Junio C Hamano wrote:
> > Signed-off-by: Peter Stuge <peter@xxxxxxxx>
>
> Care to elaborate a bit more please?
Okey. I thought subject together with change would be clear enough. :)
> Explanation of what you are fixing is totally lacking.
The subject sums it up, if briefly.
> What happens with the current code, why it is wrong, and how the
> updated pattern improves the result in what way?
Current code generates links to line numbers like ../file.c#l1234;js=1
It is wrong because a URI fragment always goes at the end.
The updated pattern improves this by treating # like end of string,
to detect js=1 also before # and not only at end of string.
The updated code improves this by injecting [?;]js=1 before # if
one exists, or at end of string (like before) otherwise.
> > -var jsExceptionsRe = /[;?]js=[01]$/;
> > +var jsExceptionsRe = /[;?]js=[01](#.*)?$/;
..
> > - if (!jsExceptionsRe.test(link)) { // =~ /[;?]js=[01]$/;
> > - link.href +=
> > - (link.href.indexOf('?') === -1 ? '?' : ';') + 'js=1';
> > + if (!jsExceptionsRe.test(link)) { // =~ /[;?]js=[01](#.*)?$/;
> > + link.href = link.href.replace(/(#|$)/,
> > + (link.href.indexOf('?') === -1 ? '?' : ';') + 'js=1$1');
To test how this works you can try it on
http://git.libusb.org/?p=libusb.git;a=blob;f=COPYING
where the change is in production. Compare the line number links with
those generated by another gitweb with javascript-actions enabled.
//Peter
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html