PyPI 官网下载 | pympi-ling-1.39.tar.gz

# -*- coding: utf-8 -*- from xml.etree import cElementTree as etree import os import re import sys import time VERSION = '1.39' class Eaf: """Read and write Elan's Eaf files. .. note:: All times are in milliseconds and can't have decimals. :var dict adocument: Annotation document TAG entries. :var list licenses: Licences included in the file of the form: ``(name, url)``. :var dict header: XML header. :var list media_descriptors: Linked files, where every file is of the form: ``{attrib}``. :var list properties: Properties, where every property is of the form: ``(key, value)``. :var list linked_file_descriptors: Secondary linked files, where every linked file is of the form: ``{attrib}``. :var dict timeslots: Timeslot data of the form: ``{id -> time(ms)}``. :var dict tiers: Tiers, where every tier is of the form: ``{tier_name -> (aligned_annotations, reference_annotations, attributes, ordinal)}``, aligned_annotations of the form: ``[{id -> (begin_ts, end_ts, value, svg_ref)}]``, reference annotations of the form: ``[{id -> (reference, value, previous, svg_ref)}]``. :var list linguistic_types: Linguistic types, where every type is of the form: ``{id -> attrib}``. :var dict locales: Locales, of the form: ``{lancode -> (countrycode, variant)}``. :var dict languages: Languages, of the form: ``{langid -> (langdef, langlabel)}``. :var dict constraints: Constraints, every constraint is of the form: ``{stereotype -> description}``. :var dict controlled_vocabularies: Controlled vocabulary, where every controlled vocabulary is of the form: ``{id -> (descriptions, entries, ext_ref)}``, descriptions of the form: ``[(value, lang_ref, description)]``, entries of the form: ``{id -> (values, ext_ref)}``, values of the form: ``[(lang_ref, description, text)]``. :var list external_refs: External references of the form: ``{id -> (type, value)}``. :var list lexicon_refs: Lexicon references, where every reference is of the form: ``{id -> {attribs}}``. :var dict annotations: Dictionary of annotations of the form: ``{id -> tier}``, this is only used internally. """ ETYPES = {'iso12620', 'ecv', 'cve_id', 'lexen_id', 'resource_url'} CONSTRAINTS = { 'Time_Subdivision': "Time subdivision of parent annotation's time inte" 'rval, no time gaps allowed within this interval', 'Symbolic_Subdivision': 'Symbolic subdivision of a parent annotation. ' 'Annotations refering to the same parent are ordered', 'Symbolic_Association': '1-1 association with a parent annotation', 'Included_In': 'Time alignable annotations within the parent annotatio' "n's time interval, gaps are allowed"} MIMES = {'wav': 'audio/x-wav', 'mpg': 'video/mpeg', 'mpeg': 'video/mpg', 'xml': 'text/xml'} def __init__(self, file_path=None, author='pympi'): """Construct either a new Eaf file or read on from a file/stream. :param str file_path: Path to read from, - for stdin. If ``None`` an empty Eaf file will be created. :param str author: Author of the file. """ ctz = -time.altzone if time.localtime(time.time()).tm_isdst and\ time.daylight else -time.timezone self.maxts = None self.maxaid = None self.adocument = { 'AUTHOR': author, 'DATE': time.strftime('%Y-%m-%dT%H:%M:%S{:0=+3d}:{:0=2d}').format( ctz // 3600, ctz % 3600), 'VERSION': '2.8', 'FORMAT': '2.8', 'xmlns:xsi': 'http://www.w3.org/2001/XMLSchema-instance', 'xsi:noNamespaceSchemaLocation': 'http://www.mpi.nl/tools/elan/EAFv2.8.xsd'} self.annotations = {} self.constraints = {} self.controlled_vocabularies = {} self.external_refs = {} self.header = {} self.languages = {} self.lexicon_refs = {} self.linguistic_types = {} self.locales = {} self.tiers = {} self.timeslots = {} self.licenses = [] self.linked_file_descriptors = [] self.media_descriptors = [] self.properties = [] if file_path is None: self.add_linguistic_type('default-lt') self.constraints = self.CONSTRAINTS.copy() self.properties.append(('lastUsedAnnotation', 0)) self.add_tier('default') else: parse_eaf(file_path, self) def add_annotation(self, id_tier, start, end, value='', svg_ref=None): """Add an annotation. :param str id_tier: Name of the tier. :param int start: Start time of the annotation. :param int end: End time of the annotation. :param str value: Value of the annotation. :param str svg_ref: Svg reference. :raises KeyError: If the tier is non existent. :raises ValueError: If one of the values is negative or start is bigger then end or if the tiers already contains ref annotations. """ if self.tiers[id_tier][1]: raise ValueError('Tier already contains ref annotations...') if start == end: raise ValueError('Annotation length is zero...') if start > end: raise ValueError('Annotation length is negative...') if start < 0: raise ValueError('Start is negative...') start_ts = self.generate_ts_id(start) end_ts = self.generate_ts_id(end) aid = self.generate_annotation_id() self.annotations[aid] = id_tier self.tiers[id_tier][0][aid] = (start_ts, end_ts, value, svg_ref) def add_controlled_vocabulary(self, cv_id, ext_ref=None): """Add a controlled vocabulary. This will initialize the controlled vocabulary without entries. :param str cv_id: Name of the controlled vocabulary. :param str ext_ref: External reference. """ self.controlled_vocabularies[cv_id] = ([], {}, ext_ref) def add_cv_entry(self, cv_id, cve_id, values, ext_ref=None): """Add an entry to a controlled vocabulary. :param str cv_id: Name of the controlled vocabulary to add an entry. :param str cve_id: Name of the entry. :param list values: List of values of the form: ``(value, lang_ref, description)`` where description can be ``None``. :param str ext_ref: External reference. :throws KeyError: If there is no controlled vocabulary with that id. :throws ValueError: If a language in one of the entries doesn't exist. """ for value, lang_ref, description in values: if lang_ref not in self.languages: raise ValueError('Language not present: {}'.format(lang_ref)) self.controlled_vocabularies[cv_id][1][cve_id] = (values, ext_ref) def add_cv_description(self, cv_id, lang_ref, description=None): """Add a description to a controlled vocabulary. :param str cv_id: Name of the controlled vocabulary to add the description. :param str lang_ref: Language reference. :param str description: Description, this can be none. :throws KeyError: If there is no controlled vocabulary with that id. :throws ValueError: If the language provided doesn't exist. """ if lang_ref not in self.languages: raise ValueError('Language not present: {}'.format(lang_ref)) self.controlled_vocabularies[cv_id][0].append((lang_ref, description)) def add_external_ref(self, eid, etype, value): """Add an external reference. :param str eid: Name of the external reference. :param str etype: Type of the external