Compromising Python if statement

Question

I have to work with a bunch of if statement in one code. They are all the same with slight changes. Is there any way how I can compromise all this code and make it more elegant and shorter ?

Code below:

if con_name == 'coh':
    coh = my_coherence(n_freqs, Rxy_mean, Rxx_mean, Ryy_mean)
    coh_surro = my_coherence(n_freqs, Rxy_s_mean, Rxx_s_mean, Ryy_s_mean)
    return coh, coh_surro, freqs, freqs_surro

if con_name == 'imcoh':
    imcoh = my_imcoh(n_freqs, Rxy_mean, Rxx_mean, Ryy_mean)
    imcoh_surro = my_imcoh(n_freqs, Rxy_s_mean, Rxx_s_mean, Ryy_s_mean)
    return imcoh, imcoh_surro, freqs, freqs_surro

if con_name == 'cohy':
    cohy = my_cohy(n_freqs, Rxy_mean, Rxx_mean, Ryy_mean)
    cohy_surro = my_cohy(n_freqs, Rxy_s_mean, Rxx_s_mean, Ryy_s_mean)
    return cohy, cohy_surro, freqs, freqs_surro

if con_name == 'plv':
    plv = my_plv(n_freqs, Rxy, Rxy_mean)
    plv_surro = my_plv(n_freqs, Rxy_s, Rxy_s_mean)
    return plv, plv_surro, freqs, freqs_surro

if con_name == 'pli':
    pli = my_pli(n_freqs, Rxy, Rxy_mean)
    pli_surro = my_pli(n_freqs, Rxy_s, Rxy_s_mean)
    return pli, pli_surro, freqs, freqs_surro

if con_name == 'wpli':
    wpli = my_wpli(n_freqs, Rxy, Rxy_mean)
    wpli_surro = my_wpli(n_freqs, Rxy_s, Rxy_s_mean)
    return wpli, wpli_surro, freqs, freqs_surro

I am sorry if this is to easy, but I tried and tried and can't figure out a way.

elif is useless since there is a return in all statements. — Hacketo
– Hacketo, Commented Jun 29, 2015 at 11:40

Wonjung Kim · Accepted Answer · 2015-06-29 11:51:16Z

5

Without reflection

func, flag = {
    "coh": (my_coherence, True),
    "imcoh": (my_imcoh, True)
    "cohy": (my_cohy, True),
    "ply": (my_plv, False),
    "pli": (my_pli, False),
    "wpli": (my_wpli, False)
}[con_name]

args = (n_freqs, Rxy_mean, Rxx_mean, Ryy_mean) if flag else (n_freqs, Rxy, Rxy_mean)
surro_args = (n_freqs, Rxy_s_mean, Rxx_s_mean, Ryy_s_mean) if flag else (n_freqs, Rxy_s Rxy_s_mean)

val = func(*args)
surro = func(*surro_args)
return val, surro, freqs, freqs_surro

OR This is also possible

...
args = (Rxy_mean, Rxx_mean, Ryy_mean) if flag else (Rxy, Rxy_mean)
surro_args = (Rxy_s_mean, Rxx_s_mean, Ryy_s_mean) if flag else (Rxy_s Rxy_s_mean)

val = func(n_freqs, *args)
surro = func(n_freqs *surro_args)
...

Maybe there is a cool name for flag for classifying those functions. Use it instead.

edited Jun 29, 2015 at 11:51

answered Jun 29, 2015 at 11:45

Wonjung Kim

1,93315 silver badges18 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Renatius · Accepted Answer · 2015-06-29 11:45:45Z

