You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Stefan Küng <to...@gmail.com> on 2006/01/10 17:28:47 UTC
Crash with url-handling
Hi,
the following command makes the Subversion client crash:
svn ls httpS://something/somethingmore
tested with 1.3.0 on Windows, but earlier clients have the same fault.
The problem is the capital S in httpS.
Stefan
--
___
oo // \\ "De Chelonian Mobile"
(_,\/ \_/ \ TortoiseSVN
\ \_/_\_/> The coolest Interface to (Sub)Version Control
/_/ \_\ http://tortoisesvn.tigris.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: Crash with url-handling
Posted by Garrett Rooney <ro...@electricjellyfish.net>.
On 10 Jan 2006 12:23:59 -0600, kfogel@collab.net <kf...@collab.net> wrote:
> Garrett Rooney <ro...@electricjellyfish.net> writes:
> > Oops, I'm wrong, it's not fixed in trunk, it's just that my trunk
> > build didn't have SSL support turned on. The problem appears to be
> > that we use a strcasecmp to check for https, while neon uses strcmp,
> > and once we think we've got an SSL session, we assume that the SSL
> > parts of the neon session get set up properly... I imagine the
> > correct fix is just to make svn use strcmp like neon does, since even
> > if we're right, it's never worked, and we'll need to be more strict to
> > interoperate with existing neon installs...
>
> I've got a better fix:
>
> In svn_ra_dav__open(), where we run this code:
>
> is_ssl_session = (strcasecmp(uri.scheme, "https") == 0);
>
> If 'is_ssl_session' is true after that, then let's just hard-set
> uri.scheme to "https". After all, it doesn't matter whether the user
> typed "httpS" or whatever. We know it's supposed to be
> case-insensitive, and if we just change it to "https", we'll be
> compatible with Neon, and everything will work.
>
> Don't have time to code & test that right now, but I thought I'd post
> the suggestion here.
I actually did something very similar to this, setting scheme to all
lower case before passing it on to Neon.
-garrett
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: Crash with url-handling
Posted by kf...@collab.net.
Garrett Rooney <ro...@electricjellyfish.net> writes:
> Oops, I'm wrong, it's not fixed in trunk, it's just that my trunk
> build didn't have SSL support turned on. The problem appears to be
> that we use a strcasecmp to check for https, while neon uses strcmp,
> and once we think we've got an SSL session, we assume that the SSL
> parts of the neon session get set up properly... I imagine the
> correct fix is just to make svn use strcmp like neon does, since even
> if we're right, it's never worked, and we'll need to be more strict to
> interoperate with existing neon installs...
I've got a better fix:
In svn_ra_dav__open(), where we run this code:
is_ssl_session = (strcasecmp(uri.scheme, "https") == 0);
If 'is_ssl_session' is true after that, then let's just hard-set
uri.scheme to "https". After all, it doesn't matter whether the user
typed "httpS" or whatever. We know it's supposed to be
case-insensitive, and if we just change it to "https", we'll be
compatible with Neon, and everything will work.
Don't have time to code & test that right now, but I thought I'd post
the suggestion here.
-Karl
--
www.collab.net <> CollabNet | Distributed Development On Demand
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: Crash with url-handling
Posted by Garrett Rooney <ro...@electricjellyfish.net>.
On 1/10/06, Garrett Rooney <ro...@electricjellyfish.net> wrote:
> On 1/10/06, Stefan Küng <to...@gmail.com> wrote:
> > Hi,
> >
> > the following command makes the Subversion client crash:
> >
> > svn ls httpS://something/somethingmore
> >
> > tested with 1.3.0 on Windows, but earlier clients have the same fault.
> > The problem is the capital S in httpS.
>
> Reproduced here, seems to have been fixed on trunk. I'll try and
> figure out what the fix was, and get something ready for backport to
> 1.3.1.
Oops, I'm wrong, it's not fixed in trunk, it's just that my trunk
build didn't have SSL support turned on. The problem appears to be
that we use a strcasecmp to check for https, while neon uses strcmp,
and once we think we've got an SSL session, we assume that the SSL
parts of the neon session get set up properly... I imagine the
correct fix is just to make svn use strcmp like neon does, since even
if we're right, it's never worked, and we'll need to be more strict to
interoperate with existing neon installs...
-garrett
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: Crash with url-handling
Posted by Garrett Rooney <ro...@electricjellyfish.net>.
On 1/10/06, Stefan Küng <to...@gmail.com> wrote:
> Hi,
>
> the following command makes the Subversion client crash:
>
> svn ls httpS://something/somethingmore
>
> tested with 1.3.0 on Windows, but earlier clients have the same fault.
> The problem is the capital S in httpS.
Reproduced here, seems to have been fixed on trunk. I'll try and
figure out what the fix was, and get something ready for backport to
1.3.1.
-garrett
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: Crash with url-handling
Posted by Garrett Rooney <ro...@electricjellyfish.net>.
On 1/10/06, Stefan Küng <to...@gmail.com> wrote:
> Hi,
>
> the following command makes the Subversion client crash:
>
> svn ls httpS://something/somethingmore
>
> tested with 1.3.0 on Windows, but earlier clients have the same fault.
> The problem is the capital S in httpS.
Fixed in trunk and backported to 1.3.x, thanks for the report!
-garrett
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org