def format_person_names(self, person_names, reverse=False):
"""
From each tei:name/tei:persName/... in person_names, extracts a full name of the form
'surname, forename [name link] [& additional name]' or (if reverse=True) 'forename [name link] surname [<additional name>]'.
However, if there is an @key, this will be used as the full name.
"""
names = []
for person in person_names:
name = ''
if exists(person, '@key'):
name = person.xpath('@key')[0]
elif exists(person, 'tei:surname and tei:forename'):
surname = person.xpath('tei:surname/text()', namespaces=xml_ns)[0]
forename = person.xpath('tei:forename/text()', namespaces=xml_ns)[0]
name_link = ''
if exists(person, 'tei:nameLink'):
name_link += ' ' + person.xpath('tei:nameLink/text()', namespaces=xml_ns)[0]
add_name = ''
if exists(person, 'tei:addName'):
if reverse:
add_name += ' <' + person.xpath('tei:addName/text()', namespaces=xml_ns)[0] + '>'
else:
add_name += ' & (' + person.xpath('tei:addName/text()', namespaces=xml_ns)[0] + ')'
if reverse:
name = forename + ' ' + surname + name_link + add_name
else:
name = surname + ', ' + forename + name_link + add_name
else:
name = str(person.xpath('string(.)')[0])
names.append(name)
return names
def get_publish_date_range(self, date: etree._Element):
if exists(date, '@from'):
date = {'start': date.xpath('@start')[0]}
if exists(date, '@to'):
date['to'] = date.xpath('@to')[0]
return date
else:
return date.xpath('text()')[0]
def transform_p(self, node, mode):
text = self.passthru(node, mode)
if exists(node, 'ancestor::tei:note'):
if exists(node, 'following-sibling::tei:p'):
return text + '\n'
else:
return text
else:
return '\n' + text + '\n'
def make_title(self, node: etree._Element) -> str:
if exists('self::tei:charDecl'):
return str(
node.xpath(
'ancestor::tei:teiheader/tei:fileDesc/tei:titleStmt/tei:title[@xml:lang = "en"]',
namespaces=xml_ns)[0]) # TODO i18n
elif exists('self::tei:char/tei:desc'):
return str(node.xpath('tei:desc/text()', namespaces=xml_ns)[0])
else:
return None
def is_passagetrail_node(self, node):
"""
Determines if a node constitutes a 'passagetrail' part.
Note: assumes that get_elem_type(node) == True.
"""
name = etree.QName(node).localname
return bool(exists(node, 'self::tei:text[@type = "work_volume"]') \
or (exists(node, 'self::tei:div') and citation_labels[node.get('type')].get('isCiteRef')) \
or (exists(node, 'self::tei:milestone') and citation_labels[node.get('unit')].get('isCiteRef')) \
or exists(node, 'self::tei:pb[not(@sameAs or @corresp)]') \
or (citation_labels.get(name) and citation_labels.get(name).get('isCiteRef')))
def transform_item(
self, node,
mode): # TODO test this, esp. with more complicated/nested lists
text = ''
if self.analysis.is_basic_node(
node) or self.analysis.has_basic_ancestor(node):
text = self.passthru(node, mode)
leading = '- '
if exists(node, 'parent::tei:list/@type = "numbered"'):
leading = '# '
elif exists(node, 'parent::tei:list/@type = "simple"'):
leading = ' '
return leading + text + '\n'
def get_citable_ancestors(self, node: etree._Element, node_type: str, mode: str):
"""
Gets all citetrail or passagetrail ancestors of a node (switch modes: 'citetrail' vs 'passagetrail').
"""
tei_ancestors = node.xpath('ancestor::*')
ancestors = []
if node_type == 'marginal' or node_type == 'anchor':
# marginals and anchors must not have p (or some other "main" node) as their parent
for anc in tei_ancestors:
if (mode == 'citetrail' or (mode == 'passagetrail' and self.is_passagetrail_node(anc))) \
and self.is_structural_node(anc):
ancestors.append(anc)
elif node_type == 'page':
# within front, back, and single volumes, citable parent resolves to one of those elements for avoiding
# collisions with identically named pb in other parts
for anc in tei_ancestors:
if (mode == 'citetrail' or (mode == 'passagetrail' and self.is_passagetrail_node(anc))) \
and exists(anc, 'self::tei:front or self::tei:back'
+ ' or self::tei:text[1][not(@xml:id = "completeWork" or @type = "work_part")]'):
ancestors.append(anc)
# note: this makes all other pb appear outside of any structural hierarchy, but this should be fine
else:
for anc in tei_ancestors:
if (mode == 'citetrail' or (mode == 'passagetrail' and self.is_passagetrail_node(anc))) \
and self.get_node_type(anc):
ancestors.append(anc)
return ancestors[::-1] # ancestors.reverse() is not working here
def transform_bibl(self, node, mode):
if mode == 'edit' and exists(node, '@sortKey'):
text = self.passthru(node, mode)
return text + ' [' + re.sub(
r'_', ', ',
node.get('sortKey')) + ']' # TODO revision of bibl/@sortKey
else:
return self.passthru(node, mode)
def transform_orig_elem(self, node, mode):
if mode == 'orig' or not exists(
node,
'parent::tei:choice/*[self::tei:expan or self::tei:corr or self::tei:reg]'
):
return self.passthru(node, mode)
else:
return ''
def get_citetrail_prefix(self, node: etree._Element, node_type: str):
"""
Citetrails for certain node types are always prefixed with a 'categorical' keyword/string.
"""
prefix = ''
name = etree.QName(node).localname
if node_type == 'page':
prefix = 'p'
elif node_type == 'marginal':
prefix = 'n'
elif node_type == 'anchor' and exists(node, 'self::tei:milestone[@unit]'):
prefix = node.get('unit')
elif node_type == 'structural':
if name == 'front':
prefix = 'frontmatter'
elif name == 'back':
prefix = 'backmatter'
elif exists(node, 'self::tei:text[@type = "work_volume"]'):
prefix = 'vol'
elif node_type == 'main':
if exists(node, 'self::tei:head'):
prefix = 'heading'
elif exists(node, 'self::tei:titlePage'):
prefix = 'titlepage'
elif node_type == 'list':
if exists(node, 'self::tei:list[@type = "dict" or @type = "index"]'):
prefix = node.get('type')
elif exists(node, 'self::tei:item[ancestor::tei:list[@type = "dict"]]'):
prefix = 'entry'
return prefix
def make_resource_metadata(self, tei_header: etree._Element, config, wid: str):
"""Translates data from the teiHeader of a work to DTS+DC metadata for a DTS textual Resource"""
# 1.) gather metadata
# a) digital edition
id = id_server + '/texts/' + config.get_wid()
# TODO @id shouldn't be the same as the @id of the parent collection, but sth more specific
title = tei_header.xpath('tei:fileDesc/tei:titleStmt/tei:title[@type = "short"]/text()', namespaces=xml_ns)[0]
alt_title = tei_header.xpath('tei:fileDesc/tei:titleStmt/tei:title[@type = "main"]/text()', namespaces=xml_ns)[0] # or short title here?
