def _populate_uri_to_validator(
validator: AValidator,
root_base_uri,
uri_to_validator: Dict,
):
"""
This needs to be run after _set_to_canonical_uri because _set_to_canonical_uri propagates
the base_URIs through the subschemas.
returns a dictionary of the uri to validator mapping
the uris are of three types.
- canonical id
- canonical id + location
- canonical id + anchor
"""
if (isinstance(validator, (Keyword, AValidator))
and not isinstance(validator, KeywordGroup)
and validator.location):
uri_to_validator[urijoin(root_base_uri,
"#" + validator.location)] = validator
if validator.id is not None and isinstance(validator, Validator):
validator_id = validator.id.rstrip("/")
uri_to_validator[validator_id] = validator
if validator.anchor:
uri_to_validator[(urijoin(validator.base_uri or "",
validator.anchor))] = validator
for sub_validator in validator.sub_validators():
_populate_uri_to_validator(
validator=sub_validator,
root_base_uri=root_base_uri,
uri_to_validator=uri_to_validator,
)
def _make_podcast_uri(self, xml, ep_guid=None):
if not xml:
return None
# Feeds use main URL, not localized
baseurl = ivooxapi.get_baseurl()
program = uritools.urijoin(baseurl, xml)
episode = '#' + uritools.urijoin(baseurl, ep_guid) if ep_guid else ''
return 'podcast+{}{}'.format(program, episode)
def check(self, base, ref, expected, strict=False):
self.assertEqual(expected, urijoin(base, ref, strict))
# base as bytes, ref as str
self.assertEqual(expected, urijoin(base.encode(), ref, strict))
# base as str, ref as bytes
self.assertEqual(expected, urijoin(base, ref.encode(), strict))
# both base and ref as bytes
self.assertEqual(expected.encode(),
urijoin(base.encode(), ref.encode(), strict))
def wrap_response(url, data, method):
title = ("title" in data and data["title"]) or ""
image = ("image" in data and data["image"]) or ""
favicon = ("favicon" in data and data["favicon"]) or ""
description = ("description" in data and data["description"]) or ""
url = ("url" in data and data["url"]) or url
css = ("css" in data and data["css"]) or ""
if image:
image = urijoin(url, image)
if favicon:
favicon = urijoin(url, favicon)
domain = urisplit(url).authority
if "domain" in data and data["domain"]:
domain = data["domain"]
site = data["title"]
if "site_name" in data and data["site_name"]:
site = data["site_name"]
if "provider_name" in data and data["provider_name"]:
site = data["provider_name"]
if method == "oembed" or (method == "custom" and "html" in data and data["html"]):
html = data["html"]
else:
if image:
html = template("unfurl-image", url, image, title, description, domain)
elif favicon:
html = template(
"unfurl-image unfurl-favicon", url, favicon, title, description, domain
)
else:
html = template(
"unfurl-image unfurl-default",
url,
"https://i.imgur.com/wQ37ilJ.png",
title,
description,
domain,
)
return {
"method": method,
"site": site,
"domain": domain,
"title": title,
"description": description,
"image": image,
"favicon": favicon,
"url": url,
"type": "rich",
"html": html,
"css": css,
}
def check(self, base, ref, expected, strict=False):
self.assertEqual(expected, urijoin(base, ref, strict))
# base as bytes, ref as str
self.assertEqual(expected, urijoin(base.encode(), ref, strict))
# base as str, ref as bytes
self.assertEqual(expected, urijoin(base, ref.encode(), strict))
# both base and ref as bytes
self.assertEqual(
expected.encode(),
urijoin(base.encode(), ref.encode(), strict)
)
def url_fix(self, url):
if url == "": # Null case
rv = url
elif url.startswith('data:'): # data uri, not actually a link, leave it alone
# e.g. data:text/html,<script>alert('hi');</script>
rv = url
elif urisplit(url)[0]: # absolute / has scheme
# e.g http://example.com/path
rv = self.proxy +'/' + url
elif urisplit(url)[1]: # protocol relatives prefixed with '//' / no scheme but has authority
