ENH: partial string match in query #26027
Conversation
…tring-match-in-query
Codecov Report
@@ Coverage Diff @@ ## master #26027 +/- ## =========================================== - Coverage 91.82% 40.72% -51.11% =========================================== Files 175 175 Lines 52539 52554 +15 =========================================== - Hits 48246 21404 -26842 - Misses 4293 31150 +26857
Continue to review full report at Codecov.
|
Codecov Report
@@ Coverage Diff @@ ## master #26027 +/- ## ========================================== + Coverage 91.82% 91.95% +0.12% ========================================== Files 175 175 Lines 52539 52427 -112 ========================================== - Hits 48246 48211 -35 + Misses 4293 4216 -77
Continue to review full report at Codecov.
|
| partial_str_match : bool, optional, default False | ||
| If this is True, an `expr` like "string_query in list_like_of_strings" | ||
| is interpreted as partial string match (the default behavior is exact | ||
| matching). |
There was a problem hiding this comment.
Add a "versionadded" tag in the docstring.
| # equality | ||
| res1 = df.query('color == "red"', parser=parser, engine=engine) | ||
| res2 = df.query('"red" == color', parser=parser, engine=engine) | ||
| res1 = df.query('color == "red"', **kwargs) |
There was a problem hiding this comment.
Can you just be explicit about the parameter being passed instead of using kwargs?
pandas/core/computation/eval.py Outdated
| def eval(expr, parser='pandas', engine=None, truediv=True, | ||
| local_dict=None, global_dict=None, resolvers=(), level=0, | ||
| target=None, inplace=False): | ||
| target=None, inplace=False, partial_str_match=False): |
There was a problem hiding this comment.
Can you add a type annotation for the new keyword?
pandas/core/computation/expr.py Outdated
| preparser=preparser) | ||
| def __init__(self, env, engine, parser, preparser=lambda x: x, | ||
| partial_str_match=False): | ||
| super(PythonExprVisitor, self).__init__( |
There was a problem hiding this comment.
Don't need Py2 compat call any more so just super() should be fine
| return request.param | ||
| | ||
| | ||
| @pytest.fixture(params=[False, True]) |
There was a problem hiding this comment.
Does this need to be a fixture? I think it would be better served to have a generic True / False fixture instead of duplicating this for particular parameters (if one doesn't already exist)
There was a problem hiding this comment.
Thank you for your comment. I could not find such a generic boolean fixture in tests/**/*.py. How do you think about adding, say, def boolean fixture in tests/frame/common.py? Am I missing your point?
| a = np.random.choice(['red', 'green'], size=10) | ||
| b = np.random.choice(['eggs', 'ham'], size=10) | ||
| a = np.random.choice(['red', 'a_red', 'a_red_a', | ||
| 'red_a', 'Red', 'green'], size=30) |
| res1 = df.query('["red"] in color', **kwargs) | ||
| res2 = df.query('"red" in color', **kwargs) | ||
| exp1 = df[ind.isin(['red'])] | ||
| if partial_str_match: |
There was a problem hiding this comment.
This test is pretty huge already. I think it would be better to create a separate test for this
…np2/pandas into feature/partial-string-match-in-query
| I am -1 on this idea generally. This is very easily implemented using the |
4 participants
git diff upstream/master -u -- "*.py" | flake8 --diffThis PR proposes to add partial string matching functionality to
querymethod. In the proposed implementation, a query likedf.query('"alice" in person_name', partial_str_match=True)returns rows with 'person_name' containing 'alice' in them. For other kind of queries (e.g.'["alice", "bob"] in person_name'or'age == 20') withpartial_str_match=Trueor those withpartial_str_match=False(which is set to default for backward compatibility), results as before are returned.