def arxiv_fft_get(obj, eng):
"""Get FFT from arXiv, if arXiv ID is provided."""
deposition = Deposition(obj)
sip = deposition.get_latest_sip(sealed=False)
metadata = sip.metadata
if 'arxiv_id' in metadata and metadata['arxiv_id']:
arxiv_pdf_url = cfg.get("ARXIV_PDF_URL", "http://arxiv.org/pdf/") + \
"{0}.{1}"
from invenio.config import CFG_TMPSHAREDDIR
arxiv_file, arxiv_file_path = mkstemp(
prefix="%s_" % (metadata['arxiv_id'].replace("/", "_")),
suffix='.pdf',
dir=CFG_TMPSHAREDDIR,
)
os.close(arxiv_file)
download_url(url=arxiv_pdf_url.format(metadata['arxiv_id'], "pdf"),
content_type="pdf",
download_to_file=arxiv_file_path)
# To get 1111.2222.pdf as filename.
filename = "{0}.pdf".format(metadata['arxiv_id'].replace("/", "_"))
try:
try:
save_deposition_file(deposition,
filename,
arxiv_file_path)
except FilenameAlreadyExists:
obj.log.error("PDF file not saved: filename already exists.")
except Exception as e:
obj.log.error("PDF file not saved: {}.".format(e.message))
def test_content_type(self):
"""Test simple calls to download_url."""
from invenio_utils.filedownload import (download_url,
InvenioFileDownloadError)
tmpdoc = download_url("http://duckduckgo.com", content_type="html")
self.assertTrue(tmpdoc)
fun = lambda: download_url("http://google.com", content_type="pdf")
self.assertRaises(InvenioFileDownloadError, fun)
def output_keywords_for_sources(
input_sources, taxonomy_name, output_mode="text",
output_limit=cfg['CLASSIFIER_DEFAULT_OUTPUT_NUMBER'], spires=False,
match_mode="full", no_cache=False, with_author_keywords=False,
rebuild_cache=False, only_core_tags=False, extract_acronyms=False,
**kwargs):
"""Output the keywords for each source in sources."""
# Inner function which does the job and it would be too much work to
# refactor the call (and it must be outside the loop, before it did
# not process multiple files)
def process_lines():
if output_mode == "text":
print("Input file: %s" % source)
line_nb = len(text_lines)
word_nb = 0
for line in text_lines:
word_nb += len(re.findall("\S+", line))
current_app.logger.info(
"Remote file has %d lines and %d words.".format(
line_nb, word_nb
)
)
return get_keywords_from_text(
text_lines,
taxonomy_name,
output_mode=output_mode,
output_limit=output_limit,
spires=spires,
match_mode=match_mode,
no_cache=no_cache,
with_author_keywords=with_author_keywords,
rebuild_cache=rebuild_cache,
only_core_tags=only_core_tags,
extract_acronyms=extract_acronyms
)
# Get the fulltext for each source.
for entry in input_sources:
current_app.logger.info("Trying to read input file %s." % entry)
text_lines = None
source = ""
if os.path.isdir(entry):
for filename in os.listdir(entry):
if filename.startswith('.'):
continue
filename = os.path.join(entry, filename)
if os.path.isfile(filename):
text_lines, dummy = get_plaintext_document_body(filename)
if text_lines:
source = filename
process_lines()
elif os.path.isfile(entry):
text_lines, dummy = get_plaintext_document_body(entry)
if text_lines:
source = os.path.basename(entry)
process_lines()
else:
# Treat as a URL.
local_file = download_url(entry)
text_lines, dummy = get_plaintext_document_body(local_file)
if text_lines:
source = entry.split("/")[-1]
process_lines()
def output_keywords_for_sources(input_sources,
taxonomy_name,
output_mode="text",
output_limit=None,
spires=False,
match_mode="full",
no_cache=False,
with_author_keywords=False,
rebuild_cache=False,
only_core_tags=False,
extract_acronyms=False,
**kwargs):
"""Output the keywords for each source in sources."""
if output_limit is None:
output_limit = cfg['CLASSIFIER_DEFAULT_OUTPUT_NUMBER']
# Inner function which does the job and it would be too much work to
# refactor the call (and it must be outside the loop, before it did
# not process multiple files)
def process_lines():
if output_mode == "text":
print("Input file: %s" % source)
line_nb = len(text_lines)
word_nb = 0
for line in text_lines:
word_nb += len(re.findall("\S+", line))
current_app.logger.info(
"Remote file has %d lines and %d words.".format(line_nb, word_nb))
return get_keywords_from_text(
text_lines,
taxonomy_name,
output_mode=output_mode,
output_limit=output_limit,
spires=spires,
match_mode=match_mode,
no_cache=no_cache,
with_author_keywords=with_author_keywords,
rebuild_cache=rebuild_cache,
only_core_tags=only_core_tags,
extract_acronyms=extract_acronyms)
# Get the fulltext for each source.
for entry in input_sources:
current_app.logger.info("Trying to read input file %s." % entry)
text_lines = None
source = ""
if os.path.isdir(entry):
for filename in os.listdir(entry):
if filename.startswith('.'):
continue
filename = os.path.join(entry, filename)
if os.path.isfile(filename):
text_lines, dummy = get_plaintext_document_body(filename)
if text_lines:
source = filename
process_lines()
elif os.path.isfile(entry):
text_lines, dummy = get_plaintext_document_body(entry)
if text_lines:
source = os.path.basename(entry)
process_lines()
else:
# Treat as a URL.
from invenio_utils.filedownload import download_url
local_file = download_url(entry)
text_lines, dummy = get_plaintext_document_body(local_file)
if text_lines:
source = entry.split("/")[-1]
process_lines()
