def _find_near_matches_substitutions_ngrams(subsequence, sequence,
max_substitutions):
subseq_len = len(subsequence)
seq_len = len(sequence)
def make_match(start, end, dist):
return Match(start, end, dist, matched=sequence[start:end])
ngram_len = subseq_len // (max_substitutions + 1)
if ngram_len == 0:
raise ValueError(
"The subsequence's length must be greater than max_substitutions!")
for ngram_start in range(0, len(subsequence) - ngram_len + 1, ngram_len):
ngram_end = ngram_start + ngram_len
subseq_before = subsequence[:ngram_start]
subseq_after = subsequence[ngram_end:]
for index in search_exact(
subsequence[ngram_start:ngram_end],
sequence,
ngram_start,
seq_len - (subseq_len - ngram_end),
):
n_substitutions = 0
seq_before = sequence[index - ngram_start:index]
if subseq_before != seq_before:
n_substitutions += count_differences_with_maximum(
seq_before, subseq_before,
max_substitutions - n_substitutions + 1)
if n_substitutions > max_substitutions:
continue
seq_after = sequence[index + ngram_len:index - ngram_start +
subseq_len]
if subseq_after != seq_after:
if n_substitutions == max_substitutions:
continue
n_substitutions += count_differences_with_maximum(
seq_after, subseq_after,
max_substitutions - n_substitutions + 1)
if n_substitutions > max_substitutions:
continue
yield make_match(
start=index - ngram_start,
end=index - ngram_start + subseq_len,
dist=n_substitutions,
)
def find_near_matches_substitutions_ngrams(subsequence, sequence,
max_substitutions):
if not (
isinstance(subsequence, text_type) or
isinstance(sequence, text_type)
):
try:
results = _subs_only_fnm_ngram_byteslike(
subsequence, sequence, max_substitutions)
except TypeError:
pass
else:
matches = [
Match(
index,
index + len(subsequence),
count_differences_with_maximum(
sequence[index:index+len(subsequence)],
subsequence,
max_substitutions + 1,
),
)
for index in results
]
return [
get_best_match_in_group(group)
for group in group_matches(matches)
]
return py_find_near_matches_substitutions_ngrams(
subsequence, sequence, max_substitutions)
def _find_near_matches_substitutions_ngrams(subsequence, sequence,
max_substitutions):
subseq_len = len(subsequence)
seq_len = len(sequence)
ngram_len = subseq_len // (max_substitutions + 1)
if ngram_len == 0:
raise ValueError(
"The subsequence's length must be greater than max_substitutions!"
)
for ngram_start in range(0, len(subsequence) - ngram_len + 1, ngram_len):
ngram_end = ngram_start + ngram_len
subseq_before = subsequence[:ngram_start]
subseq_after = subsequence[ngram_end:]
for index in search_exact(
subsequence[ngram_start:ngram_end], sequence,
ngram_start, seq_len - (subseq_len - ngram_end),
):
n_substitutions = 0
seq_before = sequence[index - ngram_start:index]
if subseq_before != seq_before:
n_substitutions += count_differences_with_maximum(
seq_before, subseq_before,
max_substitutions - n_substitutions + 1)
if n_substitutions > max_substitutions:
continue
seq_after = sequence[index + ngram_len:index - ngram_start + subseq_len]
if subseq_after != seq_after:
if n_substitutions == max_substitutions:
continue
n_substitutions += count_differences_with_maximum(
seq_after, subseq_after,
max_substitutions - n_substitutions + 1)
if n_substitutions > max_substitutions:
continue
yield Match(
start=index - ngram_start,
end=index - ngram_start + subseq_len,
dist=n_substitutions,
)
def search(self, subsequence, sequence, max_subs):
results = fnm_subs_lp_byteslike(subsequence, sequence, max_subs)
matches = [
Match(index,
index + len(subsequence),
count_differences_with_maximum(
sequence[index:index + len(subsequence)],
subsequence,
max_subs + 1,
),
matched=sequence[index:index + len(subsequence)])
for index in results
]
return matches
def search(self, subsequence, sequence, max_subs):
results = fnm_subs_lp_byteslike(subsequence, sequence,
max_subs)
matches = [
Match(
index,
index + len(subsequence),
count_differences_with_maximum(
sequence[index:index+len(subsequence)],
subsequence,
max_subs + 1,
),
)
for index in results
]
return matches
def search(self, subsequence, sequence, max_subs):
results = fnm_subs_ngrams_byteslike(subsequence, sequence,
max_subs)
matches = [
Match(index,
index + len(subsequence),
count_differences_with_maximum(
sequence[index:index + len(subsequence)],
subsequence,
max_subs + 1,
),
matched=sequence[index:index + len(subsequence)])
for index in results
]
return [
get_best_match_in_group(group)
for group in group_matches(matches)
]
def search(self, subsequence, sequence, max_subs):
results = fnm_subs_ngrams_byteslike(subsequence, sequence,
max_subs)
matches = [
Match(
index,
index + len(subsequence),
count_differences_with_maximum(
sequence[index:index+len(subsequence)],
subsequence,
max_subs + 1,
),
)
for index in results
]
return [
get_best_match_in_group(group)
for group in group_matches(matches)
]
def count_diffs(self, seq1, seq2, max_diffs):
return count_differences_with_maximum(seq1, seq2, max_diffs)
def count_diffs(self, seq1, seq2, max_diffs):
return count_differences_with_maximum(seq1, seq2, max_diffs)
