number of spectrum is different for each sample

[hadas-equinom]

Hi @nzhagen

First, thank you for this much needed package.

I have been trying to run it, but I got different headers [x] values for each samples. not only different on value, but also different in size. instead of getting 1400 spectrum points I get a different number for each samples.

Can you please check this issue? or if there is already an existing solution, please let me know,

Thank you,
Hadas

[nzhagen]

@hadas-equinom,
Actually, that behavior sounds correct to me. The code is designed to read a JCAMP format file, so if two files sample spectra at different wavelengths, then the code will exactly reflect that. Or am I possibly misreading the issue that you are seeing? If so, please send some example code so I can try to replicate the issue...

[hadas-equinom]

@nzhagen thank you for your response.
Unfortunately I cannot send an example of the data I have, but the only code I ran is:
file = jcamp.jcamp_readfile('\\Data\\iqx_file.dx')

as you can see the data was not extracted correctly. you can see the npoints is not matching the length of x and y [see figure below] . After some debugging it seems that it might be related to how you interpret the duplicates of the data [by taking only one letter] . somewhere in here:

jcamp/jcamp.py

Line 515 in 86daf5b

mul = DUP_digits[c]

[hadas-equinom]

in addition the version I get after pip install --upgrade jcamp , is different from what I see in git.
here is what I see in my pycharm:

[nzhagen]

Ah, now I see what you mean -- that difference in the number of points is definitely an issue. And, yeah, it looks like the PyPI version of the package has fallen behind the GitHub version -- I must have forgotten to do an update. I will make an update soon. But first, could you send to me the JCAMP file you are using? I don't have a good example file that uses the DUP codes, so my ability to test without that is limited.

[hadas-equinom]

@nzhagen well, I cannot share the files, but someone from my work was able to fix it. see the following code:

if any(c in DUP_set for c in line):
        ## Split the line into individual characters so that you can check for coded characters one-by-one.
        newline = ''
        mul_str = ''
        cont = 0
        for (i,c) in enumerate(line):
            while cont > 0:
                cont -= 1
                continue
            if (c in DUP_digits):
                ## Check for last DIF_digit which is start of last y-value by default, so that all
                ## characters belonging to last value is fully decompressed by DUP compression.
                back = 1
                while line[i-back] not in DIF_digits:
                    back += 1
                prev_c = line[i-back:i]
                fwd = 1
                cont = 0
                mul_str = ''
                while line[i+fwd] not in DIF_digits:
                    mul_str += line[i+fwd]
                    cont += 1
                    fwd += 1
                if mul_str:
                    mul = (DUP_digits[c] * (10 ** cont)) + int(mul_str)
                else:
                    mul = DUP_digits[c]
                newline += prev_c * (mul-1)
            else:
                mul = ''
                newline += c
        line = "".join(newline)