def __init__(self, resource, name=None, logger=None, dispatcher=None):
""" Initiate the XMLResolver
"""
if dispatcher is None:
inventory_collection = CtsTextInventoryCollection(identifier="defaultTic")
ti = XmlCtsTextInventoryMetadata("default")
ti.parent = inventory_collection
ti.set_label("Default collection", "eng")
self.dispatcher = CollectionDispatcher(inventory_collection)
else:
self.dispatcher = dispatcher
self.__inventory__ = self.dispatcher.collection
self.name = name
self.logger = logger
if not logger:
self.logger = logging.getLogger(name)
if not name:
self.name = "repository"
self.TEXT_CLASS = type(self).TEXT_CLASS
self.works = []
self.parse(resource)
def test_dispatching_error(self):
tic = CtsTextInventoryCollection()
latin = CtsTextInventoryMetadata("urn:perseus:latinLit", parent=tic)
latin.set_label("Classical Latin", "eng")
dispatcher = CollectionDispatcher(tic)
# We remove default dispatcher
dispatcher.__methods__ = []
@dispatcher.inventory("urn:perseus:latinLit")
def dispatchLatinLit(collection, path=None, **kwargs):
if collection.id.startswith("urn:cts:latinLit:"):
return True
return False
NautilusCTSResolver.RAISE_ON_UNDISPATCHED = True
with self.assertRaises(Exception):
resolver = NautilusCTSResolver(["./tests/testing_data/latinLit2"],
dispatcher=dispatcher)
resolver.logger.disabled = True
resolver.parse()
NautilusCTSResolver.RAISE_ON_UNDISPATCHED = False
try:
resolver = NautilusCTSResolver(["./tests/testing_data/latinLit2"],
dispatcher=dispatcher)
resolver.logger.disabled = True
resolver.REMOVE_EMPTY = False
resolver.parse()
except UndispatchedTextError as E:
self.fail("UndispatchedTextError should not have been raised")
def __init__(self, resource, name=None, logger=None, cache=None, dispatcher=None):
""" Initiate the XMLResolver
"""
if dispatcher is None:
inventory_collection = TextInventoryCollection(identifier="defaultTic")
ti = TextInventory("default")
ti.parent = inventory_collection
ti.set_label("Default collection", "eng")
self.dispatcher = CollectionDispatcher(inventory_collection)
else:
self.dispatcher = dispatcher
self.__inventory__ = None
self.__texts__ = []
self.name = name
self.logger = logger
if not logger:
self.logger = logging.getLogger(name)
if not name:
self.name = "repository"
if cache is None:
cache = NullCache()
self.__cache__ = cache
self.__resources__ = resource
self.inventory_cache_key = _cache_key("Nautilus", self.name, "Inventory", "Resources")
self.texts_parsed_cache_key = _cache_key("Nautilus", self.name, "Inventory", "TextsParsed")
def test_dispatching_latin_greek(self):
tic = CtsTextInventoryCollection()
latin = CtsTextInventoryMetadata("urn:perseus:latinLit", parent=tic)
latin.set_label("Classical Latin", "eng")
farsi = CtsTextInventoryMetadata("urn:perseus:farsiLit", parent=tic)
farsi.set_label("Farsi", "eng")
gc = CtsTextInventoryMetadata("urn:perseus:greekLit", parent=tic)
gc.set_label("Ancient Greek", "eng")
gc.set_label("Grec Ancien", "fre")
dispatcher = CollectionDispatcher(tic)
@dispatcher.inventory("urn:perseus:latinLit")
def dispatchLatinLit(collection, path=None, **kwargs):
if collection.id.startswith("urn:cts:latinLit:"):
return True
return False
@dispatcher.inventory("urn:perseus:farsiLit")
def dispatchfFarsiLit(collection, path=None, **kwargs):
if collection.id.startswith("urn:cts:farsiLit:"):
return True
return False
@dispatcher.inventory("urn:perseus:greekLit")
def dispatchGreekLit(collection, path=None, **kwargs):
if collection.id.startswith("urn:cts:greekLit:"):
