The version of _cart.html.erb shown on this page does not have Ajax for the “empty cart” button, where as most users following the example would have the Ajax version at this point.

(Sam says: Iteration D4 says “So let’s not use this code” and describes a change to store.html.erb instead)

A first attempt to hide the flash might involve adding the following line to
add_to_cart.rjs

should be

A first attempt to hide the flash might involve adding the following line to
add_to_cart.js.rjs

Instead of…

<%= image_tag product.image_url, :class => ‘list-image’ %>

Not a typo but a question

Why do I get a name like this

:depot michelbrosseau$ ruby script/generate migration add_test_data
exists db/migrate
create db/migrate/20080603231948_add_test_data.rb
:depot michelbrosseau$

Thanks

(Sam says: You clearly are running rails 2.1; the book has now been updated to match this release.)

The view pages are currently listed as rhtml and the demo code has all rhtml pages. The new Rails 2 convention is .html.erb

(sam says: This should not be the case for any chapters that have been marked as updated)

“An .js.rjs template is a way of getting JavaScript on the browser to do what you want”: “An” should be “A”

“<%= button_to ”Checkout“, :action => :checkout %>”: Put a “this line changed” triangle in front of this line, to show the reader which line they need to add for this step.

“We have just limited access to the store iteself to administrators.”: “iteself” should be “itself”

“Now we’ll use respond_to to vector to the correct template” … “:xml => product.to_xml(:include => :orders)": By including this ":xml => product.to_xml()” part at this point, it changes the XML generated so that it isn’t using the builder template any more, and therefore the XML generated doesn’t match the curl output shown just a few paragraphs down the page. The “:xml => @product.to_xml()” part should be omitted here (p. 186), and shouldn’t be added in until the “Autogenerating the XML” section (p. 187)

“Be warned: the only data you should load in migrations is data that you’ll also want to see in production: lookup tables, predefined users, and the like. Do not load test data into your application this way.”:

This question has been on my mind ever since you had me do exactly that on pages 88-90. By creating the “add_test_data” migration for the Depot sample application, it looks like you’ve set the stage for my development test data to ultimately end up on the production server, right?

I think it would be good to add a footnote or other explanation on pp. 88-90 about not doing it this way for a real application, and/or showing the reader here a different method to populate data for the “development” environment that will not affect the “production” environment. (What would that method be? I’m still wondering.)

3.3 Installing on Mac OS X

Leopard, but not to difficult either way.

not to -> not too

The function for cart.total_price is incorrect, because it doesn’t take quantity into account. It should be:
def total_price
\t@items.sum { |item| item.price * item.quantity }
end

Instead of:
def total_price
\t@items.sum { |item| item.price }
end

(Sam: says: look at the definition of CartItem.price in app/models/cart_item.rb; item.price is defined as product.price * quantity)

“Be warned: the only data you should load in migrations is data that you’ll also want to see in production: lookup tables, predefined users, and the like. Do not load test data into your application this way.”: Is it possible to put an if statement into up function to determine the environment (development or production) and act according?

(Sam says: Yes, but it is not recommended.)

the code for the login index view is incorrect, it is showing the code for the products index

(Sam says: Page 166 in B1.3 contains the correct code for the users index view. No mention of products on that page.)

“As well as freezing the current gem version of Rails into your application, you
can also link your application to a version of Rails from Rails’ own Subversion
repository…” Hasn’t rails moved away from subversion?

The view naming table gives the name of the file to be app/views/store/list.rhtml Shouldn’t it be list.html.erb?

It is not clear why, in the addition to new.html.erb, the “Price” label is added using literal HTML, whereas in the already existing template code, and in the following edit.html.erb modification, you use the f.label function vie ERb.

In the example we explicitly add “:scale => 2” to store prices with two digits after the decimal. I am following along using sqlite3 as the back end, and the app seems to store at least four digits after the decimal. Perhaps sqlite does not use this directive? If so, the reader might be warned, or a different example chosen.

(Sam says: added a warning)

The top of this page shows code :

class LineItem < ActiveRecord::Base
belongs_to :order
belongs_to :product
def self.from_cart_item(cart_item)
li = self.new
li.product = cart_item.product
li.quantity = cart_item.quantity
li.total_price = cart_item.price
li
end
end

However, at this stage the from_cart_item helper method has not been introduced. It is introduced and code supplied on page 152.

