==== common pandas data types ====

^ data type ^ description ^ supports missing values ^
| float | The NumPy float type | Yes |
| int | The NumPy integer type | No |
| 'Int64' | pandas nullable integer type | Yes |
| object | The NumPy type for storing strings (and mixed types) | |
| 'category' | pandas categorical type | Yes |
| bool | The NumPy Boolean type | No. \\ None becomes False, np.nan becomes True. |
| 'boolean' | pandas nullable Boolean type | Yes |
| datetime64[ns] | The NumPy date type | Yes (NaT) |

Ref:- (Pandas 1.x Cookbook, by Matt Harrison and Theodore Petrou, second edition, published in 2020) -> Chapter 1 -> page-7


==== What packages does pandas depend on? ====
  * Dependencies - https://pandas.pydata.org/pandas-docs/stable/getting_started/install.html#dependencies
  * Recommended dependencies - https://pandas.pydata.org/pandas-docs/stable/getting_started/install.html#recommended-dependencies
  * Optional dependencies - https://pandas.pydata.org/pandas-docs/stable/getting_started/install.html#optional-dependencies

==== links I came across ====
  * https://github.com/pandas-dev/pandas/releases - pandas release history
  * http://pandas.pydata.org/pandas-docs/stable/getting_started/install.html - pandas installation page. Contains instructions to install pandas in various ways.
  * http://pandas.pydata.org/pandas-docs/stable/ - pandas official documentation
  * https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html - good reference to learn about iloc, slicing ranges
  * DuckDB can run SQL queries directly on Parquet files and automatically take advantage of the advanced features of the Parquet format.
    * https://duckdb.org/2021/06/25/querying-parquet.html - gives more details
    * tags | pandas, streaming

==== pandas.errors.EmptyDataError ====
Sample code to generate EmptyDataError exception
<code>
>>> import pandas as pd
>>> from io import StringIO
>>> empty = StringIO()
>>> pd.read_csv(empty)
Traceback (most recent call last):
...
pandas.errors.EmptyDataError: No columns to parse from file
</code>

Ref: https://pandas.pydata.org/pandas-docs/dev/reference/api/pandas.errors.EmptyDataError.html - Got the initial version of the code from here.

==== pandas.io.common.EmptyDataError is deprecated ====
Instead of
  pandas.io.common.EmptyDataError
use
  pandas.errors.EmptyDataError

data point | as of pandas 1.1.2, pandas.io.common.EmptyDataError does not work.

Ref:-
  * https://pandas.pydata.org/docs/whatsnew/v0.20.0.html#pandas-errors
  * <WRAP indent>
> We are adding a standard public module for all pandas exceptions & warnings pandas.errors. (GH14800). Previously these exceptions & warnings could be imported from pandas.core.common or pandas.io.common. These exceptions and warnings will be removed from the *.common locations in a future release. (GH15541)
</WRAP>
  * https://pandas.pydata.org/docs/reference/api/pandas.errors.EmptyDataError.html - documentation from latest stable version
  * https://pandas.pydata.org/pandas-docs/version/1.4/reference/api/pandas.errors.EmptyDataError.html - documentation from version 1.4
  * https://pandas.pydata.org/pandas-docs/version/0.20/generated/pandas.errors.EmptyDataError.html - documentation from version 0.20

==== assert_frame_equal ====
Instead of
  from pandas.util.testing import assert_frame_equal
use
  from pandas.testing import assert_frame_equal

data point:

Using ''from pandas.util.testing import assert_frame_equal'' in pandas 1.1.2, I get

  FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.