def update_activities_count(self):
# This module to give us imformation the count activity in database
# and in the XML
try:
# Activity count in the XML
file_grabber = FileGrabber()
response = file_grabber.get_the_file(self.source_url)
# Parse to XML tree
tree = etree.fromstring(response.content)
# Get version from the XML
if not self.iati_version:
parser = etree.XMLParser(huge_tree=True, encoding='utf-8')
parser_tree = etree.parse(BytesIO(response.content), parser)
root = parser_tree.getroot()
# Continue parsing if version is 2.01 or above
iati_version = root.xpath('@version')
if len(iati_version) > 0:
iati_version = iati_version[0]
self.iati_version = iati_version
count = len(tree.findall('iati-activity'))
self.activities_count_in_xml = count
# Activity count in the Database
self.activities_count_in_database = self.activity_set.all().count()
self.save(process=False)
except Exception as e:
logger.error(e)
def get_xml_activity_amount(self, url):
try:
file_grabber = FileGrabber()
xml_file = file_grabber.get_the_file(url)
occurences = 0
for line in xml_file:
if "</iati-identifier>" in line:
amount = line.count("</iati-identifier>")
occurences += amount
del xml_file
gc.collect()
return occurences
except Exception as e:
if e.args:
print(e.args[0])
print("ERROR IN GET_XML_ACTIVITY_AMOUNT, FILE URL " + url)
def handle_response(tar, source):
file_grabber = FileGrabber()
response = file_grabber.get_the_file(source.source_url)
if not response or response.code != 200:
print("source url {} down or doesn't exist".format(source.source_url))
return
modified_time = response.info().get('Last-Modified')
iati_file = StringIO.StringIO(response.read())
info = tarfile.TarInfo(
name="{}.xml".format(source.ref)
)
info.size = len(iati_file.buf)
if modified_time:
info.mtime = _parse_http_datetime(modified_time)
tar.addfile(tarinfo=info, fileobj=iati_file)
def get_xml_activity_amount(self, url):
try:
file_grabber = FileGrabber()
xml_file = file_grabber.get_the_file(url)
occurences = 0
for line in xml_file:
if "</iati-identifier>" in line:
amount = line.count("</iati-identifier>")
occurences += amount
del xml_file
gc.collect()
return occurences
except Exception as e:
if e.args:
print(e.args[0])
print("ERROR IN GET_XML_ACTIVITY_AMOUNT, FILE URL " + url)
def update_activities_count(self):
# This module to give us imformation the count activity in database
# and in the XML
try:
# Activity count in the XML
file_grabber = FileGrabber()
response = file_grabber.get_the_file(self.source_url)
# Parse to XML tree
tree = etree.fromstring(response.content)
count = len(tree.getchildren())
self.activities_count_in_xml = count - 1 if count > 0 else count
# Activity count in the Database
self.activities_count_in_database = self.activity_set.all().count()
self.save(process=False)
except Exception as e:
logger.error(e)
def parse_url(self, source):
"""
Parses the source with url
"""
url = source.source_url
xml_source_ref = source.ref
# last_hash = source.last_hash
try:
file_grabber = FileGrabber()
iati_file = file_grabber.get_the_file(url)
if iati_file:
# delete old activities
# TODO: determine this in the parser based on last-updated-datetime
# TODO: also, throw away all narratives
# try:
# deleter = Deleter()
# deleter.delete_by_source(xml_source_ref)
# except Exception as e:
# exception_handler(e, "parse url", "delete by source")
data = iati_file.read()
root = etree.fromstring(str(data))
parser = self.prepare_parser(root, source)
parser.load_and_parse(root)
# Throw away query logs when in debug mode to prevent memory from overflowing
if settings.DEBUG:
from django import db
db.reset_queries()
except Exception as e:
exception_handler(e, "parse url", "parse_url")
def __init__(self, dataset, root=None, force_reparse=False):
"""
Given a IATI dataset, prepare an IATI parser
"""
if settings.IATI_PARSER_DISABLED:
raise ParserDisabledError(
"The parser is disabled on this instance of OIPA")
self.dataset = dataset
self.url = dataset.source_url
self.force_reparse = force_reparse
self.hash_changed = True
self.valid_dataset = True
if root is not None:
self.root = root
self.parser = self._prepare_parser(self.root, dataset)
return
file_grabber = FileGrabber()
response = file_grabber.get_the_file(self.url)
from iati_synchroniser.models import DatasetNote
if not response or response.status_code != 200:
self.valid_dataset = False
note = DatasetNote(
dataset=self.dataset,
iati_identifier="n/a",
model="n/a",
field="n/a",
message="Cannot access the URL",
exception_type='UrlError',
line_number=None
)
note.save()
self.dataset.note_count = 1
# If not a XML file them sha1 should blank
self.dataset.sha1 = ''
self.dataset.save()
return
# 1. Turn bytestring into string (treat it using specified encoding):
try:
iati_file = smart_text(response.content, 'utf-8')
# XXX: some files contain non utf-8 characters:
# FIXME: this is hardcoded:
except UnicodeDecodeError:
iati_file = smart_text(response.content, 'latin-1')
# 2. Encode the string to use for hashing:
hasher = hashlib.sha1()
hasher.update(iati_file.encode('utf-8'))
