[boarpig]

This commit fixes a crash on read_html(html, flavor="bs4") if table only has one column.

closes #9178

[jreback]

pls add a test

[cpcloud]

Please add a test

[boarpig]

The test should probably test that if it succesfully parses one column table but in which file should I put it? I've never written nose tests.

[rockg]

Usually tests go in the tests sub-directory and in the module you want to test. For example, you should put your test in pandas.io.tests.test_html. And in there probably add a test in the TestReadHtml class. There are plenty of examples there to get you started. And then you can run your test using nosetests (you probably just want to run the module: nosetests pandas.io.tests.test_html).

[jreback]

@boarpig needs a test.

[jreback]

@boarpig this needs a test pls.

[jreback]

@boarpig pls add a test

[jreback]

closing, if/when updated pls reopen

[hnykda]

I have just encountered this bug in 0.18 and the fix from this PR fixed that. Is the only limiting factor to merge this PR adding tests? I could probably do it (and will appreciate any help).

[jreback]

yep needs tests

[hnykda]

But I can see there are already tests for this part here. What else should be added? This change seems to be sufficiently tested by current tests. Thanks

[hnykda]

I am willing to finish this, but I need an answer to the previous post.

[jreback]

you need to replicate the example in the issue - that is the test

[hnykda]

I have done it in a new PR (so I don't have to do a rebase).