linesep 0.2.0

Installation

linesep requires Python 3.6 or higher. Just use pip for Python 3 (You have pip, right?) to install:

python3 -m pip install linesep

Examples

Reading paragraphs separated by a blank line:

with open('text.txt') as fp:
    for entry in linesep.read_separated(fp, '\n\n'):
        ...

Parsing output from find -print0:

find = subprocess.Popen(
    ['find', '/', '-some', '-complicated', '-condition', '-print0'],
    stdout=subprocess.PIPE,
)
for filepath in linesep.read_terminated(find.stdout, '\0'):
    ...

A poor man’s JSON Sequence parser:

for entry in linesep.read_preceded(fp, '\x1E'):
    try:
        obj = json.loads(entry)
    except ValueError:
        pass
    else:
        yield obj

API

Note: Strings, filehandles, and regexes passed to this library’s functions may be either binary or text. However, the arguments to a single invocation of a function must be either all binary or all text, and the return type will match.

Note: Using the read_* functions with a variable-length regular expression is unreliable. The only truly foolproof way to split on such regexes is to first read the whole file into memory and then call one of the split_* functions.

linesep.read_preceded(
    fp: IO[AnyStr],
    sep: Union[AnyStr, re.Pattern[AnyStr]],
    retain: bool = False,
    chunk_size: int = 512,
) -> Iterator[AnyStr]

Read segments from a file-like object fp in which the beginning of each segment is indicated by the string or compiled regex sep. A generator of segments is returned; an empty file will always produce an empty generator.

If retain is true, the delimiters are included at the beginning of each segment; otherwise, they are discarded.

Data is read from the filehandle chunk_size characters at a time. If sep is a variable-length compiled regex and a delimiter in the file crosses a chunk boundary, the results are undefined.

linesep.read_separated(
    fp: IO[AnyStr],
    sep: Union[AnyStr, re.Pattern[AnyStr]],
    retain: bool = False,
    chunk_size: int = 512,
) -> Iterator[AnyStr]

Read segments from a file-like object fp in which segments are separated by the string or compiled regex sep. A generator of segments is returned; an empty file will always produce a generator with one element, the empty string.

If retain is true, the delimiters are included in the output, with the elements of the generator alternating between segments and separators, starting with a (possible empty) segment. If retain is false, the delimiters will be discarded.

Data is read from the filehandle chunk_size characters at a time. If sep is a variable-length compiled regex and a delimiter in the file crosses a chunk boundary, the results are undefined.

linesep.read_terminated(
    fp: IO[AnyStr],
    sep: Union[AnyStr, re.Pattern[AnyStr]],
    retain: bool = False,
    chunk_size: int = 512,
) -> Iterator[AnyStr]

Read segments from a file-like object fp in which the end of each segment is indicated by the string or compiled regex sep. A generator of segments is returned; an empty file will always produce an empty generator.

If retain is true, the delimiters are included at the end of each segment; otherwise, they are discarded.

Data is read from the filehandle chunk_size characters at a time. If sep is a variable-length compiled regex and a delimiter in the file crosses a chunk boundary, the results are undefined.

linesep.split_preceded(
    s: AnyStr,
    sep: Union[AnyStr, re.Pattern[AnyStr]],
    retain: bool = False,
) -> List[AnyStr]

Split a string s into zero or more segments starting with/preceded by the string or compiled regex sep. A list of segments is returned; an empty input string will always produce an empty list.

If retain is true, the delimiters are included at the beginning of each segment; otherwise, they are discarded.

linesep.split_separated(
    s: AnyStr,
    sep: Union[AnyStr, re.Pattern[AnyStr]],
    retain: bool = False,
) -> List[AnyStr]

Split a string s into one or more segments separated by the string or compiled regex sep. A list of segments is returned; an empty input string will always produce a list with one element, the empty string.

If retain is true, the delimiters are included in the output, with the elements of the list alternating between segments and separators, starting with a (possible empty) segment. If retain is false, the delimiters will be discarded.

linesep.split_terminated(
    s: AnyStr,
    sep: Union[AnyStr, re.Pattern[AnyStr]],
    retain: bool = False,
) -> List[AnyStr]

Split a string s into zero or more segments terminated by the string or compiled regex sep. A list of segments is returned; an empty input string will always produce an empty list.

If retain is true, the delimiters are included at the end of each segment; otherwise, they are discarded.

linesep.join_preceded(iterable: Iterable[AnyStr], sep: AnyStr) -> AnyStr

Join the elements of iterable together, preceding each one with sep.

linesep.join_separated(iterable: Iterable[AnyStr], sep: AnyStr) -> AnyStr

Join the elements of iterable together, separating consecutive elements with sep.

linesep.join_terminated(iterable: Iterable[AnyStr], sep: AnyStr) -> AnyStr

Join the elements of iterable together, appending sep to each one.

linesep.write_preceded(
    fp: IO[AnyStr],
    iterable: Iterable[AnyStr],
    sep: AnyStr,
) -> None

Write the elements of iterable to the filehandle fp, preceding each one with sep.

linesep.write_separated(
    fp: IO[AnyStr],
    iterable: Iterable[AnyStr],
    sep: AnyStr,
) -> None

Write the elements of iterable to the filehandle fp, separating consecutive elements with sep.

linesep.write_terminated(
    fp: IO[AnyStr],
    iterable: Iterable[AnyStr],
    sep: AnyStr,
) -> None

Write the elements of iterable to the filehandle fp, appending sep to each one.