def read_patent(file):
count = 0
storage_dir = "data"
storage_path = os.path.join(os.getcwd(), storage_dir)
path = os.path.join(storage_path, file.name.values[0])
#declaring XMLPullParser with end event
if file.name.values[0].startswith("pgb"):
pat_xml = etree.XMLPullParser(tag='PATDOC',
events=['end'],
recover=True)
elif file.name.values[0].startswith("ipgb"):
pat_xml = etree.XMLPullParser(tag='us-patent-grant',
events=['end'],
recover=True)
with open(path, 'r') as lines:
for line in lines:
if line.startswith("<?xml"):
if (patent_ends(pat_xml, file.name.values[0])):
pat_xml.close()
if file.name.values[0].startswith("pgb"):
pat_xml = etree.XMLPullParser(tag='PATDOC',
events=['end'],
recover=True)
elif file.name.values[0].startswith("ipgb"):
pat_xml = etree.XMLPullParser(tag='us-patent-grant',
events=['end'],
recover=True)
#Removing unwanted nodes
elif line.startswith("<!DOCTYPE") or line.startswith(
"]>") or line.startswith('<!ENTITY'):
pass
else:
#pudb.set_trace()
#Using feed Parser for line by line parsing
pat_xml.feed(line)
def parse_file(file_to_parse):
"""
Parse a file with multiple junos outputs in xml format
Args: file_to_parse (file): an open file to parse
Returns: A list of lxml trees with <rpc-reply> root elements
"""
f = open("tmp.xml", 'w')
f.close()
#getting rid of everything not related to xml (\n, promts and other crap)
for line in file_to_parse:
line1 = line.strip()
if line1.startswith('</rpc-reply>'):
f = open("tmp.xml", 'a')
f.write(line)
f.write('\n')
f.close()
elif line1.startswith('<'):
f = open("tmp.xml", 'a')
f.write(line)
f.close()
else:
continue
#File tmp.xml is a clear XML file with set of "rpc-reply" elements
f = open("tmp.xml")
list_of_xml_trees = []
parser = etree.XMLPullParser(events=['end'], recover=True)
for line in f:
parser.feed(line)
for action, element in parser.read_events():
if (action == 'end') and (element.tag == 'rpc-reply'):
list_of_xml_trees.append(parser.close())
parser = etree.XMLPullParser(events=('start', 'end'),
recover=True)
return list_of_xml_trees
def init_parser(self):
"""init the XML parser. The parser must always be reset for each new
connexion
"""
self.xml_depth = 0
self.xml_root = None
self.parser = ET.XMLPullParser(("start", "end"))
def parse_file(file_to_parse):
#"""Parse a file with multiple junos outputs in xml format
#Args:
#file_to_parse (file): an open file to parse
#Returns:
#A list of lxml trees with <rpc-reply> root elements
#"""
list_of_xml_trees = []
parser = etree.XMLPullParser(events=['end'], recover=True)
for line in file_to_parse:
parser.feed(line)
for action, element in parser.read_events():
if (action == 'end') and (element.tag == 'rpc-reply'):
list_of_xml_trees.append(parser.close())
parser = etree.XMLPullParser(events=('start', 'end'), recover=True)
return list_of_xml_trees
def readAll(self):
self.first_element = None
self.parser = etree.XMLPullParser(('start', 'end'))
read_bytes = 0
garbage_bytes = len(self.cmd) + 1
# Remove command string from result
while not read_bytes == garbage_bytes:
read_bytes += len(self.channel.recv(garbage_bytes - read_bytes))
response = b''
while True:
data = self.channel.recv(BUF_SIZE)
# Connection was closed by server
if not data:
break
self.parser.feed(data)
response += data
if self.valid_xml():
break
return response.decode('utf-8')
async def __aiter__(self) -> _AsyncGenerator[NSElement, None]:
if not Api.agent:
raise RuntimeError("The API's user agent is not yet set.")
url = self.value
# pylint: disable=E1101
tag = self.name.upper().rstrip("S")
parser = etree.XMLPullParser(["end"],
base_url=url,
remove_blank_text=True,
tag=tag)
parser.set_element_class_lookup(
etree.ElementDefaultClassLookup(element=NSElement))
events = parser.read_events()
dobj = zlib.decompressobj(16 + zlib.MAX_WBITS)
async with Api.session.request("GET",
url,
headers={"User-Agent":
Api.agent}) as response:
async for data, _ in response.content.iter_chunks():
parser.feed(dobj.decompress(data))
for _, element in events:
yield element
element.clear()
while element.getparent(
) is not None and element.getprevious() is not None:
del element.getparent()[0]
async def __aiter__(
self) -> _AsyncGenerator[objectify.ObjectifiedElement, None]:
if not Api.agent:
raise RuntimeError("The API's user agent is not yet set.")
