Implement Table/Column.cast

[radeusgd]

Known in the specification as set_value_type.

• The conversions in the in-memory backend should happen as described in the design.
  • We should report Lossy_Conversion if casting a non-text value to text and its text representation ahs to be truncated.
    • But if casting a text to fixed-length text, it is kind of explicit, so I think the warning can be skipped.
  • We may implement this before Extend in-memory storage to support Integers other than 64-bit, and fixed-length/length-capped strings #5159, in that case this logic will not be included in this task, and it will have to be updated as part of Extend in-memory storage to support Integers other than 64-bit, and fixed-length/length-capped strings #5159.
  • We should report a warning if some integers do not fit a smaller bit-size.
• In the Database backend, this should compile to an SQL CAST operator and work according to each database's logic.
• Add tests for in-memory casts.
• Add common (in-memory and database) tests for the most common portable cases.
  • Likely, only int->text will be portable enough.
• Add database tests for some specific and less-portable but useful cases.
  • For example, text->int, int<->float.

[enso-bot]

Radosław Waśko reports a new STANDUP for yesterday (2023-05-15):

Progress: Added test cases for cast, some design on how it could play with parse. Implementing common paths. Started implementing casting in-memory. It should be finished by 2023-05-18.

Next Day: Next day I will be working on the same task. Add remaining in-memory casts. Testing.

[enso-bot]

Radosław Waśko reports a new STANDUP for today (2023-05-16):

Progress: Implemented rest of in-memory casting logic (for now). Fixed issues ensuring (almost) all tests will be passing. Discussion about imports and encapsulation. Did today's bookclub (still need to catchup with Friday's). Got a draft PR up. It should be finished by 2023-05-18.

Next Day: Next day I will be working on the same task. Implement Lossy_Conversion facilities for the float->int conversion case in in-memory. Get the full PR ready. Catch up with bookclubs. Maybe start add_row_number.

[enso-bot]

Radosław Waśko reports a new STANDUP for today (2023-05-17):

Progress: Added the Lossy_Conversion warning. PR ready for review. Working on adding simple parse scenarios to Database bakcend replacing disabled cast paths. Basically got it done but got stuck at a weird warning propagation bug - will need to investigate. Caught up with bookclubs. It should be finished by 2023-05-18.

Next Day: Next day I will be working on the same task. Get the PR merged. Investigate the warnings bug - report it and find a workaround/fix. Create a PR for the parse improvements. Start work on add_row_numbers

[enso-bot]

Radosław Waśko reports a new STANDUP for yesterday (2023-05-18):

Progress: Applying suggestions on the PR. Got the PR merged. Found out the losing warnings bug and reported a repro. It should be finished by 2023-05-18.

Next Day: Next day I will be working on the #6410 task. Add workarounds for the bug in the parse PR. Prepare it to be merged. Start work on Mixed->? casts