delete comments in .c file

[Timex]

I want to delete all comments in .c file.

Size of .c file is very big.

Any good idea to do this?

Please show me example code.

[Keith Thompson]

Irrwahn Grausewitz <ir*******@free net.de> writes:

Stephen Samuel <st************ @telus.net> wrote:

Here's a perl script which will handle *MOST* sane C code...

<snip>

Since when is perl topical in c.l.c?

This is an interesing edge case with respect to topicality. One could
argue that we're talking *about* C (which is clearly topical), but
we're using a mixture of Perl and English to discuss it. Think of the
Perl regular expression as a description of how to strip comments from
C source code.

On the other hand, not everyone here can be expected to speak Perl
regexps fluently.

--
Keith Thompson (The_Other_Keit h) ks*@cts.com <http://www.ghoti.net/~kst>
San Diego Supercomputer Center <*> <http://www.sdsc.edu/~kst>
Schroedinger does Shakespeare: "To be *and* not to be"

[Irrwahn Grausewitz]

Stephen Samuel <st************ @telus.net> wrote:

CBFalconer wrote:

*** rude top-posting fixed ***
Hmm.. This must be a relatively recent addition to usenet
ettiquete (i.e. in the last decade or so).

It's a convention in comp.lang.c (and several other technical
newsgroups) to place your comments after the part of the original
post you are responding to, in order to retain context. Thus
top-posting is discouraged in c.l.c.
Appologies. I'm an old foggie, and it's probably been an decade
since I've posted here.

Again, please do not send email copies of your replies; thank you.

Regards
--
Irrwahn
(ir*******@free net.de)

[Irrwahn Grausewitz]

Keith Thompson <ks*@cts.com> wrote:

Irrwahn Grausewitz <ir*******@free net.de> writes:

Since when is perl topical in c.l.c?
This is an interesing edge case with respect to topicality. One could
argue that we're talking *about* C (which is clearly topical), but
we're using a mixture of Perl and English to discuss it. Think of the
Perl regular expression as a description of how to strip comments from
C source code.

That would make any solution to manipulate C sources implemented in
any language other than C topical in c.l.c. IMHO that would not be a
Good Thing[tm].
On the other hand, not everyone here can be expected to speak Perl
regexps fluently.

Indeed.

Regards
--
Irrwahn
(ir*******@free net.de)

[Arthur J. O'Dwyer]

On Sun, 2 Nov 2003, CBFalconer wrote:

Stephen Samuel wrote:

Timex wrote:

I want to delete all comments in .c file.
#!/usr/bin/perl
$s=join("",<>);
# printf "[[%s]]\n\n",$s;
$s=~ s/("(\\\\|\\"|[^"])*")|(\/\*([^*]|\*(?=[^\/]))*\*\/)|(\/\/.*)/[[$1 ]]/g;
printf "[[%s]]\n\n",$s;

/* File uncmntc.c - demo of a text filter
Strips C comments. Tested to strip itself
by C.B. Falconer. 2002-08-15
Public Domain. Attribution appreciated
report bugs to <mailto:cb***** ***@worldnet.at t.net>
*/

<snip code>

I ran your program through some hurdles, and found that
it couldn't handle multibyte character constants for some
reason. I didn't bother to track down why; I just re-wrote
the filter from scratch. ;-) Here's my version, whose
algorithm may be completely different from yours.
This algorithm, on the other hand, completely fails to
handle line-splicing in the middle of comment delimiters: /\
* this is a comment */ does not work, nor does /* this either *\
/. Comment removal really is tricky in the most general case!
Proper error-checking on getc() and putc(), and a good
command-line interface, left as exercises for the interested
reader.
/* File uncmntc2.c - demo of a different text filter
Strips C comments. Tested to strip itself
Improves on CBFalconer's design by correctly handling '/*'
and by having a C89/C99 switch, but doesn't handle the /\
* delimiter correctly.
by Arthur O'Dwyer, 2002-11-03
Public Domain. Attribution appreciated
don't bother reporting bugs, just fix 'em...
*/

#include <stdio.h>
#include <stdlib.h>

/* Strip C99-style end-of-line comments? */
int AllowEOLComment s = 1;

int strip_comments( FILE *fp, FILE *outfp);
static int put_carefully(i nt lastch, int ch, FILE *outfp);
int main(void)
{
strip_comments( stdin, stdout);
return 0;
}
int strip_comments( FILE *fp, FILE *outfp)
{
int ch;
int lastch;
int inchotes = 0;
int inquotes = 0;
int incomment = 0;
int ineolcomment = 0;

for (lastch = ' '; (ch = getc(fp)) != EOF; lastch = ch)
{
if (!incomment && !ineolcomment)
{
if (inquotes || inchotes)
putc(ch, outfp);
else
put_carefully(l astch, ch, outfp);
}

