Re: [PATCH] hc: don't segfault if DISPLAY can't be opened

[Thorsten Wißmann <re06huxa _at_ cip _dot_ cs _dot_ fau _dot_ de>]

On Mon, Nov 05, 2012 at 01:29:19PM +0100, Florian Bruhin wrote:
> Yes, I know, I'm bombarding this ML with patches. I'm done soon with
> the smaller things I think.

No Problem :) That's what the ML was created for!

> While developing I accidentally did run herbstclient over an SSH
> connection without DISPLAY set and it segfaulted, I thought it might
> be better to output some error message and exit gracefully.
> 
> If there is any better way of contributing (e.g. you pulling from my
> git repo or whatever) that's fine for me as well, just let me know.

I personally prefer patches, because in my opinion it's unclean to amend
commits that are available in a public repository. And amending commits
or changing the order of patches often is necessary. (Both is in this
case).

>  HCConnection* hc_connect() {
>      Display* display = XOpenDisplay(NULL);
> +    if (display == NULL) {
> +        return NULL;
> +    } 
        ^
        |
I removed that trailing whitespace and merged it as:

252d48b Don't segfault when DISPLAY can't be opened in hc


Thorsten