Retry once (or more?) on any TransportError

[simonw]

Got this exception:

  File "/Users/simon/Dropbox/Development/google-drive-to-sqlite/google_drive_to_sqlite/utils.py", line 79, in get
    response = httpx.get(url, params=params, headers=headers, timeout=self.timeout)
  File "/Users/simon/.local/share/virtualenvs/google-drive-to-sqlite-Wr1nXkpK/lib/python3.10/site-packages/httpx/_api.py", line 189, in get
    return request(
  File "/Users/simon/.local/share/virtualenvs/google-drive-to-sqlite-Wr1nXkpK/lib/python3.10/site-packages/httpx/_api.py", line 100, in request
    return client.request(
  File "/Users/simon/.local/share/virtualenvs/google-drive-to-sqlite-Wr1nXkpK/lib/python3.10/site-packages/httpx/_client.py", line 802, in request
    return self.send(request, auth=auth, follow_redirects=follow_redirects)
  File "/Users/simon/.local/share/virtualenvs/google-drive-to-sqlite-Wr1nXkpK/lib/python3.10/site-packages/httpx/_client.py", line 889, in send
    response = self._send_handling_auth(
  File "/Users/simon/.local/share/virtualenvs/google-drive-to-sqlite-Wr1nXkpK/lib/python3.10/site-packages/httpx/_client.py", line 917, in _send_handling_auth
    response = self._send_handling_redirects(
  File "/Users/simon/.local/share/virtualenvs/google-drive-to-sqlite-Wr1nXkpK/lib/python3.10/site-packages/httpx/_client.py", line 954, in _send_handling_redirects
    response = self._send_single_request(request)
  File "/Users/simon/.local/share/virtualenvs/google-drive-to-sqlite-Wr1nXkpK/lib/python3.10/site-packages/httpx/_client.py", line 990, in _send_single_request
    response = transport.handle_request(request)
  File "/Users/simon/.local/share/virtualenvs/google-drive-to-sqlite-Wr1nXkpK/lib/python3.10/site-packages/httpx/_transports/default.py", line 217, in handle_request
    with map_httpcore_exceptions():
  File "/Users/simon/.pyenv/versions/3.10.0/lib/python3.10/contextlib.py", line 153, in __exit__
    self.gen.throw(typ, value, traceback)
  File "/Users/simon/.local/share/virtualenvs/google-drive-to-sqlite-Wr1nXkpK/lib/python3.10/site-packages/httpx/_transports/default.py", line 77, in map_httpcore_exceptions
    raise mapped_exc(message) from exc
httpx.RemoteProtocolError: Server disconnected without sending a response.

Would be good to retry once if this happens.

[simonw]

https://www.python-httpx.org/exceptions/#the-exception-hierarchy shows the exception hierarchy:

[simonw]

I want to retry on any form of TransportError i think - no point retrying a DecodingError or a TooManyRedirects error.

[simonw]

Sadly it looks like httpx itself has decided not to implement retry logic, so I need to build this myself:

• Retry requests encode/httpx#108

[simonw]

While testing this I'm going to want to see if any transport errors have occurred - I think I'll add a -v/--verbose flag to the google-drive-to-sqlite files command.

[simonw]

I'm only going to retry GET, I won't retry POST.

[simonw]

Now manually testing this by running:

google-drive-to-sqlite files --folder 1E6Zg2X2bjjtPzVfX8YqdXZDCoB3AVA7i --nl --verbose > all-files.json-nl.txt

And keeping an eye on it while it runs with:

watch 'wc -l all-files.json-nl.txt && ls -lah all-files.json-nl.txt'

Started it running at 4:31pm.

[simonw]

It's at 37223 all-files.json-nl.txt and 49MB now, 25 minutes after starting.

[simonw]

That actually worked! 162M file resulted, with no errors.

[simonw]

Now running this to see what happens:

 time google-drive-to-sqlite files all-files.db --import-nl all-files.json-nl.txt
43.24s user 94.07s system 71% cpu 3:13.06 total

Produced a 80MB SQLite file, thanks presumably to the owners data being de-duplicated.

[simonw]

I'm suspicious of the 14,100 rows in the drive_users table.

[simonw]

Confirmed, something went very wrong there:

88 rows where permissionId is not null, 14,012 rows where permissionId is null.

[simonw]

Fixed that bug: