Re: [PATCH] config: add "hostname" condition to includeIf

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, Aug 27, 2025 at 9:53 AM Ayush Sharma <ayushoffinfo17@xxxxxxxxx> wrote:
> On Fri, Aug 22, 2025 at 2:19 PM monarch <ayushoffinfo17@xxxxxxxxx> wrote:
> > Teach "includeIf" to include configuration based on the machine's hostname,
> > as returned by gethostname(2).
> >
> > Example:
> >     [includeIf "hostname:work-laptop"]
> >         path = ~/.gitconfig.work
> >     [includeIf "hostname:home-pc"]
> >         path = ~/.gitconfig.home
> >
> > This allows users to write host-specific configuration without separate branches.
> >
> > Signed-off-by: monarch <ayushoffinfo17@xxxxxxxxx>
> > ---
> > +static int include_by_hostname(const char *cond, size_t cond_len)
> > +{
> > +    char actual_hostname[1024];
> > +    struct strbuf target_hostname = STRBUF_INIT;
> > +    int ret = 0;
> > +
> > +    // Make sure the call to gethostname is correct and its return value is checked.
> > +    if (gethostname(actual_hostname, sizeof(actual_hostname)) != 0)
> > +        return 0; // If it fails, the condition is false.
> > +
> > +    strbuf_add(&target_hostname, cond, cond_len);
> > +
> > +    // The core of the logic: strcmp returns 0 when strings are equal.
> > +    if (strcmp(actual_hostname, target_hostname.buf) == 0)
> > +        ret = 1; // Success, the hostnames match!
> > +
> > +    strbuf_release(&target_hostname);
> > +    return ret;
> > +}
>
> Just a gentle ping to see if there is any feedback on this patch.

You can increase the likelihood of feedback by:

* Studying the related email thread pointed out by Junio in his
response[1] to your initial inquiry, and taking all the open
discussion points into consideration and providing answers for them in
your proposed submission. (For instance, one of the open discussion
points[2] was how a user would know the exact hostname to use with
`includeIf [hostname "..."]`; a solution needs to be provided in order
to move forward with the proposal.)

* Including with your patch the necessary documentation update (so
users will be able to discover this new capability), as well as new
tests.

* Studying Documentation/CodingGuidelines (for instance, this project
uses `/*...*/` comments, not `//`), and
Documentation/SubmittingPatches (for instance, avoid changes, such as
inserting unnecessary blank lines or arbitrarily reformatting code,
unrelated to the purpose of your patch; use your proper name in the
Signed-off-by: footer).

* Consult the Git source code to see whether there is a better way to
obtain the hostname than rolling your own (for instance, don't fall
prey to gethostname() potentially omitting the terminating NUL from
the buffer; instead use xgethostname() from Git's wrapper.c).

[1]: https://lore.kernel.org/git/xmqqqzx8k258.fsf@gitster.g/
[2]: https://lore.kernel.org/git/CAPig+cT4fpX7Kczu0+H5TZnmpVqqq0h8nBafj4UqDs7Xv2Nf4A@xxxxxxxxxxxxxx/





[Index of Archives]     [Linux Kernel Development]     [Gcc Help]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [V4L]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Fedora Users]

  Powered by Linux