Source code for betty.gramps.loader

"""
Provide an API to load `Gramps <https://gramps-project.org>`_ family trees into Betty ancestries.
"""

from __future__ import annotations

import gzip
import re
import tarfile
from collections import defaultdict
from contextlib import suppress
from dataclasses import dataclass
from enum import Enum
from logging import getLogger
from pathlib import Path
from typing import Iterable, Any, IO, cast, TYPE_CHECKING
from xml.etree import ElementTree

import aiofiles
from aiofiles.tempfile import TemporaryDirectory
from geopy import Point
from typing_extensions import override

from betty.gramps.error import GrampsError
from betty.locale import DateRange, Datey, Date, Localizer
from betty.locale.localizable import _, plain
from betty.media_type import MediaType, InvalidMediaType
from betty.model import Entity, EntityGraphBuilder, AliasedEntity, AliasableEntity
from betty.model.ancestry import (
    Note,
    File,
    Source,
    Citation,
    Place,
    Event,
    Person,
    PersonName,
    Subject,
    Witness,
    Beneficiary,
    Attendee,
    Presence,
    PlaceName,
    Enclosure,
    HasLinks,
    Link,
    HasFiles,
    HasCitations,
    HasPrivacy,
    Speaker,
    Celebrant,
    Organizer,
    HasNotes,
)
from betty.model.event_type import (
    Birth,
    Baptism,
    Adoption,
    Cremation,
    Death,
    Funeral,
    Burial,
    Will,
    Engagement,
    Marriage,
    MarriageAnnouncement,
    Divorce,
    DivorceAnnouncement,
    Residence,
    Immigration,
    Emigration,
    Occupation,
    Retirement,
    Correspondence,
    Confirmation,
    Missing,
    UnknownEventType,
    EventType,
    Conference,
)
from betty.path import rootname

if TYPE_CHECKING:
    from betty.project import Project
    from collections.abc import MutableMapping, Mapping


[docs] class GrampsLoadFileError(GrampsError, RuntimeError): """ An error occurred when loading a Gramps family tree file. """ pass # pragma: no cover
[docs] class GrampsFileNotFoundError(GrampsError, FileNotFoundError): """ A Gramps family tree file could not be file. """ pass # pragma: no cover
[docs] class XPathError(GrampsError, RuntimeError): """ An error occurred when evaluating an XPath selector on Gramps XML. """ pass # pragma: no cover
[docs] class GrampsEntityType(Enum): """ The supported Gramps entity types. """ CITATION = "citation" EVENT = "event" OBJECT = "object" PERSON = "person" SOURCE = "source"
[docs] @dataclass(frozen=True) class GrampsEntityReference: """ A reference to an entity in a Gramps family tree. """ entity_type: GrampsEntityType entity_id: str @override def __str__(self) -> str: return f"{self.entity_type.value} ({self.entity_id})"
[docs] class GrampsLoader: """ Load Gramps family history data into a project. """
[docs] def __init__( self, project: Project, *, localizer: Localizer, ): super().__init__() self._project = project self._ancestry_builder = EntityGraphBuilder() self._added_entity_counts: dict[type[Entity], int] = defaultdict(lambda: 0) self._tree: ElementTree.ElementTree | None = None self._gramps_tree_directory_path: Path | None = None self._loaded = False self._localizer = localizer
[docs] async def load_file(self, file_path: Path) -> None: """ Load family history data from any of the supported Gramps file types. """ file_path = file_path.resolve() logger = getLogger(__name__) logger.info( self._localizer._('Loading "{file_path}"...').format( file_path=str(file_path), ) ) with suppress(GrampsLoadFileError): await self.load_gpkg(file_path) return with suppress(GrampsLoadFileError): await self.load_gramps(file_path) return try: async with aiofiles.open(file_path) as f: xml = await f.read() except FileNotFoundError: raise GrampsFileNotFoundError( _('Could not find the file "{file_path}".').format( file_path=str(file_path) ) ) from None with suppress(GrampsLoadFileError): await self.load_xml(xml, Path(file_path.anchor)) return raise GrampsLoadFileError( _( 'Could not load "{file_path}" as a *.gpkg, a *.gramps, or an *.xml family tree.' ).format(file_path=str(file_path)) )
[docs] async def load_gramps(self, gramps_path: Path) -> None: """ Load family history data from a Gramps *.gramps file. """ gramps_path = gramps_path.resolve() try: with gzip.open(gramps_path, mode="r") as f: xml: str = f.read() # type: ignore[assignment] await self.load_xml( xml, rootname(gramps_path), ) except OSError as error: raise GrampsLoadFileError(plain(error)) from error
[docs] async def load_gpkg(self, gpkg_path: Path) -> None: """ Load family history data from a Gramps *.gpkg file. """ gpkg_path = gpkg_path.resolve() try: tar_file: IO[bytes] = gzip.open(gpkg_path) # type: ignore[assignment] try: async with TemporaryDirectory() as cache_directory_path_str: tarfile.open( fileobj=tar_file, ).extractall(cache_directory_path_str, filter="data") await self.load_gramps( Path(cache_directory_path_str) / "data.gramps" ) except tarfile.ReadError as error: raise GrampsLoadFileError( _( "Could not extract {file_path} as a tar (*.tar) file after extracting the outer gzip (*.gz) file." ).format(file_path=str(gpkg_path)) ) from error except OSError as error: raise GrampsLoadFileError( _("Could not extract {file_path} as a gzip (*.gz) file.").format( file_path=str(gpkg_path) ) ) from error
[docs] async def load_xml(self, xml: str | Path, gramps_tree_directory_path: Path) -> None: """ Load family history data from XML. :param xml: The raw XML or the path to an XML file. """ if isinstance(xml, Path): async with aiofiles.open(xml) as f: xml = await f.read() try: tree = ElementTree.ElementTree( ElementTree.fromstring( xml, ) ) except ElementTree.ParseError as error: raise GrampsLoadFileError(plain(error)) from error await self.load_tree(tree, gramps_tree_directory_path)
[docs] async def load_tree( self, tree: ElementTree.ElementTree, gramps_tree_directory_path: Path ) -> None: """ Load family history data from a Gramps XML tree. """ if self._loaded: raise RuntimeError("This loader has been used up.") self._loaded = True self._tree = tree self._gramps_tree_directory_path = gramps_tree_directory_path.resolve() logger = getLogger(__name__) database = self._tree.getroot() self._load_notes(database) logger.info( self._localizer._("Loaded {note_count} notes.").format( note_count=self._added_entity_counts[Note] ) ) self._load_objects(database, self._gramps_tree_directory_path) logger.info( self._localizer._("Loaded {file_count} files.").format( file_count=self._added_entity_counts[File] ) ) self._load_repositories(database) repository_count = self._added_entity_counts[Source] logger.info( self._localizer._( "Loaded {repository_count} repositories as sources." ).format(repository_count=repository_count) ) self._load_sources(database) logger.info( self._localizer._("Loaded {source_count} sources.").format( source_count=self._added_entity_counts[Source] - repository_count ) ) self._load_citations(database) logger.info( self._localizer._("Loaded {citation_count} citations.").format( citation_count=self._added_entity_counts[Citation] ) ) self._load_places(database) logger.info( self._localizer._("Loaded {place_count} places.").format( place_count=self._added_entity_counts[Place] ) ) self._load_events(database) logger.info( self._localizer._("Loaded {event_count} events.").format( event_count=self._added_entity_counts[Event] ) ) self._load_people(database) logger.info( self._localizer._("Loaded {person_count} people.").format( person_count=self._added_entity_counts[Person] ) ) self._load_families(database) self._project.ancestry.add_unchecked_graph(*self._ancestry_builder.build())
[docs] def add_entity(self, entity: AliasableEntity[Entity]) -> None: """ Add entities to the ancestry. """ self._ancestry_builder.add_entity(entity) self._added_entity_counts[entity.type] += 1
[docs] def add_association(self, *args: Any, **kwargs: Any) -> None: """ Add an association between two entities to the ancestry. """ self._ancestry_builder.add_association(*args, **kwargs)
_NS = { "ns": "http://gramps-project.org/xml/1.7.1/", } def _xpath( self, element: ElementTree.Element, selector: str ) -> list[ElementTree.Element]: return element.findall(selector, namespaces=self._NS) def _xpath1( self, element: ElementTree.Element, selector: str ) -> ElementTree.Element: found_element = element.find(selector, namespaces=self._NS) if found_element is None: raise XPathError( plain(f'Cannot find an element "{selector}" within {str(element)}.') ) return found_element _DATE_PATTERN = re.compile(r"^.{4}((-.{2})?-.{2})?$") _DATE_PART_PATTERN = re.compile(r"^\d+$") def _load_date(self, element: ElementTree.Element) -> Datey | None: with suppress(XPathError): dateval_element = self._xpath1(element, "./ns:dateval") if dateval_element.get("cformat") is None: dateval_type = dateval_element.get("type") if dateval_type is None: return self._load_dateval(dateval_element, "val") dateval_type = str(dateval_type) if dateval_type == "about": date = self._load_dateval(dateval_element, "val") if date is None: return None date.fuzzy = True return date if dateval_type == "before": return DateRange( None, self._load_dateval(dateval_element, "val"), end_is_boundary=True, ) if dateval_type == "after": return DateRange( self._load_dateval(dateval_element, "val"), start_is_boundary=True, ) with suppress(XPathError): datespan_element = self._xpath1(element, "./ns:datespan") if datespan_element.get("cformat") is None: return DateRange( self._load_dateval(datespan_element, "start"), self._load_dateval(datespan_element, "stop"), ) with suppress(XPathError): daterange_element = self._xpath1(element, "./ns:daterange") if daterange_element.get("cformat") is None: return DateRange( self._load_dateval(daterange_element, "start"), self._load_dateval(daterange_element, "stop"), start_is_boundary=True, end_is_boundary=True, ) return None def _load_dateval( self, element: ElementTree.Element, value_attribute_name: str ) -> Date | None: dateval = str(element.get(value_attribute_name)) if self._DATE_PATTERN.fullmatch(dateval): date_parts: tuple[int | None, int | None, int | None] = tuple( # type: ignore[assignment] ( int(part) if self._DATE_PART_PATTERN.fullmatch(part) and int(part) > 0 else None ) for part in dateval.split("-") ) date = Date(*date_parts) dateval_quality = element.get("quality") if dateval_quality == "estimated": date.fuzzy = True return date return None def _load_notes(self, database: ElementTree.Element) -> None: for element in self._xpath(database, "./ns:notes/ns:note"): self._load_note(element) def _load_note(self, element: ElementTree.Element) -> None: note_handle = element.get("handle") note_id = element.get("id") assert note_id is not None text_element = self._xpath1(element, "./ns:text") assert text_element is not None text = str(text_element.text) note = Note( id=note_id, text=text, ) if element.get("priv") == "1": note.private = True self.add_entity(AliasedEntity(note, note_handle)) def _load_noteref( self, owner: AliasableEntity[HasNotes & Entity], element: ElementTree.Element ) -> None: note_handles = self._load_handles("noteref", element) for note_handle in note_handles: self.add_association(owner.type, owner.id, "notes", Note, note_handle) def _load_objects( self, database: ElementTree.Element, gramps_tree_directory_path: Path ) -> None: for element in self._xpath(database, "./ns:objects/ns:object"): self._load_object(element, gramps_tree_directory_path) def _load_object( self, element: ElementTree.Element, gramps_tree_directory_path: Path ) -> None: file_handle = element.get("handle") file_id = element.get("id") file_element = self._xpath1(element, "./ns:file") src = file_element.get("src") assert src is not None file_path = gramps_tree_directory_path / src file = File( id=file_id, path=file_path, ) mime = file_element.get("mime") assert mime is not None file.media_type = MediaType(mime) description = file_element.get("description") if description: file.description = description if element.get("priv") == "1": file.private = True aliased_file = AliasedEntity(file, file_handle) self._load_attributes_for( file, GrampsEntityReference(GrampsEntityType.OBJECT, file.id), element, "attribute", ) self.add_entity( aliased_file, # type: ignore[arg-type] ) for citation_handle in self._load_handles("citationref", element): self.add_association( File, file_handle, "citations", Citation, citation_handle ) self._load_noteref( aliased_file, # type: ignore[arg-type] element, ) def _load_people(self, database: