def test_should_always_preserve_specified_existing_tag_when_merging_front(
self, test_helper: SingleFileAutoAnnotateEndToEndTestHelper):
test_helper.tei_raw_file_path.write_bytes(
etree.tostring(
get_segmentation_tei_node([
E.note(TITLE_1),
E.lb(),
E.page(TOKEN_2),
E.lb(),
E.note(ABSTRACT_1),
E.lb()
])))
test_helper.xml_file_path.write_bytes(
etree.tostring(
get_target_xml_node(title=TITLE_1,
abstract_node=E.abstract(
E.p(ABSTRACT_1)))))
main(dict_to_args({
**test_helper.main_args_dict, 'always-preserve-fields':
'page'
}),
save_main_session=False)
tei_auto_root = test_helper.get_tei_auto_root()
assert get_xpath_text_list(tei_auto_root, '//text/page') == [TOKEN_2]
assert get_xpath_text_list(tei_auto_root,
'//text/front') == [TITLE_1, ABSTRACT_1]
def test_should_not_merge_front_interrupted_by_body_tag(
self, test_helper: SingleFileAutoAnnotateEndToEndTestHelper):
test_helper.tei_raw_file_path.write_bytes(
etree.tostring(
get_segmentation_tei_node(
get_tei_nodes_for_text('\n'.join([
TITLE_1, 'After title', TEXT_1, 'Before abstract',
ABSTRACT_1
])))))
test_helper.xml_file_path.write_bytes(
etree.tostring(
get_target_xml_node(title=TITLE_1,
abstract_node=E.abstract(E.p(ABSTRACT_1)),
body_nodes=[E.sec(E.p(TEXT_1))])))
main(dict_to_args({
**test_helper.main_args_dict, 'fields':
','.join(['title', 'abstract', 'body_section_paragraph'])
}),
save_main_session=False)
tei_auto_root = test_helper.get_tei_auto_root()
assert get_xpath_text_list(tei_auto_root,
'//text/front') == [TITLE_1, ABSTRACT_1]
assert get_xpath_text_list(tei_auto_root, '//text/body') == [
'\n'.join(['After title', TEXT_1, 'Before abstract'])
]
def test_should_auto_annotate_body_and_back_list_item_section_paragraphs(
self, test_helper: SingleFileAutoAnnotateEndToEndTestHelper):
target_body_content_nodes = [E.list(E('list-item', E.p(TEXT_1)))]
target_back_content_nodes = [E.list(E('list-item', E.p(TEXT_2)))]
tei_text = '\n'.join([
get_nodes_text(target_body_content_nodes),
get_nodes_text(target_back_content_nodes)
])
test_helper.tei_raw_file_path.write_bytes(
etree.tostring(
get_training_tei_node(get_tei_nodes_for_text(tei_text))))
test_helper.xml_file_path.write_bytes(
etree.tostring(
get_target_xml_node(body_nodes=target_body_content_nodes,
back_nodes=target_back_content_nodes)))
main(dict_to_args({
**test_helper.main_args_dict, 'fields':
','.join(['body_section_paragraph', 'back_section_paragraph'])
}),
save_main_session=False)
tei_auto_root = test_helper.get_tei_auto_root()
assert get_xpath_text_list(tei_auto_root, '//body') == [TEXT_1]
assert get_xpath_text_list(tei_auto_root,
'//div[@type="annex"]') == [TEXT_2]
def test_should_auto_annotate_body_and_back_section(
self, test_helper: SingleFileAutoAnnotateEndToEndTestHelper):
target_body_content_nodes = [
E.sec(E.title(SECTION_TITLE_1), '\n', E.p(TEXT_1))
]
target_back_content_nodes = [
E.sec(E.title(SECTION_TITLE_2), '\n', E.p(TEXT_2))
]
body_tei_text = get_nodes_text(target_body_content_nodes)
back_tei_text = get_nodes_text(target_back_content_nodes)
tei_text = '\n'.join([body_tei_text, back_tei_text])
LOGGER.debug('tei_text: %s', tei_text)
test_helper.tei_raw_file_path.write_bytes(
etree.tostring(
get_training_tei_node(get_tei_nodes_for_text(tei_text))))
test_helper.xml_file_path.write_bytes(
etree.tostring(
get_target_xml_node(body_nodes=target_body_content_nodes,
back_nodes=target_back_content_nodes)))
main(dict_to_args({
**test_helper.main_args_dict, 'fields':
','.join([
'body_section_title', 'body_section_paragraph',
'back_section_title', 'back_section_paragraph'
])
}),
save_main_session=False)
tei_auto_root = test_helper.get_tei_auto_root()
assert get_xpath_text_list(tei_auto_root, '//body') == [body_tei_text]
assert get_xpath_text_list(tei_auto_root,
'//div[@type="annex"]') == [back_tei_text]
def test_should_not_preserve_exclude_existing_tag_and_use_body_by_default(
self, test_helper: SingleFileAutoAnnotateEndToEndTestHelper):
test_helper.tei_raw_file_path.write_bytes(
etree.tostring(get_segmentation_tei_node([E.page(TOKEN_1)])))
test_helper.xml_file_path.write_bytes(
etree.tostring(E.article(E.front())))
main([*test_helper.main_args, '--no-preserve-fields=page'],
save_main_session=False)
tei_auto_root = test_helper.get_tei_auto_root()
assert get_xpath_text_list(tei_auto_root, '//text/page') == []
assert get_xpath_text_list(tei_auto_root, '//text/body') == [TOKEN_1]
def test_should_auto_annotate_and_not_merge_multiple_references(
self, test_helper: SingleFileAutoAnnotateEndToEndTestHelper):
test_helper.tei_raw_file_path.write_bytes(
etree.tostring(
get_segmentation_tei_node(
get_tei_nodes_for_lines([
TOKEN_1, LABEL_1 + ' ' + REFERENCE_TEXT_1,
LABEL_2 + ' ' + REFERENCE_TEXT_2
]))))
test_helper.xml_file_path.write_bytes(
etree.tostring(
get_target_xml_node(reference_nodes=[
get_jats_reference_node(LABEL_1, REFERENCE_TEXT_1),
get_jats_reference_node(LABEL_2, REFERENCE_TEXT_2)
])))
main(dict_to_args({
**test_helper.main_args_dict, 'no-preserve-tags':
True,
'no-merge-references':
True,
'fields':
'reference',
'xml-mapping-overrides':
'reference.use-raw-text=true'
}),
save_main_session=False)
tei_auto_root = test_helper.get_tei_auto_root()
assert get_xpath_text_list(tei_auto_root, '//text/listBibl') == [
LABEL_1 + ' ' + REFERENCE_TEXT_1, LABEL_2 + ' ' + REFERENCE_TEXT_2
]
def test_should_always_preserve_reference_tag(
self, test_helper: SingleFileAutoAnnotateEndToEndTestHelper):
_common_tokens = [TOKEN_2, TOKEN_3]
_reference_text = ' '.join(_common_tokens) + ' this is a reference 1'
test_helper.tei_raw_file_path.write_bytes(
etree.tostring(
get_segmentation_tei_node([
E.note(TOKEN_1),
E.lb(),
E.listBibl(_reference_text),
E.lb()
])))
test_helper.xml_file_path.write_bytes(
etree.tostring(
get_target_xml_node(title=' '.join([TOKEN_1] +
_common_tokens))))
main(dict_to_args({
**test_helper.main_args_dict, 'no-preserve-tags': True,
'always-preserve-fields': 'reference'
}),
save_main_session=False)
tei_auto_root = test_helper.get_tei_auto_root()
assert get_xpath_text_list(tei_auto_root,
'//text/listBibl') == [_reference_text]
def test_should_auto_annotate_body_and_back_table_label_title_caption_as_body(
self, test_helper: SingleFileAutoAnnotateEndToEndTestHelper):
target_body_content_nodes = [
E(
'table-wrap', *[
E.label(FIGURE_LABEL_1), ' ',
E.caption(E.title(SECTION_TITLE_1), '\n', E.p(TEXT_1))
])
]
target_back_content_nodes = [
E(
'table-wrap', *[
E.label(FIGURE_LABEL_2), ' ',
E.caption(E.title(SECTION_TITLE_2), '\n', E.p(TEXT_2))
])
]
body_tei_text = get_nodes_text(target_body_content_nodes)
back_tei_text = get_nodes_text(target_back_content_nodes)
tei_text = '\n'.join([body_tei_text, back_tei_text])
test_helper.tei_raw_file_path.write_bytes(
etree.tostring(
get_training_tei_node(get_tei_nodes_for_text(tei_text))))
test_helper.xml_file_path.write_bytes(
etree.tostring(
get_target_xml_node(body_nodes=target_body_content_nodes,
back_nodes=target_back_content_nodes)))
main(dict_to_args({
**test_helper.main_args_dict, 'fields':
','.join(['table'])
}),
save_main_session=False)
tei_auto_root = test_helper.get_tei_auto_root()
assert get_xpath_text_list(tei_auto_root, '//body') == [tei_text]
def test_should_auto_annotate_appendix_section_as_annex(
self, test_helper: SingleFileAutoAnnotateEndToEndTestHelper):
target_back_content_nodes = [
E(
'app-group', *[
E.title('Appendix'), '\n',
