mirror of
https://github.com/kennethreitz/dive-into-python3.git
synced 2026-06-05 15:00:18 +00:00
finished httplib2-redirects section
This commit is contained in:
Mark Pilgrim
+56
-38
@@ -527,28 +527,32 @@ reply: 'HTTP/1.1 200 OK'</samp>
|
||||
<p><abbr>HTTP</abbr> defines <a href=#redirects>two kinds of redirects</a>: temporary and permanent. There’s nothing special to do with temporary redirects except follow them, which <code>httplib2</code> does automatically.
|
||||
|
||||
<pre class=screen>
|
||||
<samp class=p>>>> </samp><kbd class=pp>import httplib2</kbd>
|
||||
<samp class=p>>>> </samp><kbd class=pp>h = httplib2.Http('.cache')</kbd>
|
||||
<a><samp class=p>>>> </samp><kbd class=pp>response, content = h.request('http://diveintopython3.org/examples/feed-302.xml')</kbd> <span class=u>①</span></a>
|
||||
<samp>connect: (diveintopython3.org, 80)
|
||||
<a>send: b'GET /examples/feed-302.xml HTTP/1.1 <span class=u>②</span></a>
|
||||
<a>send: b'GET /examples/feed-302.xml HTTP/1.1 <span class=u>②</span></a>
|
||||
Host: diveintopython3.org
|
||||
accept-encoding: deflate, gzip
|
||||
user-agent: Python-httplib2/$Rev: 259 $'
|
||||
<a>reply: 'HTTP/1.1 302 Found' <span class=u>③</span></a>
|
||||
<a>send: b'GET /examples/feed.xml HTTP/1.1 <span class=u>④</span></a>
|
||||
<a>reply: 'HTTP/1.1 302 Found' <span class=u>③</span></a>
|
||||
<a>send: b'GET /examples/feed.xml HTTP/1.1 <span class=u>④</span></a>
|
||||
Host: diveintopython3.org
|
||||
accept-encoding: deflate, gzip
|
||||
user-agent: Python-httplib2/$Rev: 259 $'
|
||||
reply: 'HTTP/1.1 200 OK'</samp></pre>
|
||||
<ol>
|
||||
<li>
|
||||
<li>
|
||||
<li>
|
||||
<li>
|
||||
<li>There is no feed at this <abbr>URL</abbr>. I’ve set up my server to issue a temporary redirect to the correct address.
|
||||
<li>There’s the request.
|
||||
<li>And there’s the response: <code>302 Found</code>. Not shown here, this response also includes a <code>Location</code> header that points to the real <abbr>URL</abbr>.
|
||||
<li><code>httplib2</code> immediately turns around and “follows” the redirect by issuing another request for the <abbr>URL</abbr> given in the <code>Location</code> header: <code>http://diveintopython3.org/examples/feed.xml</code>
|
||||
</ol>
|
||||
|
||||
<p>“Following” a redirect is nothing more than this example shows. <code>httplib2</code> sends a request for the <abbr>URL</abbr> you asked for. The server comes back with a response that says “No no, look over there instead.” <code>httplib2</code> sends another request for the new <abbr>URL</abbr>.
|
||||
|
||||
<pre class=screen>
|
||||
# continued from the previous example
|
||||
<a><samp class=p>>>> </samp><kbd class=pp>print(dict(response.items()))</kbd> <span class=u>①</span></a>
|
||||
<a><samp class=p>>>> </samp><kbd class=pp>print(dict(response.items()))</kbd> <span class=u>①</span></a>
|
||||
<samp class=pp>{'status': '200',
|
||||
'content-length': '3070',
|
||||
<a> 'content-location': 'http://diveintopython3.org/examples/feed.xml', <span class=u>②</span></a>
|
||||
@@ -560,60 +564,74 @@ reply: 'HTTP/1.1 200 OK'</samp></pre>
|
||||
'connection': 'close',
|
||||
<a> '-content-encoding': 'gzip', <span class=u>③</span></a>
|
||||
'etag': '"bfe-4cbbf5c0"',
|
||||
'cache-control': 'max-age=86400',
|
||||
<a> 'cache-control': 'max-age=86400', <span class=u>④</span></a>
|
||||
'date': 'Wed, 03 Jun 2009 02:21:41 GMT',
|
||||
'content-type': 'application/xml'}</samp></pre>
|
||||
<ol>
|
||||
<li>
|
||||
<li>
|
||||
<li>
|
||||
<li>The <var>response</var> you get back from this single call to the <code>request()</code> method is the response from the final <abbr>URL</abbr>.
