generate code for Boolean expressions

[Nico]

With `codegen`, one can generate code from a sympy expression. Typically, it is assumed that the return value is a scalar.

In one of my use cases, the return value is Boolean. (The answer to "Is x inside a particular domain?")

The hack I employed so far is to generate the code as usual

```

from sympy import *

from sympy.utilities.codegen import codegen

x = Symbol('x')

expr = x**2 - x + sin(x) < x**3 - 2*x - cos(5*x)  # ok

[(c_name, c_code), (h_name, c_header)] = codegen(("f", expr), "C")

print(c_code)

```

and then regex the relevant part out of `c_code`. (If you know of a better way to handle this, I'll be happy to hear about it.)

This works alright, but fails in the most simple cases, e.g.,

```

expr = True

```

with

```

TypeError: The first argument must be a sympy expression.

```

Quite understandable.

Is there a better way/hack to support this kind of code generation?

Cheers,

Nico

[Sudhanshu Mishra]

Hi

I can't comment on the first part but for the boolean expression, it's a design issue. Booleans have not been derived from Expr class in sympy which is why it fails.

​

Sudhanshu Mishra

--
You received this message because you are subscribed to the Google Groups "sympy" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sympy+un...@googlegroups.com.
To post to this group, send email to sy...@googlegroups.com.
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/21baed7f-e7e6-4d1e-979c-6d30dee206e9%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Björn Dahlgren]

On Monday, 24 August 2015 23:09:11 UTC+2, Nico wrote:

Is there a better way/hack to support this kind of code generation?

Not that I know of, we should teach codegen how to handle Boolean and Relational (and make it emit the
apropriate boolean data type).