Parse flat-file (positional text-file) to read the wavelength

Question

I have the next txt with data:

FI R       83.0000m   34.960    1.1262      Fe 2      1.32055m   33.626    0.0522      N  
2      5754.61A   33.290    0.0241
TI R       1800.00m   33.092    0.0153      Fe 2      1.24854m   32.645    0.0054      N  
2      915.612A   31.997    0.0012
NI Ra      2.85000m   36.291   24.1132      Fe 2      7637.54A   33.077    0.0147      

what I want is to obtain the third column, is the wavelength of the emergent line, but my problem is when I put the condition in the if.

Name1,ion1,wavelength1,da1,de1,name2,ion2,
wavelength2,da2,de2,name3,ion3,wavelength3,da3,de3=np.genfromtxt('Emergent_line.txt', 
skip_header=3, delimiter="", unpack=True)

if(Name1=="Fe" and ion1==2):
    print(wavelength1)
elif(name2=="Fe" and ion2==2):
    print(wavelength2)
elif(name3=="Fe" and ion3==2):
    print(wavelength3)

In the txt I want to find the wavelength for Fe 2, but I think the problem is that the wavelength have a letter in the end, I don't want to delete, because I have a large list like that. I tried another froms, but I haven't solved it.

Answer 1

I think you are better off using regex

Example:

import re


text = '''FI R       83.0000m   34.960    1.1262      Fe 2      1.32055m   33.626    0.0522      N  
2      5754.61A   33.290    0.0241
TI R       1800.00m   33.092    0.0153      Fe 2      1.24854m   32.645    0.0054      N  
2      915.612A   31.997    0.0012
NI Ra      2.85000m   36.291   24.1132      Fe 2      7637.54A   33.077    0.0147'''

find_this = re.findall('(Fe 2.*?[0-9].*?)\s', text)
print(find_this)

Output:

['Fe 2      1.32055m', 'Fe 2      1.24854m', 'Fe 2      7637.54A']

[Program finished]

Or if you only want the values.

find_this = re.findall('Fe 2.*?([0-9].*?)\s', text)

Output:

['1.32055m', '1.24854m', '7637.54A']

[Program finished]

ANSWER TO NEW QUESTION

Here's an example of how you could achieve picking out values between 1.35 - 1.40 using a for loop and converting var into float. Now we can use conditions as in this line:

if (float_value >= 1.35) and (float_value <= 1.40):
    print(value)

if matched it prints the untouched string, keeping the ending letter.

Here's the full code: (I shortened the text for easier read)

import re


text = '''Fe 2      1.405A   33.077    0.0147
Fe 2      1.305A   33.077    0.0147
Fe 2      1.345A   33.077    0.0147
Fe 2      1.35A   33.077    0.0147
Fe 2      1.35623A   33.077    0.0147
Fe 2      1.40A   33.077    0.0147
Fe 2      1.37A   33.077    0.0147
Fe 2      1.41A   33.077    0.0147'''


find_this = re.findall('Fe 2.*?([0-9].*?)\s', text)

for value in find_this:
    del_letters = re.sub('[A-Za-z]', '', value)
    float_value = float(del_letters)
    if (float_value >= 1.35) and (float_value <= 1.40):
        print(value)

Output:

1.35A
1.35623A
1.40A
1.37A

Answer 2

The text-file you presented seems a flat-file or fixed-with file where data (columns) are layed out

• as positional text (each column starting at a predefined position)
• in a fixed-width format (each column having a fixed-width)

Pandas has a method for reading fixed-width file

You could use pandas and their IO tools method read_fwf.

import io  # just for demonstration without needing a file
import pandas


text = '''FI R       83.0000m   34.960    1.1262      Fe 2      1.32055m   33.626    0.0522      N  
2      5754.61A   33.290    0.0241
TI R       1800.00m   33.092    0.0153      Fe 2      1.24854m   32.645    0.0054      N  
2      915.612A   31.997    0.0012
NI Ra      2.85000m   36.291   24.1132      Fe 2      7637.54A   33.077    0.0147'''

buffer = io.StringIO(text)  # just a helper to read from text as from file

filepath_or_buffer = buffer  # can also be the file-path directly
df =  pandas.read_fwf(filepath_or_buffer, colspecs='infer', widths=None, infer_nrows=100, header=None)
print(df)  # df represented as complete table read

wave_lengths = df.loc[(df[3] == 'Fe') & (df[4] == 2)][5]
print("== Wavelengths:")
print(wave_lengths)

buffer.close()

Prints:

    0    1                            2    3    4         5       6       7    8
0  FI    R  83.0000m   34.960    1.1262   Fe  2.0  1.32055m  33.626  0.0522    N
1   2  NaN  5754.61A   33.290    0.0241  NaN  NaN       NaN     NaN     NaN  NaN
2  TI    R  1800.00m   33.092    0.0153   Fe  2.0  1.24854m  32.645  0.0054    N
3   2  NaN  915.612A   31.997    0.0012  NaN  NaN       NaN     NaN     NaN  NaN
4  NI   Ra  2.85000m   36.291   24.1132   Fe  2.0  7637.54A  33.077  0.0147  NaN
== Wavelengths:
0    1.32055m
2    1.24854m
4    7637.54A

Note:

• Python's io.StringIO was used as helper to simulate a buffer instead the file.
• Panda's loc method to locate or filter the Fe 2 rows, where we printed the 5th column with wavelength

See also

• Towards Data Science: Parsing Fixed Width Text Files with Pandas | by Amy Rask
• Pandas' IO tools (text, CSV, HDF5, …)
• Parsing out data from a flat file
• python - Parsing a text file into a pandas DataFrame - Code Review Stack Exchange
• How to efficiently parse fixed width files?

Answer 3

Here is another idea, without using the re module:

someText ='FI R       83.0000m   34.960    1.1262'
someText.split()
#>> ['FI', 'R', '83.0000m', '34.960', '1.1262']
name1,ion1, lambda1, *other = someText.split()
lambda1 = float(lambda1[0:-1])
print(lambda1, other)
#>> 83.0 ['34.960', '1.1262']

You can use the .split() str method, which splits on the whitespace separating your data without the need for regex.

Regex is great for extracting more complex text formatting cases usually when the input varies. Since the input is not really varying in this case, you could also use simpler str methods instead.