Reading LAS file using lasio library doesn't handle "-" in datetime column

Question

I have a las file and I am trying to read it in python using lasio library, one of the columns is TIME which is in the following format: 00:00:00.22-04-23

Sample of data copied from las file:

TIME               col1 col2
00:00:00.22-06-23  1010  20
00:00:05.22-06-23  1020  25
00:00:10.22-06-23  1015  32

My code to read the data:

df = lasio.read(file_path).df().reset_index()

This returns the df in the following format:

TIME               col1 col2 UNKNOWN:1  UNKNOWN:2
00:00:00.22         -06 -23    1010       20
00:00:05.22         -06 -23    1020       25
00:00:10.22         -06 -23    1015       32

As you can see, my TIME column has been split into three columns at every -. The data from col1 and col2 have been shifted to UNKNOWN:1 and UNKNOWN:2 (probably these columns are created by lasio during reading). I need it to return the TIME column as in the original form and avoid shifting the values of col1 and col2, so I can strip, split and manipulate TIME using pandas once it is read into a dataframe.

Any advice is appreciated.

I need to read the data into dataframe so I can use pandas to manipulate the columns as I need — serdar_bay, May 09 '23 at 16:46

Andrej Kesely · Accepted Answer · 2023-05-09T16:46:40.243

0

You can try to use pd.read_csv with correct delimiter. For example:

df = pd.read_csv('your_file.txt', sep=r"\s+", engine="python")
print(df)

Prints:

                TIME  col1  col2
0  00:00:00.22-06-23  1010    20
1  00:00:05.22-06-23  1020    25
2  00:00:10.22-06-23  1015    32

EDIT: With updated file:

import re
import pandas as pd
from io import StringIO

with open('your_file.txt', 'r') as f_in:
    data = re.sub(r'\A.*~A', '', f_in.read(), count=1, flags=re.S)
    df = pd.read_csv(StringIO(data), sep=r"\s+", engine="python")

print(df)

Prints:

                TIME     col1  col2   col3
0  00:00:00.23-04-23  1977.47   160  160.5

edited May 09 '23 at 16:46

answered May 09 '23 at 16:06

Andrej Kesely

168,389
15
48
91

This means I will need to change the extension of my las files to txt? Also, I will need to manually delete the header information on top of las file – serdar_bay May 09 '23 at 16:18
@serdar_bay No, just change `your_file.txt` for your filename. – Andrej Kesely May 09 '23 at 16:19
My file is las file and has lots of header info – serdar_bay May 09 '23 at 16:22
@serdar_bay I suggest to post how your file looks into the question (with correct formatting, not all - only the relevant parts). – Andrej Kesely May 09 '23 at 16:23
Please see sample of header, I only left the first row of data – serdar_bay May 09 '23 at 16:39
Thanks this worked perfectly well. Can you explain a bit your code! – serdar_bay May 09 '23 at 16:49
@serdar_bay I remove the header (from the beginning of the file to the `~A` string) and then use `pd.read_csv` to read the rest to the dataframe. – Andrej Kesely May 09 '23 at 16:50
Let us [continue this discussion in chat](https://chat.stackoverflow.com/rooms/253570/discussion-between-serdar-bay-and-andrej-kesely). – serdar_bay May 09 '23 at 16:50

Reading LAS file using lasio library doesn't handle "-" in datetime column

1 Answers1