E.app(E.title(SECTION_TITLE_2), '\n', E.p(TEXT_2))
])
]
tei_text = get_nodes_text(target_back_content_nodes)
test_helper.tei_raw_file_path.write_bytes(
etree.tostring(
get_training_tei_node(get_tei_nodes_for_text(tei_text))))
test_helper.xml_file_path.write_bytes(
etree.tostring(
get_target_xml_node(back_nodes=target_back_content_nodes)))
main(dict_to_args({
**test_helper.main_args_dict, 'fields':
','.join([
'body_section_title', 'body_section_paragraph',
'back_section_title', 'back_section_paragraph',
'appendix_group_title', 'appendix'
])
}),
save_main_session=False)
tei_auto_root = test_helper.get_tei_auto_root()
assert get_xpath_text_list(tei_auto_root,
'//div[@type="annex"]') == [tei_text]
def test_should_not_preserve_existing_front_body_tag_front_and_use_headnote_for_repeated_text(
self, test_helper: SingleFileAutoAnnotateEndToEndTestHelper):
test_helper.tei_raw_file_path.write_bytes(
etree.tostring(
get_segmentation_tei_node([
E.front(*get_tei_nodes_for_lines([
'Page header', 'Before title', TITLE_1, 'After title'
])),
E.body(*get_tei_nodes_for_lines([
'Page header',
'Before paragraph',
TEXT_1,
'After paragraph',
]))
])))
test_helper.xml_file_path.write_bytes(
etree.tostring(
get_target_xml_node(title=TITLE_1,
body_nodes=[E.sec(E.p(TEXT_1))])))
main(dict_to_args({
**test_helper.main_args_dict, 'no-preserve-fields':
'front,body',
'fields':
','.join(['title', 'abstract', 'body_section_paragraph'])
}),
save_main_session=False)
tei_auto_root = test_helper.get_tei_auto_root()
assert get_xpath_text_list(tei_auto_root,
'//text/note[@place="headnote"]') == [
'Page header', 'Page header'
]
assert get_xpath_text_list(tei_auto_root, '//text/front') == [
'\n'.join(['Before title', TITLE_1])
]
assert get_xpath_text_list(tei_auto_root, '//text/body') == [
'After title',
'\n'.join(['Before paragraph', TEXT_1, 'After paragraph'])
]
def test_should_preserve_page_numbers_while_detecting_headnote(
self, test_helper: SingleFileAutoAnnotateEndToEndTestHelper):
test_helper.tei_raw_file_path.write_bytes(
etree.tostring(
get_segmentation_tei_node([
E.front('Page header', E.lb()),
'\n',
E.page('123', E.lb()),
'\n',
E.front(TITLE_1, E.lb()),
'\n',
E.body('Page header', E.lb()),
'\n',
E.page('123', E.lb()),
'\n',
E.body(TEXT_1, E.lb()),
'\n',
])))
test_helper.xml_file_path.write_bytes(
etree.tostring(
get_target_xml_node(title=TITLE_1,
body_nodes=[E.sec(E.p(TEXT_1))])))
main(dict_to_args({
**test_helper.main_args_dict, 'always-preserve-fields':
'page',
'fields':
','.join(['title', 'abstract', 'body_section_paragraph'])
}),
save_main_session=False)
tei_auto_root = test_helper.get_tei_auto_root()
assert get_xpath_text_list(tei_auto_root,
'//text/page') == ['123', '123']
assert get_xpath_text_list(tei_auto_root,
'//text/note[@place="headnote"]') == [
'Page header', 'Page header'
]
assert get_xpath_text_list(tei_auto_root, '//text/front') == [TITLE_1]
assert get_xpath_text_list(tei_auto_root, '//text/body') == [TEXT_1]
def test_should_merge_front_tags_and_include_preceeding_text(
self, test_helper: SingleFileAutoAnnotateEndToEndTestHelper):
front_tei_text = '\n'.join(['Before', TITLE_1, 'Other', ABSTRACT_1])
tei_text = '\n'.join([front_tei_text, 'After'])
test_helper.tei_raw_file_path.write_bytes(
etree.tostring(
get_training_tei_node(get_tei_nodes_for_text(tei_text))))
test_helper.write_xml_root(
get_target_xml_node(title=TITLE_1,
abstract_node=E.abstract(E.p(ABSTRACT_1))))
main(dict_to_args({
**test_helper.main_args_dict, 'fields':
','.join(['title', 'abstract'])
}),
save_main_session=False)
tei_auto_root = test_helper.get_tei_auto_root()
assert get_xpath_text_list(tei_auto_root,
'//front') == [front_tei_text]