return True
return False
resolver = CtsCapitainsLocalResolver(
["./tests/testing_data/latinLit2"],
dispatcher=dispatcher
)
latin_stuff = resolver.getMetadata("urn:perseus:latinLit")
greek_stuff = resolver.getMetadata("urn:perseus:greekLit")
farsi_stuff = resolver.getMetadata("urn:perseus:farsiLit")
self.assertEqual(
len(latin_stuff.readableDescendants), 20,
"There should be 20 readable descendants in Latin"
)
self.assertIsInstance(
latin_stuff, CtsTextInventoryMetadata, "should be textinventory"
)
self.assertEqual(
len(greek_stuff.readableDescendants), 6,
"There should be 6 readable descendants in Greek [6 only in __cts__.xml]"
)
self.assertEqual(
len(farsi_stuff.descendants), 0,
"There should be nothing in FarsiLit"
)
self.assertEqual(
str(greek_stuff.get_label("fre")), "Grec Ancien",
"Label should be correct"
)
with self.assertRaises(KeyError):
_ = latin_stuff["urn:cts:greekLit:tlg0003"]
def make_dispatcher():
tic = CtsTextInventoryCollection()
latin = CtsTextInventoryMetadata("urn:perseus:latinLit", parent=tic)
latin.set_label("Classical Latin", "eng")
latin.set_label("Latin Classique", "fre")
dispatcher = CollectionDispatcher(tic)
@dispatcher.inventory("urn:perseus:latinLit")
def dispatchLatinLit(collection, path=None, **kwargs):
if collection.id.startswith("urn:cts:latinLit:"):
return True
return False
return dispatcher
def test_post_work_dispatching_active(self):
""" Dispatching is working after editions, we dispatch based on citation scheme"""
tic = CtsTextInventoryCollection()
poetry = CtsTextInventoryMetadata("urn:perseus:poetry", parent=tic)
prose = CtsTextInventoryMetadata("urn:perseus:prose", parent=tic)
dispatcher = CollectionDispatcher(tic, default_inventory_name="urn:perseus:prose")
@dispatcher.inventory("urn:perseus:poetry")
def dispatchPoetry(collection, **kwargs):
for readable in collection.readableDescendants:
for citation in readable.citation:
if citation.name == "line":
return True
return False
resolver = CtsCapitainsLocalResolver(
["./tests/testing_data/latinLit2"],
dispatcher=dispatcher
)
all = resolver.getMetadata().export(Mimetypes.XML.CTS)
poetry_stuff = resolver.getMetadata("urn:perseus:poetry").export(Mimetypes.XML.CTS)
prose_stuff = resolver.getMetadata("urn:perseus:prose").export(Mimetypes.XML.CTS)
get_graph().remove((None, None, None))
del poetry, prose
poetry, prose = XmlCtsTextInventoryMetadata.parse(poetry_stuff), XmlCtsTextInventoryMetadata.parse(prose_stuff)
self.assertEqual(
len(poetry.textgroups), 3,
"There should be 3 textgroups in Poetry (Martial, Ovid and Juvenal)"
)
self.assertIsInstance(poetry, CtsTextInventoryMetadata, "should be textinventory")
self.assertEqual(
len(prose.textgroups), 1,
"There should be one textgroup in Prose (Greek texts)"
)
get_graph().remove((None, None, None))
del poetry, prose
all = XmlCtsTextInventoryMetadata.parse(all)
self.assertEqual(
len(all.readableDescendants), 26,
"There should be all 26 readable descendants in the master collection"
)
self.nemo = Nemo(resolver=NemoResource.endpoint, app=Flask(__name__))
tic = CtsTextInventoryCollection()
latin = XmlCtsTextInventoryMetadata("urn:perseus:latinLit")
latin.parent = tic
latin.set_label("Classical Latin", "eng")
farsi = XmlCtsTextInventoryMetadata("urn:perseus:farsiLit")
farsi.parent = tic
farsi.set_label("Farsi", "eng")
gc = XmlCtsTextInventoryMetadata("urn:perseus:greekLit")
gc.parent = tic
gc.set_label("Ancient Greek", "eng")
gc.set_label("Grec Ancien", "fre")
dispatcher = CollectionDispatcher(tic)
