def get_all_equal_edits(self):
edits = []
for op, group in groupby(self.align_seq, lambda x: x[0]):
if op != "M":
merged = self.merge_edits(list(group))
edits.append(Edit(self.orig, self.cor, merged[0][1:]))
return edits
def get_all_merge_edits(self):
edits = []
for op, group in groupby(self.align_seq, lambda x: True
if x[0] == "M" else False):
if not op:
merged = self.merge_edits(list(group))
edits.append(Edit(self.orig, self.cor, merged[0][1:]))
return edits
def import_edit(self, orig, cor, edit, min=True, old_cat=False):
# Undefined error type
if len(edit) == 4:
edit = Edit(orig, cor, edit)
# Existing error type
elif len(edit) == 5:
edit = Edit(orig, cor, edit[:4], edit[4])
# Unknown edit format
else:
raise Exception("Edit not of the form: "
"[o_start, o_end, c_start, c_end, (cat)]")
# Minimise edit
if min:
edit = edit.minimise()
# Classify edit
if not old_cat:
edit = self.classify(edit)
return edit
def get_rule_edits(alignment):
edits = []
# Split alignment into groups of M, T and rest. (T has a number after it)
for op, group in groupby(alignment.align_seq,
lambda x: x[0][0] if x[0][0] in {"M", "T"} else False):
group = list(group)
# Ignore M
if op == "M": continue
# T is always split
elif op == "T":
for seq in group:
edits.append(Edit(alignment.orig, alignment.cor, seq[1:]))
# Process D, I and S subsequence
else:
processed = process_seq(group, alignment)
# Turn the processed sequence into edits
for seq in processed:
edits.append(Edit(alignment.orig, alignment.cor, seq[1:]))
return edits
def get_all_split_edits(self):
edits = []
for align in self.align_seq:
if align[0] != "M":
edits.append(Edit(self.orig, self.cor, align[1:]))
return edits