def test_should_not_create_person_name_for_note(self):
label_and_layout_line_list_list = [[
('O', get_next_layout_line_for_text('before')),
('<forename>', get_next_layout_line_for_text('John')),
('<surname>', get_next_layout_line_for_text('Smith')),
('<marker>', get_next_layout_line_for_text('1')),
('O', get_next_layout_line_for_text(',')),
('<forename>', get_next_layout_line_for_text('Maria')),
('<surname>', get_next_layout_line_for_text('Madison')),
('<marker>', get_next_layout_line_for_text('2')),
('O', get_next_layout_line_for_text('after')),
]]
labeled_model_data_list_list = get_labeled_model_data_list_list(
label_and_layout_line_list_list,
data_generator=get_data_generator())
training_data_generator = get_tei_training_data_generator()
xml_root = training_data_generator.get_training_tei_xml_for_multiple_model_data_iterables(
labeled_model_data_list_list)
LOGGER.debug('xml: %r', etree.tostring(xml_root))
nodes = tei_xpath(xml_root, AUTHOR_XPATH)
assert len(nodes) == 2
assert get_tei_xpath_text_content_list(nodes[0],
'./tei:forename') == ['John']
assert get_tei_xpath_text_content_list(nodes[1],
'./tei:forename') == ['Maria']
def test_should_add_bold_and_italics_text(self):
block = LayoutBlock.for_tokens([
LayoutToken(TOKEN_1),
LayoutToken(TOKEN_2, font=BOLD_ITALICS_FONT_1),
LayoutToken(TOKEN_3)
])
node = TEI_E.node(*iter_layout_block_tei_children(block))
LOGGER.debug('xml: %r', etree.tostring(node))
assert get_tei_xpath_text_content_list(
node, './/tei:hi[@rend="bold"]') == [TOKEN_2]
assert get_tei_xpath_text_content_list(
node, './/tei:hi[@rend="italic"]') == [TOKEN_2]
assert get_text_content(node) == ' '.join([TOKEN_1, TOKEN_2, TOKEN_3])
def test_should_generate_tei_for_most_labels(self):
label_and_layout_line_list = [
('<figure_head>', get_next_layout_line_for_text('Figure Head 1')),
('<figDesc>', get_next_layout_line_for_text('Figure Desc 1')),
]
labeled_model_data_list = get_labeled_model_data_list(
label_and_layout_line_list, data_generator=get_data_generator())
xml_root = get_training_tei_xml_for_model_data_iterable(
labeled_model_data_list)
assert get_tei_xpath_text_content_list(
xml_root, f'{FIGURE_XPATH}/head') == ['Figure Head 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{FIGURE_XPATH}/figDesc') == ['Figure Desc 1']
def test_should_add_label_to_head_element_without_additional_text(self):
label_and_layout_line_list = [
('<label>', get_next_layout_line_for_text('Figure Label 1')),
('<figDesc>', get_next_layout_line_for_text('Figure Desc 1'))
]
labeled_model_data_list = get_labeled_model_data_list(
label_and_layout_line_list, data_generator=get_data_generator())
xml_root = get_training_tei_xml_for_model_data_iterable(
labeled_model_data_list)
assert get_tei_xpath_text_content_list(
xml_root, f'{FIGURE_XPATH}/head/label') == ['Figure Label 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{FIGURE_XPATH}/head') == ['Figure Label 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{FIGURE_XPATH}/figDesc') == ['Figure Desc 1']
def test_should_add_label_at_the_end_inside_head_element(self):
label_and_layout_line_list = [
('<figure_head>', get_next_layout_line_for_text('Figure Head 1')),
('<label>', get_next_layout_line_for_text('Figure Label 1'))
]
labeled_model_data_list = get_labeled_model_data_list(
label_and_layout_line_list, data_generator=get_data_generator())
xml_root = get_training_tei_xml_for_model_data_iterable(
labeled_model_data_list)
assert get_tei_xpath_text_content_list(
xml_root, f'{FIGURE_XPATH}/head/label') == ['Figure Label 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{FIGURE_XPATH}/head') == [
'\n'.join(['Figure Head 1', 'Figure Label 1'])
]
def test_should_generate_tei_from_multiple_model_data_lists_using_model_labels(
self):
label_and_layout_line_list_list = [[
