Bug #31620 for WWW-Facebook-API: FAQ suggestion

Mon Dec 17 19:36:18 2007 PERRIN [...] cpan.org - Ticket created

Subject:

FAQ suggestion

After fumbling with this for a while, I discovered that Facebook doesn't work unless you make your forms POST. Using a GET form causes duplicate values that make WWW::Facebook::API choke. See this PerlMonks node for more: http://perlmonks.org/index.pl?node_id=657556 I think this would be a useful thing to have in the FAQ.

Mon Dec 17 22:53:42 2007 unobe@cpan.org - Correspondence added

Subject:	Re: [rt.cpan.org #31620] FAQ suggestion
Date:	Mon, 17 Dec 2007 19:52:41 -0800
To:	PERRIN via RT <bug-WWW-Facebook-API [...] rt.cpan.org>
From:	David Romano <unobe [...] cpan.org>

Hi Perrin, PERRIN via RT wrote on Mon, Dec 17, 2007 at 04:36:20PM PST: Show quoted text

> After fumbling with this for a while, I discovered that Facebook doesn't > work unless you make your forms POST. Using a GET form causes duplicate > values that make WWW::Facebook::API choke. See this PerlMonks node for > more: http://perlmonks.org/index.pl?node_id=657556 > > I think this would be a useful thing to have in the FAQ.

Thanks for letting me know, and I'm terribly sorry for the inconvenience! However, wouldn't it be better to change WWW::Facebook::API to account for this problem rather than document it in the FAQ? My understanding is that tests need to be written that make sure the parameters which must be single values actually are single values (and, of course, implement the logic to make the tests pass ;-) I'll work on it over winter break, but don't have much time now. Of course, patches are welcome :-) - David -- '"Why--damn it--it's medieval," I exclaimed; for I still had all the chronological snobbery of my period and I used names of earlier periods as terms of abuse.' -- C.S. Lewis, Miracles

Mon Dec 17 22:53:46 2007 The RT System itself - Status changed from 'new' to 'open'

Mon Dec 17 23:51:39 2007 perrin [...] elem.com - Correspondence added

Subject:	Re: [rt.cpan.org #31620] FAQ suggestion
Date:	Mon, 17 Dec 2007 23:51:14 -0500
To:	bug-WWW-Facebook-API [...] rt.cpan.org
From:	"Perrin Harkins" <perrin [...] elem.com>

On Dec 17, 2007 10:53 PM, unobe@cpan.org via RT <bug-WWW-Facebook-API@rt.cpan.org> wrote: Show quoted text

> Thanks for letting me know, and I'm terribly sorry for the > inconvenience!

It's not your fault. The Facebook docs don't seem to mention this. Show quoted text

> However, wouldn't it be better to change > WWW::Facebook::API to account for this problem rather than document it > in the FAQ?

If you mean try to work around it, I don't think so. It's much simpler to just make forms use POST and there's no reason not to if Facebook is going to turn them into POST anyway. However, the failure behavior could be improved. Right now it just gives mysterious internal errors about the wrong number of values in a hash assignment. If it finds multiple values for any fb_sig_* parameter, it's almost certainly caused by this and could issue a warning about making sure to use POST. - Perrin

Tue Jan 29 22:41:49 2008 unobe [...] cpan.org - Taken

Wed Jan 30 02:11:18 2008 unobe@cpan.org - Correspondence added

Subject:	Re: [rt.cpan.org #31620] FAQ suggestion
Date:	Tue, 29 Jan 2008 23:10:50 -0800
To:	"perrin [...] elem.com via RT" <bug-WWW-Facebook-API [...] rt.cpan.org>
From:	David Romano <unobe [...] cpan.org>

Hi Perrin, perrin@elem.com via RT wrote on Mon, Dec 17, 2007 at 08:51:43PM PST: Show quoted text

> If you mean try to work around it, I don't think so. It's much > simpler to just make forms use POST and there's no reason not to if > Facebook is going to turn them into POST anyway. However, the failure > behavior could be improved. Right now it just gives mysterious > internal errors about the wrong number of values in a hash assignment. > If it finds multiple values for any fb_sig_* parameter, it's almost > certainly caused by this and could issue a warning about making sure > to use POST.

Yeah, so it took a bit longer than I planned for me to get to this :-( Anyway, I have a new version in the googlecode repo that has a FAQ added on the topic, and I've changed a couple methods in WFA::Canvas to croak if any of the hash values are references. I've attached the relevant patch to this message, and if I don't hear otherwise, I'll make a release next week. - David -- "Beware of the inclination to dictate to God as to what you will allow to happen if you obey Him." -- Oswald Chambers, My Utmost for His Highest, January 11

Message body is not shown because sender requested not to inline it.

Mon May 26 03:04:26 2008 unobe [...] cpan.org - Status changed from 'open' to 'resolved'