def analyze(cls, something):
""" Accept a given input (e.g. a URL, file path, or file handle
and determine how to normalize it into an ``Ingestor`` while
generating metadata. """
if isinstance(something, cls):
return (something, )
if isinstance(something, basestring):
# Treat strings as paths or URLs
url = urlparse(something)
if url.scheme.lower() in ['http', 'https']:
something = requests.get(something)
elif url.scheme.lower() in ['file', '']:
finalpath = url.path
if osname == 'nt':
finalpath = finalpath[1:]
upath = fullpath(finalpath)
if path.isdir(upath):
return (cls(file_name=f) for f in directory_files(upath))
return (cls(file_name=upath), )
# Python requests
if isinstance(something, requests.Response):
fd = StringIO(something.content)
return (cls(file_obj=fd,
meta={
'http_status': something.status_code,
'http_headers': clean_headers(something.headers),
'source_url': something.url
}), )
if isinstance(something, HTTPResponse):
# Can't tell the URL for HTTPResponses
return (cls(file_obj=something,
meta={
'http_status': something.status,
'http_headers':
clean_headers(something.getheaders()),
'source_url': something.url
}), )
elif hasattr(something, 'geturl') and hasattr(something, 'info'):
# assume urllib or urllib2
return (cls(file_obj=something,
meta={
'http_status': something.getcode(),
'http_headers': clean_headers(something.headers),
'source_url': something.url
}), )
elif hasattr(something, 'read'):
# Fileobj will be a bit bland
return (cls(file_obj=something), )
return []
def analyze(cls, something):
""" Accept a given input (e.g. a URL, file path, or file handle
and determine how to normalize it into an ``Ingestor`` while
generating metadata. """
if isinstance(something, cls):
return (something, )
if isinstance(something, basestring):
# Treat strings as paths or URLs
url = urlparse(something)
if url.scheme.lower() in ['http', 'https']:
something = requests.get(something)
elif url.scheme.lower() in ['file', '']:
finalpath = url.path
if osname == 'nt':
finalpath = finalpath[1:]
upath = fullpath(finalpath)
if path.isdir(upath):
return (cls(file_name=f) for f in directory_files(upath))
return (cls(file_name=upath),)
# Python requests
if isinstance(something, requests.Response):
fd = StringIO(something.content)
return (cls(file_obj=fd, meta={
'http_status': something.status_code,
'http_headers': clean_headers(something.headers),
'source_url': something.url
}), )
if isinstance(something, HTTPResponse):
# Can't tell the URL for HTTPResponses
return (cls(file_obj=something, meta={
'http_status': something.status,
'http_headers': clean_headers(something.getheaders()),
'source_url': something.url
}), )
elif hasattr(something, 'geturl') and hasattr(something, 'info'):
# assume urllib or urllib2
return (cls(file_obj=something, meta={
'http_status': something.getcode(),
'http_headers': clean_headers(something.headers),
'source_url': something.url
}), )
elif hasattr(something, 'read'):
# Fileobj will be a bit bland
return (cls(file_obj=something), )
return []
def __init__(self, path=None, **kwargs):
self.path = fullpath(path)
if os.path.exists(path) and not os.path.isdir(path):
raise ValueError('Not a directory: %s' % path)
def __init__(self, path=None, **kwargs):
self.path = fullpath(path)
if os.path.exists(path) and not os.path.isdir(path):
raise ValueError('Not a directory: %s' % path)
