A set of Python utilities for reading JCAMP-DX files.

The JCAMP_reader() function takes a filename as input, and returns a dictionary containing the data found in the file. Specifically, the keys contained in the dictionary are: (1) the field names found in the file's header, with values being int- or float-type if the corresponding field is a numerical type, or a string-type otherwise. (2) two arrays x and y, giving the scaled values of the data points (scaled according to the xfactor and yfactor fields in the header, if they exist. The units of x and y are whatever are indicated in the header fields xunits and yunits, if these exist.

If the input is a compound file, then the returned dictionary will contain a children field. This field is an array of dictionaries that each represent a block.

The JCAMP_calc_xsec() function is intended to takes as input the result of the JCAMP_reader() function and to convert the x data to wavelength in microns, and the y data to cross-section in units of m^2 for gas concentration of 1ppm at standard atmospheric pressure and temperature, across a path length of 1 meter. The JCAMP_calc_xsec() function takes as input the data dictionary jcamp_dict, and manipulates that dictionary directly without having a separate return value. Changes to the dictionary may including adding the fields::

wavelengths: the array of wavelength values (in microns) for each data point
wavenumbers: the array of wavenumber values (in cm^-1) for each data point
xsec: the array of cross-section values (in units of m^2 at 1ppm.m)

and modifying the fields::

xunits: micron
yunits: m^2 at 1ppm.m

The optional arguments wavemin, wavemax are used if the user wishes to truncate the data to only a desired spectral range. For example, setting wavemin=8.0 and wavemax=12.0 means that the returned data arrays will only contain data corresponding to those wavlengths. If the skip_nonquant optional input argument is used, then any input spectrum that does not have the complete path_length and partial_pressure fields in its dictionary will be passed without modification. (That is, no conversion to quantitative cross-section will be attempted.) If this option is set to True, then if the code finds missing data, it will attempt to generate a quantitative cross-section by guessing the missing values. Based upon NIST's infrared database, typical values for guessing here are partial_pressure = 150.0 mmHg and path length = 0.1 m.

You can view a notebook demo in the doc folder to see how you can produce a series of plots showing spectra.

The repository comes with four folders containing JCAMP-format files: infrared_spectra/, mass_spectra/, raman_spectra/, and uvvis_spectra. These were downloaded from freely-available internet databases, and can be used as example format files.

In order to use jcamp for online queries rather than static text files, we can use the following procedure with the requests package:

response = requests.get(something)
content = response.content.splitlines()
content = [line.decode("utf-8") for line in content]
data_dict = jcamp_read(content)