def post_media(request):
"""
Post the media.
Args:
request (HTTPRequest): The HTTP request
Return:
a HTTPResponse
"""
LOGGING.info("Post media")
_doi = get_doi_from_request(request, "media")
if _doi is None:
return get_response("Bad Request - doi not found in URL", 400)
LOGGING.debug("Post media, doi: %s", _doi)
try:
doi_suffix = _doi.split(DOI_PREFIX, 1)[1]
except IndexError:
return get_response(
"Bad Request - wrong prefix, doi should start "
"with %s" % DOI_PREFIX, 400)
if not is_authorized(request, doi_suffix):
return get_response("Unauthorized - insufficient privileges", 403)
url = urljoin(DATACITE_URL, request.get_full_path())
return _post(url, request.body, _get_content_type_header(request))
def post_metadata(request):
"""
Post the metadata.
Args:
request (HTTPRequest): The HTTP request
Return:
a HTTPResponse
"""
LOGGING.info('Post metadata')
try:
_doi = _get_doi_from_xml_body(request.body)
except ET.ParseError as ex:
LOGGING.info('Error parsing xml from users request: %s', ex)
return get_response("Bad Request - error parsing xml: %s" % ex, 400)
if _doi == None:
return get_response("Bad Request - doi not found in XML", 400)
LOGGING.debug('Post metadata, doi: %s', _doi)
try:
doi_suffix = _doi.split(DOI_PREFIX, 1)[1]
except IndexError:
return get_response("Bad Request - wrong prefix, doi should start " \
"with %s" % DOI_PREFIX, 400)
if not is_authorized(request, doi_suffix):
return get_response("Unauthorized - insufficient privileges", 403)
url = urljoin(DATACITE_URL, request.get_full_path())
return _post(url, request.body, _get_content_type_header(request))
def post_doi(request):
"""
Post the DOI.
Args:
request (HTTPRequest): The HTTP request
Return:
a HTTPResponse
"""
LOGGING.info('Post doi')
try:
_doi = _get_doi_from_text_body(request.body)
except IndexError:
return get_response("Bad Request - request body must be exactly two " \
"lines: DOI and URL", 400)
LOGGING.debug('Post doi, doi: %s', _doi)
try:
doi_suffix = _doi.split(DOI_PREFIX, 1)[1]
except IndexError:
return get_response("Bad Request - wrong prefix, doi should start " \
"with %s" % DOI_PREFIX, 400)
if not is_authorized(request, doi_suffix):
return get_response("Unauthorized - insufficient privileges", 403)
url = urljoin(DATACITE_URL, request.get_full_path())
return _post(url, request.body, _get_content_type_header(request))
def _post(url, body, headers):
"""
Send a post request to DataCite.
Args:
url (str): The URL to call
body (str): The data
headers (dict): A dictionary of headers to use
Return:
a HTTPResponse
"""
_set_timeout()
opener = get_opener()
auth_string = (base64.encodestring(DATACITE_USER_NAME + ':'
+ DATACITE_PASSWORD)).rstrip()
headers.update({'Authorization':'Basic ' + auth_string})
