def test_from_filepath(self, neb_assembled_data):
filepath = neb_assembled_data / 'm46882.xhtml'
# Hit the target
doc = Document.from_filepath(filepath)
# Verify the metadata
assert doc.id == 'm46882'
expected_metadata = copy(M46882_METADATA)
# cnx-epub metadata is mutable, so sequences are lists rather than
# tuples.
expected_metadata['keywords'] = list(expected_metadata['keywords'])
expected_metadata['subjects'] = list(expected_metadata['subjects'])
assert doc.metadata == expected_metadata
# Verify the content is content'ish
assert doc._xml.xpath(
"/xhtml:body/*[@data-type='metadata']",
namespaces=HTML_DOCUMENT_NAMESPACES,
) == []
assert len(
doc._xml.xpath(
"//*[@id='fs-idm20141232']",
namespaces=HTML_DOCUMENT_NAMESPACES,
)) == 1
# Verify the resources are attached to the object
expected_filenames = []
filenames = [r.filename for r in doc.resources]
assert sorted(filenames) == expected_filenames
# Verify the references have been rewritten
ref = '{}/CNX_Stats_C01_M10_003.jpg'.format(REFERENCE_MARKER).encode()
assert ref in doc.content
def test_from_git_filepath(self, git_assembled_data):
filepath = git_assembled_data / 'm46882.xhtml'
# Hit the target
doc = Document.from_filepath(filepath)
# Verify the metadata
assert doc.id == 'm46882'
expected_metadata = copy(M46882_GIT_METADATA)
# cnx-epub metadata is mutable, so sequences are lists rather than
# tuples.
expected_metadata['keywords'] = list(expected_metadata['keywords'])
expected_metadata['subjects'] = list(expected_metadata['subjects'])
# Document.from_index_cnxml uses cnxml to parse metadata and then does
# some conversions while Document.from_filepath uses cnx-epub. These
# generate slightly different values
del expected_metadata['uuid']
expected_metadata.update({
'license_text': None,
})
assert doc.metadata == expected_metadata
# Verify the content is content'ish
assert doc._xml.xpath(
"/xhtml:body/*[@data-type='metadata']",
namespaces=HTML_DOCUMENT_NAMESPACES,
) == []
assert len(
doc._xml.xpath(
"//*[@id='fs-idm20141232']",
namespaces=HTML_DOCUMENT_NAMESPACES,
)) == 1
def test_sanatize_content(self, request, assembled):
assembled_data = request.getfixturevalue(assembled)
with (assembled_data / 'm46913.xhtml').open('rb') as fb:
html = etree.parse(fb)
# And parse a second copy for verification
fb.seek(0)
expected_html = etree.parse(fb)
# Hit the target
results = Document._sanatize_content(html)
# Construct expected results
body = expected_html.xpath(
"//xhtml:body",
namespaces=HTML_DOCUMENT_NAMESPACES,
)[0]
metadata_elm = body.xpath(
"//xhtml:div[@data-type='metadata']",
namespaces=HTML_DOCUMENT_NAMESPACES,
)[0]
body.remove(metadata_elm)
body.attrib.pop('itemtype')
body.attrib.pop('itemscope')
expected_results = etree.tostring(expected_html)
assert results == expected_results
def test_find_resources(self, request, neb_collection_data):
loc = neb_collection_data / 'm46909'
# Hit the target
resources = Document._find_resources(loc)
# Verify we discovered the resource files
expected_filenames = [
'Prev_m16020_DotPlot.png',
'fig-ch01_02_01n.png',
'm16020_DotPlot_description.html',
'm16020_DotPlot_download.pdf',
]
assert sorted([r.id for r in resources]) == expected_filenames
assert sorted([r.filename for r in resources]) == expected_filenames
def test_from_git_index_cnxml(self, git_collection_data):
filepath = git_collection_data / 'm46882' / 'index.cnxml'
# Hit the target
doc = Document.from_index_cnxml(filepath, mock_reference_resolver)
# Verify the metadata
assert doc.id == 'm46882'
expected_metadata = copy(M46882_GIT_METADATA)
assert doc.metadata == expected_metadata
# Verify the content is content'ish
assert doc._xml.xpath(
"/xhtml:body/*[@data-type='metadata']",
namespaces=HTML_DOCUMENT_NAMESPACES,
) == []
assert len(
doc._xml.xpath(
"//*[@id='fs-idm20141232']",
namespaces=HTML_DOCUMENT_NAMESPACES,
)) == 1
def test_from_index_cnxml(self, neb_collection_data):
filepath = neb_collection_data / 'm46882' / 'index.cnxml'
# Hit the target
doc = Document.from_index_cnxml(filepath, mock_reference_resolver)
# Verify the metadata
assert doc.id == 'm46882'
expected_metadata = copy(M46882_METADATA)
# When parsing from index.cnxml, neb uses the cnxml metadata parser
expected_metadata['uuid'] = None
assert doc.metadata == expected_metadata
# Verify the content is content'ish
assert doc._xml.xpath(
"/xhtml:body/*[@data-type='metadata']",
namespaces=HTML_DOCUMENT_NAMESPACES,
) == []
assert len(
doc._xml.xpath(
"//*[@id='fs-idm20141232']",
namespaces=HTML_DOCUMENT_NAMESPACES,
)) == 1
# Verify the resources are attached to the object
expected_filenames = [
'CNX_Stats_C01_M10_001.jpg',
'CNX_Stats_C01_M10_002.jpg',
'CNX_Stats_C01_M10_003.jpg',
]
filenames = [r.filename for r in doc.resources]
assert sorted(filenames) == expected_filenames
# Verify the references have been rewritten
ref = '{}/CNX_Stats_C01_M10_003.jpg'.format(REFERENCE_MARKER).encode()
assert ref in doc.content
# Verify external and non-existent resource references remain
assert b'src="foobar.png"' in doc.content
assert b'ef="/[email protected]"' in doc.content # rewritten in cnxml->html
assert b'ef="http://en.wikibooks.org/"' in doc.content
def main():
# Transform the modules from cnxml to html
for id in ('m46882', 'm46909', 'm46913'):
filepath = input_dir / id / 'index.cnxml'
doc = Document.from_index_cnxml(filepath, mock_reference_resolver)
# Write the html to file
html_filepath = (output_dir / '{}.xhtml'.format(id))
html_filepath.unlink()
with html_filepath.open('wb') as fb:
fb.write(bytes(formatters.HTMLFormatter(doc)))
# Create a symbolic link back to the module's source directory
link_to_source_dir = output_dir / id
link_to_source_dir.unlink()
source_dir = input_dir / id
link_to_source_dir.symlink_to(relative_path(source_dir, output_dir))
# Create the single-page-html
binder = Binder.from_collection_xml(input_dir / 'collection.xml')
with (output_dir / 'collection.assembled.xhtml').open('wb') as fb:
fb.write(bytes(formatters.SingleHTMLFormatter(binder)))