I have stumble upon several PHP-scripts for fetching my feedburner
subscriber count. But no one of them works when using the new;
feedproxy.google.com.
Is there a solution or are you going to fix this soon? :)
> I have stumble upon several PHP-scripts for fetching my feedburner
> subscriber count. But no one of them works when using the new;
> feedproxy.google.com.
> Is there a solution or are you going to fix this soon? :)
He will not find in the documentation that feedproxy activates with
google ads in feeds, and after it is once activated some of the
aggregators will stop seeing actual data because of some header
misinterpretations. They will see 10+ days OLD data.
> > I have stumble upon several PHP-scripts for fetching my feedburner
> > subscriber count. But no one of them works when using the new;
> > feedproxy.google.com.
> > Is there a solution or are you going to fix this soon? :)
We are only documenting and supporting the new APIs as we are actively
transitioning all feeds to Google accounts. We cannot automatically
forward requests because of authentication differences.
A correction to the statement below that is simply incorrect: No ads
are ever activated in feeds unless they were already activated on
feedburner.com or the publisher has activated them.
> He will not find in the documentation that feedproxy activates with
> google ads in feeds, and after it is once activated some of the
> aggregators will stop seeing actual data because of some header
> misinterpretations. They will see 10+ days OLD data.
> > > I have stumble upon several PHP-scripts for fetching my feedburner
> > > subscriber count. But no one of them works when using the new;
> > > feedproxy.google.com.
> > > Is there a solution or are you going to fix this soon? :)
OK, maybe I am wrong, but I don't like these changes.
I had never any problems with feedburner api before google start
messing with FeedBurner. I have problems now and I told you what they
are. In my websites I have 9 feeds with feedproxy to check. All nine
are using feed ads.
If feedburner was my job, I would not change api url or if I was
forced to change http to https, I would make new URL to work for all
feeds, not only for feedproxy ones. 1000s of developers must now store
information whether or not a feed is feedproxy and to use different
url and protocol to get data. 1000s of developers are wondering how to
cheat feedproxy to return them NEW posts, not 10 days old.
> We are only documenting and supporting the new APIs as we are actively
> transitioning all feeds to Google accounts. We cannot automatically
> forward requests because of authentication differences.
> A correction to the statement below that is simply incorrect: No ads
> are ever activated in feeds unless they were already activated on
> feedburner.com or the publisher has activated them.
> On Dec 24, 6:48 am, dzver wrote:
> > I disagree. He will not find in the documentation that
> > He will not find in the documentation that feedproxy activates with
> > google ads in feeds, and after it is once activated some of the
> > aggregators will stop seeing actual data because of some header
> > misinterpretations. They will see 10+ days OLD data.
> > > > I have stumble upon several PHP-scripts for fetching my feedburner
> > > > subscriber count. But no one of them works when using the new;
> > > > feedproxy.google.com.
> > > > Is there a solution or are you going to fix this soon? :)
wrong, it returns status code 200 OK in both cases.
if you try google api first, you'll have to check 100s of wrong
addresses a single feedproxy address.
if you try feedburner api first, you'll have to check 10s of false-0
results. I mean, feedburner api returns 0 in approximately 5% of the
cases and 0 means "feedburner is currently returning zeroes", not
"feedburner has just migrated this account to google" :-)
other problem: if you have an aggregator and you are following
redirects, you'll know which is the last redirect, so you'll raise a
flag "this url is a feedproxy". But if you get data from feedproxy
you're already lost the battle, because the data is old. So you need
to get data with feedburner user-agent and this case you'll never know
if this specific account uses feedproxy.
On Dec 25, 8:24 pm, Franklin Tse [Community Expert] wrote:
Franklin: you see now what I mean :-) I am doing this with python once
per week and I'm storing a flag in DB for each feed, so I could check
valid api url later. But this is an ugly solution for an ugly
problem.
I think FeedBurner must keep old api available or make new one to
serve all feeds. I am not sure how easy is to solve this with PHP.
Eric might use curl
> He will not find in the documentation thatfeedproxyactivates with
> google ads in feeds, and after it is once activated some of the
> aggregators will stop seeing actual data because of some header
> misinterpretations. They will see 10+ days OLD data.
> > > I have stumble upon several PHP-scripts for fetching my feedburner
> > > subscriber count. But no one of them works when using the new;
> > >feedproxy.google.com.
> > > Is there a solution or are you going to fix this soon? :)
> > He will not find in the documentation thatfeedproxyactivates with
> > google ads in feeds, and after it is once activated some of the
> > aggregators will stop seeing actual data because of some header
> > misinterpretations. They will see 10+ days OLD data.
> > > > I have stumble upon several PHP-scripts for fetching my feedburner
> > > > subscriber count. But no one of them works when using the new;
> > > >feedproxy.google.com.
> > > > Is there a solution or are you going to fix this soon? :)
Hi Franklin, that URL was just an example, I didn't check wich type
was for that specific account(I have many), I know the 2 URL that I
need to use for it, to the old feedburner accounts and the new one
after changed to feedproxy, the problem is that the new URL often is
down, or given the wrong subscribers number.
Today seems fine, but who knows tomorrow? How can I trust and use it
on my application if I never know if it will work tomorrow or in the
next hours?
On Jan 2, 3:16 am, Franklin Tse [Community Expert] wrote:
