numpy - Extract data from a txt file that contains data blocks and strings using python -

i need postprocess output file model using python. output file has combination of data , strings. first, want separate strings data , save columns 0,1 , 2 each output time (only data, no strings) in separate text file. so, below example, have 3 text files (for time=0, time=0.01, time=0.04) each containing data each output time without header or other strings in them. short form of output file model looks this:

 ******* program ******    ******* program ******    ******* program ******                                                     date:  26. 4.    time:  15:40:32  units: l = cm   , t = days , m = mmol     time:     0.000000   node    depth    head   moisture   headf  moisturef      flux          [l]      [l]     [-]       [l]       [-]       [l/t]        1     0.00   -1000.00 0.1088    -1000.00 0.002508  -0.562e-03     2    -0.04   -1000.00 0.1088    -1000.00 0.002508  -0.562e-03     3    -0.08   -1000.00 0.1088    -1000.00 0.002508  -0.562e-03  end    time:     0.010000   node    depth    head   moisture   headf  moisturef      flux              [l]      [l]     [-]       [l]       [-]       [l/t]        1     0.00    -666.06 0.1304      -14.95 0.139033  -0.451e-02     2    -0.04    -666.11 0.1304      -15.01 0.138715  -0.887e-02     3    -0.08    -666.35 0.1304      -15.06 0.138394  -0.174e-01  end    time:     0.040000   node    depth    head   moisture   headf  moisturef      flux               [l]      [l]     [-]       [l]       [-]       [l/t]         1     0.00    -324.87 0.1720      -12.30 0.157799  -0.315e-02      2    -0.04    -324.84 0.1720      -12.31 0.157724  -0.628e-02      3    -0.08    -324.83 0.1720      -12.32 0.157649  -0.125e-01   end 

i found following code question posted in stackoverflow earlier. here link question: enter link description here

that problem similar mine; however, have problems modifying solve problem. how should modify problem? or should use strategy approach problem?

def parse_dpt(lines):     dpt = []     while lines:         line = lines.pop(0).lstrip()         if line == ' ' or line.startswith('*'):             continue         if line.startswith('*'):             lines.insert(0, line)             break         data = line.split(' ')         # pick columns 0, 1, 2 ,         # convert appropiate numeric format         # , append list current dpt , step         dpt.append([int(data[0]), float(data[1]), float(data[2])])     return dpt  raw = [] open('nod_inftest.txt') nit:     lines = nit.readlines() while lines: line = lines.pop(0)  if line.startswith(''):     if line.find('time:') > -1:         raw.append(parse_dpt(lines))  pprint import pprint raw_step in zip(raw):     print 'raw:'     pprint(raw_step) 

here error message python:

'import sitecustomize' failed; use -v traceback traceback (most recent call last):   file "c:\users\desktop\python test\p-test3.py", line 58, in <module>     raw.append(parse_dpt(lines))   file "c:\users\desktop\python test\p-test3.py", line 35, in parse_dpt     dpt.append([int(data[0]), float(data[1]), float(data[2])]) valueerror: invalid literal int() base 10: 'units:' 

if understood question code should trick:

import re  open('in.txt', 'r') in_file:     file_content = in_file.read()     blocks = re.findall(         'time:\s*\d+\.\d*(.*?)end',         file_content,         re.dotall         )      file_number = 1     block in blocks:         open('out%s.txt'%str(file_number), 'w') out_file:             row in re.findall(                     '\s*(-?\d+.?\d*)\s*(-?\d+.?\d*)\s*(-?\d+.?\d*).*',                     block):                 out_file.write(row[0] + ' ' + row[1] + ' ' + row[2] + '\n')         file_number += 1 

the code assumes files containing text called in.txt