svn commit: r354497 - head/sysutils/pear-Cache_Lite

Alexey Dokuchaev danfe at FreeBSD.org
Tue Jun 3 16:11:51 UTC 2014


On Tue, Jun 03, 2014 at 10:44:38AM -0500, Drewery, Bryan wrote:
> On Mon Jun  2 20:27:26 2014, Alexey Dokuchaev wrote:
> >I'm afraid this change is wrong:
> >
> >- pear.php.net itself treats package entry page as a directory: Download,
> >   Documentation, Bugs, Trackbacks links are implemented as subdirectories;
> >   Main entry points to "http://pear.php.net/package/HTTP_Request2/" (note
> >   the trailing slash)
> >
> >- PH section 3.2: "Include a URL to the official WWW homepage [...] with
> >   WWW: (followed by single space) so that automated tools will work
> >   correctly. If the URI is the root of the website or directory, it should
> >   be terminated with a slash."
> >
> >These changes present in the next six commits of yours.  I think these
> >parts should be backed out, and Mr. Rutkowski kindly advised to refresh
> >his PH knowledge.
> 
> This is a bit extreme and pedantic. The URL is fine as is. There

It is pedantic; however, this is 1) not a newly added file, it was correct
before; 2) it contradicts pear.php.net's own output; 3) it contradicts PHB.

> was little need to change it, and even less need to revert it
> and cause grief to a contributor.

Contributor should not have to get upset, he should reread the PHB.

> The modern web more and more makes trailing / more incorrect
> than correct.

Right, but not in this case.

> You have absolutely no way to tell if a URL is the "root of a directory"
> with so many internal rewrites and dynamic URLs around the web now.

Of course; that's why, in every case, manual checking is in order.  I did
verify it in this particular case before hitting "send" button.

Then again: we have tons of submissions that contain hunks that stir the
repo for no real use, clutter the diff, harm "svn blame"; yet every time
someone who points it out gets cursed.  Shouldn't we discourage meaningless
parts of the submitted patches get landed in the first place instead?

Trust me, I hate to be "that guy", but every time I get a break and sit
silently watching, very few actually speak up.  Oh boy how do I miss sat@
and his reviews.

./danfe


More information about the svn-ports-head mailing list