regex for selectors

[ozgurkalan]

Hi,
Regex for selectors would be very helpful. Can be injected in contains.

My case: columns having Z1 to Z6 and Zcode, Hcode, Zdesc....
When use s.startswith("Z") then I cannot distinguish 'Zdesc' from 'Z1'.

my idea would be:

# search: None for t.columns, a list, or selectors expr 
s.reg(pattern="^Z\d",  search=search ) 

[cpcloud]

Hi @ozgurkalan, regular expression based selectors are already implemented as s.matches:

In [3]: from ibis.interactive import *

In [4]: t = ibis.examples.penguins.fetch()

In [5]: t.columns
Out[5]:
['species',
 'island',
 'bill_length_mm',
 'bill_depth_mm',
 'flipper_length_mm',
 'body_mass_g',
 'sex',
 'year']

In [6]: t.select(s.matches("^bill_"))
Out[6]:
┏━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ bill_length_mm ┃ bill_depth_mm ┃
┡━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ float64        │ float64       │
├────────────────┼───────────────┤
│           39.1 │          18.7 │
│           39.5 │          17.4 │
│           40.3 │          18.0 │
│            nan │           nan │
│           36.7 │          19.3 │
│           39.3 │          20.6 │
│           38.9 │          17.8 │
│           39.2 │          19.6 │
│           34.1 │          18.1 │
│           42.0 │          20.2 │
│              … │             … │
└────────────────┴───────────────┘

[ozgurkalan]

How did I miss that!
Many thanks, that's what I was looking for.🖖🏽

[cpcloud]

No worries! I'm going to leave the discussion open so that folks can find it later!