Fix replace error when regex has only zero match quantifiers

[davidwendt]

Closes #10753

Fixes cudf::strings::replace_re logic that was reading past the end of a string when given a regex that contained net zero match quantifier pattern (e.g. 'D*' or 'D?s?' both can match to nothing).

[codecov]

Codecov Report

Merging #10760 (adaee7b) into branch-22.06 (84f88ce) will increase coverage by 0.02%.
The diff coverage is n/a.

@@               Coverage Diff                @@
##           branch-22.06   #10760      +/-   ##
================================================
+ Coverage         86.40%   86.43%   +0.02%     
================================================
  Files               143      143              
  Lines             22444    22444              
================================================
+ Hits              19393    19399       +6     
+ Misses             3051     3045       -6     

Impacted Files	Coverage Δ
python/cudf/cudf/comm/gpuarrow.py	79.76% <ø> (ø)
python/cudf/cudf/core/column/string.py	89.21% <ø> (+0.12%)	⬆️
python/cudf/cudf/core/frame.py	93.41% <ø> (ø)
python/cudf/cudf/core/series.py	95.16% <ø> (ø)
python/custreamz/custreamz/kafka.py	29.16% <ø> (ø)
python/cudf/cudf/_fuzz_testing/utils.py	0.00% <0.00%> (ø)
python/cudf/cudf/core/dataframe.py	93.74% <0.00%> (+0.04%)	⬆️
python/cudf/cudf/core/groupby/groupby.py	91.79% <0.00%> (+0.22%)	⬆️
python/cudf/cudf/core/tools/datetimes.py	84.49% <0.00%> (+0.30%)	⬆️
... and 1 more

---

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 9b8d26f...adaee7b. Read the comment docs.

[bdice]

The fix looks good. Can we add a test that covers the case that previously failed in #10753?

[bdice]

After reading #10753, I see that the behavior may not be well defined across all regex implementations. This page documents some of those differences: https://www.regular-expressions.info/zerolength.html

However, it seems there is some consensus on how to proceed after a zero-length match:

The simplest solution, which is used by most regex engines, is to start the next match attempt one character after the end of the previous match, if the previous match was zero-length.

I would suggest that we write a test matching this behavior (which is the same as the current behavior, as far as I can tell), and claim that is our "defined" behavior for now. From my understanding, it seems that we have three pieces of evidence that this behavior is reasonable: it matches the user's expectation in #10753, it matches the Python re.sub behavior, and it aligns with the guidance from that webpage on what is "used by most regex engines."

[bdice]

This test looks great. Thanks @davidwendt! For the record, I checked and the behavior of this test matches that of Python's re.sub. Might be worth adding a Python test for zero-match quantifiers if we already have some tests there.

[davidwendt]

@gpucibot merge