Commented out includes

[Zitrax]

Noticed that we had some commented out includes that still were included in the graph. Would be nice if the regexp could take that into account.

(Although I know it would be a bit tricky to handle all kind of situations like multi line comments using /* */ and #ifdef, however a simple // comment should be easy to handle).

[pvigier]

Hi,

Yes, a simple solution would be to use the regex '^\s*#include\s+["<](.*)[">]' in multiline mode but it will only detect at most one #include by line while the previous version can detect several. But I don't think anyone is doing this anyway.

The ideal solution is to write a custom parser that detects #include and comments, it is not that hard. Maybe a bit slow in Python.

[gsc2001]

@pvigier this regex will not work, better do '[^\/\/]#[ ]{0,1}include\s+["<](.*)[">]'. It also includes for space between # and include

[pvigier]

Hey,

Thanks for the suggestion, but your regex has several issues:

• it does not work if there is no character before #
• the regex should allow for spaces between the comment token // and #
• [^\/\/] is the same thing as [^\/] so it does not detect // but / so it will fail to match valid includes such as /**/#include "foo.h"
• Are you sure that it is legal to put a space between # and include? I have never seen that.

[gsc2001]

• It does work, I ran it on a source code
• Yeah I got it! sorry for this
• The company I am working with has the whole code formating like this (a space between # and include)

[pvigier]

import re
regex = re.compile('[^\/\/]#[ ]{0,1}include\s+["<](.*)[">]')
regex.findall('#include "foo.h"')

returns []

I suspect there are some characters before the first #include in your files (e.g. a license preamble, author list, etc.). Otherwise, the first #include is skipped.

[Kochise]

import re

# Accept commented out includes
#regex = re.compile('^\s*[\/\/|\/\*]*\s*#\s*include\s+["<](.*)[">]\s*[\*\/]*\s*$')

# Reject commented out includes (but not on multi lines)
regex = re.compile('^\s*#\s*include\s+["<](.*)[">]\s*$')

print(regex.findall('     #   include "foo.h"  */'))