- FIELD_SIZE_COMPAT_RE = r'(?<!%)%\((?P<field>autonumber|playlist_index)\)s'
- mobj = re.search(FIELD_SIZE_COMPAT_RE, outtmpl)
- if mobj:
- outtmpl = re.sub(
- FIELD_SIZE_COMPAT_RE,
- r'%%(\1)0%dd' % field_size_compat_map[mobj.group('field')],
- outtmpl)
-
- numeric_fields = list(self._NUMERIC_FIELDS)
-
- # Format date
- FORMAT_DATE_RE = FORMAT_RE.format(r'(?P<key>(?P<field>\w+)>(?P<format>.+?))')
- for mobj in re.finditer(FORMAT_DATE_RE, outtmpl):
- conv_type, field, frmt, key = mobj.group('type', 'field', 'format', 'key')
- if key in template_dict:
- continue
- value = strftime_or_none(template_dict.get(field), frmt, na)
- if conv_type in 'crs': # string
- value = sanitize(field, value)
- else: # number
- numeric_fields.append(key)
- value = float_or_none(value, default=None)
- if value is not None:
- template_dict[key] = value
-
- # Missing numeric fields used together with integer presentation types
- # in format specification will break the argument substitution since
- # string NA placeholder is returned for missing fields. We will patch
- # output template for missing fields to meet string presentation type.
- for numeric_field in numeric_fields:
- if numeric_field not in template_dict:
- outtmpl = re.sub(
- FORMAT_RE.format(re.escape(numeric_field)),
- r'%({0})s'.format(numeric_field), outtmpl)
-
- return outtmpl, template_dict
+
+ TMPL_DICT = {}
+ EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljqBU]'))
+ MATH_FUNCTIONS = {
+ '+': float.__add__,
+ '-': float.__sub__,
+ }
+ # Field is of the form key1.key2...
+ # where keys (except first) can be string, int or slice
+ FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
+ MATH_FIELD_RE = r'''{field}|{num}'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
+ MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
+ INTERNAL_FORMAT_RE = re.compile(r'''(?x)
+ (?P<negate>-)?
+ (?P<fields>{field})
+ (?P<maths>(?:{math_op}{math_field})*)
+ (?:>(?P<strf_format>.+?))?
+ (?P<alternate>(?<!\\),[^|)]+)?
+ (?:\|(?P<default>.*?))?
+ $'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
+
+ def _traverse_infodict(k):
+ k = k.split('.')
+ if k[0] == '':
+ k.pop(0)
+ return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
+
+ def get_value(mdict):
+ # Object traversal
+ value = _traverse_infodict(mdict['fields'])
+ # Negative
+ if mdict['negate']:
+ value = float_or_none(value)
+ if value is not None:
+ value *= -1
+ # Do maths
+ offset_key = mdict['maths']
+ if offset_key:
+ value = float_or_none(value)
+ operator = None
+ while offset_key:
+ item = re.match(
+ MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
+ offset_key).group(0)
+ offset_key = offset_key[len(item):]
+ if operator is None:
+ operator = MATH_FUNCTIONS[item]
+ continue
+ item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
+ offset = float_or_none(item)
+ if offset is None:
+ offset = float_or_none(_traverse_infodict(item))
+ try:
+ value = operator(value, multiplier * offset)
+ except (TypeError, ZeroDivisionError):
+ return None
+ operator = None
+ # Datetime formatting
+ if mdict['strf_format']:
+ value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
+
+ return value
+
+ na = self.params.get('outtmpl_na_placeholder', 'NA')
+
+ def _dumpjson_default(obj):
+ if isinstance(obj, (set, LazyList)):
+ return list(obj)
+ raise TypeError(f'Object of type {type(obj).__name__} is not JSON serializable')
+
+ def create_key(outer_mobj):
+ if not outer_mobj.group('has_key'):
+ return outer_mobj.group(0)
+ key = outer_mobj.group('key')
+ mobj = re.match(INTERNAL_FORMAT_RE, key)
+ initial_field = mobj.group('fields').split('.')[-1] if mobj else ''
+ value, default = None, na
+ while mobj:
+ mobj = mobj.groupdict()
+ default = mobj['default'] if mobj['default'] is not None else default
+ value = get_value(mobj)
+ if value is None and mobj['alternate']:
+ mobj = re.match(INTERNAL_FORMAT_RE, mobj['alternate'][1:])
+ else:
+ break
+
+ fmt = outer_mobj.group('format')
+ if fmt == 's' and value is not None and key in field_size_compat_map.keys():
+ fmt = '0{:d}d'.format(field_size_compat_map[key])
+
+ value = default if value is None else value
+
+ str_fmt = f'{fmt[:-1]}s'
+ if fmt[-1] == 'l': # list
+ delim = '\n' if '#' in (outer_mobj.group('conversion') or '') else ', '
+ value, fmt = delim.join(variadic(value)), str_fmt
+ elif fmt[-1] == 'j': # json
+ value, fmt = json.dumps(value, default=_dumpjson_default), str_fmt
+ elif fmt[-1] == 'q': # quoted
+ value, fmt = compat_shlex_quote(str(value)), str_fmt
+ elif fmt[-1] == 'B': # bytes
+ value = f'%{str_fmt}'.encode('utf-8') % str(value).encode('utf-8')
+ value, fmt = value.decode('utf-8', 'ignore'), 's'
+ elif fmt[-1] == 'U': # unicode normalized
+ opts = outer_mobj.group('conversion') or ''
+ value, fmt = unicodedata.normalize(
+ # "+" = compatibility equivalence, "#" = NFD
+ 'NF%s%s' % ('K' if '+' in opts else '', 'D' if '#' in opts else 'C'),
+ value), str_fmt
+ elif fmt[-1] == 'c':
+ if value:
+ value = str(value)[0]
+ else:
+ fmt = str_fmt
+ elif fmt[-1] not in 'rs': # numeric
+ value = float_or_none(value)
+ if value is None:
+ value, fmt = default, 's'
+
+ if sanitize:
+ if fmt[-1] == 'r':
+ # If value is an object, sanitize might convert it to a string
+ # So we convert it to repr first
+ value, fmt = repr(value), str_fmt
+ if fmt[-1] in 'csr':
+ value = sanitize(initial_field, value)
+
+ key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
+ TMPL_DICT[key] = value
+ return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
+
+ return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
+
+ def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
+ outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
+ return self.escape_outtmpl(outtmpl) % info_dict
projects / yt-dlp.git / blobdiff