ansaurus

Question

How to extract comment out of header file using python, perl, or sed?

Answer 1

A:

sed -i -r "s/[\/\ ]{1}\*[\/\ ]?//g" YOURFILENAME

This replaces trims comments from your file, keeping the content. This will modify YOURFILENAME file though. If you don't want that remove -i from the line

Alex N. 2010-05-22 21:14:14

That is close, but it does not strip all the C++ code following the first comment.

WilliamKF 2010-05-22 21:16:07

Also, don't wish to side effect the original file, but instead create a new one.

WilliamKF 2010-05-22 21:17:49

Answer 2

+5 A:

This should work for you:

sed -n '/\*\//q; /^\/\*/d; s/^ \* \?//p' <file.h >comment.txt

Here's an explanation: sed (as you may know) is a command that goes through a file applying a list of rules to each line. Each rule consists of a "selector" and commands that are applied to that line only if the selector matches.

The first rule has the selector /\*\//. This is a regular expression selector; it matches any line that contains the characters */. Both of these need to be backslash-escaped since they have special meanings in a regexp. (I've assumed that this will only match the closing line of the comment in your case and that this entire line should be deleted.) The command is q which means "quit." sed just stops. Ordinarily it would print out the line, but I provided the -n option which means "don't print unless explicitly instructed to."

The second rule has the selector /^\/\*/ which is again a regexp selector that matches the characters /* at the start of the line. Again, I've assumed this line will not contain part of the comment. The d command tells sed to delete this line and move on.

The final rule has no selector, so it applies to all lines (unless a previous command prevented processing from reaching the final rule). The command in this last rule is a substitution command, s/PATTERN/REPLACEMENT/, which finds text in the line that matches some pattern and replaces it with a replacement text. The pattern here is ^ \* \?, which matches a space, an asterisk, and either 0 or 1 spaces, but only at the beginning of the line. And the replacement is nothing. So sed simply deletes the leading space-asterisk-(space)? sequence. The p is actually a flag to the substitution command that tells sed to print out the result of the substitution. It's needed because of the -n option.

David Zaslavsky 2010-05-22 21:15:50

This is close, it leaves behind empty lines of "/*", " *", " */". The /* and */ lines should be deleted and the " *" lines left as blanks.

WilliamKF 2010-05-22 21:19:27

BTW, if you can explain how the command line is interpreted that would be helpful so I can perform this magic myself.

WilliamKF 2010-05-22 21:21:58

Yeah, I know, those are the tweaks I was working out. A full solution and explanation are forthcoming.

David Zaslavsky 2010-05-22 21:23:39

Well we could always pipe it through an extra sed, but that would be cheating ;-)

WilliamKF 2010-05-22 21:25:33

yeah ;-) this one wasn't too hard though. Notice that I simplified the command a bit in my latest edit.

David Zaslavsky 2010-05-22 21:35:56

I had to change the sed script slightly to not eat extra whitespace: "sed -n '\|\*/|q; \|^/\*|d; s/^ \*\|^\/\*//p' file.h > comment.txt"

WilliamKF 2010-05-22 22:14:45

Oh, you mean to only consume one space after the *, not unlimited spaces? Good call, I didn't notice that error, I'll edit that in. (Although the command you just posted in that comment quits on the first line...)

David Zaslavsky 2010-05-22 23:27:47

Answer 3

+3 A:

Pyparsing includes a built-in pattern for matching comment formats from various languages. Using cStyleComment and scanString to find the first comment in the source file makes the rest just string functions:

c_src = open(c_source_file).read()

from pyparsing import cStyleComment
cmt = cStyleComment.scanString(c_src).next()[0][0]
lines = [l[3:] for l in cmt.splitlines()]
print '\n'.join(lines)

scanString is a generator that returns each match before going to the next instance, so only the first comment gets processed. With your sample code, this returns:

APP 180-2 ALG-254/258/772 implementation 
Last update: 03/01/2006 
Issue date:  08/22/2004 

Copyright (C) 2006 Somebody's Name here 
All rights reserved. 

Redistribution and use in source and binary forms, with or without 
modification, are permitted provided that the following conditions 
are met: 
1. Redistributions of source code must retain the above copyright 
   notice, this list of conditions and the following disclaimer. 
2. Redistributions in binary form must reproduce the above copyright 
   notice, this list of conditions and the following disclaimer in the 
   documentation and/or other materials provided with the distribution. 
3. Neither the name of the project nor the names of its contributors 
   may be used to endorse or promote products derived from this software 
   without specific prior written permission. 

THIS SOFTWARE IS PROVIDED BY THE PROJECT AND CONTRIBUTORS ``AS IS'' AND 
ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 
ARE DISCLAIMED.  IN NO EVENT SHALL THE PROJECT OR CONTRIBUTORS BE LIABLE 
FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 
DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 
OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 
HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 
LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 
OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 
SUCH DAMAGE.

Paul McGuire 2010-05-23 01:58:56

ansaurus

tags:

views:

answers:

How to extract comment out of header file using python, perl, or sed?

related questions