# If the request body is a string, urllib2 attempts to concatenate the url,
# body and headers. If the url is unicode, the request body can get
# converted unicode. This has resulted in issues where there are characters
# with diacritic marks in the request body. To avoid these issues the url is
# UTF-8 encoded.
url_encode = url.encode('utf-8')
req = urllib2.Request(url_encode, data=body, headers=headers)
try:
response = opener.open(req)
except (urllib2.HTTPError) as ex:
msg = ex.readlines()
LOGGING.warn('HTTPError error getting %s. %s', url, msg)
return get_response(msg, ex.code)
except (socket.timeout, urllib2.URLError) as ex:
LOGGING.warn('Timeout or URLError error getting %s. %s', url, ex.reason)
return get_response(ex.reason, 500)
except (SSLError) as ex:
LOGGING.warn('SSLError error getting %s. %s', url, ex)
return get_response(ex, 500)
except UnicodeDecodeError as ex:
LOGGING.info('UnicodeDecodeError error getting %s. %s', url, ex)
return get_response(ex, 500)
finally:
_close(opener)
if response.headers.has_key('Content-Type'):
ret_response = HttpResponse(content_type=
response.headers.get('Content-Type'))
else:
ret_response = HttpResponse()
ret_response.status_code = response.code
ret_response.reason_phrase = response.msg
# pylint: disable=maybe-no-member
ret_response.writelines(response.readlines())
if response.headers.has_key('location'):
ret_response.setdefault('Location', response.headers.get('location'))
return ret_response
def get(request_method, url, headers):
"""
Send a get or head request to DataCite.
Args:
request_method (str): This should be 'GET' or 'HEAD'
url (str): The URL to call
headers (dict): A dictionary of headers to use
Return:
a HTTPResponse
"""
LOGGING.info("get(%s,%s,%s)", request_method, url, headers)
_set_timeout()
opener = get_opener()
auth_string = ((base64.encodebytes(
(DATACITE_USER_NAME + ":" +
DATACITE_PASSWORD).encode())).decode("utf-8").rstrip())
headers.update({"Authorization": "Basic " + auth_string})
req = urllib.request.Request(url, data=None, headers=headers)
if request_method == "HEAD":
req.get_method = lambda: "HEAD"
try:
response = opener.open(req)
except urllib.error.HTTPError as ex:
msg = ex.readlines()
if ex.code in [404, 410]:
LOGGING.info("HTTPError error getting %s. %s", url, msg)
else:
LOGGING.warning("HTTPError error getting %s. %s", url, msg)
return get_response(msg, ex.code)
except (socket.timeout, urllib.error.URLError) as ex:
LOGGING.warning("Timeout or URLError error getting %s. %s", url,
ex.reason)
if isinstance(ex.reason, Exception):
ex = ex.reason
LOGGING.warning("Nested exception %s. %s", ex, ex.reason)
return get_response(ex.reason, 500)
except SSLError as ex:
LOGGING.warning("SSLError error getting %s. %s", url, ex)
return get_response(ex, 500)
finally:
_close(opener)
if "Content-Type" in response.headers:
ret_response = HttpResponse(
content_type=response.headers.get("Content-Type"))
else:
ret_response = HttpResponse()
ret_response.status_code = response.code
ret_response.reason_phrase = response.msg
# pylint: disable=maybe-no-member
ret_response.writelines(response.readlines())
return ret_response
def get(request_method, url, headers):
"""
Send a get or head request to DataCite.
Args:
request_method (str): This should be 'GET' or 'HEAD'
url (str): The URL to call
headers (dict): A dictionary of headers to use
Return:
a HTTPResponse
"""
LOGGING.info('get(%s,%s,%s)', request_method, url, headers)
_set_timeout()
opener = get_opener()
auth_string = (base64.encodestring(DATACITE_USER_NAME + ':' +
DATACITE_PASSWORD)).rstrip()
headers.update({'Authorization': 'Basic ' + auth_string})
req = urllib2.Request(url, data=None, headers=headers)
if request_method == "HEAD":
req.get_method = lambda: 'HEAD'
try:
response = opener.open(req)
except (urllib2.HTTPError) as ex:
msg = ex.readlines()
if ex.code in [404, 410]:
LOGGING.info('HTTPError error getting %s. %s', url, msg)
else:
LOGGING.warn('HTTPError error getting %s. %s', url, msg)
return get_response(msg, ex.code)
except (socket.timeout, urllib2.URLError) as ex:
LOGGING.warn('Timeout or URLError error getting %s. %s', url,
ex.reason)
return get_response(ex.reason, 500)
except (SSLError) as ex:
LOGGING.warn('SSLError error getting %s. %s', url, ex)
return get_response(ex, 500)
finally:
_close(opener)
if response.headers.has_key('Content-Type'):
ret_response = HttpResponse(
content_type=response.headers.get('Content-Type'))
else:
ret_response = HttpResponse()
ret_response.status_code = response.code
ret_response.reason_phrase = response.msg
# pylint: disable=maybe-no-member
ret_response.writelines(response.readlines())
return ret_response
def get(request_method, url, headers):
"""
Send a get or head request to DataCite.