if (inchotes) {
if (ch == '\'' && lastch != '\\')
inchotes = 0;
} else if (inquotes) {
if (ch == '"' && lastch != '\\')
inquotes = 0;
} else if (incomment) {
if (ch == '/' && lastch == '*')
incomment = 0, ch = ' ';
} else if (ineolcomment) {
if (ch == '\n' && lastch != '\\')
ineolcomment = 0;
} else {
if (ch == '\'')
inchotes = 1;
else if (ch == '"')
inquotes = 1;
else if (lastch == '/' && ch == '*') {
putc(' ', outfp);
incomment = 1;
}
else if (AllowEOLCommen ts && lastch == '/' && ch == '/')
ineolcomment = 1;
}
}

if (lastch == '/')
putc(lastch, outfp);

return 0;
}
static int put_carefully(i nt lastch, int ch, FILE *outfp)
{
/* Print out 'ch', but be very careful not to print
* any characters that might be part of a comment
* delimiter. Contrariwise, if 'lastch' is now
* definitely *not* a comment delimiter, we must now
* print it, too.
*/

if (AllowEOLCommen ts) {
if (lastch == '/' && ch == '/')
return 0;
}
if (lastch == '/' && ch == '*')
return 0;
if (lastch == '/')
putc(lastch, outfp);
if (ch != '/')
putc(ch, outfp);
return 0;
}

[CBFalconer]

"Arthur J. O'Dwyer" wrote:

On Sun, 2 Nov 2003, CBFalconer wrote:

.... snip ...

/* File uncmntc.c - demo of a text filter
Strips C comments. Tested to strip itself
by C.B. Falconer. 2002-08-15
Public Domain. Attribution appreciated
report bugs to <mailto:cb***** ***@worldnet.at t.net>
*/ <snip code>

I ran your program through some hurdles, and found that
it couldn't handle multibyte character constants for some
reason. I didn't bother to track down why; I just re-wrote
the filter from scratch. ;-) Here's my version, whose
algorithm may be completely different from yours.

.... snip ...

A known failing. It also fails miserably with trigraphs. The
multibyte char is probably easily handled analogously to handling
quoted strings.

/* File uncmntc2.c - demo of a different text filter
Strips C comments. Tested to strip itself
Improves on CBFalconer's design by correctly handling '/*'
and by having a C89/C99 switch, but doesn't handle the /\
* delimiter correctly.
by Arthur O'Dwyer, 2002-11-03

^^^^
That is the year I wrote mine :-)

All of which shows that there are multiple ways to implement a
black box. I omitted any reference to cats because I happen to
like them.

--
Chuck F (cb********@yah oo.com) (cb********@wor ldnet.att.net)
Available for consulting/temporary embedded and systems.
<http://cbfalconer.home .att.net> USE worldnet address!

[Patrick Foley]

In <Pi************ *************** ********@unix42 .andrew.cmu.edu > "Arthur J. O'Dwyer" <aj*@nospam.and rew.cmu.edu> writes:

[snip] Comment removal really is tricky in the most general case!

Since this is exercise 1-23 in K&R2, there are several solutions
available at Richard's site:

http://users.powernet.co.uk/eton/kandr2/index.html

including a 556-line entry from Chris Torek that I think also brews
coffee...

Pat

BTW, Richard: Would you consider adding a plaintext version of the
"naming conventions" page to the zipfile as a sort of "README"?

[Richard Heathfield]

Patrick Foley wrote:

BTW, Richard: Would you consider adding a plaintext version of the
"naming conventions" page to the zipfile as a sort of "README"?

I am currently re-evaluating the Answers section of my site. I'll get back
to you when I have a bit more time.

--
Richard Heathfield : bi****@eton.pow ernet.co.uk
"Usenet is a strange place." - Dennis M Ritchie, 29 July 1999.
C FAQ: http://www.eskimo.com/~scs/C-faq/top.html
K&R answers, C books, etc: http://users.powernet.co.uk/eton

[Chris Torek]

>In <Pi************ *************** ********@unix42 .andrew.cmu.edu > "Arthur J. O'Dwyer" <aj*@nospam.and rew.cmu.edu> writes:

[snip] Comment removal really is tricky in the most general case!

In article <nc************ @myname.my.doma in>
Patrick Foley <pf****@earthli nk.net> writes:Since this is exercise 1-23 in K&R2, there are several solutions
available at Richard's site:

http://users.powernet.co.uk/eton/kandr2/index.html

including a 556-line entry from Chris Torek that I think also brews
coffee...

But it has (gasp!) a *bug*. :-) The "level 2 state machine" for
handling comments fails to reconsider characters in a few cases.
I think the main (only?) problem can be fixed without too much
fuss:

case L2_SLASH:
if (c == '*')
l2state = L2_COMM;
else if (c99 && c == '/')
l2state = L2_SLASHSLASH;
else {
SYNCLINES();
OUTPUT('/', 0);
--> if (c != '/') {
--> if (c != EOF)
--> COPY();
--> l2state = L2_NORMAL;
--> }
}
break;

The bug is in the marked lines, which output the first slash
and then change the level-2 state. But the new state should
be "that which results in seeing character c as if the initial
state had been L2_NORMAL", so we could replace all of them with:

l2state = L2_NORMAL;
goto l2_normal_case;

and add an "l2_normal_case " label under case L2_NORMAL: above.
Alternatively, the assignment to l2state can be changed to:

l2state = c == '\'' ? L2_CC :
c == '"' ? L2_SC : L2_NORMAL;

which avoids the dreaded "goto", and simply duplicates what would
have happened in L2_NORMAL state (except of course that instead of
replacing l2state with L2_SLASH for '/', we have to replace it with
L2_NORMAL for characters that are not in [/'"]).
--
In-Real-Life: Chris Torek, Wind River Systems
Salt Lake City, UT, USA (40°39.22'N, 111°50.29'W) [phone]
email: forget about it http://67.40.109.61/torek/index.html (for the moment)
Reading email is like searching for food in the garbage, thanks to spammers.

