$.post and $.get are probably not handling the abcense of data well

[adsmart]

I was looking for the post and get code today (I'm planning on replicating them for the DELETE and PUT verbs... those should be added) and I noticed a funny problem with the following (GET has the same problem):

	post: function( url, data, callback, type ) {

		if ( jQuery.isFunction( data ) ) {
			callback = data;
			data = {};
		}

		return jQuery.ajax({
			type: "POST",
			url: url,
			data: data,
			success: callback,
			dataType: type
		});
	},

You make a check to see if the second argument is a function, indicating that no data is being sent. That's fine but it ignores the fact that the type might also be passed. As a result, the following would have unintended effects:

$.post("/sample/test", do_stuff_function, "text/html"); 

It also doesn't take into account

$.post("/sample/test", "text/html"); 

Which, while probably silly in most cases, is legitimate within the published docs for the functions.

I suspect the following would be a better solution (depending on how you feel about using arguments.length, you could use other methods to identify things):

if (arguments.length == 2 ) {
    type = data;
    callback = function () {};
    data = {};
} else if (arguments.length == 3) {
    type = callback;
    callback = data;
    data = {};
};

[ygirouard]

A fullproof solution would probably be to force users to pass an array of options (array object or json), much like you do for $.ajax(); That way it avoids having to deal with omited arguments since you reference each as an object property.

For example:

$.get({url: "/sample/test", callback: do_stuff_function, type: "html"});

Then you can test each parameter individually, and if set to 'undefined' you can ignore it or set a default.

i.e.:

  post: function(params) {
    if (params.data == undefined || params.data == ""){params.data = {};}
    // Do other testing here if needed...

    return jQuery.ajax({
      type: "POST",
      url: params.url,
      data: params.data,
      success: params.callback,
      dataType: params.type
    });
  },

[nathanhammond]

Makes sure the argument shift is complete.

[nathanhammon]

I've attached a patch that can close this issue.

[flesler]

Ok, thanks for the patch. I'll review this asap.

[paul.irish]

Ticket #3775 has gotten some attention on this issue, too. I think we can close it as a dupe.

Ariel, can we drop in Nathan's non-intrusive patch?

[JDay]

A patch is included with dupe #4384 with a test case and a fix for the comment typo.

[flesler]

Fixed at [6564]

[rwaldron]

#4785 is a duplicate of this ticket.