|
||||
<li><code>httplib2</code> adds the final <abbr>URL</abbr> to the <var>response</var> dictionary, as <code>content-location</code>. This is not a header that came from the server; it’s specific to <code>httplib2</code>.
|
||||
<li>Apropos of nothing, this feed is <a href=#httplib2-compression>compressed</a>.
|
||||
<li>And cacheable. (This is important, as you’ll see in the next example.)
|
||||
</ol>
|
||||
|
||||
<p>What happens if you request the same <abbr>URL</abbr> again?
|
||||
|
||||
<pre class=screen>
|
||||
# continued from the previous example
|
||||
<a><samp class=p>>>> </samp><kbd class=pp>response, content = h.request('http://diveintopython3.org/examples/feed-302.xml')</kbd> <span class=u>①</span></a>
|
||||
<a><samp class=p>>>> </samp><kbd class=pp>response2, content2 = h.request('http://diveintopython3.org/examples/feed-302.xml')</kbd> <span class=u>①</span></a>
|
||||
<samp>connect: (diveintopython3.org, 80)
|
||||
<a>send: b'GET /examples/feed-302.xml HTTP/1.1 <span class=u>②</span></a>
|
||||
<a>send: b'GET /examples/feed-302.xml HTTP/1.1 <span class=u>②</span></a>
|
||||
Host: diveintopython3.org
|
||||
accept-encoding: deflate, gzip
|
||||
user-agent: Python-httplib2/$Rev: 259 $'
|
||||
<a>reply: 'HTTP/1.1 302 Found' <span class=u>③</span></a></samp></pre>
|
||||
<a>reply: 'HTTP/1.1 302 Found' <span class=u>③</span></a></samp>
|
||||
<a><samp class=p>>>> </samp><kbd class=pp>content2 == content</kbd> <span class=u>④</span></a>
|
||||
<samp class=pp>True</samp></pre>
|
||||
<ol>
|
||||
<li>
|
||||
<li>
|
||||
<li>
|
||||
<li>Same <abbr>URL</abbr>, same <code>httplib2.Http</code> object (and therefore the same cache).
|
||||
<li>The <code>302</code> response was not cached, so <code>httplib2</code> sends another request for the same <abbr>URL</abbr>.
|
||||
<li>Once again, the server responds with a <code>302</code>. But notice what <em>didn’t</em> happen: there wasn’t ever a second request for the final <abbr>URL</abbr>, <code>http://diveintopython3.org/examples/feed.xml</code>. That response was cached (remember the <code>Cache-Control</code> header that you saw in the previous example). Once <code>httplib2</code> received the <code>302 Found</code> code, <em>it checked its cache before issuing another request</em>. The cache contained a fresh copy of <code>http://diveintopython3.org/examples/feed.xml</code>, so there was no need to re-request it.
|
||||
<li>By the time the <code>request()</code> method returns, it has read the feed data from the cache and returned it. Of course, it’s the same as the data you received last time.
|
||||
</ol>
|
||||
|
||||
<p>In other words, you don’t have to do anything special for temporary redirects. <code>httplib2</code> will follow them automatically, and the fact that one <abbr>URL</abbr> redirects to another has no bearing on <code>httplib2</code>’s support for compression, caching, <code>ETags</code>, or any of the other features of <abbr>HTTP</abbr>.
|
||||
|
||||
<p>Permanent redirects are just as simple.
