def parse_op_collab_react(op_array, editor):
"""
Parse a op in OT type into a list of elementary operations.
There will be missing the author, timestamp...
https://github.com/ottypes/text
:param editor: FROG or collab-react-components
:param op_array: string to parse
:return : List of elem_ops contained in op
:rtype: list[ElementaryOperation]
:type op_array: [{'p': list[int], str: str}]
"""
elem_ops = []
""":type: list[ElementaryOperation]"""
for op in op_array:
if editor == 'collab-react-components':
assert len(op['p']) == 1
position = op['p'][0] # Collab
else:
if len(op['p']) != 2:
raise AssertionError(
"len(op['p']) != 2. Might be because you"
"are treating ill-formatted logs from the DB.")
assert op['p'][0] == 'text'
position = op['p'][1] # FROG
if 'sd' in op.keys():
# Deleting some letters
length_to_delete = len(op['sd'])
elem_ops.append(
ElementaryOperation(operation_type="del",
abs_position=position,
length_to_delete=length_to_delete,
changeset=op))
if 'si' in op.keys():
# Inserting some letters
text_to_add = op['si']
elem_ops.append(
ElementaryOperation(operation_type="add",
abs_position=position,
text_to_add=text_to_add,
changeset=op))
return elem_ops
def generateElemOps(self):
'''
used to sort the elem_operation and define the end time of the window
:return: None
'''
for op in self.operations:
if self.endTime < op.timestamp_end:
self.endTime = op.timestamp_end
if self.start_time > op.timestamp_start:
self.start_time = op.timestamp_start
self.elemOps.extend(op.elem_ops)
self.elemOps = ElementaryOperation.sort_elem_ops(self.elemOps)
def get_elem_ops(self, sorted_):
"""
Get the list of ElementaryOperation from all the Operation. The result is ordered by timestamp. Good for
building a representation of the text. Note that each ElementaryOperation know to which Operation it belongs
:return: list of ElementaryOperation
:rtype: list[ElementaryOperation]
"""
# Recover all the elementary ops
elem_ops = []
for op in self.operations:
for elem_op in op.elem_ops:
elem_ops.append(elem_op)
if sorted_:
return ElementaryOperation.sort_elem_ops(elem_ops)
else:
return elem_ops
def parse_changeset_etherpad(changeset):
"""
Parse a changeset of type etherpad into a list of elementary operations. There will be missing the author,
timestamp... http://policypad.readthedocs.io/en/latest/changesets.html
:param changeset: string to parse
:type changeset: str
:return : List of elem_ops contained in the changeset
:rtype: list[ElementaryOperation]
"""
def find_next_symbol_idx(string, start):
"""
Find the next symbol from start
"""
symbols = ['|', '$', '+', '-', '=', '*']
for i in range(start, len(string)):
if string[i] in symbols:
return i
# We reached the end of the text
return len(string)
line_number = 0
line_abs_position = 0
position = 0
position_inline = 0
elementary_operations = []
used_databank = 0
# Finding the first operations. It's always a |
idx = find_next_symbol_idx(changeset, 0)
while idx < len(changeset):
if changeset[idx] == '|':
# It's going to be taken care of by elif changeset[idx].isalnum()
idx += 1
elif changeset[idx] == '$':
# Should have already been added with the '+'
return elementary_operations
elif changeset[idx].isalnum():
# Format is |L+N,|L-N or |L=N
symbol_idx = find_next_symbol_idx(changeset, idx)
if changeset[symbol_idx] == '=':
# |L=N
# Keep N characters from the source text, containing L newlines.
# The last character kept MUST be a newline, and the final newline
# of the document is allowed.
# L
line_number += int(changeset[idx:symbol_idx], 36)
next_symbol_position = find_next_symbol_idx(changeset, symbol_idx + 1)
# N
line_abs_position += int(changeset[symbol_idx + 1:next_symbol_position], 36)
position += int(changeset[symbol_idx + 1:next_symbol_position], 36)
idx = next_symbol_position
elif changeset[symbol_idx] == '+':
# |L+N
# Insert N characters from the source text, containing L newlines.
# The last character inserted MUST be a newline, but not the (new)
# document's final newline.
next_symbol_position = find_next_symbol_idx(changeset, symbol_idx + 1)
# We care about N (size of the addition) since we will date N chars from the databank.
# noinspection PyPep8Naming
N = int(changeset[symbol_idx + 1:next_symbol_position], 36)
data_bank = changeset[changeset.find('$') + 1:]
text_to_add = data_bank[used_databank:used_databank + N]
elementary_operations.append(ElementaryOperation("add",
position,
text_to_add=text_to_add,
line_number=line_number,
position_inline=position_inline,
changeset=changeset))
used_databank += N
position += N
idx = next_symbol_position
else:
assert (changeset[symbol_idx] == '-')
# |L-N
# Delete N characters from the source text, containing L newlines.
# The last character inserted MUST be a newline, but not the (old)
# document's final newline.
next_symbol_position = find_next_symbol_idx(changeset, symbol_idx + 1)
# N
chars_to_delete = int(changeset[symbol_idx + 1:next_symbol_position], 36)
idx = next_symbol_position
elementary_operations.append(ElementaryOperation("del",
position,
length_to_delete=chars_to_delete,
line_number=line_number,
position_inline=position_inline,
changeset=changeset))
elif changeset[idx] == '=':
# Keep N characters from the source text, none of them newlines
# (position inline)
next_symbol_position = find_next_symbol_idx(changeset, idx + 1)
# We add the inline offset
position += int(changeset[idx + 1:next_symbol_position], 36)
position_inline += int(changeset[idx + 1:next_symbol_position], 36)
idx = next_symbol_position
elif changeset[idx] == '+':
next_symbol_position = find_next_symbol_idx(changeset, idx + 1)
# We care about N (size of the addition). We only take N chars from the databank (not counting the
# already used ones)
N = int(changeset[idx + 1:next_symbol_position], 36)
data_bank = changeset[changeset.find('$') + 1:]
text_to_add = data_bank[used_databank:used_databank + N]
elementary_operations.append(ElementaryOperation("add",
position,
text_to_add=text_to_add,
line_number=line_number,
position_inline=position_inline,
changeset=changeset))
used_databank += N
position += N
idx = next_symbol_position
elif changeset[idx] == '-':
# Remove the next n symbols
next_symbol_position = find_next_symbol_idx(changeset, idx + 1)
chars_to_delete = int(changeset[idx + 1:next_symbol_position], 36)
idx = next_symbol_position
elementary_operations.append(ElementaryOperation("del",
position,
length_to_delete=chars_to_delete,
line_number=line_number,
position_inline=position_inline,
changeset=changeset))
else:
# TODO: Format with '*'
assert changeset[idx] == '*'
idx = find_next_symbol_idx(changeset, idx + 1)