def test_empty_file(self):
"""
make sure an empty file does not throw an error when reverse_readline
is called this was a problem with an earlier implementation
"""
for idx, line in enumerate(
reverse_readfile(os.path.join(test_dir, "empty_file.txt"))):
raise ValueError("an empty file is being read!")
def test_empty_file(self):
"""
make sure an empty file does not throw an error when reverse_readline
is called this was a problem with an earlier implementation
"""
for idx, line in enumerate(reverse_readfile(
os.path.join(test_dir, "empty_file.txt"))):
raise ValueError("an empty file is being read!")
def test_reverse_readfile_gz(self):
"""
We are making sure a file containing line numbers is read in reverse
order, i.e. the first line that is read corresponds to the last line.
number
"""
fname = os.path.join(test_dir, "3000lines.txt.gz")
for idx, line in enumerate(reverse_readfile(fname)):
self.assertEqual(
int(line), self.NUMLINES - idx,
"read_backwards read {} whereas it should "
"have read {}".format(int(line), self.NUMLINES - idx))
def test_reverse_readfile_gz(self):
"""
We are making sure a file containing line numbers is read in reverse
order, i.e. the first line that is read corresponds to the last line.
number
"""
fname = os.path.join(test_dir, "3000_lines.txt.gz")
for idx, line in enumerate(reverse_readfile(fname)):
self.assertEqual(int(line), self.NUMLINES - idx,
"read_backwards read {} whereas it should "
"have read {}".format(
int(line), self.NUMLINES - idx))
def regrep(filename,
patterns,
reverse=False,
terminate_on_match=False,
postprocess=str):
r"""
A powerful regular expression version of grep.
Args:
filename (str): Filename to grep.
patterns (dict): A dict of patterns, e.g.,
{"energy": r"energy\\(sigma->0\\)\\s+=\\s+([\\d\\-\\.]+)"}.
reverse (bool): Read files in reverse. Defaults to false. Useful for
large files, especially when used with terminate_on_match.
terminate_on_match (bool): Whether to terminate when there is at
least one match in each key in pattern.
postprocess (callable): A post processing function to convert all
matches. Defaults to str, i.e., no change.
Returns:
A dict of the following form:
{key1: [[[matches...], lineno], [[matches...], lineno],
[[matches...], lineno], ...],
key2: ...}
For reverse reads, the lineno is given as a -ve number. Please note
that 0-based indexing is used.
"""
compiled = {k: re.compile(v) for k, v in patterns.items()}
matches = collections.defaultdict(list)
gen = reverse_readfile(filename) if reverse else zopen(filename, "rt")
for i, l in enumerate(gen):
for k, p in compiled.items():
m = p.search(l)
if m:
matches[k].append([[postprocess(g) for g in m.groups()],
-i if reverse else i])
if terminate_on_match and all(
len(matches.get(k, [])) for k in compiled.keys()):
break
try:
# Try to close open file handle. Pass if it is a generator.
gen.close()
except Exception:
pass
return matches
