chore: remove python2

[dorianj]

Summary of Changes

Remove all direct usage of six (must stay in requirements because other packages still use it), remove python 2 CI.

Ref: amundsen-io/amundsen#560

Tests

Removed Python 2 from CI suite, but no tests modified.

Documentation

README updated to reflect new requirements.

CheckList

Make sure you have checked all steps below to ensure a timely review.

• PR title addresses the issue accurately and concisely. Example: "Updates the version of Flask to v1.0.2"
  • In case you are adding a dependency, check if the license complies with the ASF 3rd Party License Policy.
• PR includes a summary of changes.
• PR adds unit tests, updates existing unit tests, OR documents why no test additions or modifications are needed.
• In case of new functionality, my PR adds documentation that describes how to use it.
  • All the public functions and the classes in the PR contain docstrings that explain what it does
• PR passes make test

cc @feng-tao

[codecov-commenter]

Codecov Report

Merging #310 into master will decrease coverage by 0.11%.
The diff coverage is 87.50%.

@@            Coverage Diff             @@
##           master     #310      +/-   ##
==========================================
- Coverage   74.30%   74.19%   -0.12%     
==========================================
  Files         105      105              
  Lines        4492     4449      -43     
  Branches      419      405      -14     
==========================================
- Hits         3338     3301      -37     
+ Misses       1049     1041       -8     
- Partials      105      107       +2     

Impacted Files	Coverage Δ
databuilder/extractor/db2_metadata_extractor.py	0.00% <0.00%> (ø)
databuilder/extractor/mysql_metadata_extractor.py	0.00% <0.00%> (ø)
...abuilder/extractor/snowflake_metadata_extractor.py	95.00% <ø> (-0.32%)	⬇️
databuilder/callback/call_back.py	92.00% <100.00%> (-0.60%)	⬇️
databuilder/extractor/mssql_metadata_extractor.py	95.23% <100.00%> (-0.29%)	⬇️
...tabuilder/extractor/postgres_metadata_extractor.py	94.82% <100.00%> (-0.34%)	⬇️
databuilder/loader/file_system_neo4j_csv_loader.py	88.46% <100.00%> (-0.83%)	⬇️
databuilder/models/neo4j_csv_serde.py	80.00% <100.00%> (-0.86%)	⬇️
databuilder/models/table_metadata.py	91.78% <100.00%> (-0.04%)	⬇️
databuilder/publisher/neo4j_csv_publisher.py	84.21% <100.00%> (-0.23%)	⬇️
... and 6 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update d24cba9...044eeb3. Read the comment docs.

[feng-tao]

lgtm, thanks @dorianj , Also I think we should bump the dep to indicate we remove py2. In terms of typing, I would suggest waiting for a bit as a few soon to be reviewed PR is still open (e.g https://github.com/lyft/amundsendatabuilder/pull/303/files). If we change the mypy rule, it may cause some churns for the open pr. WDYT?

[jfhbrook]

Out of curiosity: how urgent is this change? I know it's 2020 and all, but I, uh, have use cases for potentially running a custom data builder on top of python 2 code. I totally understand if six is a maintenance burden, but wanted to ask anyway.

[feng-tao]

@jfhbrook, all the prevision pypi should support both py2 and py3.

[dorianj]

@jfhbrook the biggest improvement is in the ability to upgrade mypy and enforce type annotations. After doing that, I found dozens of legitimate logic errors already in the code (and hundreds of harmless type violations)... so the more we wait, the harder it will be do to get that upgraded

@feng-tao I bumped version from 2.6.5 -> 3.0.0. I believe by semvar this is correct because this is an API breaking change. But happy to do 2.7.0 if you disagree, not sure what norm here is.

With regards to holding the typings PR, I'd rather get it landed, rather than wait. There will be more PRs opened during the time it's being reviewed, which could cause the typing PR to get delayed indefinitely (and require continuous follow-up work). I will personally rebase and add typings to the existing PRs that are out there to make it seamless. wdyt?

[jfhbrook]

@dorianj yeah I have a lot of sympathy for that. Personally not stoked about maintaining a py2 codebase, just more trying to plan accordingly.

[dorianj]

Thanks Tao!