Args:
request_method (str): This should be 'GET' or 'HEAD'
url (str): The URL to call
headers (dict): A dictionary of headers to use
Return:
a HTTPResponse
"""
LOGGING.info('get(%s,%s,%s)', request_method, url, headers)
_set_timeout()
opener = get_opener()
auth_string = (base64.encodestring(DATACITE_USER_NAME + ':'
+ DATACITE_PASSWORD)).rstrip()
headers.update({'Authorization':'Basic ' + auth_string})
req = urllib2.Request(url, data=None, headers=headers)
if request_method == "HEAD":
req.get_method = lambda: 'HEAD'
try:
response = opener.open(req)
except (urllib2.HTTPError) as ex:
msg = ex.readlines()
if ex.code in [404, 410]:
LOGGING.info('HTTPError error getting %s. %s', url, msg)
else:
LOGGING.warn('HTTPError error getting %s. %s', url, msg)
return get_response(msg, ex.code)
except (socket.timeout, urllib2.URLError) as ex:
LOGGING.warn('Timeout or URLError error getting %s. %s', url, ex.reason)
return get_response(ex.reason, 500)
except (SSLError) as ex:
LOGGING.warn('SSLError error getting %s. %s', url, ex)
return get_response(ex, 500)
finally:
_close(opener)
if response.headers.has_key('Content-Type'):
ret_response = HttpResponse(content_type=
response.headers.get('Content-Type'))
else:
ret_response = HttpResponse()
ret_response.status_code = response.code
ret_response.reason_phrase = response.msg
# pylint: disable=maybe-no-member
ret_response.writelines(response.readlines())
return ret_response
def post_doi(request, method="POST"):
"""
Post the DOI.
Args:
request (HTTPRequest): The HTTP request
Return:
a HTTPResponse
"""
LOGGING.info("Post doi")
try:
_doi = _get_doi_from_text_body(request.body.decode("utf-8"))
except IndexError:
return get_response(
"Bad Request - request body must be exactly two "
"lines: DOI and URL", 400)
LOGGING.debug("Post doi, doi: %s", _doi)
try:
doi_suffix = _doi.split(DOI_PREFIX, 1)[1]
except IndexError:
return get_response(
"Bad Request - wrong prefix, doi should start "
"with %s" % DOI_PREFIX, 400)
try:
# The URL can contain the DOI - check that it matches
url_doi = request.get_full_path().split("doi/", 1)[1]
if len(url_doi) > 0 and url_doi != _doi:
return get_response(
"Bad Request - DOI in URL does not match DOI in request body\n",
400)
except IndexError:
# There is no DOI in the URL, which is fine
pass
if not is_authorized(request, doi_suffix):
return get_response("Unauthorized - insufficient privileges", 403)
url = urljoin(DATACITE_URL, request.get_full_path())
return _post(url,
request.body,
_get_content_type_header(request),
method=method)
def post_metadata(request, method="POST"):
"""
Post the metadata.
Args:
request (HTTPRequest): The HTTP request
Return:
a HTTPResponse
"""
LOGGING.info("Post metadata")
try:
_doi = _get_doi_from_xml_body(request.body)
except ET.ParseError as ex:
LOGGING.info("Error parsing xml from users request: %s", ex)
return get_response("Bad Request - error parsing xml: %s" % ex, 400)
if _doi is None:
return get_response("Bad Request - doi not found in XML", 400)
LOGGING.debug("Post metadata, doi: %s", _doi)
try:
doi_suffix = _doi.split(DOI_PREFIX, 1)[1]
except IndexError:
return get_response(
"Bad Request - wrong prefix, doi should start "
"with %s" % DOI_PREFIX, 400)
try:
# The URL can contain the DOI - check that it matches the metadata
url_doi = request.get_full_path().split("metadata/", 1)[1]
if len(url_doi) > 0 and url_doi != _doi:
return get_response(
"Bad Request - DOI in URL does not match "
"DOI in metadata\n", 400)
except IndexError:
# There is no DOI in the URL, which is fine
pass
if not is_authorized(request, doi_suffix):
return get_response("Unauthorized - insufficient privileges", 403)
url = urljoin(DATACITE_URL, request.get_full_path())
return _post(url,
request.body,
_get_content_type_header(request),
method=method)
def _delete(url):
"""
Send a delete request to DataCite.