|
||||
|
||||
<pre class=screen>
|
||||
<samp class=p>>>> </samp><kbd class=pp>response, content = h.request('http://diveintopython3.org/examples/feed-301.xml')</kbd>
|
||||
# continued from the previous example
|
||||
<a><samp class=p>>>> </samp><kbd class=pp>response, content = h.request('http://diveintopython3.org/examples/feed-301.xml')</kbd> <span class=u>①</span></a>
|
||||
<samp>connect: (diveintopython3.org, 80)
|
||||
send: b'GET /examples/feed-301.xml HTTP/1.1
|
||||
Host: diveintopython3.org
|
||||
accept-encoding: deflate, gzip
|
||||
user-agent: Python-httplib2/$Rev: 259 $'
|
||||
reply: 'HTTP/1.1 301 Moved Permanently'</samp>
|
||||
<samp class=p>>>> </samp><kbd class=pp>print(dict(response.items()))</kbd>
|
||||
<samp class=pp>{'status': '200',
|
||||
'content-length': '3070',
|
||||
'content-location': 'http://diveintopython3.org/examples/feed.xml',
|
||||
'accept-ranges': 'bytes',
|
||||
'expires': 'Thu, 04 Jun 2009 02:21:41 GMT',
|
||||
'vary': 'Accept-Encoding',
|
||||
'server': 'Apache',
|
||||
'last-modified': 'Wed, 03 Jun 2009 02:20:15 GMT',
|
||||
'connection': 'close',
|
||||
'-content-encoding': 'gzip',
|
||||
'etag': '"bfe-4cbbf5c0"',
|
||||
'cache-control': 'max-age=86400',
|
||||
'date': 'Wed, 03 Jun 2009 02:21:41 GMT',
|
||||
'content-type': 'application/xml'}</samp>
|
||||
<samp class=p>>>> </samp><kbd class=pp>response2, content2 = h.request('http://diveintopython3.org/examples/feed-301.xml')</kbd>
|
||||
<samp class=p>>>> </samp><kbd class=pp>response2.fromcache</kbd>
|
||||
<a>reply: 'HTTP/1.1 301 Moved Permanently' <span class=u>②</span></samp>
|
||||
<a><samp class=p>>>> </samp><kbd class=pp>response.fromcache</kbd> <span class=u>③</span></a>
|
||||
<samp class=pp>True</samp></pre>
|
||||
<ol>
|
||||
<li>FIXME
|
||||
<li>Once again, this <abbr>URL</abbr> doesn’t really exist. I’ve set up my server to issue a permanent redirect to <code>http://diveintopython3.org/examples/feed.xml</code>.
|
||||
<li>And here it is: status code <code>301</code>. But again, notice what <em>didn’t</em> happen: there was no request to the redirect <abbr>URL</abbr>. Why not? Because it’s already cached locally.
|
||||
<li><code>httplib2</code> “followed” the redirect right into its cache.
|
||||
</ol>
|
||||
|
||||
<p>But wait! There’s more!
|
||||
|
||||
<pre class=screen>
|
||||
# continued from the previous example
|
||||
<a><samp class=p>>>> </samp><kbd class=pp>response2, content2 = h.request('http://diveintopython3.org/examples/feed-301.xml')</kbd> <span class=u>①</span></a>
|
||||
<a><samp class=p>>>> </samp><kbd class=pp>response2.fromcache</kbd> <span class=u>②</span></a>
|
||||
<a><samp class=p>>>> </samp><kbd class=pp>content2 == content</kbd> <span class=u>③</span></a>
|
||||
<samp class=pp>True</samp>
|
||||
</pre>
|
||||
<ol>
|
||||
<li>Here’s the difference between temporary and permanent redirects: once <code>httplib2</code> follows a permanent redirect, all further requests for that <abbr>URL</abbr> will transparently be rewritten to the target <abbr>URL</abbr> <em>without hitting the network for the original <abbr>URL</abbr></em>. Remember, debugging is still turned on, yet there is no output of network activity whatsoever.
|
||||
<li>Yep, this response was retrieved from the local cache.
|
||||
<li>Yep, you got the entire feed (from the cache).
|
||||
</ol>
|
||||
|
||||
<p><abbr>HTTP</abbr>. It works.
|
||||
|
||||
<p class=a>⁂
|
||||
|
||||
<h2 id=beyond-get>Beyond HTTP GET</h2>
|
||||
|
||||
@@ -272,7 +272,7 @@ body{counter-reset:h1 4}
|
||||
<samp class=pp><_sre.SRE_Match object at 0x008EEB48></samp>
|
||||
<a><samp class=p>>>> </samp><kbd class=pp>re.search(pattern, 'MCMLXXXIX', re.VERBOSE)</kbd> <span class=u>②</span></a>
|
||||
<samp class=pp><_sre.SRE_Match object at 0x008EEB48></samp>
|
||||
<a><samp class=p>>>> </samp><kbd class=pp>re.search(pattern, 'MMMDCCCLXXXVIII', re.VERBOSE)</kbd> <span class=u>③</span></a>
|
||||
<a><samp class=p>>>> </samp><kbd class=pp>re.search(pattern, 'MMMDCCCLXXXVIII', re.VERBOSE)</kbd> <span class=u>③</span></a>
|
||||
<samp class=pp><_sre.SRE_Match object at 0x008EEB48></samp>
|
||||
<a><samp class=p>>>> </samp><kbd class=pp>re.search(pattern, 'M')</kbd> <span class=u>④</span></a></pre>
|
||||
<ol>
|
||||
|
||||
Reference in New Issue
Block a user