def __getitem__(self, identifier):
"""
Access the item with id 'identifier' in the file by iterating the xml-tree.
Arguments:
identifier (str): native id of the item to access
Returns:
data (str): text associated with the given identifier
"""
old_pos = self.file_handler.tell()
self.file_handler.seek(0, 0)
mzml_iter = iter(iterparse(self.file_handler, events=['end']))
while True:
event, element = next(mzml_iter)
if event == 'end':
if element.tag.endswith('}spectrum'):
if int(
regex_patterns.SPECTRUM_ID_PATTERN.search(
element.get('id')).group(0)) == identifier:
self.file_handler.seek(old_pos, 0)
return spec.Spectrum(element, measured_precision=5e-6)
elif element.tag.endswith('}chromatogram'):
if element.get('id') == identifier:
self.file_handler.seek(old_pos, 0)
return spec.Chromatogram(element,
measured_precision=5e-6)
def __getitem__(self, identifier):
"""
Access the item with id 'identifier'.
Either use linear, binary or interpolated search.
Arguments:
identifier (str): native id of the item to access
Returns:
data (str): text associated with the given identifier
"""
#############################################################################
# DOES NOT HOLD IF NUMBERS DONT START WITH ONE AND/OR DONT INCREASE BY ONE #
# TODO FIXME #
#############################################################################
self.file_handler.seek(0)
spectrum = None
if str(identifier).upper() == 'TIC':
# print(str(identifier).upper())
found = False
mzmliter = iter(iterparse(self.file_handler, events=['end']))
while found is False:
event, element = next(mzmliter, ('STOP', 'STOP'))
if event == 'end':
if element.tag.endswith('}chromatogram'):
if element.get('id') == 'TIC':
found = True
spectrum = spec.Chromatogram(
element, measured_precision=5e-6)
elif event == 'STOP':
raise StopIteration
elif identifier in self.offset_dict:
start = self.offset_dict[identifier]
with open(self.path, 'rb') as seeker:
seeker.seek(start[0])
start, end = self._read_to_spec_end(seeker)
self.file_handler.seek(start, 0)
data = self.file_handler.read(end - start)
spectrum = spec.Spectrum(XML(data), measured_precision=5e-6)
elif type(identifier) == str:
return self._search_string_identifier(identifier)
else:
spectrum = self._interpol_search(identifier)
return spectrum
def _search_string_identifier(self, search_string, chunk_size=8):
with open(self.path, 'rb') as seeker:
data = None
total_chunk_size = chunk_size * 512
spec_start = None
# NOTE: This needs to go intp regex_patterns.py
regex_string = re.compile(
"<\s*spectrum[^>]*index=\"[0-9]+\"\sid=\"({0})\"\sdefaultArrayLength=\"[0-9]+\">".format(
"".join(
['.*', search_string, '.*']
)
).encode()
)
search_string = search_string.encode()
while True:
file_pointer = seeker.tell()
data = seeker.read(total_chunk_size)
string, seeker = self._read_until_tag_end(seeker, byte_mode=True)
data += string
spec_start = regex_string.search(data)
chrom_start = regex_patterns.CHROMO_OPEN_PATTERN.search(data)
if spec_start:
spec_start_offset = file_pointer + spec_start.start()
current_index = spec_start.group(1)
if search_string in current_index:
seeker.seek(spec_start_offset)
start, end = self._read_to_spec_end(seeker)
seeker.seek(start)
spec_string = seeker.read(end-start)
xml_string = XML(spec_string)
return spec.Spectrum(
xml_string,
measured_precision=5e-6
)
elif chrom_start:
chrom_start_offset = file_pointer + chrom_start.start()
if search_string == chrom_start.group(1):
seeker.seek(chrom_start_offset)
start, end = self._read_to_spec_end(seeker)
seeker.seek(start)
chrom_string = seeker.read(end-start)
xml_string = XML(chrom_string)
return spec.Chromatogram(xml_string)
elif len(data) == 0:
raise Exception('cant find specified string')
def __getitem__(self, key):
"""
Execute a SQL request, process the data and return a spectrum object.
Args:
key (str or int): unique identifier for the given spectrum in the
database
"""
self.cursor.execute("SELECT * FROM spectra WHERE id=?", key)
ID, element = self.cursor.fetchone()
element = et.XML(element)
if "spectrum" in element.tag:
spectrum = spec.Spectrum(element)
elif "chromatogram" in element.tag:
spectrum = spec.Chromatogram(element)
return spectrum
def __getitem__(self, identifier):
"""
Access the item with id 'identifier' in the file.
Arguments:
identifier (str): native id of the item to access
Returns:
data (str): text associated with the given identifier
"""
#TODO more elegant way to add NameSpace (.register_namespace maybe??)
ns_prefix = '<mzML xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://psi.hupo.org/ms/mzml http://psidev.info/files/ms/mzML/xsd/mzML1.1.0.xsd" id="test_Creinhardtii_QE_pH8" version="1.1.0" xmlns="http://psi.hupo.org/ms/mzml">'
ns_suffix = '</mzML>'
data = self.Reader.read_block(identifier)
element = XML(ns_prefix + data.decode('utf-8') + ns_suffix)
if 'chromatogram' in element[0].tag:
return spec.Chromatogram(list(element)[0], measured_precision=5e-6)
else:
return spec.Spectrum(list(element)[0], measured_precision=5e-6)