fix: convert column type to str during dual read/write

[hughhhh]

SUMMARY

Users are blocked when trying to save a dataset on a query/table that has an ARRAY column in it. Due the python_type returning a list instead of a str for the type. To fix this i've added a util function to check if the type is a list and convert it to a string like this ARRAY

BEFORE/AFTER SCREENSHOTS OR ANIMATED GIF

TESTING INSTRUCTIONS

ADDITIONAL INFORMATION

• Has associated issue:
• Required feature flags:
• Changes UI
• Includes DB Migration (follow approval process in SIP-59)
  • Migration is atomic, supports rollback & is backwards-compatible
  • Confirm DB migration upgrade and downgrade tested
  • Runtime estimates and downtime expectations provided
• Introduces new feature or API
• Removes existing feature or API

[eschutho]

This is great @hughhhh. Can you write a quick unit test for it?

[codecov]

Codecov Report

Merging #19701 (8269591) into master (c8304a2) will decrease coverage by 0.00%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master   #19701      +/-   ##
==========================================
- Coverage   66.51%   66.51%   -0.01%     
==========================================
  Files        1684     1684              
  Lines       64559    64584      +25     
  Branches     6626     6626              
==========================================
+ Hits        42941    42955      +14     
- Misses      19923    19934      +11     
  Partials     1695     1695              

Flag	Coverage Δ
hive	52.67% <16.66%> (-0.02%)	⬇️
javascript	51.15% <ø> (ø)
mysql	81.93% <100.00%> (-0.03%)	⬇️
postgres	81.97% <100.00%> (-0.03%)	⬇️
presto	52.52% <16.66%> (-0.02%)	⬇️
python	82.41% <100.00%> (-0.03%)	⬇️
sqlite	81.75% <100.00%> (-0.03%)	⬇️
unit	47.73% <100.00%> (-0.02%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
superset/connectors/sqla/utils.py	91.34% <100.00%> (+0.52%)	⬆️
superset/sql_lab.py	78.90% <0.00%> (-2.74%)	⬇️

---

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 c8304a2...8269591. Read the comment docs.

[ktmud]

Should we save the original database type here? The type field specifically used Text as storage so the originally intention is probably to allow storing the full schema of complex types such as ARRAY and STRUCT?

What does the old column model save?

[hughhhh]

Should we save the original database type here? The type field specifically used Text as storage so the originally intention is probably to allow storing the full schema of complex types such as ARRAY and STRUCT?

What does the old column model save?

I'm pretty sure it was still ARRAY, but this is only a patch i'm pretty sure there other types we'd have to map to strings later

[ktmud]

Can it also be a STRUCT? I guess it probably depends on the database.

[hughhhh]

so i'm following the pattern from the sqlalchemy.sql.sqltypes

[villebro]

Should we save the original database type here? The type field specifically used Text as storage so the originally intention is probably to allow storing the full schema of complex types such as ARRAY and STRUCT?

I agree, using the original native type is the ideal solution here.

[villebro]

@hughhhh @ktmud I've opened up an alternative solution for this issue that reuses existing code that's used when creating physical datasets: #19714

[sadpandajoe]

🏷️ preset:2022.15