Running the SQLite3 SQL

depot> sqlite3 -line db/development.sqlite3
SQLite version 3.4.0
Enter “.help” for instructions
sqlite> select * from orders;

Shows many fields that would not be there for those following along in the book (e.x ship_class, created_at, updated_at, customer_email, etc.)

UPDATE: I see the timestamps updated_at and created_at are actually created by the application (and thus those fields are in our table as well — it helps to scroll completely across when viewing). However, fields like customer_email and ship_class are not.

The example used for create table may be updated to Rails 2 syntax. (t.integer instead of t.column)

In the sentence that begins “The Rails generators automatically…”, the word “produts” should be “products”.

In the sentence that begins “Our Depot application, for instance”, the word “produts” should be “products”.

>redirect_to(:controller => “login” , :action => “login” )

I think this should be
redirect_to(:controller => “admin” , :action => “login” )

Footnote 2 states that “:id=>product is idiomatic shorthand for :id=>product.id”. It is very helpful that you point this out, and it would be great if there was one of those references to a section in Appendix A that would explain what language feature makes this happen (for those of us learning Ruby as we are learning Rails). Also, it might be better if this footnote appeared on page 102 (where this snippet of code first appears).

(Dave says: it’s not a language feature. It’s just Rails saying "oh, I got a model, so I’ll ask for its id)

The server in the URL in footnote 4 (wiki.script.aculo.us) does not resolve…
$ host wiki.script.aculo.us
Host wiki.script.aculo.us not found: 2(SERVFAIL)

Should add “changes this” arrows to the <% if cart > .. <% end %> addition as well…

line 3: in schema.rb there is no table schema_migrations, nevertheless the database has such a table, which contains the timestamp

I get an error in the application log saying rails is looking for add_to_cart.js.erb instead of add_to_cart.js.rjs. As a result the cart is not updated. ruby ver 1.8.6 rails 2.1.0 on OS X 10.5.

(sam says: make sure the highlighted “respond_to { |format| format.js }” line is present in store_controller.rb)

The second full bullet on this page reads “We also used the h method to ensure that any remaining the HTML in the product title and description is escaped.” The ‘the’ before HTML does not seem to fit. If that ‘the’ is removed, the sentence flows much better.

ActionController::Routing::Routes is nil for me. Is there another step to initialize it?

I would have been helped by a sentence or two about routing to the root directory of a [sub]domain - e.g.
map.connect “/”, :controller => “mycontroller”, :action => “index”
with the hint that you must delete [or rename] public\\index.html for this to work.

[sorry; should have posted this here first. I posted it over in the forums section before realizing there was a separate errata system. No need to reply to the forum thread on this.]

The business about making the store layout the default layout for the application is sequenced oddly, in such a way as to generate some confusing errors that aren’t mentioned in the text.

If the instructions are followed in the order that the book lists them, then when the user visits localhost:3000/users immediately after putting the layout “store” instruction in application.rb, you get a route error (or at least, I got a route error doing this on Windows).

The text should say to remove to unnecessary products.html.erb and users.html.erb BEFORE telling the user to go look at either of those controllers.

Also-although I’m sure you know this-the illustrations for that section need to be fixed because the browser screenshots in them show the wrong URLs, which was also quite confusing for me until I remembered the note at the front of the book saying not to trust the illustrations.

“Note that it isn’t possible to set attributes in the intermediate table using this
approach.” This leaves a huge gap. So how do you set the intermediate table? Maybe refer to another page???

(Sam says: This is describing a shortcut “equivalent to the previous example”. It isn’t possible to set attributes using the shortcut, one one would need to follow the approach in the previous example.)

In the paragraph starting with “As is usual with Rails…” it is written “… added to to-level application controller…” instead of “top-level application controller”

How does Rake know which migrations have and have not been applied to your
database? Have a look at your schema after running a migration. You’ll find a
table called schema_mirations that it uses to keep track of the version number.7

should be

How does Rake know which migrations have and have not been applied to your
database? Have a look at your schema after running a migration. You’ll find a
table called schema_migrations that it uses to keep track of the version number.7

Creating a new rails application with a specific database, the syntax is —database=TYPE (i.e. it needs an = between —database and the type). So the line

work> rails —database mysql depot

should read

work> rails —database=mysql depot

Instructions created a Rake task in lib/task named db_schema_migrations.rake and then demonstrate running it with the command “rake db:schema_version”. The correct command to run the task should be “rake db:schema_migrations”.