@dispatcher.inventory("urn:perseus:latinLit")
def dispatchLatinLit(collection, path=None, **kwargs):
if collection.id.startswith("urn:cts:latinLit:"):
return True
return False
@dispatcher.inventory("urn:perseus:farsiLit")
def dispatchfFarsiLit(collection, path=None, **kwargs):
if collection.id.startswith("urn:cts:farsiLit:"):
return True
return False
def build_resolver(configuration_file):
"""
:param configuration_file:
:return: Organizer, Resolver and Cache handler
"""
with open(configuration_file) as f:
xml = etree.parse(f)
directories = [
# Compute path relative to the configuration files
relative_folder(configuration_file, directory)
for directory in xml.xpath("//corpora/corpus/text()")
]
default_collection = None
general_collection = CtsTextInventoryCollection()
filters_to_register = []
for collection in xml.xpath("//collections/collection"):
identifier = collection.xpath("./identifier/text()")[0]
if collection.get("default") == "true":
default_collection = identifier
current_collection = CtsTextInventoryMetadata(
identifier, parent=general_collection)
for name in collection.xpath("./name"):
current_collection.set_label(name.text, name.get("lang"))
# We look at dispatching filters in the collection
for filters in collection.xpath("./filters"):
# We register prefix filters
prefix_filters = []
for prefix in filters.xpath("./id-starts-with/text()"):
prefix_filters.append(
lambda collection: str(collection.id).startswith(prefix))
# We register citation filters
citation_filters = []
for citation_name in filters.xpath("./citation-contains/text()"):
citation_filters.append(
lambda collection: citation_contain_filter(
collection, citation_name))
# We register path based filters
directory_filters = []
for target_directory in filters.xpath("./folder/text()"):
directory_filters.append(
lambda collection, path=None: path.startswith(
relative_folder(configuration_file, target_directory)))
filters_to_register += [
(identifier,
collection_dispatcher_builder(collection, prefix_filters,
citation_filters,
directory_filters))
]
# Create the dispatcher
organizer = CollectionDispatcher(general_collection,
default_inventory_name=default_collection)
for destination_collection, anonymous_dispatching_function in filters_to_register:
organizer.add(anonymous_dispatching_function, destination_collection)
# Set-up the cache folder
# ToDO : Add a system for redis ?
cache = None
for cache_folder in xml.xpath("//cache-folder/text()"):
cache = FileSystemCache(cache_folder)
if cache is None:
cache = SimpleCache()
resolver = NautilusCTSResolver(resource=directories,
dispatcher=organizer,
cache=cache)
return organizer, resolver, cache
str(chapter_number), # First the reference for the URI as string
"Pratum Spirituale " +
str(chapter_number) # Then the readable format for humans
))
return chapters
# Setting up the collections
general_collection = CtsTextInventoryCollection()
greek_texts = CtsTextInventoryMetadata("greek_texts",
parent=general_collection)
greek_texts.set_label("Greek Texts", "eng")
organizer = CollectionDispatcher(general_collection,
default_inventory_name="id:misc")
@organizer.inventory("greek_texts")
def organize_my_meadow(collection, path=None, **kwargs):
if collection.id.startswith("urn:cts:greekLit"):
return True
return False
flask_app = Flask("Flask Application for Nemo")
resolver = NautilusCTSResolver(["corpora/meadow"], dispatcher=organizer)
resolver.parse()
nautilus_api = FlaskNautilus(prefix="/api", app=flask_app, resolver=resolver)
