All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jeff King <peff@peff.net>
To: "brian m. carlson" <sandals@crustytoothpaste.net>,
	Patrick Sharp <jakanapes@gmail.com>,
	Johannes Schindelin <johannes.schindelin@gmx.de>,
	git@vger.kernel.org
Subject: Re: [BUG] having 'plink' anywhere in the GIT_SSH environment variables sets putty = true
Date: Wed, 22 Apr 2015 18:00:54 -0400	[thread overview]
Message-ID: <20150422220053.GA32268@peff.net> (raw)
In-Reply-To: <20150422214445.GC827497@vauxhall.crustytoothpaste.net>

On Wed, Apr 22, 2015 at 09:44:45PM +0000, brian m. carlson wrote:

> On Wed, Apr 22, 2015 at 05:29:04PM -0400, Jeff King wrote:
> > > Perhaps it would be worthwhile to check instead if the text "plink" is
> > > the beginning of string or is preceded by a path separator.  That would
> > > give us a bit more confidence that the user is looking for plink, but
> > > would still allow people to use "plink-0.63" if they like.
> > 
> > Yeah, I think that is a reasonable approach. Note that it needs to
> > handle the "tortoiseplink" case from below, too (you can still use your
> > strategy, you just need to look for either string).
> 
> So maybe something like this?
> 
> diff --git a/connect.c b/connect.c
> index 391d211..ba3ab34 100644
> --- a/connect.c
> +++ b/connect.c
> @@ -749,10 +749,15 @@ struct child_process *git_connect(int fd[2], const char *url,
>  					conn->use_shell = 1;
>  					putty = 0;
>  				} else {
> +					char *plink, *tplink;
> +
>  					ssh = getenv("GIT_SSH");
>  					if (!ssh)
>  						ssh = "ssh";
> -					putty = !!strcasestr(ssh, "plink");
> +					plink = strcasestr(ssh, "plink");
> +					tplink = strcasestr(ssh, "tortoiseplink");
> +					putty = plink == ssh || (plink && is_dir_sep(plink[-1])) ||
> +						tplink == ssh || (tplink && is_dir_sep(tplink[-1]));

Yeah, that looks right to me. You might want to represent the "are we
tortoise" check as a separate flag, though, and reuse it a few lines
later.

Also, not related to your patch, but I notice the "putty" declaration is
in a different scope than I would have expected, which made me wonder if
it gets initialized in all code paths. I think is from the recent
addition of CONNECT_DIAG_URL, which pushes the bulk of the code into its
own else clause, even though the first part of the "if" always returns
early.  I wonder if it would be simpler to read like:

diff --git a/connect.c b/connect.c
index 391d211..749a07b 100644
--- a/connect.c
+++ b/connect.c
@@ -743,28 +743,28 @@ struct child_process *git_connect(int fd[2], const char *url,
 				free(path);
 				free(conn);
 				return NULL;
-			} else {
-				ssh = getenv("GIT_SSH_COMMAND");
-				if (ssh) {
-					conn->use_shell = 1;
-					putty = 0;
-				} else {
-					ssh = getenv("GIT_SSH");
-					if (!ssh)
-						ssh = "ssh";
-					putty = !!strcasestr(ssh, "plink");
-				}
-
-				argv_array_push(&conn->args, ssh);
-				if (putty && !strcasestr(ssh, "tortoiseplink"))
-					argv_array_push(&conn->args, "-batch");
-				if (port) {
-					/* P is for PuTTY, p is for OpenSSH */
-					argv_array_push(&conn->args, putty ? "-P" : "-p");
-					argv_array_push(&conn->args, port);
-				}
-				argv_array_push(&conn->args, ssh_host);
 			}
+
+			ssh = getenv("GIT_SSH_COMMAND");
+			if (ssh) {
+				conn->use_shell = 1;
+				putty = 0;
+			} else {
+				ssh = getenv("GIT_SSH");
+				if (!ssh)
+					ssh = "ssh";
+				putty = !!strcasestr(ssh, "plink");
+			}
+
+			argv_array_push(&conn->args, ssh);
+			if (putty && !strcasestr(ssh, "tortoiseplink"))
+				argv_array_push(&conn->args, "-batch");
+			if (port) {
+				/* P is for PuTTY, p is for OpenSSH */
+				argv_array_push(&conn->args, putty ? "-P" : "-p");
+				argv_array_push(&conn->args, port);
+			}
+			argv_array_push(&conn->args, ssh_host);
 		} else {
 			/* remove repo-local variables from the environment */
 			conn->env = local_repo_env;