I think there is a typo in the third line of new code for the “destroy” method:

#{user.name}

should be:

#{@user.name}

You have the default template generated as well as the one that you create in the suggested code snippet. Note the two lines marked with >> as not necessary:

>>

Info#who_bought

>>

Find me in app/views/info/who_bought.html.erb

People Who Bought <%= product.title %></h3> <ul> <% for order in orders -%>

<%= mail_to order.email, order.name %>

<% end -%>

Thanks!
-Irene

In the def for who_bought you’ve changed the respond_to from |format| 2 pages prior to |accepts| without any changes as far as I can tell. If the change was merely cosmetic, I found it to be confusing.
Thanks,
-Irene

When authoring a migration for the purpose of “creating data that’s a convenience for the developer when they play with the application”, it seems that the migration should be written so that it does not affect a production database. A comment here about how to check what environment the migration is running under (or perhaps some other way to do this) might be appropriate here.

(Sam says: Per the book “Be warned: the only data you should load in migrations is data that you’ll also want to see in production: lookup tables, predefined users, and the like. Do not load test data into your application this way.”)

The product_test.rb generated automatically does not contain the fixtures :products as indicated on this page.
This is using rails 2.1.0.

Thanks!
-Irene

Why the switch to mysql from sqlite3? The depot application has been using sqlite3 and it doesn’t quite seem consistent (or necessary) to switch without at least an explanation.

Thanks!
-Irene

We never created a LoginController… we created an Admin Controller with the methods: login, logout and index.

Also, the automatic admin_controller_test.rb that is generated (rails gem ver 2.1.0) does not contain the sample you give, it only contains:

require ‘test_helper’

class AdminControllerTest < ActionController::TestCase
# Replace this with your real tests.
def test_truth
assert true
end
end

Thanks!

You seem to have pasted the code twice for the ApplicationController separated by a single line… it doesn’t really make sense. It starts on 207 and goes to 208.
Thanks!

Suggest you describe why using :id => product is okay as the “:id” of the item, as opposed to :id => product.id.

“return salable items.” is poor wording choice, as in the real world, returns happen when we don’t like an item and take it back. Since you are describing the return value of a method, the wording is technically correct. Perhaps “provides a list of items for sale” might be better.

Pagination as described in the section “Pagination Helpers” was deprecated and removed from Rails as of 2.0 (as per wiki.rubyonrails.org). They suggest the will_paginate plugin.

in footnote #7 on page 77, you refer to the migrations table as “schema_mirations” instead of “schema_migrations”

This line
create db/migrate/20080601000002_add_price_to_product.rb
should not have .db at the end.

(Sam says: It has a .rb at the end)

I would like a better explanation of the version number in the database. I can see that part of the version number is the date but the rest is gibberish to me. This is what one of mine looks like:
20080617021145

rake db:migrate not ruby db:migrate

The illustrations for the One-to-One and One-to-Many relationships are correct, but the readability could be tweaked.

I tend to read from left to right and I usually assume the parent comes before the child. You may want to re-arrange your illustrations on the One-to-One relationships and the One-to-Many relationships, so that the parent table is to the left of the child tables (with the arrow connector pointing to the left).

(Dave says: I like my arrows pointing inthe direction I read, so I think I prefer the diagrams this way)

in validates_inclusion_of, at options, :allow_nil appears twice. i think that the second is allow_blank

This section talks about pagination helpers and how they may be split off in Rails 2.0. Rails 2 is here and pagination is gone.

The word “divine” in this sentence should be “define”?

“…we can see why it’s hard for Active Record to divine the relationships between tables..”

after adding the storecontroller, i received an error “No route matches ”/login/index" with {:method=>:get}".

restarting mongrel cleared the problem.

might be providing this as a hint.
why do i need to restart mongrel is a secondary question that the authors might like to shed some light on.

i am running 2.0

(Sam says: this is a non-deterministic bug in Rails. DHH is aware)

“respond_to { |format| format.js }”

I saw this line of code and had trouble understanding why “format.js” and not “add_to_cart.js”. Only after I googled it I found out that it’s a convention and that the name of the variable didn’t matter, only the “.js” did.