author = '; '.join(self.format_person_names(tei_header.xpath('tei:fileDesc/tei:titleStmt/tei:author/tei:persName',
namespaces=xml_ns), reverse=False))
scholarly_editors = self.format_person_names(tei_header.xpath('tei:fileDesc/tei:titleStmt/' +
'tei:editor[contains(@role, "#scholarly")]/tei:persName',
namespaces=xml_ns))
technical_editors = self.format_person_names(tei_header.xpath('tei:fileDesc/tei:titleStmt/' +
'tei:editor[contains(@role, "#technical")]/tei:persName',
namespaces=xml_ns))
editors = list(set(scholarly_editors + technical_editors))
pub_date = self.get_publish_date(tei_header)
version = tei_header.xpath('tei:fileDesc/tei:editionStmt/tei:edition/@n', namespaces=xml_ns)[0]
series_volume = tei_header.xpath('tei:fileDesc/tei:seriesStmt/tei:biblScope[@unit = "volume"]/@n',
namespaces=xml_ns)[0]
rights_holder = {
'@id': 'https://id.salamanca.school',
'name': {
'@language': 'en',
'@value': 'The School of Salamanca'
}
} # TODO provisional values
bibliographic_citation = self.make_bibliographic_citations(tei_header, wid)
# b) print source
source_title = tei_header.xpath('tei:fileDesc/tei:sourceDesc/tei:biblStruct/tei:monogr/tei:title[@type = "main"]/text()',
namespaces=xml_ns)[0]
source_publishers = self.format_person_names(self.get_source_publishers(tei_header), reverse=False)
source_extents = tei_header.xpath('tei:fileDesc/tei:sourceDesc/tei:biblStruct/tei:monogr/tei:extent',
namespaces=xml_ns)
source_extents_i18n = []
for extent in source_extents:
if exists(extent, '@xml:lang'):
extent_i18n = {
'@language': extent.xpath('@xml:lang', namespaces=xml_ns)[0],
'@value': extent.xpath('text()')[0]
}
source_extents_i18n.append(extent_i18n)
else:
source_extents_i18n.append(extent.xpath('text()')[0])
source_lang = []
for lang in tei_header.xpath('tei:profileDesc/tei:langUsage/tei:language/@ident', namespaces=xml_ns):
source_lang.append(lang)
source_pub_date = self.get_source_publish_date(tei_header)
source_pub_place = self.get_source_publish_place(tei_header)
source_repositories = self.get_source_repositories(tei_header)
# c) other dts metadata
total_items = 0 # TODO
dts_total_children = 0 # TODO
dts_total_parents = 1 # resource is part only of the parent collection that represents the work
dts_cite_depth = config.get_cite_depth() # TODO multivol vs singlevol?
# 2.) construct metadata object
resource_metadata = {
'@context': context,
'@id': id,
'@type': 'Resource',
'title': title,
'totalItems': total_items,
'dts:totalParents': dts_total_parents,
'dts:totalChildren': dts_total_children,
'dts:citeDepth': dts_cite_depth,
'dts:dublincore': {
'dc:title': title,
'dc:alternative': alt_title,
'dc:contributor': editors, # TODO editors as "contributors"? information can be found also in sal:...Editors
'dc:type': [
'http://purl.org/spar/fabio/work',
'dc:Text'
],
'dc:created': pub_date,
'dc:bibliographicCitation': bibliographic_citation,
'dc:rightsHolder': rights_holder,
'dc:license': 'http://creativecommons.org/licenses/by/4.0/',
'dc:source': {
'dc:title': source_title,
'dc:creator': author,
'dc:publisher': source_publishers,
'dc:format': source_extents_i18n,
'dc:language': source_lang,
'dc:created': source_pub_date
}
},
'dts:extensions': { # supplementary information that doesn't easily fit into dts/dc elements
'sal:version': version, # the version of this edition of the text
'sal:scholarlyEditors': scholarly_editors,
'sal:technicalEditors': technical_editors,
# omitting "#additional" editors here
'sal:seriesVolume': series_volume,
'sal:sourcePublishPlace': source_pub_place, # there seem to be no dc elements for this type of information...