sha1 = hasher.hexdigest()
if dataset.sha1 == sha1:
# dataset did not change, no need to reparse normally
self.hash_changed = False
else:
dataset.sha1 = sha1
# Save a sha1 in the first time of the process parse
dataset.save()
try:
self.root = etree.fromstring(response.content)
self.parser = self._prepare_parser(self.root, dataset)
if settings.ERROR_LOGS_ENABLED:
self.xsd_validate()
# TODO: when moving error messages to frontend, create a separate error
# for wrong file type:
except etree.XMLSyntaxError as e:
self.valid_dataset = False
DatasetNote.objects.filter(dataset=self.dataset).delete()
note = DatasetNote(
dataset=self.dataset,
iati_identifier="n/a",
model="n/a",
field="n/a",
message="This file contains XML syntax errors or it's not an "
"XML file",
exception_type='XMLSyntaxError',
line_number=None
)
note.save()
self.dataset.note_count = 1
# If not the XML should not have a sha1
self.dataset.sha1 = ''
self.dataset.save()
return
def __init__(self, dataset, root=None, force_reparse=False):
"""
Given a IATI dataset, prepare an IATI parser
"""
if settings.IATI_PARSER_DISABLED:
raise ParserDisabledError(
"The parser is disabled on this instance of OIPA")
self.dataset = dataset
self.url = dataset.source_url
self.force_reparse = force_reparse
self.hash_changed = True
self.valid_dataset = True
if root is not None:
self.root = root
self.parser = self._prepare_parser(self.root, dataset)
return
file_grabber = FileGrabber()
response = file_grabber.get_the_file(self.url)
from iati_synchroniser.models import DatasetNote
if not response or response.code != 200:
self.valid_dataset = False
DatasetNote.objects.filter(dataset=self.dataset).delete()
note = DatasetNote(dataset=self.dataset,
iati_identifier="n/a",
model="n/a",
field="n/a",
message="URL down or does not exist",
exception_type='UrlError',
line_number=None)
note.save()
self.dataset.note_count = 1
self.dataset.save()
return
iati_file = response.read()
iati_file_str = str(iati_file)
hasher = hashlib.sha1()
hasher.update(iati_file_str)
sha1 = hasher.hexdigest()
if dataset.sha1 == sha1:
# dataset did not change, no need to reparse normally
self.hash_changed = False
else:
dataset.sha1 = sha1
try:
self.root = etree.fromstring(iati_file_str)
self.parser = self._prepare_parser(self.root, dataset)
except etree.XMLSyntaxError as e:
self.valid_dataset = False
DatasetNote.objects.filter(dataset=self.dataset).delete()
note = DatasetNote(dataset=self.dataset,
iati_identifier="n/a",
model="n/a",
field="n/a",
message="This file contains XML syntax errors",
exception_type='XMLSyntaxError',
line_number=None)
note.save()
self.dataset.note_count = 1
self.dataset.save()
return
def __init__(self, dataset, root=None, force_reparse=False):
"""
Given a IATI dataset, prepare an IATI parser
"""
if settings.IATI_PARSER_DISABLED:
raise ParserDisabledError(
"The parser is disabled on this instance of OIPA")
self.dataset = dataset
self.url = dataset.source_url
self.force_reparse = force_reparse
self.hash_changed = True
self.valid_dataset = True
if root is not None:
self.root = root
self.parser = self._prepare_parser(self.root, dataset)
return
file_grabber = FileGrabber()
response = file_grabber.get_the_file(self.url)
from iati_synchroniser.models import DatasetNote
if not response or response.status_code != 200:
self.valid_dataset = False
note = DatasetNote(dataset=self.dataset,
iati_identifier="n/a",
model="n/a",
field="n/a",
message="Cannot access the URL",
exception_type='UrlError',
line_number=None)
note.save()
self.dataset.note_count = 1
# If not a XML file them sha1 should blank
self.dataset.sha1 = ''
self.dataset.save()
return
# 1. Turn bytestring into string (treat it using specified encoding):
try:
iati_file = smart_text(response.content, 'utf-8')
# XXX: some files contain non utf-8 characters:
# FIXME: this is hardcoded:
except UnicodeDecodeError:
iati_file = smart_text(response.content, 'latin-1')
# 2. Encode the string to use for hashing:
hasher = hashlib.sha1()
hasher.update(iati_file.encode('utf-8'))
sha1 = hasher.hexdigest()
if dataset.sha1 == sha1:
# dataset did not change, no need to reparse normally
self.hash_changed = False
else:
dataset.sha1 = sha1
# Save a sha1 in the first time of the process parse
dataset.save()
try:
parser = etree.XMLParser(huge_tree=True)
tree = etree.parse(BytesIO(response.content), parser)
self.root = tree.getroot()
self.parser = self._prepare_parser(self.root, dataset)
if settings.ERROR_LOGS_ENABLED:
self.xsd_validate()
# TODO: when moving error messages to frontend, create a separate error
# for wrong file type:
except etree.XMLSyntaxError as e:
self.valid_dataset = False
DatasetNote.objects.filter(dataset=self.dataset).delete()
note = DatasetNote(
dataset=self.dataset,
iati_identifier="n/a",
model="n/a",
field="n/a",
message="This file contains XML syntax errors or it's not an "
"XML file",
exception_type='XMLSyntaxError',
line_number=None)
note.save()
self.dataset.note_count = 1
# If not the XML should not have a sha1
self.dataset.sha1 = ''
self.dataset.save()
return
