use first 10 non null values only to infer format

FBruzzesi · FBruzzesi · commit 7d7398126172 · 2024-10-29T11:47:58.000+01:00
diff --git a/narwhals/_arrow/utils.py b/narwhals/_arrow/utils.py
@@ -362,7 +362,7 @@ def parse_datetime_format(arr: pa.StringArray) -> str:
     import pyarrow.compute as pc  # ignore-banned-import
 
     matches = pa.concat_arrays(  # converts from ChunkedArray to StructArray
-        pc.extract_regex(arr, pattern=FULL_RE).chunks
+        pc.extract_regex(pc.drop_null(arr).slice(0, 10), pattern=FULL_RE).chunks
     )
 
     if not pc.all(matches.is_valid()).as_py():

Original file line number	Diff line number	Diff line change
`@@ -362,7 +362,7 @@ def parse_datetime_format(arr: pa.StringArray) -> str:`
`362`	`362`	`import pyarrow.compute as pc # ignore-banned-import`
`363`	`363`
`364`	`364`	`matches = pa.concat_arrays( # converts from ChunkedArray to StructArray`
`365`		`- pc.extract_regex(arr, pattern=FULL_RE).chunks`
	`365`	`+ pc.extract_regex(pc.drop_null(arr).slice(0, 10), pattern=FULL_RE).chunks`
`366`	`366`	`)`
`367`	`367`
`368`	`368`	`if not pc.all(matches.is_valid()).as_py():`