[pacman-dev] [PATCH 2/2] Bail out on NULL destfile in curl_download_internal()

[Lukas Fleischer]

On Wed, Aug 17, 2011 at 09:40:22AM -0400, Dave Reisner wrote:
> On Wed, Aug 17, 2011 at 10:15:17AM +0200, Lukas Fleischer wrote:
> > Avoid a potential segfault that may occur if we use a temporary file and
> > fail to build the destination file name from the effective URL.
> > 
> > Signed-off-by: Lukas Fleischer <archlinux at cryptocrack.de>
> > ---
> > Another approach is to use a random default destfile name. Since the URL
> > parsing is very unlikely to fail, I took the easy option here.
> > 
> >  lib/libalpm/dload.c |    5 +++++
> >  1 files changed, 5 insertions(+), 0 deletions(-)
> > 
> > diff --git a/lib/libalpm/dload.c b/lib/libalpm/dload.c
> > index 731d807..c552d2b 100644
> > --- a/lib/libalpm/dload.c
> > +++ b/lib/libalpm/dload.c
> > @@ -373,6 +373,11 @@ static int curl_download_internal(struct dload_payload *payload,
> >  				destfile = get_fullpath(localpath, effective_filename, "");
> >  			}
> >  		}
> > +		else {
> > +			_alpm_log(handle, ALPM_LOG_ERROR, _("could not parse file name from url (%s)"),
> > +					effective_url);
> > +			goto cleanup;
> > +		}
> >  	}
> >  
> >  	ret = 0;
> > -- 
> > 1.7.6
> > 
> > 
> 
> This is after the file has already been downloaded. We shouldn't bail
> out so easily since we already have the file. It looks like in this
> case, we wouldn't even unlink after jumping to cleanup.

Oh, right. I missed that part.

> 
> I'm not sure that this is the right thing to do. Worst case scenario, we
> should probably just skip the rename.

Yeah, either use some default for destfile (some random name or just set
"destfile = strdup(tempfile);") or skip the rename (I prefer the
latter).