DEPR: Deprecate tupleize_cols in Index constructor

[gfyoung]

Title is self-explanatory.

xref #17060.

[jreback]

why are you changing this?

[gfyoung]

Because you said that we should deprecate this parameter back in #17060. If that's no longer the case, I can just close this.

[codecov]

Codecov Report

Merging #17899 into master will decrease coverage by 0.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #17899      +/-   ##
==========================================
- Coverage   91.23%   91.22%   -0.02%     
==========================================
  Files         163      163              
  Lines       50105    50111       +6     
==========================================
- Hits        45715    45712       -3     
- Misses       4390     4399       +9

Flag	Coverage Δ
#multiple	89.03% <100%> (ø)	⬆️
#single	40.32% <85.71%> (-0.06%)	⬇️

Impacted Files	Coverage Δ
pandas/core/dtypes/dtypes.py	95.14% <100%> (ø)	⬆️
pandas/core/indexes/base.py	96.43% <100%> (+0.01%)	⬆️
pandas/io/gbq.py	25% <0%> (-58.34%)	⬇️
pandas/core/frame.py	97.75% <0%> (-0.1%)	⬇️

---

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 5bf7f9a...613499c. Read the comment docs.

[codecov]

Codecov Report

Merging #17899 into master will decrease coverage by 0.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #17899      +/-   ##
==========================================
- Coverage   91.23%   91.22%   -0.02%     
==========================================
  Files         163      163              
  Lines       50105    50111       +6     
==========================================
- Hits        45715    45712       -3     
- Misses       4390     4399       +9

Flag	Coverage Δ
#multiple	89.03% <100%> (ø)	⬆️
#single	40.32% <85.71%> (-0.06%)	⬇️

Impacted Files	Coverage Δ
pandas/core/indexes/base.py	96.43% <100%> (+0.01%)	⬆️
pandas/core/dtypes/dtypes.py	95.14% <100%> (ø)	⬆️
pandas/io/gbq.py	25% <0%> (-58.34%)	⬇️
pandas/core/frame.py	97.75% <0%> (-0.1%)	⬇️

---

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 5bf7f9a...613499c. Read the comment docs.

[jorisvandenbossche]

I don't think we can deprecate this. When we implement the "dtype=objcet disables all inference" (and thus preserves the tuples as tuples instead of returning MultiIndex) (#17246), we have an alternative way to do this, and we might consider deprecating tupleize_cols. But #17246 hasn't been solved yet.

[jorisvandenbossche]

This is also what Tom mentioned in #17060 (comment)

Including the Index constructor one (just to confirm)?

I think so, with the idea the Index([('a', 1)]) -> MultiIndex, while Index([('a', 1)], dtype=object) returns a regular Index with tuples for values. xref #17246

The main thing that would not be possible with dtype=object is to "infer the dtype, but never return MI (so don't infer tuples to MI)", which is possible now with tupleize_cols=False

[gfyoung]

Ah, right...thanks for reminding me about that. I guess the deprecation does seem premature in light of the ongoing discussion in #17246.

I'll close the PR but can always revisit once it gets resolved.

[jorisvandenbossche]

the ongoing discussion in #17246.

I think we agree on the idea, it just needs somebody doing the work.