# e.g. //example.com/path
# PRs should be able to be either http or https. In practice some sites are turning off
# support for http, so just force everything through https
rv = self.proxy + '/https://' + url.lstrip('/')
elif not urisplit(url)[1] and urisplit(url)[2]: # relative / no authority but path
# e.g. ../path
rv = self.proxy + '/' + urijoin(self.host, url)
elif url == '#' or (not urisplit(url)[2] and not urisplit(url)[3] and urisplit(url)[4]): # fragment
# e.g. #id-12345
# fragments are left alone
rv = url
else:
click.echo('\n\nUnknown url protocol with url: %s' % url)
rv = url
return rv
def get_redirect(url):
# simple case: it's already in the dict
url = absolute_from_relative_url(url)
if url in redirects:
return redirects[url]
# Try looking it up without the fragment
defrag_url = uritools.uridefrag(url).uri
fragment = uritools.uridefrag(url).uri
if fragment:
if defrag_url in redirects:
return uritools.urijoin(redirects[defrag_url], '#'+fragment)
# Try fixing http/https to match the TOC
url_parts = uritools.urisplit(url)
base_url_parts = uritools.urisplit(redirects[base_url])
fixed_scheme_url = uritools.uriunsplit(
list(base_url_parts)[:1] + list(url_parts)[1:])
if fixed_scheme_url in redirects:
return redirects[fixed_scheme_url]
# if same domain, try scraping it
if url_parts.host == base_url_parts.host:
try:
print(f"Scraping url for get_redirect: {url}")
scraper_result = scraper.scrape(
url, wait_for_selector=config['post_body_selector'])
redirects[url] = scraper_result['final_url']
# TODO: Make store this scraped result in the book as well?
return redirects[url]
except (urllib.error.URLError, ssl.SSLError):
return url # TODO: Could return '' or something but for now leaving it seems fine
# else, couldn't find it, so leave it alone.
return url
def dereference(self, uri: str) -> dict:
"""Return JSON object corresponding to resolved URI reference
:param uri: URI string
"""
reference_path = urijoin(self.scope_uri, uri)
return self.registry.load_uri(reference_path)
def follow_uri(self, uri: str) -> 'Context':
"""Return new Context corresponding to scope after following uri
:param uri: URI string
"""
new_uri = urijoin(self.scope_uri, uri)
return self.__class__(new_uri, self.registry)
def _get_or_create_image_file(self, path, data=None):
what = imghdr.what(path, data)
if not what:
raise ValueError("Unknown image type")
if not data:
with open(path, "rb") as f:
data = f.read()
digest, width, height = hashlib.md5(data).hexdigest(), None, None
try:
if what == "png":
width, height = get_image_size_png(data)
elif what == "gif":
width, height = get_image_size_gif(data)
elif what == "jpeg":
width, height = get_image_size_jpeg(data)
except Exception as e:
logger.error("Error getting image size for %r: %r", path, e)
if width and height:
name = "%s-%dx%d.%s" % (digest, width, height, what)
else:
name = f"{digest}.{what}"
image_path = self._image_dir / name
if not image_path.is_file():
logger.info(f"Creating file {image_path.as_uri()}")
image_path.write_bytes(data)
return uritools.urijoin(self._base_uri, name)
def _get_or_create_image_file(self, path, data=None):
what = imghdr.what(path, data)
if not what:
raise ValueError('Unknown image type')
if not data:
data = open(path).read()
digest, width, height = hashlib.md5(data).hexdigest(), None, None
try:
if what == 'png':
width, height = get_image_size_png(data)
elif what == 'gif':
width, height = get_image_size_gif(data)
elif what == 'jpeg':
width, height = get_image_size_jpeg(data)
except Exception as e:
logger.error('Error getting image size for %r: %r', path, e)
if width and height:
name = '%s-%dx%d.%s' % (digest, width, height, what)
else:
name = '%s.%s' % (digest, what)
dest = os.path.join(self.image_dir, name)
if not os.path.isfile(dest):
logger.info('Creating file %s', dest)
with open(dest, 'wb') as fh:
fh.write(data)
return uritools.urijoin(self.base_uri, name)
def bootstrap(self, refSchemaTuple = tuple()):
(id2ElemId , keyRefs , refSchemaCache) = refSchemaTuple
keyList = keyRefs[self.triggerAttribute]
errors = []
# Saving the unique locations
# based on information from FeatureLoc elems
for loc in keyList:
fk_defs = loc.context[self.triggerAttribute]
fk_defs_gid = str(id(loc.context))
#fk_defs_gid = loc.path
for fk_loc_i, p_FK_decl in enumerate(fk_defs):
fk_loc_id = fk_defs_gid + '_' + str(fk_loc_i)
ref_schema_id = p_FK_decl['schema_id']
if uritools.isabsuri(self.schemaURI):
abs_ref_schema_id = uritools.urijoin(self.schemaURI,ref_schema_id)
else:
abs_ref_schema_id = ref_schema_id
if abs_ref_schema_id not in refSchemaCache:
errors.append({
'reason': 'fk_no_schema',
'description': "No schema with {0} id, required by {1} ({2})".format(abs_ref_schema_id,self.jsonSchemaSource,self.schemaURI)
})
fk_members = p_FK_decl.get('members',[])
fkLoc = FKLoc(schemaURI=self.schemaURI,refSchemaURI=abs_ref_schema_id,path=loc.path+'/'+str(fk_loc_i),values=list())
fk_id = abs_ref_schema_id
fkDefH = self.FKWorld.setdefault(fk_id,{})
# This control is here for same primary key referenced from multiple cases
fkDefH[fk_loc_id] = FKDef(fkLoc=fkLoc,members=fk_members)
return errors
def recursive_traversal(
schema_object: Union[Dict[str, Any],
List[Any]]) -> Dict[str, Any]:
retrieved_object = None
if isinstance(schema_object, dict):
for k, v in schema_object.items():
if isinstance(v, dict) or isinstance(v, list):
recursive_traversal(v)
elif k == "$ref":
# Processes case
if v.startswith("processes/"):
pass
else:
if not uritools.urisplit(v).gethost():
v = uritools.urijoin(url, v)
retrieved_object = self._resolve_schema(v)
if retrieved_object is not None:
schema_object.update(retrieved_object)
del schema_object["$ref"]
if isinstance(schema_object, list):
for i in schema_object:
if isinstance(i, dict) or isinstance(i, list):
recursive_traversal(i)
def _cleanup_images(self):
logger.info("Cleaning up image directory")
with self._connect() as c:
uris = set(schema.get_image_uris(c))
for image_path in self._image_dir.glob("**/*"):
if uritools.urijoin(self._base_uri, image_path.name) not in uris:
logger.info(f"Deleting file {image_path.as_uri()}")
image_path.unlink()
def get_or_create_image_file(self, path, data=None):
what = imghdr.what(path, data)
if not what:
raise ValueError('Unknown image type')
if not data:
data = open(path).read()
name = hashlib.md5(data).hexdigest() + '.' + what
path = os.path.join(self.image_dir, name)
get_or_create_file(str(path), True, data)
return uritools.urijoin(self.base_uri, name)
def _get(self, path, params=None):
url = uritools.urijoin(self.base_url, path)
retries = self.retries
timeout = self.timeout
while True:
try:
return self.session.get(url, params=params, timeout=timeout)
except requests.exceptions.ConnectionError as e:
if not retries:
raise e
logger.warn('Error connecting to the Internet Archive: %s', e)
retries -= 1
def cleanup(self):
logger.info('Cleaning up image directory')
uris = set()
for track in self.library.begin():
if track.album and track.album.images:
uris.update(track.album.images)
self.library.close()
for root, _, files in os.walk(self.image_dir):
for name in files:
if uritools.urijoin(self.base_uri, name) not in uris:
path = os.path.join(root, name)
logger.info('Deleting file %s', path)
os.remove(path)
def _cleanup(self):
logger.info('Cleaning up image directory')
uris = set()
for track in self.library.begin():
if track.album and track.album.images:
uris.update(track.album.images)
self.library.close()
for root, _, files in os.walk(self.image_dir):
for name in files:
if uritools.urijoin(self.base_uri, name) not in uris:
path = os.path.join(root, name)
logger.info('Deleting file %s', path)
os.remove(path)
def build_ref(type_entry):
if type_entry["ns"] == "common":
defs_path = common_config["defs_path"]
elif type_entry["ns"] == "cbr_process":
defs_path = cbr_process_config["defs_path"]
return uritools.urijoin(
_build_rel_path(
os.path.join(cbr_process_config["cbr_process_collection_dir"],
"dummy"),
defs_path,
build_params["root_build_dir"],
),
f"#{type_entry['def']}",
)
def findFKs(jsonSchema, jsonSchemaURI, prefix=""):
FKs = []
if isinstance(jsonSchema, dict):
# First, this level's foreign keys
isArray = False
if 'items' in jsonSchema and isinstance(jsonSchema['items'], dict):
jsonSchema = jsonSchema['items']
isArray = True
if prefix != '':
prefix += '[]'
if 'foreign_keys' in jsonSchema and isinstance(
jsonSchema['foreign_keys'], (list, tuple)):
for fk_def in jsonSchema['foreign_keys']:
# Only valid declarations are taken into account
if isinstance(
fk_def, dict
) and 'schema_id' in fk_def and 'members' in fk_def:
ref_schema_id = fk_def['schema_id']
members = fk_def['members']
if isinstance(members, (list, tuple)):
# Translating to absolute URI (in case it is relative)
abs_ref_schema_id = uritools.urijoin(
jsonSchemaURI, ref_schema_id)
# Translating the paths
components = tuple(
map(
lambda component: prefix + '.' + component
if component not in ['.', ''] else prefix,
members))
FKs.append((abs_ref_schema_id, components))
# Then, the foreign keys inside sublevels
if 'properties' in jsonSchema and isinstance(jsonSchema['properties'],
dict):
if prefix != '':
prefix += '.'
p = jsonSchema['properties']
for k, subSchema in p.items():
FKs.extend(findFKs(subSchema, jsonSchemaURI, prefix + k))
return FKs
def crawl(self, pages, depth=2):
# iterate for each depth
for i in range(depth):
# use a set to prevent repeats
newpages = set()
# for each page in pages list
for page in pages:
c = None
try:
c = urllib2.urlopen(page)
except:
print("Could not open {}".format(page))
if not c:
continue
# after retrieving the html
soup = BeautifulSoup(c.read())
# index page (as in add all the words into the words table)
self.addtoindex(page, soup)
# iterate through all the links in the page
links = soup('a')
for link in links:
if ('href' in dict(link.attrs)):
url = urijoin(page, link['href'])
# check for quotes
if url.find("'") != -1: continue
# remove location fragments
url = url.split('#')[0]
# is the result a valid url
if url[0:4] == 'http' and not self.isindexed(url):
newpages.add(url)
# create a link between the two pages
linkText = self.gettextonly(link)
self.addlinkref(page, url, linkText)
# store the db
self.dbcommit()
# recurse
pages = newpages
def custom_patterns(parser, results):
a_tags = []
elements = cxpath_href(parser.doc)
for element in elements:
href = element.get('href')
# skip links to invalid hrefs
if not href:
continue
if href.startswith('javascript:'):
continue
# canonicalize the href
href = uritools.urijoin(parser.url, href)
href = uritools.uridefrag(href).uri
href = clean_url(href)
try:
href = url_normalize.url_normalize(href)
except Exception as e:
logging.debug('failed to normalize url ' + href)
a_tag = {'href': href}
# extract the rel
if element.get('rel') is not None:
rel = element.get('rel').strip()
if len(rel) > 0:
a_tag['rel'] = rel
# extract the text_content
if element.text_content() is not None:
text_content = element.text_content().strip()
if len(text_content) > 0:
a_tag['text_content'] = text_content
# add to list
a_tags.append(a_tag)
# remove duplicate tags
a_tags.sort(key=lambda x: x['href'])
a_tags_dedup = []
prev_href = None
for a_tag in a_tags:
if a_tag['href'] != prev_href:
prev_href = a_tag['href']
a_tags_dedup.append(a_tag)
results.append({'value': a_tags_dedup})
def find_linked_extras(posts):
extra_page_urls = []
for post in posts:
for body_soup in post['body_soups']:
for element in body_soup.select('[href]'):
full_href = uritools.urijoin(
post['final_url'], element['href'])
defragged_href = uritools.uridefrag(full_href).uri
if not url_is_included(defragged_href):
href_parts = uritools.urisplit(full_href)
base_url_parts = uritools.urisplit(redirects[base_url])
if href_parts.host == base_url_parts.host: # Never try to include linked pages from other domains
if defragged_href not in extra_page_urls:
# TODO: defragged, or full? Uniqueness or is the fragment important?
extra_page_urls.append(defragged_href)
return extra_page_urls
def _traverse_dict(schemaURI, j, jp="", fragment=None):
# Pre-processing
newPartialSchemaURI = j.get('$id')
if newPartialSchemaURI:
# Computing the absolute schema URI
if uritools.isabsuri(schemaURI):
newSchemaURI , uriFragment = uritools.uridefrag(uritools.urijoin(schemaURI,newPartialSchemaURI))
else:
newSchemaURI , uriFragment = uritools.uridefrag(newPartialSchemaURI)
else:
newSchemaURI = schemaURI
# Are we jumping to a different place?
if newSchemaURI == schemaURI:
theId = id(j)
theIdStr = str(theId)
# Does the dictionary contain a '$ref'?
isRef = REF_FEATURE in j
for k,v in j.items():
# Following JSON reference standards, we have to
# ignore other keys when there is a $ref one
# https://tools.ietf.org/html/draft-pbryan-zyp-json-ref-03#section-3
if isRef and (k != REF_FEATURE):
continue
elemId = theIdStr + ':' + k
elemPath = jp + '/' + k
jp2val[elemPath] = elemId
# Is the key among the "special ones"?
if k in keySet:
# Saving the correspondence from Python address
# to unique id of the feature
id2ElemId.setdefault(theId,{})[k] = [ elemId ]
keyRefs.setdefault(k,[]).append(FeatureLoc(schemaURI=schemaURI,fragment=fragment,path=elemPath,context=j,id=elemId))
if isinstance(v,dict):
# Fragment must not be propagated to children
_traverse_dict(schemaURI,v,jp=elemPath)
elif isinstance(v,list):
_traverse_list(schemaURI,v,jp=elemPath)
else:
traverseJSONSchema(j,schemaURI=newSchemaURI,fragment=uriFragment,keys=keys,refSchemaListSet=refSchemaListSet)
def _build_abs_schema_uri(rel_path: str, cb_schemas_base_url: str) -> str:
"""
Form an absolute URI from a relative path.
It takes the :code:`common.cb_schemas_base_url` property provided by the build
configuration file.
:param str rel_path: A path relative to :code:`build_params.root_build_dir`.
:param str cb_schemas_base_url: The base URL pointing to the directory where the
Cookbase Schemas are to be located.
:return: An absolute URI intended to unambiguously refer to the given path.
:rtype: str
"""
base_url_splits = uritools.urisplit(cb_schemas_base_url)
abs_path = base_url_splits.path or "/"
return uritools.urijoin(cb_schemas_base_url,
os.path.join(abs_path, rel_path))
def meta_tags(url, timeout=15, html=None):
if not html:
html = get(url, timeout=timeout).text
d = pq(html)
favicon = d('link[rel="shortcut icon"]').attr("href")
if not favicon:
favicon_url = urijoin(url, "/favicon.ico")
r = requests.head(favicon_url)
if r.status_code == 200:
favicon = favicon_url
return {
"title": d('meta[name="title"]').attr("content") or d("title").text(),
"description": d('meta[name="description"]').attr("content"),
"image": d('meta[name="image"]').attr("content"),
"favicon": favicon,
"url": d('meta[name="canonical"]').attr("content")
or d('meta[name="url"]').attr("content"),
"keywords": d('meta[name="keywords"]').attr("content"),
}
def _delete_test_dag_ids(self, remote_uri: str,
test_dag_ids: List[str]) -> Dict[str, str]:
"""
Calls an endpoint on Airflow where the etsy-test-api service is active.
