# Find\_by\_id vs. find in postback action

**URL:** <https://www.ruby-forum.com/t/find-by-id-vs-find-in-postback-action/96304>\
**Category:** Rails\
**Created:** [June 28, 2006, 12:23am UTC](https://www.ruby-forum.com/t/find-by-id-vs-find-in-postback-action/96304 "2006-06-28T00:23:12Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![sevensoft](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/f9ae1b/32.png) [@sevensoft](https://www.ruby-forum.com/u/sevensoft)\
**Post date:** [June 28, 2006, 12:23am UTC](https://www.ruby-forum.com/t/find-by-id-vs-find-in-postback-action/96304/1 "2006-06-28T00:23:12Z")

</div>

I’m playing around with the Postback action recipe listed in the recipes  
book.  
Here is the default code given:

def edit  
@recipe = Recipe.find\_by\_id(params[:id]) || Recipe.new  
if request.post?  
@recipe.attributes = params[:recipe]  
redirect\_to :main\_url and return if @recipe.save  
end  
end

Here is what my code sort of ended up looking like, altho i’ve ripped it  
apart so many times now trying to figure it out:

def edit  
begin  
@entry = Entry.find(params[:id])  
rescue  
@entry = Entry.new  
end

# causes a weird error where on the second time in, with no params

passed in it still loads the previous entry object saved.

# @entry = Entry.find\_by\_id(params[:id]) || Entry.new

# logger.debug “!!! edit: #{@entry.inspect}”

```
if request.post?
  @entry.attributes = params[:entry]
  # pulled this out so i can inspect it
  entrysaved = @entry.save
  logger.debug "on save: #{@entry.inspect}"
  redirect_to :action => "index" and return if entrysaved
end

```

end

I get some interesting results tho.

So, the first time i click the href (simple href to :action =\> “edit”)  
the edit form pops up (blank), i type in the values and submit.

My page redirects and then loads the index page which shows me my entry  
that i just saved.

All good so far.

Now i click the same href (no params attached or anything) and the form  
pops back up, but this time its prefilled.  
When i look at my inspect above, it looks like it loaded the last entry  
saved. However if i inspect the params[:id] that shows nil.

The weirdest thing is that if i do  
Entry.find\_by\_id(nil) in place of Entry.find\_by\_id(params[:id]) it still  
loads the last entry saved. They have different addresses, so its not  
the same object that was created and then passed around a session or  
something.

So to top it off I tried using just what you see above, a  
Entry.find(params[:id]) and did my error trapping around it.

That works. The old saved entry is never pulled back up.

So what in the world am I missing. Is there some ‘magic’ in rails that  
is supposed to be a shortcut or am I just missing something in the code?  
I can’t believe that it would be a bug, that would be caught immediately  
by the team I’m sure.

Some System details:  
Mac OSX 10.4.6  
Ruby 1.8.4  
Rails 1.1.2  
Mysql 4.1.18

thx,  
k

---

<div class="post-metadata">

**Author:** ![sevensoft](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/f9ae1b/32.png) [@sevensoft](https://www.ruby-forum.com/u/sevensoft)\
**Post date:** [June 29, 2006, 4:18am UTC](https://www.ruby-forum.com/t/find-by-id-vs-find-in-postback-action/96304/2 "2006-06-29T04:18:47Z")

</div>

no ideas? ☹

ken brooks wrote:

> # causes a weird error where on the second time in, with no params
> 
> passed in it still loads the previous entry object saved.
> 
> # @entry = Entry.find\_by\_id(params[:id]) || Entry.new
> 
> # logger.debug “!!! edit: #{@entry.inspect}”
> 
> ```
> if request.post?
> @entry.attributes = params[:entry]
> # pulled this out so i can inspect it
> entrysaved = @entry.save
> logger.debug "on save: #{@entry.inspect}"
> redirect_to :action => "index" and return if entrysaved
> end
> 
> ```
> 
> end

> …snip

> So what in the world am I missing. Is there some ‘magic’ in rails that  
> is supposed to be a shortcut or am I just missing something in the code?  
> I can’t believe that it would be a bug, that would be caught immediately  
> by the team I’m sure.

---

<div class="post-metadata">

**Author:** ![sevensoft](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/f9ae1b/32.png) [@sevensoft](https://www.ruby-forum.com/u/sevensoft)\
**Post date:** [November 7, 2006, 6:01pm UTC](https://www.ruby-forum.com/t/find-by-id-vs-find-in-postback-action/96304/3 "2006-11-07T18:01:07Z")

</div>

Ken Brooks wrote:

> @recipe = Recipe.find\_by\_id(params[:id]) || Recipe.new  
> …
> 
> # causes a weird error where on the second time in, with no params
> 
> passed in it still loads the previous entry object saved.

Unfortunately, the “magic” is in MySQL. Although I wouldn’t exactly call  
it magic. The explanation may be found here:  
[http://dev.mysql.com/doc/refman/5.0/en/myodbc-usagenotes-functionality.html](http://dev.mysql.com/doc/refman/5.0/en/myodbc-usagenotes-functionality.html).  
In a nutshell, MySQL is “helping” you by replacing NULL with last  
generated ID. And as I found out, it does that even if last ID was  
generated for some other table.

My workaround for this situation was to prepend "params[:id] && " as  
follows:

@recipe = params[:id] && Recipe.find\_by\_id(params[:id]) || Recipe.new

Or, if you do not want to treat invalid ID’s as “new”, you may do this:

@recipe = params[:id] ? Recipe.find(params[:id]) : Recipe.new

Regards,  
Fyodor Golos

---

<div class="post-metadata">

**Author:** ![sevensoft](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/f9ae1b/32.png) [@sevensoft](https://www.ruby-forum.com/u/sevensoft)\
**Post date:** [November 8, 2006, 4:03am UTC](https://www.ruby-forum.com/t/find-by-id-vs-find-in-postback-action/96304/4 "2006-11-08T04:03:27Z")

</div>

Ahhhh, I got bitten by this last week and couldn’t for the life of me  
figure  
out what was going on!! I fixed it in the same way you described, but  
didn’t  
know why it worked :).

If you ask me, that’s really weird default behaviour on the part of  
mysql.

-Jonathan.