('<figure_head>', get_next_layout_line_for_text(TEXT_1))
], [('<figure_head>', get_next_layout_line_for_text(TEXT_2))]]
labeled_model_data_list_list = get_labeled_model_data_list_list(
label_and_layout_line_list_list,
data_generator=get_data_generator())
training_data_generator = get_tei_training_data_generator()
xml_root = training_data_generator.get_training_tei_xml_for_multiple_model_data_iterables(
labeled_model_data_list_list)
LOGGER.debug('xml: %r', etree.tostring(xml_root))
nodes = tei_xpath(xml_root, FIGURE_XPATH)
assert len(nodes) == 2
assert get_tei_xpath_text_content_list(nodes[0], './head') == [TEXT_1]
assert get_tei_xpath_text_content_list(nodes[1], './head') == [TEXT_2]
def test_should_generate_tei_from_multiple_model_data_lists_using_model_labels(
self):
label_and_layout_line_list_list = [[
('<institution>', get_next_layout_line_for_text(TEXT_1))
], [('<institution>', get_next_layout_line_for_text(TEXT_2))]]
labeled_model_data_list_list = get_labeled_model_data_list_list(
label_and_layout_line_list_list,
data_generator=get_data_generator())
training_data_generator = AffiliationAddressTeiTrainingDataGenerator()
xml_root = training_data_generator.get_training_tei_xml_for_multiple_model_data_iterables(
labeled_model_data_list_list)
LOGGER.debug('xml: %r', etree.tostring(xml_root))
aff_nodes = tei_xpath(xml_root, AFFILIATION_XPATH)
assert len(aff_nodes) == 2
assert get_tei_xpath_text_content_list(
aff_nodes[0], './tei:orgName[@type="institution"]') == [TEXT_1]
assert get_tei_xpath_text_content_list(
aff_nodes[1], './tei:orgName[@type="institution"]') == [TEXT_2]
def test_should_map_unknown_label_to_note(self):
label_and_layout_line_list = [('<unknown>',
get_next_layout_line_for_text(TEXT_1))]
labeled_model_data_list = get_labeled_model_data_list(
label_and_layout_line_list, data_generator=get_data_generator())
xml_root = get_training_tei_xml_for_model_data_iterable(
labeled_model_data_list)
assert get_tei_xpath_text_content_list(
xml_root, f'{AUTHOR_XPATH}/tei:note[@type="unknown"]') == [TEXT_1]
def test_should_map_pubnum_to_idno_with_type_if_detected(
self, test_input: str, expected_type: Optional[str]):
label_and_layout_line_list = [
('<pubnum>', get_next_layout_line_for_text(test_input))
]
labeled_model_data_list = get_labeled_model_data_list(
label_and_layout_line_list, data_generator=get_data_generator())
xml_root = get_training_tei_xml_for_model_data_iterable(
labeled_model_data_list)
if expected_type:
assert get_tei_xpath_text_content_list(
xml_root,
f'{BIBL_XPATH}/tei:idno[@type="{expected_type}"]') == [
test_input
]
else:
assert get_tei_xpath_text_content_list(
xml_root,
f'{BIBL_XPATH}/tei:idno[not(@type)]') == [test_input]
def test_should_add_superscript_text(self):
block = LayoutBlock.for_tokens([
LayoutToken(TOKEN_1),
LayoutToken(TOKEN_2, font=SUPERSCRIPT_FONT_1),
LayoutToken(TOKEN_3)
])
node = TEI_E.node(*iter_layout_block_tei_children(block))
assert get_tei_xpath_text_content_list(
node, './tei:hi[@rend="superscript"]') == [TOKEN_2]
assert get_text_content(node) == ' '.join([TOKEN_1, TOKEN_2, TOKEN_3])
def test_should_map_other_label_as_text_without_note(self):
label_and_layout_line_list = [('<other>',
get_next_layout_line_for_text(TEXT_1))]
labeled_model_data_list = get_labeled_model_data_list(
label_and_layout_line_list, data_generator=get_data_generator())
training_data_generator = get_tei_training_data_generator()
xml_root = training_data_generator.get_training_tei_xml_for_model_data_iterable(
labeled_model_data_list)
LOGGER.debug('xml: %r', etree.tostring(xml_root))
assert not tei_xpath(xml_root, f'{BIBL_XPATH}//tei:note')
assert get_tei_xpath_text_content_list(xml_root,
BIBL_XPATH) == [f'{TEXT_1}\n']
def test_should_generate_tei_for_most_labels(self):
