# Gracefully degrading Ajax AND Drying up actions

**URL:** <https://www.ruby-forum.com/t/gracefully-degrading-ajax-and-drying-up-actions/56589>\
**Category:** Rails\
**Created:** [April 1, 2006, 12:38pm UTC](https://www.ruby-forum.com/t/gracefully-degrading-ajax-and-drying-up-actions/56589 "2006-04-01T12:38:51Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![Chris\_T](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/c/a8b319/32.png) [@Chris\_T](https://www.ruby-forum.com/u/Chris_T)\
**Post date:** [April 1, 2006, 12:38pm UTC](https://www.ruby-forum.com/t/gracefully-degrading-ajax-and-drying-up-actions/56589/1 "2006-04-01T12:38:51Z")

</div>

Still a semi-newbie, but so far I’ve managed to solve most things with a  
bit of help from the Agile book, some googling around, and a bit of  
trial-and-error, but I’m really stuck with this one (very much an Ajax  
newbie):

I know how to gracefully degrade an AJAX newpost type call so it can  
appear inline or (for those without JS support) go to a new page (put an  
“:href =\> url\_for” in the link\_to\_remote).

I also know how to DRY up two (non-AJAX) actions so that you use the  
same newpost for both the original call and processing the form info (if  
request.get? … elsif request.post?).

I even know how to have it so the same newpost action can handle the  
original request and tell if it’s an AJAX request or not, so I can use  
just one action for AJAX and none AJAX requests  
(request…xml\_http\_request?).

What I can’t work out to do is how to have it so the processing of the  
AJAX a form info is done in the same action as the original AJAX call  
(they are both posts, right?), and it’s forcing me to have a createpost  
action, just for the AJAX route.

If I could work this out, I could then have one neat action that wrapped  
the whole thing up in a nice DRY, gracefully degrading way.

Can anyone help?

Thanks  
Chris T

---

<div class="post-metadata">

**Author:** ![Chris\_T](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/c/a8b319/32.png) [@Chris\_T](https://www.ruby-forum.com/u/Chris_T)\
**Post date:** [April 1, 2006, 5:54pm UTC](https://www.ruby-forum.com/t/gracefully-degrading-ajax-and-drying-up-actions/56589/2 "2006-04-01T17:54:31Z")

</div>

On Apr 1, 2006, at 5:34 AM, Chris T wrote:

> the same newpost for both the original call and processing the form  
> have a createpost action, just for the AJAX route.  
> Are you testing request.xhr? _before_ you test for request.post?

If not, switch them around (or say “request.post? && ! request.xhr?”)  
since all xhr? posts are, of course, posts!

-Rob

Rob B. [http://agileconsultingllc.com](http://agileconsultingllc.com)  
removed\_email\_address@domain.invalid

---

<div class="post-metadata">

**Author:** ![Chris\_T](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/c/a8b319/32.png) [@Chris\_T](https://www.ruby-forum.com/u/Chris_T)\
**Post date:** [April 1, 2006, 6:34pm UTC](https://www.ruby-forum.com/t/gracefully-degrading-ajax-and-drying-up-actions/56589/3 "2006-04-01T18:34:01Z")

</div>

Rob B. wrote:

> > the AJAX a form info is done in the same action as the original AJAX  
> > Rob B. [http://agileconsultingllc.com](http://agileconsultingllc.com)  
> > removed\_email\_address@domain.invalid

The problem is not differentiating between the non-xhr posts and xhr  
posts (the order I’ve got it currently is: test for gets, else test for  
xhr, else test for post, which I _think_ has the same effect) but  
testing between the xhr posts that are the original call (i.e. the  
equivalent of the non-Ajax GET) and those that are returning form  
values.

I suspect I’m being dense here, but can’t figure it out. It’s simple if  
I have a new post and create post action – then on each I only need to  
test xhr? – and this avoids duplicating the codes for xhr and non-xhr,  
but does mean two actions. Not the end of the world, but would be nice  
to know if it’s easy to do (and therefore I am being dense).

Cheers  
Chris

---

<div class="post-metadata">

**Author:** ![Chris\_T](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/c/a8b319/32.png) [@Chris\_T](https://www.ruby-forum.com/u/Chris_T)\
**Post date:** [April 4, 2006, 10:19am UTC](https://www.ruby-forum.com/t/gracefully-degrading-ajax-and-drying-up-actions/56589/4 "2006-04-04T10:19:57Z")

</div>

Solved this – of a fashion, though the code is somewhat ugly at the  
moment so I’m not convinced I’m doing it the best way:

def newpost # NB somewhat simplified to remove some of the non-logic  
bits  
if request.get? # tests if non-Ajax GET request in which case…  
… # generate new object  
return  
elsif request.xml\_http\_request? # tests if Ajax request  
if not params[:post] # could prob bundle this up in Ajax test  
above  
… # generate new object for Ajax calls  
end  
end  
@post=Post.new(params[:post]) #get the params, whether they came from  
Ajax or non-Ajax call  
if @post.save  
flash[:notice] = “New post submitted successfully”  
if not request.xml\_http\_request? # plain old vanilla non-Ajax  
redirect\_to :action =\> ‘index’ and return  
else  
render(:partial =\> ‘post’, :object =\> @post, :layout =\>  
false) # show the new item for Ajax  
end  
elsif request.xml\_http\_request? # if we haven’t saved successfully  
and it was an Ajax call  
render :action =\> ‘newpost’, :id =\> params[:id], :layout =\> false  
and return # re-show form  
else  
render :action =\> ‘newpost’, :id =\> params[:id] and return # ditto  
for non-Ajax call  
end  
end

Comments or suggestions for improving this?  
Thanks  
Chris T