This call will delete test_dag_ids from the airflow DB.
"""
delete_uri = uritools.urijoin(remote_uri, '/etsy-test-api/delete')
response = requests.post(delete_uri,
data=json.dumps({'test_dags': test_dag_ids}),
headers={'Content-Type': 'application/json'},
timeout=600)
results = {}
if response.status_code == 200:
try:
results = response.json()
except JSONDecodeError as e:
logging.info(
'Received JSONDecodeError error on call to {}. Full error: {}'
.format(delete_uri, e))
return results
def validate(self, validator, fp_def, value, schema):
if fp_def and isinstance(fp_def, str):
fp_loc_id = id(schema)
# Getting the absolute schema id and the route
if uritools.isabsuri(self.schemaURI):
abs_ref_schema_id, rel_json_pointer = uritools.uridefrag(
uritools.urijoin(self.schemaURI, fp_def))
else:
abs_ref_schema_id, rel_json_pointer = uritools.uridefrag(
fp_def)
fpDef = self.FPWorld.setdefault(abs_ref_schema_id,
{}).get(fp_loc_id)
# And getting the foreign property definition
if fpDef is None:
fpDef = FPDef(schemaURI=self.schemaURI,
refSchemaURI=abs_ref_schema_id,
path='(unknown {})'.format(fp_loc_id),
refPath=rel_json_pointer,
values=list())
self.FPWorld[abs_ref_schema_id][fp_loc_id] = fpDef
obtainedValues = [(value, )]
isAtomicValue = len(obtainedValues) == 1 and len(
obtainedValues[0]) == 1 and isinstance(
obtainedValues[0][0], ALLOWED_ATOMIC_VALUE_TYPES)
if isAtomicValue:
theValues = [obtainedValues[0][0]]
else:
theValues = UniqueKey.GenKeyStrings(obtainedValues)
fpVals = fpDef.values
# Second pass will do the validation
for theValue in theValues:
fpVals.append(FPVal(where=self.currentJSONFile,
value=theValue))
def bootstrap(self, refSchemaTuple=tuple()):
(id2ElemId, keyRefs, refSchemaCache) = refSchemaTuple
keyList = keyRefs[self.triggerAttribute]
errors = []
# Saving the unique locations
# based on information from FeatureLoc elems
for loc in keyList:
fp_def = loc.context[self.triggerAttribute]
fp_loc_id = id(loc.context)
# Getting the absolute schema id and the route
if uritools.isabsuri(self.schemaURI):
abs_ref_schema_id, rel_json_pointer = uritools.uridefrag(
uritools.urijoin(self.schemaURI, fp_def))
else:
abs_ref_schema_id, rel_json_pointer = uritools.uridefrag(
fp_def)
if abs_ref_schema_id not in refSchemaCache:
errors.append({
'reason':
'fp_no_schema',
'description':
"No schema with {0} id, required by {1} ({2})".format(
abs_ref_schema_id, self.jsonSchemaSource,
self.schemaURI)
})
fpDefH = self.FPWorld.setdefault(abs_ref_schema_id, {})
# This control is here for same primary key referenced from multiple cases
fpDefH[fp_loc_id] = FPDef(schemaURI=self.schemaURI,
refSchemaURI=abs_ref_schema_id,
path=loc.path,
refPath=rel_json_pointer,
values=list())
return errors
def validate(self,validator,fk_defs,value,schema):
if fk_defs and isinstance(fk_defs,(list,tuple)):
fk_defs_gid = str(id(schema))
for fk_loc_i, p_FK_decl in enumerate(fk_defs):
fk_loc_id = fk_defs_gid + '_' + str(fk_loc_i)
ref_schema_id = p_FK_decl['schema_id']
if uritools.isabsuri(self.schemaURI):
abs_ref_schema_id = uritools.urijoin(self.schemaURI,ref_schema_id)
else:
abs_ref_schema_id = ref_schema_id
fk_members = p_FK_decl.get('members',[])
if isinstance(fk_members,list):
obtainedValues = PrimaryKey.GetKeyValues(value,fk_members)
else:
obtainedValues = [(value,)]