[Thomas Matthews]

Timex wrote:

I want to delete all comments in .c file.

Size of .c file is very big.

Any good idea to do this?

Please show me example code.

Perhaps a better idea is to break the file into
smaller pieces upon better themes.

I believe that delete all the comments is crime
against programming ethics. After all, one of
the greatest ideals to achieve is to make a
program readable by a programming illiterate
person.

--
Thomas Matthews

C++ newsgroup welcome message:
http://www.slack.net/~shiva/welcome.txt
C++ Faq: http://www.parashift.com/c++-faq-lite
C Faq: http://www.eskimo.com/~scs/c-faq/top.html
alt.comp.lang.l earn.c-c++ faq:
http://www.raos.demon.uk/acllc-c++/faq.html
Other sites:
http://www.josuttis.com -- C++ STL Library book

[Gary E. Ansok]

In article <Pi************ *************** ********@unix42 .andrew.cmu.edu >,
Arthur J. O'Dwyer <aj*@nospam.and rew.cmu.edu> wrote:

> Timex wrote:
> >
> > I want to delete all comments in .c file.

I tested Arthur's program and, despite its claim, it couldn't
even strip its own comments (it left in the comment in
put_carefully() ). The bug is that it thought the backslash
meant that '\\' was not a complete character constant (nor
would it think "\\" was a complete string).

Is this a complete C99-style comment?
// \\
If it is, a similar fix may be needed in that part of the code.

Lesson: Comment removal really is tricky in the most general case!

Agreed.

-- Gary

My attempt at a bug fix:
/* File uncmntc2.c - demo of a different text filter
Strips C comments. Tested to strip itself
Improves on CBFalconer's design by correctly handling '/*'
and by having a C89/C99 switch, but doesn't handle the /\
* delimiter correctly.
by Arthur O'Dwyer, 2002-11-03
bug fix by Gary Ansok, 2003-11-06 to handle '\\' and "\\"
Public Domain. Attribution appreciated
don't bother reporting bugs, just fix 'em...
*/

#include <stdio.h>
#include <stdlib.h>

/* Strip C99-style end-of-line comments? */
int AllowEOLComment s = 1;

int strip_comments( FILE *fp, FILE *outfp);
static int put_carefully(i nt lastch, int ch, FILE *outfp);
int main(void)
{
strip_comments( stdin, stdout);
return 0;
}
int strip_comments( FILE *fp, FILE *outfp)
{
int ch;
int lastch;
int inchotes = 0;
int inquotes = 0;
int incomment = 0;
int ineolcomment = 0;
int backslashed = 0;

for (lastch = ' '; (ch = getc(fp)) != EOF; lastch = ch)
{
if (!incomment && !ineolcomment)
{
if (inquotes || inchotes)
putc(ch, outfp);
else
put_carefully(l astch, ch, outfp);
}

if (inchotes) {
if (lastch == '\\')
backslashed ^= 1;
else
backslashed = 0;
if (ch == '\'' && !backslashed)
inchotes = 0;
} else if (inquotes) {
if (lastch == '\\')
backslashed ^= 1;
else
backslashed = 0;
if (ch == '"' && !backslashed)
inquotes = 0;
} else if (incomment) {
if (ch == '/' && lastch == '*')
incomment = 0, ch = ' ';
} else if (ineolcomment) {
if (ch == '\n' && lastch != '\\')
ineolcomment = 0;
} else {
if (ch == '\'')
inchotes = 1;
else if (ch == '"')
inquotes = 1;
else if (lastch == '/' && ch == '*') {
putc(' ', outfp);
incomment = 1;
}
else if (AllowEOLCommen ts && lastch == '/' && ch == '/')
ineolcomment = 1;
}
}

if (lastch == '/')
putc(lastch, outfp);

return 0;
}
static int put_carefully(i nt lastch, int ch, FILE *outfp)
{
/* Print out 'ch', but be very careful not to print
* any characters that might be part of a comment
* delimiter. Contrariwise, if 'lastch' is now
* definitely *not* a comment delimiter, we must now
* print it, too.
*/

if (AllowEOLCommen ts) {
if (lastch == '/' && ch == '/')
return 0;
}
if (lastch == '/' && ch == '*')
return 0;
if (lastch == '/')
putc(lastch, outfp);
if (ch != '/')
putc(ch, outfp);
return 0;
}