url = self.url
tag = self._category.name.upper().rstrip("S")
dobj = zlib.decompressobj(16 + zlib.MAX_WBITS)
async with Api.session.request("GET",
url,
headers={"User-Agent":
Api.agent}) as response:
self._last_response = response
response.raise_for_status()
with contextlib.suppress(etree.XMLSyntaxError), contextlib.closing(
etree.XMLPullParser(["end"],
base_url=url,
remove_blank_text=True,
tag=tag)) as parser:
parser.set_element_class_lookup(
objectify.ObjectifyElementClassLookup())
events = parser.read_events()
async for data, _ in response.content.iter_chunks():
parser.feed(dobj.decompress(data))
for _, element in events:
yield element
element.clear(keep_tail=True)
def _start_xml(self):
self._first_element = None
# act on start and end element events and
# allow huge text data (for report content)
self._parser = etree.XMLPullParser(
events=("start", "end"), huge_tree=True
)
def get_jmeter_data(self):
try:
sample_started = False
sample_children = []
parser = et.XMLPullParser(['start', 'end'])
parser.feed(self.buf)
for event, elem in parser.read_events():
if event == 'start' and elem.tag == 'sample':
sample_started = True
sample_children = []
elif event == 'end' and elem.tag == 'httpSample':
sample = self.get_sample(elem)
if sample_started:
sample_children.append(sample)
else:
return sample
elif event == 'end' and elem.tag == 'sample':
sample = self.get_sample(elem, sample_children)
sample_started = False
return sample
except Exception as err:
collectd.error("Plugin jmeter: Exception in get_jmeter_data due to %s" % err)
def file_to_annotation_docs(fileObj):
parser = etree.XMLPullParser(events=('start', 'end'))
docs = []
currentDoc = None
currentAnnotation = None
currentMistake = None
currentText = ''
# work around multiple roots in original doc
parser.feed(b'<EVERYTHING>\n')
for line in fileObj.readlines():
parser.feed(line)
for action, element in parser.read_events():
if (action == 'start' and element.tag == 'DOC'):
currentDoc = {'currentText': '', 'nid': element.attrib['nid']}
currentText = ''
continue
if (action == 'end' and element.tag == 'DOC'):
currentDoc['currentText'] = currentText
docs.append(currentDoc)
currentDoc = None
continue
if (action == 'start' and element.tag == 'TEXT'):
continue
if (action == 'end' and element.tag in ['TITLE', 'P']):
currentText += element.text
continue
if (action == 'start' and element.tag == 'ANNOTATION'):
currentAnnotation = {
'teacher_id': element.attrib['teacher_id'],
'mistakes': []
}
currentDoc['annotation'] = currentAnnotation
continue
if (action == 'start' and element.tag == 'MISTAKE'):
currentMistake = {
'start':
str(int(element.attrib['start_par']) * 1000) +
element.attrib['start_off'],
'end':
str(int(element.attrib['end_par']) * 1000) +
element.attrib['end_off']
}
continue
if (action == 'end' and element.tag == 'MISTAKE'):
currentAnnotation['mistakes'].append(currentMistake)
continue
if (action == 'end' and element.tag == 'TYPE'):
currentMistake['type'] = element.text
continue
if (action == 'end' and element.tag == 'CORRECTION'):
currentMistake['corr'] = element.text
continue
parser.feed(b'</EVERYTHING>\n')
return docs
def xmlParse(f):
parser = etree.XMLPullParser()
events = parser.read_events()
# for line in f.readlines():
# print(line)
for line in f.readlines():
parser.feed(line)
print_events(parser)
result = parser.close()
f.close()
async def __aiter__(
self,
*,
clear: bool = True
) -> _AsyncGenerator[objectify.ObjectifiedElement, None]:
if not Api.agent:
raise RuntimeError("The API's user agent is not yet set.")
if "a" in self and self["a"].lower() == "sendtg":
raise RuntimeError(
"This API wrapper does not support API telegrams.")
