Tab completion of parameters of pandas.read_csv() fails.

[EricDepagne]

Hi all.
I've come across a strange behaviour. I'm not sure if it's pandas related or ipython. If it's pandas, i'm happy to open an issue for them.
Here's my setup:
Jupyter QtConsole 4.4.3
Python 3.7.3 (default, Mar 26 2019, 21:43:19)
Type 'copyright', 'credits' or 'license' for more information
IPython 7.5.0 -- An enhanced Interactive Python. Type '?' for help.

Here's my problem:
When I do :
plt.hist(bin<TAB> , I'm presented with the options to choose between bin and bins=

When I do :
pd.array(dt<TAB> , ipython autocompletes with dtype, as it's the only possibility

When I do :
np.histogram(ra<TAB> the options ipython shows are range and range=
This is the expected behaviour, and it works perfectly fine.

But then, if I do
pd.read_csv(de<TAB> nothing happens. I was expecting ipython to autocomplete in delimiter=.
None of the possible parameters of the pd.read_csv() method are auto completed.

It's the only function I've found to behave like this. pd.read_clipboard() autocompletes fine for instance.

[augustogoulart]

@meeseeksdev tag tab-completion