Extract Text Table From File

[Tim Chase]

On 08/27/12 04:53, Huso wrote:
> Below is just ONE block of the traffic i have in the log files. There will be more in them with different data.
> 
> ROUTES TRAFFIC RESULTS, LSR
> TRG  MP   DATE   TIME
>  37  17 120824   0000
> 
> R         TRAFF   NBIDS   CCONG   NDV  ANBLO   MHTIME  NBANSW
> AABBCCO     6.4     204     0.0   115    1.0    113.4     144
> AABBCCI     3.0     293           115    1.0     37.0     171
> DDEEFFO     0.2       5     0.0    59    0.0    107.6       3
> HHGGFFI     0.3      15            30    0.0     62.2       4
> END

In the past I've used something like the following to find columnar
data based on some found headers:

  import re
  token_re = re.compile(r'\b(\w+)\s*')
  f = file(FILENAME)
  headers = f.next() # in your case, you'd
                     # search forward until
                     # you got to a header line
                     # and use that TRAFF... line
  header_map = dict(
    # build a map of field-name to slice
    (
      matchobj.group(1).upper(),
      slice(*matchobj.span())
    )
    for matchobj
    in token_re.finditer(headers)
    )

You can then access your values as you iterate through the rest of
the rows:

  for row in f:
    if row.startswith("END"): break
    traff = float(row[header_map["TRAFF"]])
    # ...

which makes the code pretty easy to read, effectively turning it
into a CSV file.

It has the advantage that, if for some reason data in the columns
have spaces in them, it won't throw off the row as a .split() would.

-tkc