PERF: perform a unique intersection with IntervalIndex if at least one side is unique

[qwhelan]

Resolves #26709

       before           after         ratio
     [3ff4f38f]       [30bd5369]
     <datetime_iter~1>
-       749±100μs         365±70μs     0.49  index_object.IntervalIndexMethod.time_intersection_duplicate(1000)
-        59.6±3ms       8.04±0.2ms     0.13  index_object.IntervalIndexMethod.time_intersection_duplicate(100000)
           failed       2.50±0.01s      n/a  index_object.IntervalIndexMethod.time_intersection_duplicate(10000000)

• closes #xxxx
• tests added / passed
• passes git diff upstream/master -u -- "*.py" | flake8 --diff
• whatsnew entry

[makbigc]

@qwhelan Thanks for your quick respond.

Maybe adding check for duplicate nan in either side can resolve.
if self.isna().sum() > 1 or other.isna().sum() > 1:

[jreback]

can you merge master & try @makbigc suggestion

[codecov]

Codecov Report

Merging #26711 into master will decrease coverage by 0.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #26711      +/-   ##
==========================================
- Coverage   91.71%    91.7%   -0.02%     
==========================================
  Files         178      178              
  Lines       50740    50742       +2     
==========================================
- Hits        46538    46533       -5     
- Misses       4202     4209       +7

Flag	Coverage Δ
#multiple	90.3% <100%> (-0.01%)	⬇️
#single	41.21% <0%> (-0.11%)	⬇️

Impacted Files	Coverage Δ
pandas/core/indexes/interval.py	96.11% <100%> (-0.32%)	⬇️
pandas/io/gbq.py	78.94% <0%> (-10.53%)	⬇️
pandas/core/frame.py	96.88% <0%> (-0.12%)	⬇️
pandas/util/testing.py	90.84% <0%> (-0.11%)	⬇️

---

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 0f3e8e8...66d323b. Read the comment docs.

[codecov]

Codecov Report

Merging #26711 into master will decrease coverage by 0.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #26711      +/-   ##
==========================================
- Coverage   91.71%    91.7%   -0.02%     
==========================================
  Files         178      178              
  Lines       50740    50742       +2     
==========================================
- Hits        46538    46533       -5     
- Misses       4202     4209       +7

Flag	Coverage Δ
#multiple	90.3% <100%> (-0.01%)	⬇️
#single	41.21% <0%> (-0.11%)	⬇️

Impacted Files	Coverage Δ
pandas/core/indexes/interval.py	96.11% <100%> (-0.32%)	⬇️
pandas/io/gbq.py	78.94% <0%> (-10.53%)	⬇️
pandas/core/frame.py	96.88% <0%> (-0.12%)	⬇️
pandas/util/testing.py	90.84% <0%> (-0.11%)	⬇️

---

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 0f3e8e8...66d323b. Read the comment docs.

[qwhelan]

@makbigc Thanks, your suggestion works. The calls to .is_unique appear to give us that condition for free, so I've only added self.isna().sum() <= 1

@jreback Done and passing

[jreback]

thanks @qwhelan and @makbigc