As tsfromtxt like ability to convert multiple text file columns into a single date column

[wesm]

No description provided.

[wesm]

example data from @timmie

ID,date,NominalTime,ActualTime,TDew,TAir,Windspeed,Precip,WindDir
 KORD,1999027, 19:00:00, 18:56:00, 0.8100, 2.8100, 7.2000, 0.0000, 280.0000
 KORD,1999027, 20:00:00, 19:56:00, 0.0100, 2.2100, 7.2000, 0.0000, 260.0000
 KORD,1999027, 21:00:00, 20:56:00, -0.5900, 2.2100, 5.7000, 0.0000, 280.0000
 KORD,1999027, 21:00:00, 21:18:00, -0.9900, 2.0100, 3.6000, 0.0000, 270.0000
 KORD,1999027, 22:00:00, 21:56:00, -0.5900, 1.7100, 5.1000, 0.0000, 290.0000
 KORD,1999027, 23:00:00, 22:56:00, -0.5900, 1.7100, 4.6000, 0.0000, 280.0000

[changhiskhan]

Isn't this in #1174 already?

[timmie]

The following is liked and may be considered:
#854 (comment)

and after this issue here is solved, we may start with:
#1180

[timmie]

to comment:
#1186 (comment)

can both issues be merged?

[wesm]

This looks good now, we'll need to write docs prior to release

[wesm]

The only remaining question might be: should the combined columns be dropped from the resulting DataFrame?

[timmie]

The only remaining question might be: should the combined columns be dropped from the resulting DataFrame?
I think it's possible because they're in the index.

[changhiskhan]

We should add another keyword to control that actually. And perhaps also to control whether to put the new date columns first or last

[timmie]

Could you add a small documentation snippet?

I have seen that there's a test file, but something more to the front would be nice.

BTW, is the website with the dev docs updated on nightly basis?

[changhiskhan]

The docs are not updated nightly right now but will be soon.
Thanks for the suggestion. I will add a snippet to the docs.
On May 17, 2012 8:32 AM, "timmie" <
reply@reply.github.com>
wrote:

Could you add a small documentation snippet?

I have seen that there's a test file, but something more to the front
would be nice.

BTW, is the website with the dev docs updated on nightly basis?

---

Reply to this email directly or view it on GitHub:
#1186 (comment)

[timmie]

This is still not transparent to me:

This works weel:

from pandas import read_table
from cStringIO import StringIO
import os
N = 10000
K = 8
data = '''\
KORD,19990127, 19:00:00, 18:56:00, 0.8100, 2.8100, 7.2000, 0.0000, 280.0000
KORD,19990127, 20:00:00, 19:56:00, 0.0100, 2.2100, 7.2000, 0.0000, 260.0000
KORD,19990127, 21:00:00, 20:56:00, -0.5900, 2.2100, 5.7000, 0.0000, 280.0000
KORD,19990127, 21:00:00, 21:18:00, -0.9900, 2.0100, 3.6000, 0.0000, 270.0000
KORD,19990127, 22:00:00, 21:56:00, -0.5900, 1.7100, 5.1000, 0.0000, 290.0000
'''


data = data * 2000


mydata = read_table(StringIO(data), sep=',', header=None, parse_dates={'date1' : [1, 2], 'date2' : [1,3]})

import pandas as pd

myseries = pd.Series(data=mydata['X.5'], index=mydata.date1)

but changing to

mydata = read_table(StringIO(data), sep=',', header=None, parse_dates=[1, 2])

does not return the desired combined date time column

[changhiskhan]

you want parse_dates=[[1, 2]]

parse_dates=[1, 2] means you want to parse each of columns 1 and 2 as dates

I'll see if I can make the docs a little more clear before the 0.8 release.