Wouldn’t it be better if this was explained?

missing: code truncated

(Dave says: we fix the code listings just before typesetting)

Substitue “class Book” with “class BadBook”

The code says
“class Book < ActivieRecord::Base
self.primary_key = ”isbn"
end"

The next example says
“book = BadBook.new
book.id = ”0-12345-6789"

I’m assuming that BadBook is defined by the Book class listed just prior to it use.

You say :

A better way to handle the problem is to create the cart HTML but set the CSS style to display: none if the cart is empty. To do that, we need to change the store.html.erb layout in app/views/layouts. Our first attempt is something like this.

<div id=“cart”
<% if cart.items.empty? %> style="display: none" <% end %> > <%= render(:partial => "cart" , :object => cart) %>

It doesn’t work, because in the code you do :

<% hidden_div_if(cart.items.empty?, :id => "cart") do %> <%= render(:partial => "cart", :object => cart) %>
<% end %>

While this does mean that you have to edit three files, the
changes are very straightforward

should read four files rather than 3

In the third sentence of the page, there’s a missing character: “You’ll find a
table called schema_mirations that it uses…”

Well, now I paid to be a beta tester…

this command
sqlite3 db/development.sqlite3 “select version from schema_migrations”
does not work on my system (Windows, InstantRails 2.0). The answer is:
SQL error: no such table: schema_migrations

(sam says: schema_migrations is new in Rails 2.1)

protect_from_forgery :only => [:create, :update, :destroy]

Should sort it. Books needs to mention this.

You state in the line below the heading “Autogenerating the XML” that in previous examples we used the template to create the XML. However, in the example at the end of page 185 and at the top of 186 (in the PDF), you have the same code for generating the XML as you do in this example namely “:xml => @product.to_xml(:include => :orders)”

On this page, for the first time, we see the following code:
self.salt = self.object_id.to_s + rand.to_s

With this explanation:
“Without the self., Ruby would have thought we were assigning to a local variable, and our code would have no effect.”

Well, I can understand why “self.salt” instead of just “salt” but not the reason for the sudden change to “self.salt” from “@salt”. I believe that an explanation is necessary, especially because the User class doesn’t have a “salt=” accessor.

I like the yellow arrows showing what new code to add to previous code. I think it should belong on page 102 to illustrate to the reader to add this code:

<%= button_to “Add to Cart” , :action => :add_to_cart, :id => product %>

To quote Chuck Hinson:

"#31714: The statement on pg 105

‘The next time you restart your application (stopping and starting script/server), it will store its session data in the database. Why not do that now?’

suggests (to me at least) that my application will now work properly with cookies. Yet restarting the server as suggested and then trying to view a page in my browser results in an error. … it would be nice to include an indication that while a restart is necessary, there’s still one more thing to do before the cookie stuff works properly."

I agree, although now it’s on Page 107.

This paragraph on page 109 would go better here so to warn developers of the effects of writing code below the “protected” line.

From page 109:
“Be careful as you add methods to this controller as we work further on the cart—if you add them after the private declaration, they’ll be invisible outside the class. New actions must go before the private line.”

The listing of the dynamic fixture ‘test/fixtures/performance/products.yml’ appears to be incorrect.

The lead in to the listing reads ‘To create all these products, let’s use a dynamic fixture.’ What follows that appears to be a very non dynamic, very static fixture.

#START:ruby
ruby_book:
id: 1
title: Programming Ruby
description: Dummy description
price: 1234
image_url: ruby.png
#END:ruby

rails_book:
id: 2
title: Agile Web Development with Rails
description: Dummy description
price: 2345
image_url: rails.png

In addition to the listing in the book being incorrect, the accompanying online version is also static. (/titles/rails3/code/depot_r/test/fixtures/performance/products.yml).

There appears to be no difference between this fixture and the static fixture introduced on page 199 (titles/rails3/code/depot_r/test/fixtures/products.yml).

The second edition of the book contains the following listing in the same spot. (titles/rails2/code/depot_r/test/fixtures/performance/products.yml)

<% 1.upto(1000) do |i| %>
product_<%= i %>:
id: <%= i %>
title: Product Number <%= i %>
description: My description
image_url: product.gif
price: 1234
<% end %>

Happy to stand corrected if I’ve missed something obvious.