label_and_layout_line_list = [
('<marker>', get_next_layout_line_for_text('Marker 1')),
('<title>', get_next_layout_line_for_text('Title 1')),
('<forename>', get_next_layout_line_for_text('Forename 1')),
('<middlename>', get_next_layout_line_for_text('Middlename 1')),
('<surname>', get_next_layout_line_for_text('Surname 1')),
('<suffix>', get_next_layout_line_for_text('Suffix 1'))
]
labeled_model_data_list = get_labeled_model_data_list(
label_and_layout_line_list, data_generator=get_data_generator())
xml_root = get_training_tei_xml_for_model_data_iterable(
labeled_model_data_list)
assert get_tei_xpath_text_content_list(
xml_root, f'{AUTHOR_XPATH}/tei:marker') == ['Marker 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{AUTHOR_XPATH}/tei:roleName') == ['Title 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{AUTHOR_XPATH}/tei:forename') == ['Forename 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{AUTHOR_XPATH}/tei:middlename') == ['Middlename 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{AUTHOR_XPATH}/tei:surname') == ['Surname 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{AUTHOR_XPATH}/tei:suffix') == ['Suffix 1']
def test_should_map_unknown_label_to_note(self):
label_and_layout_line_list = [('<unknown>',
get_next_layout_line_for_text(TEXT_1))]
labeled_model_data_list = get_labeled_model_data_list(
label_and_layout_line_list, data_generator=get_data_generator())
training_data_generator = AffiliationAddressTeiTrainingDataGenerator()
xml_root = training_data_generator.get_training_tei_xml_for_model_data_iterable(
labeled_model_data_list, )
LOGGER.debug('xml: %r', etree.tostring(xml_root))
aff_nodes = tei_xpath(xml_root, AFFILIATION_XPATH)
assert len(aff_nodes) == 1
assert get_tei_xpath_text_content_list(
aff_nodes[0], './tei:note[@type="unknown"]') == [TEXT_1]
assert get_text_content(aff_nodes[0]) == f'{TEXT_1}\n'
def _check_tei_training_data_generator_output(
tei_training_data_generator: TeiTrainingDataGenerator,
output_path: Path, expect_raw_data: bool, tei_xml_xpath: str,
tei_expected_values: Sequence[str], **kwargs):
xml_root = _check_tei_training_data_generator_output_and_return_xml_root(
tei_training_data_generator=tei_training_data_generator,
output_path=output_path,
expect_raw_data=expect_raw_data,
**kwargs)
assert normalize_whitespace_list(
get_tei_xpath_text_content_list(xml_root, tei_xml_xpath)) == [
normalize_whitespace(tei_expected_value)
for tei_expected_value in tei_expected_values
]
def test_should_not_join_separate_labels(self):
label_and_layout_line_list = [
('<institution>', get_next_layout_line_for_text(TEXT_1)),
('<institution>', get_next_layout_line_for_text(TEXT_2))
]
labeled_model_data_list = get_labeled_model_data_list(
label_and_layout_line_list, data_generator=get_data_generator())
training_data_generator = AffiliationAddressTeiTrainingDataGenerator()
xml_root = training_data_generator.get_training_tei_xml_for_model_data_iterable(
labeled_model_data_list)
LOGGER.debug('xml: %r', etree.tostring(xml_root))
aff_nodes = tei_xpath(xml_root, AFFILIATION_XPATH)
assert len(aff_nodes) == 1
assert get_tei_xpath_text_content_list(
aff_nodes[0],
'./tei:orgName[@type="institution"]') == [TEXT_1, TEXT_2]
assert get_text_content(aff_nodes[0]) == f'{TEXT_1}\n{TEXT_2}\n'
def test_should_be_able_to_set_title_with_italic_layout_tokens(self):
title_block = LayoutBlock.for_tokens([
LayoutToken('rend'),
LayoutToken('italic1', font=ITALICS_FONT_1),
LayoutToken('test')
])
document = TeiDocument()
document.set_title_layout_block(title_block)
LOGGER.debug('xml: %r', etree.tostring(document.root))
nodes = document.root.xpath(
'//tei:fileDesc/tei:titleStmt/tei:title[@level="a"][@type="main"]',
namespaces=TEI_NS_MAP)
assert len(nodes) == 1
title_node = nodes[0]
