BUG: pd.concat doesn't recognise GeoDataFrame as expected output

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of geopandas.

• (optional) I have confirmed this bug exists on the master branch of geopandas.

---

Code Sample, a copy-pastable example

import pandas as pd
import geopandas as gpd


def df() -> pd.DataFrame:
    return pd.DataFrame(
        {
            "City": ["Buenos Aires", "Brasilia", "Santiago", "Bogota", "Caracas"],
            "Country": ["Argentina", "Brazil", "Chile", "Colombia", "Venezuela"],
            "Latitude": [-34.58, -15.78, -33.45, 4.60, 10.48],
            "Longitude": [-58.66, -47.91, -70.66, -74.08, -66.86],
        }
    )


def gdf_points_from_xy(df: pd.DataFrame) -> gpd.GeoDataFrame:
    return gpd.GeoDataFrame(df, geometry=gpd.points_from_xy(df.Longitude, df.Latitude))


def concat_gdf(gdf: gpd.GeoDataFrame) -> gpd.GeoDataFrame:
    # pycharm linter complains:
    # Expected type 'GeoDataFrame', got 'Union[DataFrame, Series]' instead 
    return pd.concat([gdf, gdf])


def test_pd_concat():
    _df = df()
    _gdf = gdf_points_from_xy(_df)
    catted = concat_gdf(_gdf)

    assert isinstance(catted, gpd.GeoDataFrame) #  No problem

Problem description

Pycharm complains about the return type of pd.concat when the method calling it expects the return type to be gpd.GeoDataFrame. See screenshot.

One workaround is to explicitly create a new GeoDataFrame object after concat as discussed in this thread.

Is there something obvious we can do with the GeoPandas library to overcome this?

Package versions

PYTHON DEPENDENCIES
-------------------
geopandas  : 0.9.0
pandas     : 1.3.2
fiona      : 1.8.20
numpy      : 1.21.1
shapely    : 1.7.1
rtree      : 0.9.7
pyproj     : 3.1.0
matplotlib : 3.4.3
mapclassify: None
geopy      : 2.2.0
psycopg2   : 2.9.1 (dt dec pq3 ext lo64)
geoalchemy2: 0.8.5
pyarrow    : 5.0.0
pygeos     : None

[martinfleis]

There’s a bit more on this in mypy docs (https://mypy.readthedocs.io/en/latest/kinds_of_types.html#the-type-of-class-objects, and https://mypy.readthedocs.io/en/latest/class_basics.html). I want to go through that in the evening to understand it a bit before reporting to pandas.

[jorisvandenbossche]

That SO question/answer might be about (sub)class objects and not (sub)class instances?