在R / Python中从非结构化数据中提取数据集

Question

We are trying to extract travel itineraries from travel requests, which are filled by standardized auditors.

Example:

EY  275   13FEB HYDAUH 0425   0715  

here the Data would imply as below

EY> Travel Type
275> Flight Number
13FEB> Date of travel
HYDAUH>Airports involved during the Departure and arrival.
0425   0715 > Boarding and landing times of the flight.

Here we would need to extract the individual data elements from the raw text fields, and later map them to their respective travel fields and compute several values.

Are there procedures in R/Python to achive them with minimal efforts.

I am looking for subsist functions/procedures for the data split/mappings.

Answer 1

If you can extract a single record, as shown in your second example, and if there is always at least one space between the fields, then pulling out the individual pieces of data is straightforward in Python:

>>> itin = 'EY  275   13FEB HYDAUH 0425   0715'
>>> ifields = itin.split()
>>> ifields[0] # travel type
'EY'
>>> ifields[1] # flight number
'275'
>>> ifields[2] # date of travel
'13FEB'
>>> ifields[3][0:3] # departure airport
'HYD'
>>> ifields[3][3:6] # destination airport
'AUH'
>>> ifields[4] # boarding time
'0425'
>>> ifields[5] # landing time
'0715'

Your first example shows a second record following directly on from the first with no space - is that correct? If so, is each record always the same number of characters in length?

>>> itinline = 'QR 529  09AUG MAADOH  0405  0600QR  67  09AUG DOHFRA  0815'
>>> itinline[0:32]
'QR 529  09AUG MAADOH  0405  0600'
>>> itinline[32:64]
'QR  67  09AUG DOHFRA  0815'

If your data has multiple records of variable length on a single line, or if there may or may not be spaces between each field, the parsing becomes more complex, but should still be fairly easy to do in Python. In this case please post a more complete example with several records and show the output you want to get.