def download_one(recid, version):
"""Download given version of the PDF from arxiv"""
write_message('fetching %s' % recid)
for count, arxiv_id in enumerate(extract_arxiv_ids_from_recid(recid)):
if count != 0:
write_message("Warning: %s has multiple arxiv #" % recid)
continue
url_for_pdf = build_arxiv_url(arxiv_id, version)
filename_arxiv_id = arxiv_id.replace('/', '_')
temp_file = NamedTemporaryFile(prefix="arxiv-pdf-checker",
dir=CFG_TMPSHAREDDIR,
suffix="%s.pdf" % filename_arxiv_id)
write_message('downloading pdf from %s' % url_for_pdf)
path = download_external_url(url_for_pdf,
temp_file.name,
content_type='pdf')
# Check if it is not an html not found page
filesize = os.path.getsize(path)
if filesize < 25000:
f = open(path)
try:
for line in f:
if 'PDF unavailable' in line:
raise PdfNotAvailable()
finally:
f.close()
docs = BibRecDocs(recid)
bibdocfiles = docs.list_latest_files(doctype="arXiv")
needs_update = False
try:
bibdocfile = bibdocfiles[0]
except IndexError:
bibdocfile = None
needs_update = True
else:
existing_md5 = calculate_md5(bibdocfile.fullpath)
new_md5 = calculate_md5(path.encode('utf-8'))
if new_md5 != existing_md5:
write_message('md5 differs updating')
needs_update = True
else:
write_message('md5 matches existing pdf, skipping')
if needs_update:
if bibdocfiles:
write_message('adding as new version')
docs.add_new_version(path, docname=bibdocfile.name)
else:
write_message('adding as new file')
docs.add_new_file(path,
doctype="arXiv",
docname="arXiv:%s" % filename_arxiv_id)
else:
raise FoundExistingPdf()
def download_one(recid, version):
"""Download given version of the PDF from arxiv"""
write_message("fetching %s" % recid)
for count, arxiv_id in enumerate(extract_arxiv_ids_from_recid(recid)):
if count != 0:
write_message("Warning: %s has multiple arxiv #" % recid)
continue
url_for_pdf = build_arxiv_url(arxiv_id, version)
filename_arxiv_id = arxiv_id.replace("/", "_")
temp_file = NamedTemporaryFile(
prefix="arxiv-pdf-checker", dir=CFG_TMPSHAREDDIR, suffix="%s.pdf" % filename_arxiv_id
)
write_message("downloading pdf from %s" % url_for_pdf)
path = download_external_url(url_for_pdf, temp_file.name, content_type="pdf")
# Check if it is not an html not found page
filesize = os.path.getsize(path)
if filesize < 25000:
f = open(path)
try:
for line in f:
if "PDF unavailable" in line:
raise PdfNotAvailable()
finally:
f.close()
docs = BibRecDocs(recid)
bibdocfiles = docs.list_latest_files(doctype="arXiv")
needs_update = False
try:
bibdocfile = bibdocfiles[0]
except IndexError:
bibdocfile = None
needs_update = True
else:
existing_md5 = calculate_md5(bibdocfile.fullpath)
new_md5 = calculate_md5(path.encode("utf-8"))
if new_md5 != existing_md5:
write_message("md5 differs updating")
needs_update = True
else:
write_message("md5 matches existing pdf, skipping")
if needs_update:
if bibdocfiles:
write_message("adding as new version")
docs.add_new_version(path, docname=bibdocfile.name)
else:
write_message("adding as new file")
docs.add_new_file(path, doctype="arXiv", docname="arXiv:%s" % filename_arxiv_id)
else:
raise FoundExistingPdf()
def test_md5_algorithms(self):
"""bibdocfile - compare md5 algorithms"""
from invenio.legacy.bibdocfile.api import calculate_md5, \
calculate_md5_external
filepath = os.path.join(self.path, 'test.txt')
open(filepath, "w").write("test")
self.assertEqual(calculate_md5(filepath, force_internal=True),
calculate_md5_external(filepath))
def test_md5_algorithms(self):
"""bibdocfile - compare md5 algorithms"""
from invenio.legacy.bibdocfile.api import calculate_md5, \
calculate_md5_external
filepath = os.path.join(self.path, 'test.txt')
open(filepath, "w").write("test")
self.assertEqual(calculate_md5(filepath, force_internal=True),
calculate_md5_external(filepath))
def test_posting_file(self):
"""webstyle - direct posting of a file"""
from invenio.legacy.bibdocfile.api import calculate_md5
path = os.path.join(CFG_PREFIX, 'lib', 'webtest', 'invenio', 'test.gif')
body = open(path).read()
md5 = calculate_md5(path)
mimetype = 'image/gif'
connection = httplib.HTTPConnection(urlparse.urlsplit(CFG_SITE_URL)[1])
connection.request('POST', '/httptest/post2', body, {'Content-MD5': md5, 'Content-Type': mimetype, 'Content-Disposition': 'filename=test.gif'})
response = connection.getresponse()
body2 = response.read()
self.assertEqual(body, body2, "Body sent differs from body received")
def test_posting_file(self):
"""webstyle - direct posting of a file"""
from invenio.legacy.bibdocfile.api import calculate_md5
path = os.path.join(cfg['CFG_PREFIX'], 'lib', 'webtest', 'invenio',
'test.gif')
body = open(path).read()
md5 = calculate_md5(path)
mimetype = 'image/gif'
connection = httplib.HTTPConnection(
urlparse.urlsplit(cfg['CFG_SITE_URL'])[1])
connection.request(
'POST', '/httptest/post2', body, {
'Content-MD5': md5,
'Content-Type': mimetype,
'Content-Disposition': 'filename=test.gif'
})
response = connection.getresponse()
body2 = response.read()
self.assertEqual(body, body2,
"Body sent differs from body received")
