设置angularjs应用进行爬网

Question

I have an angular app set up to work in html5 mode with a #! fallback, so on most browsers it works with http://example.com/foo/bar and on less cool browsers we get http://example.com/#!/foo/bar . All that seems to work.

I have been going through trying to get google to crawl the site nicely, and it doesn't seem to be working as expected. I have set up <meta property="fragment" content="!" /> <meta property="fragment" content="!" /> in the page to signify to google to recrawl with ?_escaped_fragment_= , and set up nginx to redirect to a static version of the page when it receives a request like this.

It is working for the front page - looking in the access logs I can see http://example.com/?_escaped_fragment= and can google "A sentence from the front page" and get the home page back as a result.

However it is not working for any of the interior pages, if I look in the access logs I can see a whole bunch of http://example.com/foo/bar/?_escaped_fragment_= rather than http://example.com/?_escaped_fragment_=/foo/bar/ as I might have expected.

Is there anything obvious I am missing to make google do what I want it to?

Answer 1

I think that is for AngularJS apps with HTML5 routes, and indeed, you should see requests with just ?_escaped_fragment_=, not ?_escaped_fragment_=/foo/bar/. For more info check section "3. Handle pages without hash fragments" here, https://developers.google.com/webmasters/ajax-crawling/docs/getting-started .