Django / Python RegEx-'？id ='是做什么的？

Question

I'm reading a Django tutorial and in the tutorial, the urls.py is this:

(r'^vote/$', bookmark_vote_page),

and there is a model called 'SharedBookmark':

class SharedBookmark(models.Model):
    bookmark = models.ForeignKey(Bookmark, unique=True)

but in the template, the link which leads to /vote/ is this:

{% if shared_bookmarks %}
     <ul class="bookmarks">
     {% for shared_bookmark in shared_bookmarks %}
         <li>
             <a href="/vote/?id={{ shared_bookmark.id }}" class="vote">[+]</a>

As you can see in the line:

<a href="/vote/?id={{ shared_bookmark.id }}" class="vote">[+]</a>

the link does not just go to /vote/, it goes to /vote/ with a '?id=x' where x is a number.

The view which handles the link is this:

def bookmark_vote_page(request):
     if request.GET.has_key('id'):
         try:
             id = request.GET['id']

My question is, what does '?id=x' exactly do? Because it does NOT change the URL (when I click the link, it still goes to /vote/).

I tried google'ing but all the explanations I saw explained what ?P= is but not what ?id= does. I remember seeing a django documentation which explains it but I can't find the link to the documentation anymore for some reason.

Answer 1

Sam explained how the URL is put together. Continuing from there...

In Django, the query strings are turned into a dictionary-like object. Since the request is a GET request (the kind that puts query parameters into the URL instead of the request body), this dictionary-like object is stored at request.GET . So when the view executes this line:

 if request.GET.has_key('id'):

it's checking if the URL had a query string parameter with the name "id" at all. Once it's sure, it continues by getting it with:

 id = request.GET['id']

BTW, it's recommended (for readability, mostly) to instead check for query parameter presence like this:

 if 'id' in request.GET:

Answer 2

? in URLs start a query string . Query strings are list of parameters in the form:

?name1=value1&name2=value2&name3=value3

This is an easy way to pass data to modify GET requests.

---

In most HTTP requests, like POST/PUT/DELETE/etc., you have the ability to send data along with the request. In GET requests, you should only be asking for content from a certain URL. But sometimes, you would like to request a URL but receive slightly modified contents. Think about searches or query parameters:

GET /search?q=example
GET /search?q=example&p=2
GET /search?q=new%20word

---

My final note more so applies to the design of RESTful APIs. I say you can "pass data" with query strings in GET requests, but depending on the kind of data you pass it could be bad design to do so. In general, query parameters should do something along the lines of:

• filtering
• sorting
• modifying

This would include things like ?q=example for filtering a set of results, ?sort=alphanumeric for sorting results, and ?return=json to modify the response type. If you are ever sending data to modify a DB , you should probably not be using a GET request. For example, registrations should use a POST request (since it is creating something) and modifying a user's settings should use a PUT request (since it is updating something).