def test_dispatching_output(self):
tic = CtsTextInventoryCollection()
latin = CtsTextInventoryMetadata("urn:perseus:latinLit", parent=tic)
latin.set_label("Classical Latin", "eng")
farsi = CtsTextInventoryMetadata("urn:perseus:farsiLit", parent=tic)
farsi.set_label("Farsi", "eng")
gc = CtsTextInventoryMetadata("urn:perseus:greekLit", parent=tic)
gc.set_label("Ancient Greek", "eng")
gc.set_label("Grec Ancien", "fre")
dispatcher = CollectionDispatcher(tic)
@dispatcher.inventory("urn:perseus:latinLit")
def dispatchLatinLit(collection, path=None, **kwargs):
if collection.id.startswith("urn:cts:latinLit:"):
return True
return False
@dispatcher.inventory("urn:perseus:farsiLit")
def dispatchfFarsiLit(collection, path=None, **kwargs):
if collection.id.startswith("urn:cts:farsiLit:"):
return True
return False
@dispatcher.inventory("urn:perseus:greekLit")
def dispatchGreekLit(collection, path=None, **kwargs):
if collection.id.startswith("urn:cts:greekLit:"):
return True
return False
resolver = NautilusCTSResolver(["./tests/testing_data/latinLit2"],
dispatcher=dispatcher)
resolver.logger.disabled = True
resolver.REMOVE_EMPTY = False
resolver.parse()
all = resolver.getMetadata().export(Mimetypes.XML.CTS)
latin_stuff = resolver.getMetadata("urn:perseus:latinLit").export(
Mimetypes.XML.CTS)
greek_stuff = resolver.getMetadata("urn:perseus:greekLit").export(
Mimetypes.XML.CTS)
farsi_stuff = resolver.getMetadata("urn:perseus:farsiLit").export(
Mimetypes.XML.CTS)
get_graph().remove((None, None, None))
latin_stuff, greek_stuff, farsi_stuff = XmlCtsTextInventoryMetadata.parse(latin_stuff), \
XmlCtsTextInventoryMetadata.parse(greek_stuff), \
XmlCtsTextInventoryMetadata.parse(farsi_stuff)
self.assertEqual(len(latin_stuff.readableDescendants), 19,
"There should be 19 readable descendants in Latin")
self.assertIsInstance(latin_stuff, CtsTextInventoryMetadata,
"should be textinventory")
self.assertEqual(
len(greek_stuff.readableDescendants), 6,
"There should be 6 readable descendants in Greek [6 only in __cts__.xml]"
)
self.assertEqual(len(farsi_stuff.descendants), 0,
"There should be nothing in FarsiLit")
self.assertEqual(
greek_stuff.get_label("fre"),
None, # Text inventory have no label in CTS
"Label should be correct")
get_graph().remove((None, None, None))
all = XmlCtsTextInventoryMetadata.parse(all)
self.assertEqual(
len(all.readableDescendants), 25,
"There should be all 25 readable descendants in the master collection"
)
formulae = CtsTextInventoryMetadata('formulae_collection',
parent=general_collection)
formulae.set_label('Formulae', 'ger')
formulae.set_label('Formulae', 'eng')
formulae.set_label('Formulae', 'fre')
chartae = CtsTextInventoryMetadata('other_collection',
parent=general_collection)
chartae.set_label('Andere Texte', 'ger')
chartae.set_label('Other Texts', 'eng')
chartae.set_label('Autres Textes', 'fre')
elexicon = CtsTextInventoryMetadata('lexicon_entries',
parent=general_collection)
elexicon.set_label('Lexikon', 'ger')
elexicon.set_label('Lexicon', 'eng')
elexicon.set_label('Lexique', 'fre')
organizer = CollectionDispatcher(general_collection,
default_inventory_name='other_collection')
@organizer.inventory("formulae_collection")
def organize_formulae(collection, path=None, **kwargs):
if collection.id.startswith('urn:cts:formulae:andecavensis'):
return True
return False
@organizer.inventory("lexicon_entries")
def organize_elexicon(collection, path=None, **kwargs):
if collection.id.startswith('urn:cts:formulae:elexicon'):
return True
return False