Args:
url (str): The URL to call
Return:
a HTTPResponse
"""
_set_timeout()
opener = get_opener()
auth_string = ((base64.encodebytes(
(DATACITE_USER_NAME + ":" +
DATACITE_PASSWORD).encode())).decode("utf-8").rstrip())
headers = {"Authorization": "Basic " + auth_string}
req = urllib.request.Request(url, data=None, headers=headers)
req.get_method = lambda: "DELETE"
try:
response = opener.open(req)
except urllib.error.HTTPError as ex:
msg = ex.readlines()
LOGGING.warning("HTTPError error getting %s. %s", url, msg)
return get_response(msg, ex.code)
except (socket.timeout, urllib.error.URLError) as ex:
LOGGING.warning("Timeout or URLError error getting %s. %s", url,
ex.reason)
return get_response(ex.reason, 500)
except SSLError as ex:
LOGGING.warning("SSLError error getting %s. %s", url, ex)
return get_response(ex, 500)
finally:
_close(opener)
if "Content-Type" in response.headers:
ret_response = HttpResponse(
content_type=response.headers.get("Content-Type"))
else:
ret_response = HttpResponse()
ret_response.status_code = response.code
ret_response.reason_phrase = response.msg
# pylint: disable=maybe-no-member
ret_response.writelines(response.readlines())
return ret_response
def _delete(url):
"""
Send a delete request to DataCite.
Args:
url (str): The URL to call
Return:
a HTTPResponse
"""
_set_timeout()
opener = get_opener()
auth_string = (base64.encodestring(DATACITE_USER_NAME + ':' +
DATACITE_PASSWORD)).rstrip()
headers = {'Authorization': 'Basic ' + auth_string}
req = urllib2.Request(url, data=None, headers=headers)
req.get_method = lambda: 'DELETE'
try:
response = opener.open(req)
except (urllib2.HTTPError) as ex:
msg = ex.readlines()
LOGGING.warn('HTTPError error getting %s. %s', url, msg)
return get_response(msg, ex.code)
except (socket.timeout, urllib2.URLError) as ex:
LOGGING.warn('Timeout or URLError error getting %s. %s', url,
ex.reason)
return get_response(ex.reason, 500)
except (SSLError) as ex:
LOGGING.warn('SSLError error getting %s. %s', url, ex)
return get_response(ex, 500)
finally:
_close(opener)
if response.headers.has_key('Content-Type'):
ret_response = HttpResponse(
content_type=response.headers.get('Content-Type'))
else:
ret_response = HttpResponse()
ret_response.status_code = response.code
ret_response.reason_phrase = response.msg
# pylint: disable=maybe-no-member
ret_response.writelines(response.readlines())
return ret_response
def _delete(url):
"""
Send a delete request to DataCite.
