tvalentyn commented on a change in pull request #15775:
URL: https://github.com/apache/beam/pull/15775#discussion_r739404366
##########
File path: sdks/python/apache_beam/io/textio.py
##########
@@ -348,19 +345,19 @@ def _read_record(self, file_to_read, read_buffer):
return (read_buffer.data[record_start_position_in_buffer:], -1)
if self._strip_trailing_newlines:
- # Current record should not contain the separator.
+ # Current record should not contain the delimiter.
return (
read_buffer.data[record_start_position_in_buffer:sep_bounds[0]],
sep_bounds[1] - record_start_position_in_buffer)
else:
- # Current record should contain the separator.
+ # Current record should contain the delimiter.
return (
read_buffer.data[record_start_position_in_buffer:sep_bounds[1]],
sep_bounds[1] - record_start_position_in_buffer)
@staticmethod
def _is_self_overlapping(delimiter):
- # delimiter self-overlaps if v exists such as delimiter = vu = wv
+ # A delimiter self-overlaps if it has a prefix that is also its suffix.
# with u and w non-empty
Review comment:
drop this line
##########
File path: sdks/python/apache_beam/io/textio_test.py
##########
@@ -1077,7 +1077,7 @@ def test_custom_delimiter_must_not_self_overlap_ok(self):
delimiter=delimiter,
)
- def test_custom_delimiter_must_not_self_overlap_error(self):
+ def test_self_overlapping_delimiter_is_rejected(self):
"""Self-overlapping delimiter is rejected."""
Review comment:
nit: redundant comment
##########
File path: sdks/python/apache_beam/io/textio_test.py
##########
@@ -1063,7 +1063,7 @@ def test_custom_delimiter_must_not_empty_bytes(self):
delimiter=delimiter,
)
- def test_custom_delimiter_must_not_self_overlap_ok(self):
+ def test_non_self_overlapping_delimiter_is_accepted(self):
"""Non self-overlapping delimiter is accepted."""
Review comment:
nit: redundant comment
##########
File path: sdks/python/apache_beam/io/textio.py
##########
@@ -272,29 +272,26 @@ def _find_separator_bounds(self, file_to_read,
read_buffer):
# Using find() here is more efficient than a linear scan
# of the byte array.
- next_lf = read_buffer.data.find(delimiter, current_pos)
-
- if next_lf >= 0:
- if self._delimiter is None \
- and read_buffer.data[next_lf - 1:next_lf] == b'\r':
- # Found a '\r\n' if delimiter is not define.
- # Accepting that as the next separator.
- return (next_lf - 1, next_lf + 1)
+ next_delim = read_buffer.data.find(delimiter, current_pos)
+
+ if next_delim >= 0:
+ if (self._delimiter is None and
+ read_buffer.data[next_delim - 1:next_delim] == b'\r'):
+ # Accept both '\r\n' and '\n' as a default delimiter.
+ # Accepting that as the next delimiter.
Review comment:
drop this line
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
