def JCAMP_reader(file_name: Union[str, os.PathLike]) -> GCMS_data:
"""
Generic reader for JCAMP DX files
:param file_name: Path of the file to read
:type file_name: str or os.PathLike
:return: GC-MS data object
:rtype: :class:`pyms.GCMS.Class.GCMS_data`
:authors: Qiao Wang, Andrew Isaac, Vladimir Likic, David Kainer, Dominic Davis-Foster (pathlib support)
"""
if not is_path(file_name):
raise TypeError("'file_name' must be a string or a PathLike object")
file_name = prepare_filepath(file_name, mkdirs=False)
print(f" -> Reading JCAMP file '{file_name}'")
lines_list = file_name.open('r')
data = []
page_idx = 0
xydata_idx = 0
time_list = []
scan_list = []
header_info = {} # Dictionary containing header information
for line in lines_list:
if len(line.strip()) != 0:
# prefix = line.find('#')
# if prefix == 0:
if line.startswith("##"):
# key word or information
fields = line.split('=', 1)
fields[0] = fields[0].lstrip("##").upper()
fields[1] = fields[1].strip()
if "PAGE" in fields[0]:
if "T=" in fields[1]:
# PAGE contains retention time starting with T=
# FileConverter Pro style
time = float(fields[1].lstrip(
"T=")) # rt for the scan to be submitted
time_list.append(time)
page_idx = page_idx + 1
elif "RETENTION_TIME" in fields[0]:
# OpenChrom style
time = float(fields[1]) # rt for the scan to be submitted
# Check to make sure time is not already in the time list;
# Can happen when both ##PAGE and ##RETENTION_TIME are specified
if time_list[-1] != time:
time_list.append(time)
elif fields[0] in xydata_tags:
xydata_idx = xydata_idx + 1
elif fields[0] in header_info_fields:
if fields[1].isdigit():
header_info[fields[0]] = int(fields[1])
elif is_float(fields[1]):
header_info[fields[0]] = float(fields[1])
else:
header_info[fields[0]] = fields[1]
# elif prefix == -1:
else:
# Line doesn't start with ##
# data
if page_idx > 1 or xydata_idx > 1:
if len(data) % 2 == 1:
# TODO: This means the data is not in x, y pairs
# Make a better error message
raise ValueError("data not in pair !")
mass_list = []
intensity_list = []
for i in range(len(data) // 2):
mass_list.append(data[i * 2])
intensity_list.append(data[i * 2 + 1])
if len(mass_list) != len(intensity_list):
raise ValueError(
"len(mass_list) is not equal to len(intensity_list)"
)
scan_list.append(Scan(mass_list, intensity_list))
data = []
data_sub = line.strip().split(',')
for item in data_sub:
if not len(item.strip()) == 0:
data.append(float(item.strip()))
if page_idx > 1:
page_idx = 1
if xydata_idx > 1:
xydata_idx = 1
else:
data_sub = line.strip().split(',')
for item in data_sub:
if not len(item.strip()) == 0:
data.append(float(item.strip()))
if len(data) % 2 == 1:
# TODO: This means the data is not in x, y pairs
# Make a better error message
raise ValueError("data not in pair !")
# get last scan
mass = []
intensity = []
for i in range(len(data) // 2):
mass.append(data[i * 2])
intensity.append(data[i * 2 + 1])
if len(mass) != len(intensity):
raise ValueError("len(mass) is not equal to len(intensity)")
scan_list.append(Scan(mass, intensity))
# sanity check
time_len = len(time_list)
scan_len = len(scan_list)
if time_len != scan_len:
print(time_list)
print(scan_list)
raise ValueError(
f"Number of time points ({time_len}) does not equal the number of scans ({scan_len})"
)
data = GCMS_data(time_list, scan_list)
return data
def from_jcamp(cls,
file_name: PathLike,
ignore_warnings: bool = True) -> "ReferenceData":
"""
Create a ReferenceData object from a JCAMP-DX file.
:param file_name: Path of the file to read.
:param ignore_warnings: Whether warnings about invalid tags should be shown.
:authors: Qiao Wang, Andrew Isaac, Vladimir Likic, David Kainer, Dominic Davis-Foster
"""
with warnings.catch_warnings():
if ignore_warnings:
warnings.simplefilter("ignore", JcampTagWarning)
file_name = PathPlus(file_name)
# Commented this line because it also gets printed when the MassSpectrum is created
# print(f" -> Reading JCAMP file '{file_name}'")
lines_list = file_name.read_lines()
last_tag = None
header_info: Dict[str, Any] = {
} # Dictionary containing header information
for line in lines_list:
if len(line.strip()):
if line.startswith("##"):
# key word or information
fields = line.split('=', 1)
current_tag = fields[0] = fields[0].lstrip(
"##").upper()
last_tag = fields[0]
fields[1] = fields[1].strip()
if current_tag.upper().startswith("END"):
break
elif current_tag in xydata_tags:
continue
elif current_tag in header_info_fields:
if fields[1].isdigit():
header_info[current_tag] = int(fields[1])
elif is_float(fields[1]):
header_info[current_tag] = float(fields[1])
else:
header_info[current_tag] = fields[1]
else:
warnings.warn(current_tag, JcampTagWarning)
else:
if last_tag in header_info:
header_info[last_tag] += f"{line}"
return cls(
name=header_info["TITLE"],
cas=header_info["CAS REGISTRY NO"],
nist_no=header_info["$NIST MASS SPEC NO"],
contributor=header_info["ORIGIN"],
formula=header_info["MOLFORM"],
mw=header_info["MW"],
mass_spec=MassSpectrum.from_jcamp(file_name),
)