assert get_tei_xpath_text_content_list(
title_node, './tei:hi[@rend="italic"]') == ['italic1']
assert document.get_title() == 'rend italic1 test'
def test_should_add_label_to_head_element_without_additional_text(self):
label_and_layout_line_list = [
('<label>', get_next_layout_line_for_text('Table Label 1')),
('<figDesc>', get_next_layout_line_for_text('Table Desc 1')),
('<content>', get_next_layout_line_for_text('Content 1')),
('<other>', get_next_layout_line_for_text('Other 1')),
('<note>', get_next_layout_line_for_text('Note 1'))
]
labeled_model_data_list = get_labeled_model_data_list(
label_and_layout_line_list, data_generator=get_data_generator())
xml_root = get_training_tei_xml_for_model_data_iterable(
labeled_model_data_list)
assert get_tei_xpath_text_content_list(
xml_root, f'{TABLE_XPATH}/head/label') == ['Table Label 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{TABLE_XPATH}/head') == ['Table Label 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{TABLE_XPATH}/figDesc') == ['Table Desc 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{TABLE_XPATH}/table') == ['Content 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{TABLE_XPATH}/other') == ['Other 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{TABLE_XPATH}/note') == ['Note 1']
def get_abstract(self) -> str:
return '\n'.join(
get_tei_xpath_text_content_list(
self.root,
'//tei:abstract/tei:p',
))
def get_title_text(self) -> str:
return '\n'.join(
get_tei_xpath_text_content_list(
self.element,
'//tei:head',
))
def get_paragraph_text_list(self) -> List[str]:
return get_tei_xpath_text_content_list(
self.element,
'//tei:p',
)
def get_title(self) -> str:
return '\n'.join(
get_tei_xpath_text_content_list(
self.root,
'//tei:fileDesc/tei:titleStmt/tei:title[@level="a"][@type="main"]',
))
def test_should_generate_tei_for_most_labels(self):
label_and_layout_line_list = [
('<marker>', get_next_layout_line_for_text('Marker 1')),
('<institution>', get_next_layout_line_for_text('Institution 1')),
('<department>', get_next_layout_line_for_text('Department 1')),
('<laboratory>', get_next_layout_line_for_text('Laboratory 1')),
('<addrLine>', get_next_layout_line_for_text('AddrLine 1')),
('O', get_next_layout_line_for_text(',')),
('<postCode>', get_next_layout_line_for_text('PostCode 1')),
('O', get_next_layout_line_for_text(',')),
('<postBox>', get_next_layout_line_for_text('PostBox 1')),
('O', get_next_layout_line_for_text(',')),
('<region>', get_next_layout_line_for_text('Region 1')),
('O', get_next_layout_line_for_text(',')),
('<settlement>', get_next_layout_line_for_text('Settlement 1')),
('O', get_next_layout_line_for_text(',')),
('<country>', get_next_layout_line_for_text('Country 1'))
]
labeled_model_data_list = get_labeled_model_data_list(
label_and_layout_line_list, data_generator=get_data_generator())
training_data_generator = AffiliationAddressTeiTrainingDataGenerator()
xml_root = training_data_generator.get_training_tei_xml_for_model_data_iterable(
labeled_model_data_list)
LOGGER.debug('xml: %r', etree.tostring(xml_root))
aff_nodes = tei_xpath(xml_root, AFFILIATION_XPATH)
assert len(aff_nodes) == 1
assert get_tei_xpath_text_content_list(aff_nodes[0],
'./tei:marker') == ['Marker 1']
assert get_tei_xpath_text_content_list(
aff_nodes[0],
'./tei:orgName[@type="institution"]') == ['Institution 1']
assert get_tei_xpath_text_content_list(
aff_nodes[0],
'./tei:orgName[@type="department"]') == ['Department 1']
assert get_tei_xpath_text_content_list(
aff_nodes[0],
'./tei:orgName[@type="laboratory"]') == ['Laboratory 1']
assert get_tei_xpath_text_content_list(
aff_nodes[0], './tei:address/tei:addrLine') == ['AddrLine 1']
assert get_tei_xpath_text_content_list(
aff_nodes[0], './tei:address/tei:postCode') == ['PostCode 1']
