90 lines
3.2 KiB
Python
90 lines
3.2 KiB
Python
|
from collections import OrderedDict
|
||
|
|
import re
|
||
|
|
import json
|
||
|
|
|
||
|
|
|
||
|
|
def remove_empty_lines(text):
|
||
|
|
"""remove empty lines"""
|
||
|
|
assert (len(text) > 0)
|
||
|
|
assert (isinstance(text, list))
|
||
|
|
text = [t.strip() for t in text]
|
||
|
|
if "" in text:
|
||
|
|
text.remove("")
|
||
|
|
return text
|
||
|
|
|
||
|
|
|
||
|
|
class TextGrid(object):
|
||
|
|
def __init__(self, text):
|
||
|
|
text = remove_empty_lines(text)
|
||
|
|
self.text = text
|
||
|
|
self.line_count = 0
|
||
|
|
self._get_type()
|
||
|
|
self._get_time_intval()
|
||
|
|
self._get_size()
|
||
|
|
self.tier_list = []
|
||
|
|
self._get_item_list()
|
||
|
|
|
||
|
|
def _extract_pattern(self, pattern, inc):
|
||
|
|
"""
|
||
|
|
Parameters
|
||
|
|
----------
|
||
|
|
pattern : regex to extract pattern
|
||
|
|
inc : increment of line count after extraction
|
||
|
|
Returns
|
||
|
|
-------
|
||
|
|
group : extracted info
|
||
|
|
"""
|
||
|
|
try:
|
||
|
|
group = re.match(pattern, self.text[self.line_count]).group(1)
|
||
|
|
self.line_count += inc
|
||
|
|
except AttributeError:
|
||
|
|
raise ValueError("File format error at line %d:%s" % (self.line_count, self.text[self.line_count]))
|
||
|
|
return group
|
||
|
|
|
||
|
|
def _get_type(self):
|
||
|
|
self.file_type = self._extract_pattern(r"File type = \"(.*)\"", 2)
|
||
|
|
|
||
|
|
def _get_time_intval(self):
|
||
|
|
self.xmin = self._extract_pattern(r"xmin = (.*)", 1)
|
||
|
|
self.xmax = self._extract_pattern(r"xmax = (.*)", 2)
|
||
|
|
|
||
|
|
def _get_size(self):
|
||
|
|
self.size = int(self._extract_pattern(r"size = (.*)", 2))
|
||
|
|
|
||
|
|
def _get_item_list(self):
|
||
|
|
"""Only supports IntervalTier currently"""
|
||
|
|
for itemIdx in range(1, self.size + 1):
|
||
|
|
tier = OrderedDict()
|
||
|
|
item_list = []
|
||
|
|
tier_idx = self._extract_pattern(r"item \[(.*)\]:", 1)
|
||
|
|
tier_class = self._extract_pattern(r"class = \"(.*)\"", 1)
|
||
|
|
if tier_class != "IntervalTier":
|
||
|
|
raise NotImplementedError("Only IntervalTier class is supported currently")
|
||
|
|
tier_name = self._extract_pattern(r"name = \"(.*)\"", 1)
|
||
|
|
tier_xmin = self._extract_pattern(r"xmin = (.*)", 1)
|
||
|
|
tier_xmax = self._extract_pattern(r"xmax = (.*)", 1)
|
||
|
|
tier_size = self._extract_pattern(r"intervals: size = (.*)", 1)
|
||
|
|
for i in range(int(tier_size)):
|
||
|
|
item = OrderedDict()
|
||
|
|
item["idx"] = self._extract_pattern(r"intervals \[(.*)\]", 1)
|
||
|
|
item["xmin"] = self._extract_pattern(r"xmin = (.*)", 1)
|
||
|
|
item["xmax"] = self._extract_pattern(r"xmax = (.*)", 1)
|
||
|
|
item["text"] = self._extract_pattern(r"text = \"(.*)\"", 1)
|
||
|
|
item_list.append(item)
|
||
|
|
tier["idx"] = tier_idx
|
||
|
|
tier["class"] = tier_class
|
||
|
|
tier["name"] = tier_name
|
||
|
|
tier["xmin"] = tier_xmin
|
||
|
|
tier["xmax"] = tier_xmax
|
||
|
|
tier["size"] = tier_size
|
||
|
|
tier["items"] = item_list
|
||
|
|
self.tier_list.append(tier)
|
||
|
|
|
||
|
|
def toJson(self):
|
||
|
|
_json = OrderedDict()
|
||
|
|
_json["file_type"] = self.file_type
|
||
|
|
_json["xmin"] = self.xmin
|
||
|
|
_json["xmax"] = self.xmax
|
||
|
|
_json["size"] = self.size
|
||
|
|
_json["tiers"] = self.tier_list
|
||
|
|
return json.dumps(_json, ensure_ascii=False, indent=2)
|