Args:
url (str): The URL to call
Return:
a HTTPResponse
"""
_set_timeout()
opener = get_opener()
auth_string = (base64.encodestring(DATACITE_USER_NAME + ':'
+ DATACITE_PASSWORD)).rstrip()
headers = {'Authorization':'Basic ' + auth_string}
req = urllib2.Request(url, data=None, headers=headers)
req.get_method = lambda: 'DELETE'
try:
response = opener.open(req)
except (urllib2.HTTPError) as ex:
msg = ex.readlines()
LOGGING.warn('HTTPError error getting %s. %s', url, msg)
return get_response(msg, ex.code)
except (socket.timeout, urllib2.URLError) as ex:
LOGGING.warn('Timeout or URLError error getting %s. %s', url, ex.reason)
return get_response(ex.reason, 500)
except (SSLError) as ex:
LOGGING.warn('SSLError error getting %s. %s', url, ex)
return get_response(ex, 500)
finally:
_close(opener)
if response.headers.has_key('Content-Type'):
ret_response = HttpResponse(content_type=
response.headers.get('Content-Type'))
else:
ret_response = HttpResponse()
ret_response.status_code = response.code
ret_response.reason_phrase = response.msg
# pylint: disable=maybe-no-member
ret_response.writelines(response.readlines())
return ret_response
def delete_metadata(request):
"""
Delete the metadata for the DOI.
Args:
request (HTTPRequest): The HTTP request
Return:
a HTTPResponse
"""
_doi = get_doi_from_request(request, 'metadata')
LOGGING.info('Delete metadata doi: %s', _doi)
url = urljoin(DATACITE_URL, request.get_full_path())
try:
doi_suffix = _doi.split(DOI_PREFIX, 1)[1]
except IndexError:
return get_response("Bad Request - wrong prefix, doi should start " \
"with %s" % DOI_PREFIX, 400)
if not is_authorized(request, doi_suffix):
return get_response("Unauthorized - insufficient privileges", 403)
return _delete(url)
def delete_metadata(request):
"""
Delete the metadata for the DOI.
Args:
request (HTTPRequest): The HTTP request
Return:
a HTTPResponse
"""
_doi = get_doi_from_request(request, 'metadata')
LOGGING.info('Delete metadata doi: %s', _doi)
url = urljoin(DATACITE_URL, request.get_full_path())
try:
doi_suffix = _doi.split(DOI_PREFIX, 1)[1]
except IndexError:
return get_response("Bad Request - wrong prefix, doi should start " \
"with %s" % DOI_PREFIX, 400)
if not is_authorized(request, doi_suffix):
return get_response("Unauthorized - insufficient privileges", 403)
return _delete(url)
def _post(url, body, headers, method="POST"):
"""
Send a post request to DataCite.
Args:
url (str): The URL to call
body (str): The data
headers (dict): A dictionary of headers to use
Return:
a HTTPResponse
"""
_set_timeout()
opener = get_opener()
auth_string = ((base64.encodebytes(
(DATACITE_USER_NAME + ":" +
DATACITE_PASSWORD).encode())).decode("utf-8").rstrip())
headers.update({"Authorization": "Basic " + auth_string})
# If the request body is a string, urllib2 attempts to concatenate the url,
# body and headers. If the url is unicode, the request body can get
# converted unicode. This has resulted in issues where there are characters
# with diacritic marks in the request body. To avoid these issues the url is
# UTF-8 encoded.
url_encode = url.encode("utf-8").decode()
req = urllib.request.Request(url_encode,
data=body,
headers=headers,
method=method)
try:
response = opener.open(req)
except urllib.error.HTTPError as ex:
msg = ex.readlines()
LOGGING.warning("HTTPError error getting %s. %s", url, msg)
return get_response(msg, ex.code)
except (socket.timeout, urllib.error.URLError) as ex:
LOGGING.warning("Timeout or URLError error getting %s. %s", url,
ex.reason)
return get_response(ex.reason, 500)
except SSLError as ex:
LOGGING.warning("SSLError error getting %s. %s", url, ex)
return get_response(ex, 500)
except UnicodeDecodeError as ex:
LOGGING.info("UnicodeDecodeError error getting %s. %s", url, ex)
return get_response(ex, 500)
finally:
_close(opener)
if "Content-Type" in response.headers:
ret_response = HttpResponse(
content_type=response.headers.get("Content-Type"))
else:
ret_response = HttpResponse()
ret_response.status_code = response.code
ret_response.reason_phrase = response.msg
# pylint: disable=maybe-no-member
ret_response.writelines(response.readlines())
if "location" in response.headers:
ret_response.setdefault("Location", response.headers.get("location"))
return ret_response