问题
Is there any Google Reader API that I can plug in to? I building a clean RSS/Atom reader in PHP and would love to get all the goodies from Google Reader like the history of a feed, able to add comments to each feed item, etc.
回答1:
I've built some google reader integration in python but I can share some of the api knowledge so you can get started. output=json is also available for all.
Login: https www.google.com/accounts/ClientLogin
POST &email=email&passwd=password&service=reader&source=appname&continue=http://www.google.com
from the response grab Auth=
Next hit: www.google.com/reader/api/0/token
HEADER Authorization=GoogleLogin auth=$Auth
That response becomes $token for the session.
From there it's just hitting some url's always passing that auth header and including the token in the querystring or post.
Gets a list of your subscriptions: www.google.com/reader/api/0/subscription/list?output=xml
To modify subscriptions this is the base url plus some post data for the action to perform
www.google.com/reader/api/0/subscription/edit?pos=0&client=$source
POST to add: s=$streams&t=$title&T=$token&ac=subscribe
POST to remove: s=$stream&T=$token&ac=unsubscribe
The $stream is generally feed/$feedurl like this for techcrunch, feed/http:// feeds.feedburner.com/Techcrunch
Sorry had to mangle some of the urls cause i don't have enough rep yet.
回答2:
this is a working example in python:
import urllib, urllib2
import json, pprint
email, password = 'jose@gmail.com', 'nowayjose'
clientapp, service = 'reader', 'reader'
params = urllib.urlencode({'Email': email, 'Passwd': password, 'source': clientapp, 'service': service})
req = urllib2.Request(url='https://www.google.com/accounts/ClientLogin', data=params)
f = urllib2.urlopen(req)
for line in f.readlines():
if line[0:5] == 'Auth=':
auth=line[5:]
root = "http://www.google.com/reader/api/0/"
req = urllib2.Request(root + "token")
req.add_header('Authorization', 'GoogleLogin auth=' + auth)
f = urllib2.urlopen(req)
token = f.readlines()[0]
# get user id
req = urllib2.Request(root + "user-info?output=json&token="+token)
req.add_header('Authorization', 'GoogleLogin auth=' + auth)
f = urllib2.urlopen(req)
dictUser = json.loads(f.read())
user_id = dictUser["userId"]
print "user_id",user_id
req = urllib2.Request(root + "subscription/list?output=json&token="+token)
req.add_header('Authorization', 'GoogleLogin auth=' + auth)
f = urllib2.urlopen(req)
# for line in f.readlines():
# print line
dictSubscriptions = json.loads(f.read())
# pprint.pprint(dictSubscriptions)
# print the first 3 subscription titles
for i in dictSubscriptions["subscriptions"][0:2]:
print i["title"]
req = urllib2.Request("http://www.google.com/reader/api/0/unread-count?output=json&token="+token)
req.add_header('Authorization', 'GoogleLogin auth=' + auth)
f = urllib2.urlopen(req)
dictUnread = json.loads(f.read())
# pprint.pprint(dictUnread)
# print the first 3 unread folders
for i in dictUnread["unreadcounts"][0:3]:
print i["count"], i["id"]
# this returns all starred items as xml
req = urllib2.Request("http://www.google.com/reader/atom/user/"+user_id+"/state/com.google/starred?token="+token)
req.add_header('Authorization', 'GoogleLogin auth=' + auth)
f = urllib2.urlopen(req)
starredItems = f.read()
回答3:
Google Reader has feeds for user's. I guess you could use those. Also, they're PubSubHubbub ready, so you will get comments/likes... as soon as they happen.
Also, as of July 1st 2013, Google Reader is no more. Options for replacements include Superfeedr.
来源：https://stackoverflow.com/questions/4063513/google-reader-api