I have a very messy text file that consist of both comma and space seperated data that looks like the following:
NBLOCK,3,,13
(1i9,3e20.9e3)
1 4.000000000E+01 -6.000000000E+01 0.000000000E+00
2 4.000000000E+01 6.000000000E+01 0.000000000E+00
3 4.000000000E+01 -2.000000000E+01 0.000000000E+00
4 4.000000000E+01 2.000000000E+01 0.000000000E+00
I need to clean up the file such that I get an easily searchable 2d array using the following rules:
The output should look something like the following
my_array = [['nblock','3','','13'],
['(1i9','3e20.9e3)','',''],
['1','4.000000000e+01','-6.000000000e+01','0.000000000e+00'],
['2','4.000000000e+01','6.000000000e+01','0.000000000e+00'],
['3','4.000000000e+01','-2.000000000e+01','0.000000000e+00'],
['4','4.000000000e+01','2.000000000e+01','0.000000000e+00']]
Update more realistic example of text file:
ET, 1, 42
KEYOP, 1, 2, 1
KEYOP, 1, 3, 3
RLBLOCK, 1, 1, 6, 7
N,R5.3,LOC, -1,
NBLOCK,3,,13
(1i9,3e20.9e3)
1 4.000000000E+01 -6.000000000E+01 0.000000000E+00
2 4.000000000E+01 6.000000000E+01 0.000000000E+00
3 4.000000000E+01 -2.000000000E+01 0.000000000E+00
4 4.000000000E+01 2.000000000E+01 0.000000000E+00
-1
MPTEMP,R5.0, 1, 1, 0.00000000 ,
MPDATA,R5.0, 1,EX , 1, 1, 2.100000000E+11,
MPTEMP,R5.0, 1, 1, 0.00000000 ,
MPDATA,R5.0, 1,NUXY, 1, 1, 0.300000000 ,
MPTEMP,R5.0, 1, 1, 0.00000000 ,
MPDATA,R5.0, 1,DENS, 1, 1, 7800.00000 ,
MPTEMP,R5.0, 1, 1, 0.00000000 ,
MPDATA,R5.0, 1,PRXY, 1, 1, 0.300000000 ,
EXTOPT,ATTR, 0, 0, 0
EXTOPT,ESIZE, 0, 0.0000
EXTOPT,ACLEAR, 0
BFUNIF,TEMP,_TINY
Another example:
DMPOPT,EMAT,NO
*IF,_CDRDOFF,EQ,1,THEN !if solid model was read in
_CDRDOFF= !reset flag, numoffs already performed
*ELSE !offset database for the following FE model
NUMOFF,NODE, 12
NUMOFF,ELEM, 8
NUMOFF,MAT , 1
NUMOFF,REAL, 1
NUMOFF,TYPE, 2
NUMOFF,CSYS, 12
*ENDIF
KUSE, 0
TIME, 0.00000000
NBLOCK,6,SOLID, 12, 12
(3i9,6e21.13e3)
1 0 0 4.0000000000000E+001-6.0000000000000E+001
2 0 0 4.0000000000000E+001 6.0000000000000E+001
3 0 0 4.0000000000000E+001-2.0000000000000E+001
4 0 0 4.0000000000000E+001 2.0000000000000E+001
EBLOCK,19,SOLID, 1250
(19i8)
1 1 1 1
1 1 1 1
1 1 1 1
-1
N,R5.3,LOC, -1,
MPDATA,R5.0, 1,EX , 1, 1, 2.100000000E+11,
MPTEMP,R5.0, 1, 1, 0.00000000 ,
MPDATA,R5.0, 1,NUXY, 1, 1, 0.300000000 ,
MPTEMP,R5.0, 1, 1, 0.00000000 ,
MPDATA,R5.0, 1,DENS, 1, 1, 7800.00000 ,
MPTEMP,R5.0, 1, 1, 0.00000000 ,
MPDATA,R5.0, 1,PRXY, 1, 1, 0.300000000
Consider this kind of solution:
import numpy as np
from pprint import pprint
class Block:
def __init__(self):
self.data = {}
self.array = []
def ingest(self, lines):
for line in lines:
if 'A' <= line[0] <= 'Z':
parts = [k.strip() for k in line.split(',')]
parts = [int(k) if k.isdigit() else k for k in parts]
key = parts[0].lower()
if key not in self.data:
self.data[key] = [parts[1:]]
else:
self.data[key].append( parts[1:] )
elif line[0] == '(':
self.data['nblock'].append( [line.strip().split(',')] )
self.data['nblock'].append( [] )
else:
parts = line.split()
if parts[0] != '-1':
self.data['nblock'][2].append( [float(k) for k in parts[1:]] )
blk = Block()
blk.ingest( open('x.txt') )
pprint(blk.data)
Output:
{'bfunif': [['TEMP', '_TINY']],
'et': [[1, 42]],
'extopt': [['ATTR', 0, 0, 0], ['ESIZE', 0, '0.0000'], ['ACLEAR', 0]],
'keyop': [[1, 2, 1], [1, 3, 3]],
'mpdata': [['R5.0', 1, 'EX', 1, 1, '2.100000000E+11', ''],
['R5.0', 1, 'NUXY', 1, 1, '0.300000000', ''],
['R5.0', 1, 'DENS', 1, 1, '7800.00000', ''],
['R5.0', 1, 'PRXY', 1, 1, '0.300000000', '']],
'mptemp': [['R5.0', 1, 1, '0.00000000', ''],
['R5.0', 1, 1, '0.00000000', ''],
['R5.0', 1, 1, '0.00000000', ''],
['R5.0', 1, 1, '0.00000000', '']],
'n': [['R5.3', 'LOC', '-1', '']],
'nblock': [[3, '', 13],
[['(1i9', '3e20.9e3)']],
[[40.0, -60.0, 0.0],
[40.0, 60.0, 0.0],
[40.0, -20.0, 0.0],
[40.0, 20.0, 0.0]]],
'rlblock': [[1, 1, 6, 7]]}
Note that I've provided a string lines iterator, but you could also pass a file. Anything that will iterate through a list of strings.