2

I would use a dictionary. Something like:

functions = {'coh':my_coherence,'imcoh':my_imcoh....}

Which would be called like this:

functions[con_name](n_freqs, Rxy_s_mean, Rxx_s_mean, Ryy_s_mean)

answered Jun 29, 2015 at 11:45

Renatius

5621 gold badge3 silver badges12 bronze badges

Comments

fferri · Accepted Answer · 2015-06-29 11:51:49Z

1

r = {
    'coh':   lambda: (my_coherence(n_freqs, Rxy_mean, Rxx_mean, Ryy_mean), my_coherence(n_freqs, Rxy_s_mean, Rxx_s_mean, Ryy_s_mean), freqs, freqs_surro),
    'imcoh': lambda: (my_imcoh(n_freqs, Rxy_mean, Rxx_mean, Ryy_mean), my_imcoh(n_freqs, Rxy_s_mean, Rxx_s_mean, Ryy_s_mean), freqs, freqs_surro),
    'cohy':  lambda: (my_cohy(n_freqs, Rxy_mean, Rxx_mean, Ryy_mean), my_cohy(n_freqs, Rxy_s_mean, Rxx_s_mean, Ryy_s_mean), freqs, freqs_surro),
    'plv':   lambda: (my_plv(n_freqs, Rxy, Rxy_mean), my_plv(n_freqs, Rxy_s, Rxy_s_mean), freqs, freqs_surro),
    'pli':   lambda: (my_pli(n_freqs, Rxy, Rxy_mean), my_pli(n_freqs, Rxy_s, Rxy_s_mean), freqs, freqs_surro),
    'wpli':  lambda: (my_wpli(n_freqs, Rxy, Rxy_mean), my_wpli(n_freqs, Rxy_s, Rxy_s_mean), freqs, freqs_surro),
    }
return r[con_name]()

You can further avoid repetitions, for example last two items are repeated for each case:

r = {
    'coh':   lambda: (my_coherence(n_freqs, Rxy_mean, Rxx_mean, Ryy_mean), my_coherence(n_freqs, Rxy_s_mean, Rxx_s_mean, Ryy_s_mean)),
    'imcoh': lambda: (my_imcoh(n_freqs, Rxy_mean, Rxx_mean, Ryy_mean), my_imcoh(n_freqs, Rxy_s_mean, Rxx_s_mean, Ryy_s_mean)),
    'cohy':  lambda: (my_cohy(n_freqs, Rxy_mean, Rxx_mean, Ryy_mean), my_cohy(n_freqs, Rxy_s_mean, Rxx_s_mean, Ryy_s_mean)),
    'plv':   lambda: (my_plv(n_freqs, Rxy, Rxy_mean), my_plv(n_freqs, Rxy_s, Rxy_s_mean)),
    'pli':   lambda: (my_pli(n_freqs, Rxy, Rxy_mean), my_pli(n_freqs, Rxy_s, Rxy_s_mean)),
    'wpli':  lambda: (my_wpli(n_freqs, Rxy, Rxy_mean), my_wpli(n_freqs, Rxy_s, Rxy_s_mean)),
    }
return r[con_name]() + (freqs, freqs_surro)

edited Jun 29, 2015 at 11:51

answered Jun 29, 2015 at 11:44

fferri

19k8 gold badges57 silver badges110 bronze badges

2 Comments

Eli Korvigo Over a year ago

I was about to write a similar answer, but I believe it's better to pass arguments to lambdas for them to be proper high-order functions, e.g. the call would look like r[con_name](...arguments for inner functions...)

Daniel Velden Over a year ago

@ mescalinum: Thanks. It's a nice approach. I corrected your tiny mistakes and it works fine.

Lajos Arpad · Accepted Answer · 2015-06-29 11:39:49Z

0

You could use Reflection. I would create a method, which would be responsible for calling the given methods. I would pass the two method names in each case and the parameters. Based on the method param names you could call each time the right function and return what is needed. And in the code chunk you have shown I would call the method, passing the two method names and the parameters.

answered Jun 29, 2015 at 11:39

Lajos Arpad

80.4k42 gold badges123 silver badges235 bronze badges

Collectives™ on Stack Overflow

Compromising Python if statement

4 Answers 4

Comments

Comments

2 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

Comments

Comments

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related