class Segments(object):
def __init__(self, base, path, segments, defaults):
self.base = PURL(base, path=path)
self.segments = OrderedDict(zip(segments, defaults))
def build(self):
segments = self.base.path_segments() + tuple(self.segments.values())
url = self.base.path_segments(segments)
return url
def __str__(self):
return self.build().as_string()
def _get_segment(self, segment):
return self.segments[segment]
def _set_segment(self, segment, value):
self.segments[segment] = value
@classmethod
def _segment(cls, segment):
return property(
fget=lambda x: cls._get_segment(x, segment),
fset=lambda x, v: cls._set_segment(x, segment, v),
)
class Segments(object):
def __init__(self, base, path, segments, defaults):
self.base = PURL(base, path=path)
self.segments = OrderedDict(zip(segments, defaults))
def build(self):
segments = self.base.path_segments() + tuple(self.segments.values())
url = self.base.path_segments(segments)
return url
def __str__(self):
return self.build().as_string()
def _get_segment(self, segment):
return self.segments[segment]
def _set_segment(self, segment, value):
self.segments[segment] = value
@classmethod
def _segment(cls, segment):
return property(
fget=lambda x: cls._get_segment(x, segment),
fset=lambda x, v: cls._set_segment(x, segment, v),
)
def cached_path(args, url, rel):
if url.startswith('/'):
return url
url_ = URL(url)
cached = args.data_file('edmond', url_.path_segments()[-1])
if not cached.exists():
fname, headers = urlretrieve(url, '%s' % cached)
return str(cached)
def cached_path(args, url, rel):
if url.startswith('/'):
return url
url_ = URL(url)
cached = args.data_file('edmond', url_.path_segments()[-1])
if not cached.exists():
fname, headers = urlretrieve(url, '%s' % cached)
return str(cached)
def info(self, url):
"""Interface method to be called when processing new images.
This method ties together the DataProvider workflow.
"""
url = URL(url)
return self.postprocess(
self.info_for_id(self.id_from_url(url, url.host(), url.path_segments())))
class Segments(object):
"""
URL segment handler, not intended for direct use. The URL is constructed by
joining base, path and segments.
"""
def __init__(self, base, path, segments, defaults):
# Preserve the base URL
self.base = PURL(base, path=path)
# Map the segments and defaults lists to an ordered dict
self.segments = OrderedDict(zip(segments, defaults))
def build(self):
# Join base segments and segments
segments = self.base.path_segments() + tuple(self.segments.values())
# Create a new URL with the segments replaced
url = self.base.path_segments(segments)
return url
def full_path(self):
full_path = self.build().as_string()
full_path = full_path.replace(self.base.host(), '')
full_path = full_path.replace(self.base.scheme(), '')
return full_path[4:]
def __str__(self):
return self.build().as_string()
def _get_segment(self, segment):
return self.segments[segment]
def _set_segment(self, segment, value):
self.segments[segment] = value
@classmethod
def _segment(cls, segment):
"""
Returns a property capable of setting and getting a segment.
"""
return property(
fget=lambda x: cls._get_segment(x, segment),
fset=lambda x, v: cls._set_segment(x, segment, v),
)
def get_links(instr):
"""
Given a link-value, generate individual links as dict.
"""
if instr:
for link in [h.strip() for h in link_splitter.findall(instr)]:
url, params = link.split(">", 1)
url = URL(url[1:])
fname = url.path_segments()[-1] if url.path_segments() else ''
info = {
'url': url.as_string(),
'ext': fname.split('.', 1)[1] if '.' in fname else None,
'rel': 'related',
'type': 'application/octet-stream'}
for param in _splitstring(params, PARAMETER, "\s*;\s*"):
try:
a, v = param.split("=", 1)
info[a.lower()] = _unquotestring(v)
except ValueError: # pragma: no cover
info[param.lower()] = None
yield info
def get(self, url, **kw):
url = URL(url)
content = get_resource(
url.path_segment(-1) if url.path_segments() else 'dataset', self.prefix)
if content:
return MagicMock(
mimetype='application/rdf+xml',
content=content,
links=[dict(
url='{0}.html'.format(url.as_string),
ext='.html',
rel='alternate',
type='text/html')],
canonical_url=url.as_string())
class Segments(object):
"""
URL segment handler, not intended for direct use. The URL is constructed by
joining base, path and segments.
"""
def __init__(self, base, path, segments, defaults):
# Preserve the base URL
self.base = PURL(base, path=path)
# Map the segments and defaults lists to an ordered dict
self.segments = OrderedDict(zip(segments, defaults))
def build(self):
# Join base segments and segments
segments = self.base.path_segments() + tuple(self.segments.values())
# Create a new URL with the segments replaced
url = self.base.path_segments(segments)
return url
def __str__(self):
return self.build().as_string()
def _get_segment(self, segment):
return self.segments[segment]
def _set_segment(self, segment, value):
self.segments[segment] = value
@classmethod
def _segment(cls, segment):
"""
Returns a property capable of setting and getting a segment.
"""
return property(
fget=lambda x: cls._get_segment(x, segment),
fset=lambda x, v: cls._set_segment(x, segment, v),
)
def __call__(self, index, row):
if index == 0:
self.cols = {col: i for i, col in enumerate(row)}
return row
url = URL(row[self.cols['src']])
try:
for filename in url.path_segments():
if filename in self.md:
if self.md[filename].get('source_url'):
row[self.cols['source']] = self.md[filename]['source_url']
self.count += 1
break
except IndexError:
pass
return row
def url_parts(self, url):
url = URL(url)
return url, url.host(), url.path_segments()
def get_image_info(img):
for field in ['source', 'source_url', 'id']:
for provider in PROVIDERS:
url = URL(img[field])
if provider.id_from_url(url, url.host(), url.path_segments()):
return provider.info(img[field])
inline_url = URL().scheme('https').domain('www.google.com').path(
'search').query_param('q', 'google')
print(inline_url)
print(inline_url.as_string())
u = URL('postgres://*****:*****@localhost:1234/test?ssl=true')
print(u.scheme())
print(u.host())
print(u.domain())
print(u.username())
print(u.password())
print(u.netloc())
print(u.port())
print(u.path())
print(u.query())
print(u.path_segments())
print(u.query_param('ssl'))
print(u.query_param('ssl', as_list=True))
print(u.query_params())
print(u.has_query_param('ssl'))
print(u.subdomains())
u = URL.from_string('https://github.com/minwook-shin')
print(u.path_segment(0))
new_url = u.add_path_segment('minwook-shin.github.com')
print(new_url.as_string())
from purl import expand
print(expand(u"{/path*}", {'path': ['sub', 'index']}))