'sal:sourceRepositories': source_repositories
}
}
return resource_metadata
def get_place_name(self, place_name: etree._Element):
if exists(place_name, '@key'):
return place_name.get('key')
else:
return place_name.text
def get_node_title(self, node):
name = etree.QName(node).localname
xml_id = node.xpath('@xml:id', namespaces=xml_ns)[0]
title = ''
if name == 'div':
if node.get('n') and not re.match(r'^[\d\[\]]+$', node.get('n')):
title = '"' + node.get('n') + '"'
elif exists(node, 'tei:head'):
title = self.make_node_teaser(node.xpath('tei:head[1]', namespaces=xml_ns)[0])
elif exists(node, 'tei:label'):
title = self.make_node_teaser(node.xpath('tei:label[1]', namespaces=xml_ns)[0])
elif node.get('n') and node.get('type'):
title = node.get('n')
elif exists(node, 'ancestor::tei:TEI//tei:text//tei:ref[@target = "#' + xml_id + '"]'):
title = self.make_node_teaser(node.xpath('ancestor::tei:TEI//tei:text//tei:ref[@target = "#' + xml_id
+ '"][1]')[0])
elif exists(node, 'tei:list/tei:head'):
title = self.make_node_teaser(node.xpath('tei:list/tei:head[1]', namespaces=xml_ns)[0])
elif exists(node, 'tei_list/tei:label'):
title = self.make_node_teaser(node.xpath('tei:list/tei:label[1]', namespaces=xml_ns)[0])
elif name == 'item':
#if exists(node, 'parent::tei:list[@type="dict"] and descendant::tei:term[1]/@key'):
# return '"' + node.xpath('descendant::tei:term[1]/@key')[0] + '"'
# # TODO this needs revision when we have really have such dict. lists
if node.get('n') and not re.match(r'^[\d\[\]]+$', node.get('n')):
title = '"' + node.get('n') + '"'
elif exists(node, 'tei:head'):
title = self.make_node_teaser(node.xpath('tei:head[1]', namespaces=xml_ns)[0])
elif exists(node, 'tei:label'):
title = self.make_node_teaser(node.xpath('tei:label[1]', namespaces=xml_ns)[0])
elif node.get('n'):
title = node.get('n')
elif exists(node, 'ancestor::tei:TEI//tei:text//tei:ref[@target = "#' + xml_id + '"]'):
title = self.make_node_teaser(
node.xpath('ancestor::tei:TEI//tei:text//tei:ref[@target = "#' + xml_id + '"][1]', namespaces=xml_ns)[0])
elif name == 'lg':
if exists(node, 'tei:head'):
title = self.make_node_teaser(node.xpath('tei:head[1]', namespaces=xml_ns)[0])
else:
title = self.make_node_teaser(node)
elif name == 'list':
if node.get('n') and not re.match(r'^[\d\[\]]+$', node.get('n')):
title = '"' + node.get('n') + '"'
elif exists(node, 'tei:head'):
title = self.make_node_teaser(node.xpath('tei:head[1]', namespaces=xml_ns)[0])
elif exists(node, 'tei:label'):
title = self.make_node_teaser(node.xpath('tei:label[1]', namespaces=xml_ns)[0])
elif node.get('n'):
title = node.get('n')
elif exists(node, 'ancestor::tei:TEI//tei:text//tei:ref[@target = "#'+ xml_id +'"]'):
title = self.make_node_teaser(node.xpath('ancestor::tei:TEI//tei:text//tei:ref[@target = "#'+ xml_id +'"][1]')[0])
elif name == 'milestone':
if node.get('n') and not re.match(r'^[\d\[\]]+$', node.get('n')):
title = '"' + node.get('n') + '"'
elif node.get('n'):
title = node.get('n')
elif exists(node, 'ancestor::tei:TEI//tei:text//tei:ref[@target = "#'+ xml_id +'"]'):
title = self.make_node_teaser(node.xpath('ancestor::tei:TEI//tei:text//tei:ref[@target = "#'+ xml_id +'"][1]')[0])
elif name == 'note':
if node.get('n'):
title = '"' + node.get('n') + '"'
elif name == 'pb':
if node.get('n') and re.match(r'fol\.', node.get('n')):
title = node.get('n')
else:
title = 'p. ' + node.get('n')
# one could also prepend a 'Vol. ' prefix here in case of a multivolume work
elif name == 'text':
if node.get('type') == 'work_volume':
title = node.get('n')
elif name == 'head' or name == 'label' or name == 'p' or name == 'signed' or name == 'titlePart':
title = self.make_node_teaser(node)
return title