ElementTree.Element) -> None: for element in self._xpath(database, "./ns:people/ns:person"): self._load_person(element) def _load_person(self, element: ElementTree.Element) -> None: person_handle = element.get("handle") assert person_handle is not None person = Person(id=element.get("id")) name_elements = sorted( self._xpath(element, "./ns:name"), key=lambda x: x.get("alt") == "1" ) person_names = [] for name_element in name_elements: is_alternative = name_element.get("alt") == "1" try: individual_name = self._xpath1(name_element, "./ns:first").text except XPathError: individual_name = None surname_elements = [ surname_element for surname_element in self._xpath(name_element, "./ns:surname") if surname_element.text is not None ] if surname_elements: for surname_element in surname_elements: if not is_alternative: is_alternative = surname_element.get("prim") == "0" affiliation_name = surname_element.text surname_prefix = surname_element.get("prefix") if surname_prefix is not None: affiliation_name = "%s %s" % (surname_prefix, affiliation_name) person_name = PersonName( individual=individual_name, affiliation=affiliation_name, ) self._load_citationref(person_name, name_element) person_names.append((person_name, is_alternative)) elif individual_name is not None: person_name = PersonName(individual=individual_name) self._load_citationref(person_name, name_element) person_names.append((person_name, is_alternative)) for person_name, __ in sorted(person_names, key=lambda x: x[1]): self.add_entity(person_name) self.add_association( Person, person_handle, "names", PersonName, person_name.id ) self._load_eventrefs(person_handle, element) if element.get("priv") == "1": person.private = True self._load_attributes_for( person, GrampsEntityReference(GrampsEntityType.PERSON, person.id), element, "attribute", ) aliased_person = AliasedEntity(person, person_handle) self._load_citationref( aliased_person, # type: ignore[arg-type] element, ) self._load_objref( aliased_person, # type: ignore[arg-type] element, ) self._load_noteref( aliased_person, # type: ignore[arg-type] element, ) self._load_urls(person, element) self.add_entity( aliased_person, # type: ignore[arg-type] ) def _load_families(self, database: ElementTree.Element) -> None: for element in self._xpath(database, "./ns:families/ns:family"): self._load_family(element) def _load_family(self, element: ElementTree.Element) -> None: parent_handles = [] # Load the father. father_handle = self._load_handle("father", element) if father_handle is not None: self._load_eventrefs(father_handle, element) parent_handles.append(father_handle) # Load the mother. mother_handle = self._load_handle("mother", element) if mother_handle is not None: self._load_eventrefs(mother_handle, element) parent_handles.append(mother_handle) # Load the children. child_handles = self._load_handles("childref", element) for child_handle in child_handles: for parent_handle in parent_handles: self.add_association( Person, child_handle, "parents", Person, parent_handle ) def _load_eventrefs(self, person_id: str, element: ElementTree.Element) -> None: eventrefs = self._xpath(element, "./ns:eventref") for eventref in eventrefs: self._load_eventref(person_id, eventref) _PRESENCE_ROLE_MAP = { "Primary": Subject(), "Family": Subject(), "Witness": Witness(), "Beneficiary": Beneficiary(), "Speaker": Speaker(), "Celebrant": Celebrant(), "Organizer": Organizer(), "Attendee": Attendee(), "Unknown": Attendee(), } def _load_eventref(self, person_id: str, eventref: ElementTree.Element) -> None: event_handle = eventref.get("hlink") assert event_handle is not None gramps_presence_role = cast(str, eventref.get("role")) try: role = self._PRESENCE_ROLE_MAP[gramps_presence_role] except KeyError: role = Attendee() getLogger(__name__).warning( self._localizer._( 'Betty is unfamiliar with person "{person_id}"\'s Gramps presence role of "{gramps_presence_role}" for the event with Gramps handle "{event_handle}". The role was imported, but set to "{betty_presence_role}".', ).format( person_id=person_id, event_handle=event_handle, gramps_presence_role=gramps_presence_role, betty_presence_role=role.label, ) ) presence = Presence(None, role, None) if eventref.get("priv") == "1": presence.private = True self._load_attributes_for( presence, GrampsEntityReference(GrampsEntityType.PERSON, person_id), eventref, "attribute", ) self.add_entity(presence) self.add_association(Presence, presence.id, "person", Person, person_id) self.add_association(Presence, presence.id, "event", Event, event_handle) def _load_places(self, database: ElementTree.Element) -> None: for element in self._xpath(database, "./ns:places/ns:placeobj"): self._load_place(element) def _load_place(self, element: ElementTree.Element) -> None: place_handle = element.get("handle") names = [] for name_element in self._xpath(element, "./ns:pname"): # The Gramps language is a single ISO language code, which is a valid BCP 47 locale. language = name_element.get("lang") date = self._load_date(name_element) name = name_element.get("value") assert name is not None names.append( PlaceName( name=name, locale=language, date=date, ) ) place = Place( id=element.get("id"), names=names, ) coordinates = self._load_coordinates(element) if coordinates: place.coordinates = coordinates self._load_urls(place, element) aliased_place = AliasedEntity(place, place_handle) self._load_noteref( aliased_place, # type: ignore[arg-type] element, ) self.add_entity( aliased_place, # type: ignore[arg-type] ) for enclosed_by_handle in self._load_handles("placeref", element): aliased_enclosure = AliasedEntity( Enclosure(encloses=None, enclosed_by=None) ) self.add_entity( aliased_enclosure, # type: ignore[arg-type] ) self.add_association( Enclosure, aliased_enclosure.id, "encloses", Place, place_handle ) self.add_association( Enclosure, aliased_enclosure.id, "enclosed_by", Place, enclosed_by_handle, ) def _load_coordinates(self, element: ElementTree.Element) -> Point | None: with suppress(XPathError): coord_element = self._xpath1(element, "./ns:coord") coordinates = f'{coord_element.get("lat")}; {coord_element.get("long")}' try: return Point.from_string(coordinates) except ValueError: getLogger(__name__).warning( self._localizer._( 'Cannot load coordinates "{coordinates}", because they are in an unknown format.', ).format( coordinates=coordinates, ) ) return None def _load_events(self, database: ElementTree.Element) -> None: for element in self._xpath(database, "./ns:events/ns:event"): self._load_event(element) _EVENT_TYPE_MAP = { "Birth": Birth, "Baptism": Baptism, "Adopted": Adoption, "Cremation": Cremation, "Death": Death, "Funeral": Funeral, "Burial": Burial, "Will": Will, "Engagement": Engagement, "Marriage": Marriage, "Marriage Banns": MarriageAnnouncement, "Divorce": Divorce, "Divorce Filing": DivorceAnnouncement, "Residence": Residence, "Immigration": Immigration, "Emigration": Emigration, "Occupation": Occupation, "Retirement": Retirement, "Correspondence": Correspondence, "Confirmation": Confirmation, "Missing": Missing, "Conference": Conference, } def _load_event(self, element: ElementTree.Element) -> None: event_handle = element.get("handle") event_id = element.get("id") assert event_id is not None gramps_type = self._xpath1(element, "./ns:type").text assert gramps_type is not None try: event_type: type[EventType] = self._EVENT_TYPE_MAP[gramps_type] except KeyError: event_type = UnknownEventType getLogger(__name__).warning( self._localizer._( 'Betty is unfamiliar with Gramps event "{event_id}"\'s type of "{gramps_event_type}". The event was imported, but its type was set to "{betty_event_type}".', ).format( event_id=event_id, gramps_event_type=gramps_type, betty_event_type=event_type.label().localize(self._localizer), ) ) event = Event( id=event_id, event_type=event_type, ) event.date = self._load_date(element) # Load the event place. place_handle = self._load_handle("place", element) if place_handle is not None: self.add_association(Event, event_handle, "place", Place, place_handle) # Load the description. with suppress(XPathError): event.description = self._xpath1(element, "./ns:description").text if element.get("priv") == "1": event.private = True aliased_event = AliasedEntity(event, event_handle) self._load_objref( aliased_event, # type: ignore[arg-type] element, ) self._load_citationref( aliased_event, # type: ignore[arg-type] element, ) self._load_noteref( aliased_event, # type: ignore[arg-type] element, ) self._load_attributes_for( event, GrampsEntityReference(GrampsEntityType.EVENT, event.id), element, "attribute", ) self.add_entity( aliased_event, # type: ignore[arg-type] ) def _load_repositories(self, database: ElementTree.Element) -> None: for element in self._xpath(database, "./ns:repositories/ns:repository"): self._load_repository(element) def _load_repository(self, element: ElementTree.Element) -> None: repository_source_handle = element.get("handle") source = Source( id=element.get("id"), name=self._xpath1(element, "./ns:rname").text, ) self._load_urls(source, element) aliased_source = AliasedEntity(source, repository_source_handle) self._load_noteref( aliased_source, # type: ignore[arg-type] element, ) self.add_entity( aliased_source, # type: ignore[arg-type] ) def _load_sources(self, database: ElementTree.Element) -> None: for element in self._xpath(database, "./ns:sources/ns:source"): self._load_source(element) def _load_source(self, element: ElementTree.Element) -> None: source_handle = element.get("handle") try: source_name = self._xpath1(element, "./ns:stitle").text except XPathError: source_name = None source = Source( id=element.get("id"), name=source_name, ) repository_source_handle = self._load_handle("reporef", element) if repository_source_handle is not None: self.add_association( Source, source_handle, "contained_by", Source, repository_source_handle ) # Load the author. with suppress(XPathError): source.author = self._xpath1(element, "./ns:sauthor").text # Load the publication info. with suppress(XPathError): source.publisher = self._xpath1(element, "./ns:spubinfo").text if element.get("priv") == "1": source.private = True self._load_attributes_for( source, GrampsEntityReference(GrampsEntityType.SOURCE, source.id), element, "srcattribute", ) aliased_source = AliasedEntity(source, source_handle) self._load_objref( aliased_source, # type: ignore[arg-type] element, ) self._load_noteref( aliased_source, # type: ignore[arg-type] element, ) self.add_entity( aliased_source, # type: ignore[arg-type] ) def _load_citations(self, database: ElementTree.Element) -> None: for element in self._xpath(database, "./ns:citations/ns:citation"): self._load_citation(element) def _load_citation(self, element: ElementTree.Element) -> None: citation_handle = element.get("handle") source_handle = self._xpath1(element, "./ns:sourceref").get("hlink") citation = Citation(id=element.get("id")) self.add_association(Citation, citation_handle, "source", Source, source_handle) citation.date = self._load_date(element) if element.get("priv") == "1": citation.private = True with suppress(XPathError): citation.location = plain(self._xpath1(element, "./ns:page").text) aliased_citation = AliasedEntity(citation, citation_handle) self._load_objref( aliased_citation, # type: ignore[arg-type] element, ) self._load_attributes_for( citation, GrampsEntityReference(GrampsEntityType.CITATION, citation.id), element, "srcattribute", ) self.add_entity( aliased_citation, # type: ignore[arg-type] ) def _load_citationref( self, owner: AliasableEntity[HasCitations & Entity], element: ElementTree.Element, ) -> None: for citation_handle in self._load_handles("citationref", element): self.add_association( owner.type, owner.id, "citations", Citation, citation_handle ) def _load_handles( self, handle_type: str, element: ElementTree.Element ) -> Iterable[str]: for citation_handle_element in self._xpath(element, f"./ns:{handle_type}"): hlink = citation_handle_element.get("hlink") if hlink: yield hlink def _load_handle( self, handle_type: str, element: ElementTree.Element ) -> str | None: for citation_handle_element in self._xpath(element, f"./ns:{handle_type}"): return citation_handle_element.get("hlink") return None def _load_objref( self, owner: AliasableEntity[HasFiles & Entity], element: ElementTree.Element ) -> None: file_handles = self._load_handles("objref", element) for file_handle in file_handles: self.add_association(owner.type, owner.id, "files", File, file_handle) def _load_urls(self, owner: HasLinks, element: ElementTree.Element) -> None: url_elements = self._xpath(element, "./ns:url") for url_element in url_elements: link = Link(str(url_element.get("href"))) link.relationship = "external" link.label = url_element.get("description") owner.links.append(link) def _load_attribute_privacy( self, entity: HasPrivacy & Entity, element: ElementTree.Element, tag: str ) -> None: privacy_value = self._load_attribute("privacy", element, tag) if privacy_value is None: return if privacy_value == "private": entity.private = True return if privacy_value == "public": entity.public = True return getLogger(__name__).warning( self._localizer._( 'The betty:privacy Gramps attribute must have a value of "public" or "private", but "{privacy_value}" was given for {entity_type} {entity_id} ({entity_label}), which was ignored.', ).format( privacy_value=privacy_value, entity_type=entity.entity_type_label().localize(self._localizer), entity_id=entity.id, entity_label=entity.label.localize(self._localizer), ) ) _LINK_ATTRIBUTE_PATTERN = re.compile(r"^link-([^:]+?):(.+?)$") def _load_attribute_links( self, entity: HasLinks & Entity, gramps_entity_reference: GrampsEntityReference, element: ElementTree.Element, tag: str, ) -> None: logger = getLogger(__name__) attributes = self._load_attributes(element, tag) links_attributes: MutableMapping[str, MutableMapping[str, str]] = defaultdict( dict ) for attribute_type, attribute_value in attributes.items(): match = self._LINK_ATTRIBUTE_PATTERN.fullmatch(attribute_type) if match is None: continue link_name = match.group(1) link_attribute_name = match.group(2) links_attributes[link_name][link_attribute_name] = attribute_value for link_name, link_attributes in links_attributes.items(): if "url" not in link_attributes: logger.warning( self._localizer._( 'The Gramps {gramps_entity_reference} entity requires a "betty:link-{link_name}:url" attribute. This link was ignored.', ).format( gramps_entity_reference=gramps_entity_reference, link_name=link_name, ) ) continue link = Link(link_attributes["url"]) entity.links.append(link) if "description" in link_attributes: link.description = link_attributes["description"] if "label" in link_attributes: link.label = link_attributes["label"] if "locale" in link_attributes: link.locale = link_attributes["locale"] if "media_type" in link_attributes: try: media_type = MediaType(link_attributes["media_type"]) except InvalidMediaType: logger.warning( self._localizer._( 'The Gramps {gramps_entity_reference} entity has a "betty:link-{link_name}:media_type" attribute with value "{media_type}", which is not a valid IANA media type. This media type was ignored.', ).format( gramps_entity_reference=gramps_entity_reference, link_name=link_name, media_type=link_attributes["media_type"], ) ) else: link.media_type = media_type if "relationship" in link_attributes: link.relationship = link_attributes["relationship"] def _load_attribute( self, name: str, element: ElementTree.Element, tag: str ) -> str | None: try: return self._load_attributes(element, tag)[name] except KeyError: return None def _load_attributes( self, element: ElementTree.Element, tag: str ) -> Mapping[str, str]: prefixes = ["betty"] hash(element) if self._project.configuration.name is not None: prefixes.append(f"betty-{self._project.configuration.name}") attributes: MutableMapping[str, str] = {} for prefix in prefixes: with suppress(XPathError): attribute_elements = self._xpath(element, f"./ns:{tag}") for attribute_element in attribute_elements: attribute_type = attribute_element.attrib["type"] attribute_value = attribute_element.get("value") if ( attribute_type.startswith(f"{prefix}:") and attribute_value is not None ): attributes[attribute_type[len(prefix) + 1 :]] = attribute_value return attributes def _load_attributes_for( self, entity: Entity, gramps_entity_reference: GrampsEntityReference, element: ElementTree.Element, tag: str, ) -> None: if isinstance(entity, HasPrivacy): self._load_attribute_privacy(entity, element, tag) if isinstance(entity, HasLinks): self._load_attribute_links(entity, gramps_entity_reference, element, tag)