-Peff

  reply	other threads:[~2015-04-22 22:01 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-04-22 14:36 [BUG] having 'plink' anywhere in the GIT_SSH environment variables sets putty = true Patrick Sharp
2015-04-22 17:46 ` Johannes Schindelin
2015-04-22 19:12   ` Patrick Sharp
2015-04-22 20:29     ` Jeff King
2015-04-22 21:19       ` brian m. carlson
2015-04-22 21:29         ` Jeff King
2015-04-22 21:44           ` brian m. carlson
2015-04-22 22:00             ` Jeff King [this message]
2015-04-22 22:24               ` brian m. carlson
2015-04-22 23:23                 ` Jeff King
2015-04-23  0:06                   ` [PATCH 1/2] connect: simplify SSH connection code path brian m. carlson
2015-04-23  0:06                     ` [PATCH 2/2] connect: improve check for plink to reduce false positives brian m. carlson
2015-04-23  6:50                       ` Johannes Schindelin
2015-04-23 15:53                         ` Jeff King
2015-04-23 23:14                           ` brian m. carlson
2015-04-24  6:41                             ` Johannes Schindelin
2015-04-24 22:28                             ` [PATCH v2 1/2] connect: simplify SSH connection code path brian m. carlson
2015-04-24 22:28                               ` [PATCH v2 2/2] connect: improve check for plink to reduce false positives brian m. carlson
2015-04-24 22:46                                 ` Pete Harlan
2015-04-24 22:48                                   ` brian m. carlson
2015-04-25 16:03                                 ` Torsten Bögershausen
2015-04-26 18:52                                   ` brian m. carlson
2015-04-26 20:30                               ` [PATCH v3 0/3] Improve robustness of putty detection brian m. carlson
2015-04-26 20:30                                 ` [PATCH v3 1/3] connect: simplify SSH connection code path brian m. carlson
2015-04-26 20:30                                 ` [PATCH v3 2/3] t5601: fix quotation error leading to skipped tests brian m. carlson
2015-04-26 20:30                                 ` [PATCH v3 3/3] connect: improve check for plink to reduce false positives brian m. carlson
2015-04-27  7:57                                   ` Johannes Schindelin
2015-04-28  3:53                                   ` Jeff King
2015-06-26 13:15                                   ` Jeff King
2015-06-26 16:16                                     ` Junio C Hamano
2015-06-26 16:27                                       ` Jeff King
2015-06-26 17:13                                         ` Johannes Schindelin
2015-06-26 17:23                                           ` Jeff King
2015-06-26 20:43                                     ` brian m. carlson
2015-04-26 22:04                                 ` [PATCH v3 0/3] Improve robustness of putty detection Junio C Hamano
2015-04-27 15:46                                   ` Torsten Bögershausen
2015-04-28  4:15                                   ` Jeff King
2015-04-29  1:38                                   ` brian m. carlson
2015-04-24  6:37                           ` [PATCH 2/2] connect: improve check for plink to reduce false positives Johannes Schindelin
2015-04-23  5:08     ` [BUG] having 'plink' anywhere in the GIT_SSH environment variables sets putty = true Torsten Bögershausen
2015-04-23 13:15       ` Patrick Sharp

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=20150422220053.GA32268@peff.net \
    --to=peff@peff.net \
    --cc=git@vger.kernel.org \
    --cc=jakanapes@gmail.com \
    --cc=johannes.schindelin@gmx.de \
    --cc=sandals@crustytoothpaste.net \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.