if not self:
# Preempt the request to conserve ratelimit
raise BadRequest()
url = str(self)
headers = {"User-Agent": Api.agent}
if self._password:
headers["X-Password"] = self._password
autologin = self.autologin
if autologin:
headers["X-Autologin"] = autologin
if self.get("nation") in PINS:
headers["X-Pin"] = PINS[self["nation"]]
async with Api.session.request("GET", url,
headers=headers) as response:
self._last_response = response
if "X-Autologin" in response.headers:
self._password = None
if "X-Pin" in response.headers:
PINS[self["nation"]] = response.headers["X-Pin"]
response.raise_for_status()
encoding = (response.headers["Content-Type"].split("charset=")
[1].split(",")[0])
with contextlib.suppress(etree.XMLSyntaxError), contextlib.closing(
etree.XMLPullParser(["end"],
base_url=url,
remove_blank_text=True)) as parser:
parser.set_element_class_lookup(
objectify.ObjectifyElementClassLookup())
events = parser.read_events()
async for data, _ in response.content.iter_chunks():
parser.feed(data.decode(encoding))
for _, element in events:
if clear and (element.getparent() is None
or element.getparent().getparent()
is not None):
continue
yield element
if clear:
element.clear(keep_tail=True)
async def parsexml(session: aiohttp.ClientSession) -> str:
""" Parses ~100MB gzipped XML in Chunks (for memory profiling purposes) """
import zlib
url = "http://aiweb.cs.washington.edu/research/projects/xmltk/xmldata/data/pir/psd7003.xml.gz"
d = zlib.decompressobj(zlib.MAX_WBITS | 16)
parser = etree.XMLPullParser()
async for chunk in iterread(url, session, 1024):
chunk = d.decompress(chunk)
parser.feed(chunk)
for _, elem in parser.read_events():
if elem.tag == r"ProteinEntry":
yield elem
elem.clear()
async def parsexml(session: aiohttp.ClientSession) -> str:
"""
Parses elementtree from streamed input (by chunks) in-memory
"""
url = "http://www.fit-pro.cz/export/fitpro-cf6ad8215df1f1cf993029a1684d5251.xml"
ns = {"zbozi": "http://www.zbozi.cz/ns/offer/1.0"}
parser = etree.XMLPullParser()
async for chunk in iterread(url, session, 1024):
parser.feed(chunk)
for _, elem in parser.read_events():
if elem.tag == r"{http://www.zbozi.cz/ns/offer/1.0}SHOPITEM":
url = elem.xpath('//zbozi:URL', namespaces=ns)[0]
elem.clear()
yield url.text
def _get_vtk_type(path):
'''We use the incremental event emitting parser
here since we can expect to encounter appended binary data in the xml
which lxml cannot parse.
:param path: vtk file to peek into
:return: None if no VTKFile element found, else the type attribute of the VTKFile element
'''
parser = etree.XMLPullParser(events=('start', ))
with open(path, 'rb') as xml:
for lines in xml.readlines():
parser.feed(lines)
for action, element in parser.read_events():
if element.tag == 'VTKFile':
return element.get('type')
return None
class FindElement(object):
'''
classdocs
'''
pullParser = etree.XMLPullParser()
def __init__(self, xmlDoc):
'''
Constructor
'''
self.xml = xmlDoc
# self.tree = pullP
def find(aString):
pass
def get_datapoints_from_xml():
with zipfile.ZipFile("naptandata/NaPTANxml.zip") as container:
[contentname] = container.namelist()
with container.open(contentname) as f:
parser = etree.XMLPullParser(events=("end", ), no_network=True)
while True:
data = f.read(1024)
if not data:
break
parser.feed(data)
for action, elem in parser.read_events():
if elem.tag.endswith('StopPoint'):
if elem.get("Status") == "active":
yield handle_stoppoint(elem)
cleanup(elem)
def element_iterator(stream, events=['start']):
def _iter(stream, parser):
while True:
line = stream.read(256)
if line is None:
return
parser.feed(line)
for event in parser.read_events():
yield event
parser = etree.XMLPullParser(events)
try:
yield _iter(stream, parser)
finally:
del parser
def parse_references(self, file_path, source):
if not os.path.isfile(file_path):
raise GrobidException("File {} does not exist".format(file_path))
if not file_path.endswith(".pdf"):
raise GrobidException("File {} is not pdf".format(file_path))
try:
requests.post(self.grobid_url)
except ConnectionError:
raise ConnectionError("Grobid does not answer")
# file and connection are OK
with open(file_path, "rb") as f:
ref_handler = requests.post(self.grobid_url, files={"input": f})
if ref_handler.ok is False:
self.logger.error("File {} error {} when sending to Grobid".format(
file_path, ref_handler.status_code))
return False
parser2 = etree.XMLPullParser(tag="{}biblStruct".format(
self.namespace),
load_dtd=True)
parser2.feed(ref_handler.text)
result = {}
for action, elem in parser2.read_events():
try:
for i in elem:
if i.tag.replace(self.namespace, '') == "monogr":
result["monogr"] = self.parseMonogr(i)
if i.tag.replace(self.namespace, '') == "analytic":
result["analytic"] = self.parseMonogr(i)
reference = self.get_reference(result)
self.logger.debug("Original : %s", result)
self.logger.debug("Reference: %s", reference)
except Exception as e:
self.logger.critical("Reference Error %s", e)
reference = None
if reference is None:
self.logger.debug("Invalid reference")
else:
author_list = msgpack.packb(";".join(reference["authors"]))
SingleReference.objects.create(source=source,
title=reference["title"],
authors=author_list,
date=reference["pubyear"])
def parse_self_xml(self):
current_time = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
parser = etree.XMLPullParser(events=('start', ))
records = []
def parse_tag(raw_tag):
return raw_tag.split('}')[1]
try:
parser.feed(self.page)
except XMLSyntaxError as err:
print('XMLSyntaxError: ' + str(err))
else:
for action, element in parser.read_events():
if parse_tag(element.tag) != self.entity_tag:
continue
segs = [{parse_tag(t.tag.lower()): t.text} for t in element]
record = {}
for seg in segs:
record.update(seg)
# Check if a valid record
if 'link_id' and 'region' and 'road_type' and 'road_saturation_level' \
and 'traffic_speed' and 'capture_date' in record:
record['fetch_time'] = current_time
# The time is record['CAPTURE_DATE'], to be revised
r_time = time.strptime(record['CAPTURE_DATE'.lower()],
"%Y-%m-%dT%H:%M:%S")
record['CAPTURE_DATE'.lower()] = time.strftime(
"%Y-%m-%d %H:%M:%S", r_time)
# Store the seconds rom 1970
capture_date_1970 = float(time.mktime(r_time))
current_time_1970 = float(
time.mktime(
time.strptime(current_time, "%Y-%m-%d %H:%M:%S")))
record['capture_date_1970'] = capture_date_1970
record['fetch_time_1970'] = current_time_1970
records.append(record)
else:
print("invalid record")
finally:
return records
def is_empty(self, gml_path, zip_path):
"""
Detect if the file is empty.
Cadastre empty files (usually otherconstruction) comes with a null
feature and results in a non valid layer in QGIS.
"""
fo = self.get_file_object(gml_path, zip_path)
text = fo.read(2000)
fo.close()
parser = etree.XMLPullParser(["start"])
parser.feed(text)
events = list(parser.read_events())
try:
parser.close()
except etree.XMLSyntaxError:
pass
return len([event for event, elem in events if event == "start"]) < 3
def iter_elements(xmlfile, interesting_tags):
"""Parses a large xmlfile, yielding any listed tag.
Removes tags from the tree after yielding: do not mark nested
tags as interesting.
"""
parser = etree.XMLPullParser(events=("end",), no_network=True)
while True:
data = xmlfile.read(1024)
if not data:
break
parser.feed(data)
for action, elem in parser.read_events():
tagname = elem.tag.split("}")[-1]
if tagname in interesting_tags:
yield tagname, elem
cleanup(elem)
parser.close()
def readAll(self):
self.first_element = None
self.parser = etree.XMLPullParser(('start', 'end'))
response = b''
while True:
data = self.stdout.channel.recv(BUF_SIZE)
# Connection was closed by server
if not data:
break
self.parser.feed(data)
response += data
if self.valid_xml():
break
return response.decode('utf-8')
def __init__(self, src_lang_code, dst_lang_code, input_stream):
"""Initializes `TmxParser`.
Args:
src_lang_code: String - source language code in BCP 47 spec.
dst_lang_code: String - target language code in BCP 47 spec.
input_stream: io stream - tmx stream that implemented file interface.
"""
super(TmxParser, self).__init__()
self._src_lang = _parse_locale(src_lang_code)
self._dst_lang = _parse_locale(dst_lang_code)
self._tmx_stream = input_stream
self._parser = etree.XMLPullParser(events=('start', 'end'))
self._events = self._parser.read_events()
self._buffered_parsed_pairs = []
self._buffered_pairs_index = 0
# Current stack of tag names. It should start with empty string.
self._tag_name_stack = ['']
self._header_inited = False
self._body_inited = False
self._encoding = 'utf-8'
async def __aiter__(self):
url = self.value
parser = etree.XMLPullParser(["end"],
base_url=url,
remove_blank_text=True)
parser.set_element_class_lookup(
etree.ElementDefaultClassLookup(element=_NSElement))
events = parser.read_events()
dobj = zlib.decompressobj(16 + zlib.MAX_WBITS)
async with Api.session.request("GET",
url,
headers={"User-Agent":
Api.agent}) as response:
yield parser.makeelement("HEADERS", attrib=response.headers)
async for data, _ in response.content.iter_chunks():
parser.feed(dobj.decompress(data))
for _, element in events:
yield element
element.clear()
async def __aiter__(
self,
*,
no_clear: bool = False) -> _AsyncGenerator[NSElement, None]:
if not self.agent:
raise RuntimeError("The API's user agent is not yet set.")
if not self:
# Preempt the request to conserve ratelimit
raise ValueError("Bad request")
if "a" in self and self["a"].lower() == "sendtg":
raise RuntimeError(
"This API wrapper does not support API telegrams.")
url = str(self)
parser = etree.XMLPullParser(["end"],
base_url=url,
remove_blank_text=True)
parser.set_element_class_lookup(
etree.ElementDefaultClassLookup(element=NSElement))
events = parser.read_events()
async with self.session.request("GET",
url,
headers={"User-Agent":
self.agent}) as response:
encoding = response.headers["Content-Type"].split(
"charset=")[1].split(",")[0]
async for data, _ in response.content.iter_chunks():
parser.feed(data.decode(encoding))
for _, element in events:
if not no_clear and (element.getparent() is None
or element.getparent().getparent()
is not None):
continue
yield element
if no_clear:
continue
element.clear()
while element.getprevious() is not None:
del element.getparent()[0]
def to_list(filename):
html = etree.parse(filename, etree.XMLPullParser())
all_data = html.xpath('//data')
res = []
for data in all_data:
d = {}
name = data.attrib['name']
d.update(dict(name=name))
children = data.getchildren()
for child in children:
tag = child.tag
value = child.text
d.update({tag: value})
value = d.get('value', '')
value = value.replace('"', '\'').replace("\\", r"\\")
d['value'] = value
comment = d.get('comment', '')
# update
comment = comment.replace('"', '\'')
d['comment'] = comment
res.append(d)
return res
async def ingest(self):
"""
Task that reads all incoming messages and puts them into the queue for handling.
"""
while True:
# Read the next tag from the reader
data = await self.reader.readuntil(b">")
# If we receive a new XML header, start a new parser.
if data.startswith(b"<?xml"):
parser = etree.XMLPullParser(events=['start', 'end'])
# Feed the data to the parser
parser.feed(data)
# Look for tags that we have handlers for
for event, element in parser.read_events():
if event == 'start' and element.tag == '{http://etherx.jabber.org/streams}stream':
logger.debug(f"RECEIVED {ensure_str(element)}")
await self.on_open_stream(element)
elif event == 'end' and element.tag in self.handlers:
logger.debug(f"RECEIVED {ensure_str(element)}")
asyncio.create_task(self.handlers[element.tag](element))
def generate_with_progress(response, path: str):
# dtd is resolved via base_url
stream = response.raw
parser = etree.XMLPullParser(
events=['start', 'end'],
base_url=path,
load_dtd=True,
dtd_validation=True,
)
suffix = '%(percent)d%% %(elapsed_td)s (ETA: %(eta_td)s)'
length = int(response.headers['Content-Length'])
progress = Bar(suffix=suffix, max=length, hide_cursor=False)
try:
for line in decompress_gzip_stream(stream):
parser.feed(line)
yield from parser.read_events()
current_pos = stream.tell()
if current_pos > progress.index:
# have to check, otherwise ETA is screwed up
progress.goto(current_pos)
except KeyboardInterrupt:
pass
finally:
progress.finish()
async def __aiter__(self, *, clear: bool = True):
if not self:
raise ValueError("Bad request")
url = str(self)
parser = etree.XMLPullParser(["end"],
base_url=url,
remove_blank_text=True)
parser.set_element_class_lookup(
etree.ElementDefaultClassLookup(element=_NSElement))
events = parser.read_events()
async with type(self).session.request(
"GET", url, headers={"User-Agent":
type(self).agent}) as response:
yield parser.makeelement("HEADERS", attrib=response.headers)
encoding = response.headers["Content-Type"].split(
"charset=")[1].split(",")[0]
async for data, _ in response.content.iter_chunks():
parser.feed(data.decode(encoding))
for _, element in events:
yield element
if clear:
element.clear()