assert get_tei_xpath_text_content_list(
aff_nodes[0], './tei:address/tei:postBox') == ['PostBox 1']
assert get_tei_xpath_text_content_list(
aff_nodes[0], './tei:address/tei:region') == ['Region 1']
assert get_tei_xpath_text_content_list(
aff_nodes[0], './tei:address/tei:settlement') == ['Settlement 1']
assert get_tei_xpath_text_content_list(
aff_nodes[0], './tei:address/tei:country') == ['Country 1']
assert get_tei_xpath_text_content_list(aff_nodes[0],
'./tei:address') == [
'\n,\n'.join([
'AddrLine 1',
'PostCode 1',
'PostBox 1', 'Region 1',
'Settlement 1',
'Country 1'
])
]
def test_should_generate_tei_for_most_labels(self):
label_and_layout_line_list = [
('<title>', get_next_layout_line_for_text('Title 1')),
('<author>', get_next_layout_line_for_text('Author 1')),
('<editor>', get_next_layout_line_for_text('Editor 1')),
('<institution>', get_next_layout_line_for_text('Institution 1')),
('<collaboration>',
get_next_layout_line_for_text('Collaboration 1')),
('<journal>', get_next_layout_line_for_text('Journal 1')),
('<series>', get_next_layout_line_for_text('Series 1')),
('<booktitle>', get_next_layout_line_for_text('Book Title 1')),
('<date>', get_next_layout_line_for_text('Date 1')),
('<volume>', get_next_layout_line_for_text('Volume 1')),
('<issue>', get_next_layout_line_for_text('Issue 1')),
('<pages>', get_next_layout_line_for_text('Pages 1')),
('<publisher>', get_next_layout_line_for_text('Publisher 1')),
('<location>', get_next_layout_line_for_text('Location 1')),
('<tech>', get_next_layout_line_for_text('Tech 1')),
('<pubnum>', get_next_layout_line_for_text('Pubnum 1')),
('<web>', get_next_layout_line_for_text('Web 1')),
('<note>', get_next_layout_line_for_text('Note 1'))
]
labeled_model_data_list = get_labeled_model_data_list(
label_and_layout_line_list, data_generator=get_data_generator())
xml_root = get_training_tei_xml_for_model_data_iterable(
labeled_model_data_list)
assert get_tei_xpath_text_content_list(
xml_root, f'{BIBL_XPATH}/tei:title[@level="a"]') == ['Title 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{BIBL_XPATH}/tei:author') == ['Author 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{BIBL_XPATH}/tei:editor') == ['Editor 1']
assert get_tei_xpath_text_content_list(
xml_root,
f'{BIBL_XPATH}/tei:orgName[not(@type)]') == ['Institution 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{BIBL_XPATH}/tei:orgName[@type="collaboration"]') == [
'Collaboration 1'
]
assert get_tei_xpath_text_content_list(
xml_root, f'{BIBL_XPATH}/tei:title[@level="j"]') == ['Journal 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{BIBL_XPATH}/tei:title[@level="s"]') == ['Series 1']
assert get_tei_xpath_text_content_list(
xml_root,
f'{BIBL_XPATH}/tei:title[@level="m"]') == ['Book Title 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{BIBL_XPATH}/tei:date') == ['Date 1']
assert get_tei_xpath_text_content_list(
xml_root,
f'{BIBL_XPATH}/tei:biblScope[@unit="volume"]') == ['Volume 1']
assert get_tei_xpath_text_content_list(
xml_root,
f'{BIBL_XPATH}/tei:biblScope[@unit="issue"]') == ['Issue 1']
assert get_tei_xpath_text_content_list(
xml_root,
f'{BIBL_XPATH}/tei:biblScope[@unit="page"]') == ['Pages 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{BIBL_XPATH}/tei:publisher') == ['Publisher 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{BIBL_XPATH}/tei:pubPlace') == ['Location 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{BIBL_XPATH}/tei:note[@type="report"]') == ['Tech 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{BIBL_XPATH}/tei:idno') == ['Pubnum 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{BIBL_XPATH}/tei:ptr[@type="web"]') == ['Web 1']
assert get_tei_xpath_text_content_list(
xml_root, f'{BIBL_XPATH}/tei:note[not(@type)]') == ['Note 1']