isAtomicValue = len(obtainedValues) == 1 and len(obtainedValues[0]) == 1 and isinstance(obtainedValues[0][0], ALLOWED_ATOMIC_VALUE_TYPES)
if isAtomicValue:
theValues = [ obtainedValues[0][0] ]
else:
theValues = PrimaryKey.GenKeyStrings(obtainedValues)
# Group the values to be checked
#fk_id = id(p_FK_decl) # id(schema)
fk_id = abs_ref_schema_id
# The common dictionary for this declaration where all the FK values are kept
fkDef = self.FKWorld.setdefault(fk_id,{}).setdefault(fk_loc_id,FKDef(fkLoc=FKLoc(schemaURI=self.schemaURI,refSchemaURI=abs_ref_schema_id,path='(unknown {})'.format(fk_loc_id),values=list()),members=fk_members))
fkLoc = fkDef.fkLoc
fkVals = fkLoc.values
# Second pass will do the validation
for theValue in theValues:
fkVals.append(FKVal(where=self.currentJSONFile,value=theValue))
def parse_lecture(lect_link, name):
lect_html = requests.get(urijoin(base, lect_link.get("href"))).text
lect = BeautifulSoup(lect_html, features="html.parser")
lect_times = (lect.find(
"div",
class_="tudbox").find(lambda tag: tag.name == "td" and "Zeit/Ort:" in
tag.text).find_next_sibling("td").text.strip())
dates = [
dict(
name=lect_link.text,
vert_name=name,
week=date["week"],
time=date["time"],
day=date["day"],
type="lect",
) for date in parse_dates(lect_times)
]
tuts = None
tuts_row = lect.find("div", class_="tudbox").find(
lambda tag: tag.name == "td" and "Übungen:" in tag.text)
if tuts_row:
dates += [
dict(
name=lect_link.text,
vert_name=name,
date=date,
week=date["week"],
time=date["time"],
day=date["day"],
type="tut",
) for date in parse_dates(
list(tuts_row.find_next_sibling("td").findAll("td"))[-1].text)
]
return dates
def __init__(
self,
mongodb_url: str,
db_type: str = DBTypes.MONGODB,
db_name: str = "cookbase",
):
"""Constructor method."""
if db_type == self.DBTypes.MONGODB:
self._default_db_id: str = f"{db_type}:{db_name}"
try:
client = pymongo.MongoClient(uritools.urijoin(mongodb_url, db_name))
client.admin.command("ismaster")
except pymongo.errors.PyMongoError:
import sys
raise DBClientConnectionError(self._default_db_id).with_traceback(
sys.exc_info()[2]
)
self._default_db: Any = client[db_name]
self._connections: Dict[str, Any] = {self._default_db_id: self._default_db}
else:
raise InvalidDBTypeError(db_type)
def _get_test_dag_ids(self,
remote_uri: str,
ttl_days: int = 7) -> List[str]:
"""
Calls an endpoint on Airflow where the etsy-test-api service is active.
If the service does not exist (e.g., on production airflows), this call
will return an empty list.
"""
list_uri = uritools.urijoin(
remote_uri,
'/etsy-test-api/list_test_dags?ttl_days={}'.format(ttl_days))
response = requests.get(list_uri, timeout=10)
test_dag_ids = []
if response.status_code == 200:
try:
payload = response.json()
test_dag_ids = payload.get(
'test_dags') if 'test_dags' in payload else []
except JSONDecodeError as _e:
logging.info(
'Received JSONDecodeError on call to {}.'.format(list_uri))
logging.info(
'Note: This is okay on Production airflow instances.')
return test_dag_ids
def geturi(self, filename):
return uritools.urijoin(self._base_uri, filename)
def geturl(self, identifier, filename=None):
if filename:
path = identifier + '/' + uritools.uriencode(filename)
else:
path = identifier + '/'
return uritools.urijoin(self.base_url, '/download/' + path)
def to_canonical_uri(current_base_uri, uri):
return urijoin(current_base_uri, uri)
