Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Joel Brobecker <brobecker@adacore.com>
To: Kai Tietz <ktietz@redhat.com>
Cc: Ray Donnelly <mingw.android@gmail.com>,
	gcc-patches@gcc.gnu.org,	ktietz70@gmail.com,
	"binutils@sourceware.org Development" <binutils@sourceware.org>,
	gdb-patches@sourceware.org
Subject: Re: [PATCH 1/2] Windows libibery: Don't quote args unnecessarily
Date: Mon, 21 Apr 2014 14:53:00 -0000	[thread overview]
Message-ID: <20140421145301.GA4477@adacore.com> (raw)
In-Reply-To: <1826621422.9537772.1397939013891.JavaMail.zimbra@redhat.com>

> Changelog libiberty/
>  	* pex-win32.c (argv_to_cmdline): Don't quote
>  	args unnecessarily

Some minor comments...

> > diff --git a/libiberty/pex-win32.c b/libiberty/pex-win32.c
> > index eae72c5..775b53c 100644
> > --- a/libiberty/pex-win32.c
> > +++ b/libiberty/pex-win32.c
> > @@ -340,17 +340,26 @@ argv_to_cmdline (char *const *argv)
> >    char *p;
> >    size_t cmdline_len;
> >    int i, j, k;
> > +  int needs_quotes;
> >  
> >    cmdline_len = 0;
> >    for (i = 0; argv[i]; i++)
> >      {
> > -      /* We quote every last argument.  This simplifies the problem;
> > -	 we need only escape embedded double-quotes and immediately
> > +      /* We only quote arguments that contain spaces, \n \t \v or "
> > characters
> > +	 to prevent wasting 2 chars per argument of the CreateProcess 32k char
> > +	 limit We need only escape embedded double-quotes and immediately

Period missing after "limit".  JIC, remember that we ask that periods be
followed by 2 spaces.

> >  	 preceeding backslash characters.  A sequence of backslach characters
> >  	 that is not follwed by a double quote character will not be
> >  	 escaped.  */
> > +      needs_quotes = 0;
> >        for (j = 0; argv[i][j]; j++)
> >  	{
> > +	  if (argv[i][j] == ' '  || argv[i][j] == '\n' ||
> > +	      argv[i][j] == '\t' || argv[i][j] == '"' )

GNU Coding Style requirement: Binary operators should be at the start
of the next line, not at the end. For instance:

	  if (argv[i][j] == ' '  || argv[i][j] == '\n'
              || argv[i][j] == '\t' || argv[i][j] == '"')

Also, watch out for the extra space before ')' - please remove it.


> > +	    {
> Here seems to be an intend issue.
> > +	      needs_quotes = 1;
> > +	    }
> > +
> >  	  if (argv[i][j] == '"')
> >  	    {
> >  	      /* Escape preceeding backslashes.  */
> > @@ -362,16 +371,34 @@ argv_to_cmdline (char *const *argv)
> >  	}
> >        /* Trailing backslashes also need to be escaped because they will be
> >           followed by the terminating quote.  */
> > -      for (k = j - 1; k >= 0 && argv[i][k] == '\\'; k--)
> > -	cmdline_len++;
> > +      if (needs_quotes)
> > +        {
> > +          for (k = j - 1; k >= 0 && argv[i][k] == '\\'; k--)
> > +            cmdline_len++;
> > +        }
> >        cmdline_len += j;
> > -      cmdline_len += 3;  /* for leading and trailing quotes and space */
> > +      /* for leading and trailing quotes and space */
> > +      cmdline_len += needs_quotes * 2 + 1;
> >      }
> >    cmdline = XNEWVEC (char, cmdline_len);
> >    p = cmdline;
> >    for (i = 0; argv[i]; i++)
> >      {
> > -      *p++ = '"';
> > +      needs_quotes = 0;
> > +      for (j = 0; argv[i][j]; j++)
> > +        {
> > +          if (argv[i][j] == ' '  || argv[i][j] == '\n' ||
> > +              argv[i][j] == '\t' || argv[i][j] == '"' )

Same as above.

> > +            {
> > +              needs_quotes = 1;
> > +              break;
> > +            }
> > +        }
> > +
> > +      if (needs_quotes)
> > +        {
> > +          *p++ = '"';
> > +        }
> >        for (j = 0; argv[i][j]; j++)
> >  	{
> >  	  if (argv[i][j] == '"')
> > @@ -382,9 +409,12 @@ argv_to_cmdline (char *const *argv)
> >  	    }
> >  	  *p++ = argv[i][j];
> >  	}
> > -      for (k = j - 1; k >= 0 && argv[i][k] == '\\'; k--)
> > -	*p++ = '\\';
> > -      *p++ = '"';
> > +      if (needs_quotes)
> > +        {
> > +          for (k = j - 1; k >= 0 && argv[i][k] == '\\'; k--)
> > +            *p++ = '\\';
> > +          *p++ = '"';
> > +        }
> >        *p++ = ' ';
> >      }
> >    p[-1] = '\0';
> > --
> > 1.9.2
> 
> Patch itself makes sense.  Let see if there are additional comments.
> 
> Regards,
> Kai

-- 
Joel


      parent reply	other threads:[~2014-04-21 14:53 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <1397936424-2290-1-git-send-email-mingw.android@gmail.com>
     [not found] ` <1397936424-2290-2-git-send-email-mingw.android@gmail.com>
2014-04-19 20:23   ` Kai Tietz
2014-04-20  6:07     ` Eli Zaretskii
2014-04-21 14:53     ` Joel Brobecker [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20140421145301.GA4477@adacore.com \
    --to=brobecker@adacore.com \
    --cc=binutils@sourceware.org \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=gdb-patches@sourceware.org \
    --cc=ktietz70@gmail.com \
    --cc=ktietz@redhat.com \
    --cc=mingw.android@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox