Python将单列数据转换为多列

Question

I have a .txt file with simple numerical data in it. The data reflect multiple measurements of the same thing, and are simply written out in a long column. I want a script to read through the file, recognise the delimiter separating one experiment from the next and write it all out to separate columns in a txt or csv file.

At the moment, the data is delimited by the flag ' # row = X ' where X = 0 to ~128. So I want a script that will open the file, read up to 'row = 0', and then copy the next ~1030 lines of data to some list/array as "column 0". Then when it hits 'row = 1', copy the next ~1030 lines of numbers to "column 1'...and so on. Then it should write it out as multiple columns. The input data file looks like this:

# row = 0
9501.7734375
9279.390625
[..and so on for about 1030 lines...]
8836.5
8615.1640625
# row = 1
4396.1953125
4197.1796875
[..and so on for about 1030 lines...]
3994.4296875
# row = 2
9088.046875
8680.6953125
[..and so on for about 1030 lines...]
8253.0546875

The final file should look something like this:

row0          row1         row2       row3
9501.7734375  4396.1953125 etc        etc
9279.390625   4197.1796875
[..snip...]   [...snip...]
8836.5        3994.4296875
8615.1640625  3994.4347453

Preferably python as I have some experience from some years ago! Thanks everyone, Jon

Answer 1

from io import StringIO
from collections import OrderedDict

datastring = StringIO(u"""\
# row = 0
9501.7734375
9279.390625
8615.1640625
# row = 1
4396.1953125
4197.1796875
3994.4296875
# row = 2
9088.046875
8680.6953125
8253.0546875
""")      

content = datastring.readlines()
out = OrderedDict()
final = []

for line in content:
    if line.startswith('# row'):
        header = line.strip('\n#')
        out[header] = []
    elif line not in out[header]:
        out[header].append(line.strip('\n'))


for k, v in out.iteritems():
    temp = (k + ',' + ','.join([str(item) for item in v])).split(',')
    final.append(temp)

final = zip(*final)
with open("C:/temp/output.csv", 'w') as fout:
    for item in final:
    fout.write('\t'.join([str(i) for i in item]))

Output:

 row = 0         row = 1        row = 2
9501.7734375    4396.1953125    9088.046875
9279.390625     4197.1796875    8680.6953125
8615.1640625    3994.4296875    8253.0546875