After the db:migration, there are 4 files to edit not 3

The additions for new.html.erb look wrong - not consistent with others (no in others)

The previous page states that we “clear out our database”; then we create three assets. Yet, when we query the catalog_entries table, we find four entries; there should only be three.

#5 “substituting in values”

The “in” is redundant. Recommend using “substituting values.”

in place editing has been moved to a plugin.

The migration code contained this assignment to image_url:

:image_url => ‘/images/svn.jpg’ ,

when I run the example, the generated html did not render an image. The source reveals that the images directory is doubled up:

you have to edit three files

should read

you have to edit four files

When you start talking about validates_presence_of, it would be nice, in the margin, to remind the reader that this is an example of a declaration, and link to appendix A.4, on p. 651, where this concept is defined.

I looked at it and was momentarily puzzled, coming from a C/C/Java background, as to what, if anything, the statement would do.

When my version of Rails 2.1.0 (installed from RubyGems) automatically creates test files, the first line reads:

require ‘test_helper’

Rather than

require File.dirname(FILE) + ‘/../test_helper’

Some other errata reporters have the same require statement (see, e.g., #32390).

Having the version of require without the context of FILE breaks the attempt to run product_test.rb as a single unit test using the call on page 194 of:

ruby test/unit/product_test.rb

To work with both versions, it would be smart to include the test directory explicitly:

ruby -Itest test/unit/product_test.rb

Hi,

As I really don’t know the answer to this question, I am not certain how relevant or desireable this would be but I would like to see a section on database modelling tools.

I am in the process of writing a spec for a new Rails app and it would be nice to have a section that briefly covered database modelling tools, that may fit nicely with the Rails environment.

I am at the point where I am writing a database Schema specification and would love to know if there is anything that would be advantageous to use for this process, and if not then it may be worth a mention that this is the case.

I appreciate that rails development is all about iteration and not so much planning but I have a major application that needs a visual representation of the database and or model and really needs plannoing in advance and it might fit in nicely with the Rails Is Agile 1.1 section.

I have got to the stage where this project already involves 20 tables and is growing every day and as I am my own customer as such it would be really good. Honest :-)

Just a thought

James

(Dave says: That’s really outside the scope of this book)

In the function test_two_people_buying, the function check_for_order is used, but it is neither explained nor defined anywhere until page 688 (in the Source Code appendix).

A.8 Regular Expressions

When it comes to updating this part of the book could you please explain the code supplied for the depot application on page 88

%r{\\.(gif|jpg|png)$}i particularly the $ and the i bits.

It would be most useful to understand exactly how this works.

Thanks

When you write session :session_key => ‘_depot_session_id’, I suggest you remind readers that in this context, session is a declaration, and not to be confused with the session hash mentioned just mentioned.

There are no green arrows next to the new coded added to the _cart partial at the bottom of pg 135

Latest version of ruby does not support text_field_with_auto_complete

After adding the cart and cookies to the application, I started getting “ActionController::InvalidAuthenticityToken” error messages whenever I tried to add a book to the cart. I resolved this by adding “protect_from_forgery :only => [:create, :update, :destroy]” to the top of the store_controller.rb file.

Inside the gray box, item #2, edit.rhtml should be edit.html.erb.

produts should be products

I think “version 4” in the paragraph about both migrations being applied is left over from the old rails way of doing things - the new date-based format for migration names and DB schema versions renders it useless.

The output of erb/ex2.op sample outputs
Ho!
Ho!
Ho!
Merry Cristmas!

as expected (and no blank lines in between), not

Ho!

Ho!

Ho!

Merry Christmas!

as written in PDF.

(Sam says: the blank lines are in the source, not the output)

In the commandline examples: It is sometimes hard to distinguish what is to be typed and what is actually output from the commandline (especially with line wraps). I suggest to clear this that a different colour or bold is used for the command itself.

The table “schema_migrations” (or indeed “schema_mirations”) doesn’t exist in my (postgres) database; instead it is named “schema_info”.

(Sam says: typo fixed; schema_info changed to schema_migrations in Rails 2.1)

When I got to this page and looked at the nice login form in figure 11.1, then compared it to the completely plain one in my application, I realized that we never made an admin layout. Maybe that should have been obvious, but I missed it. Just copying the users layout and editing it seemed to do the trick.

The final parenthetical statement on the page reads as follows:

“(And how will it find that table? We told it where to look when we set up the development entry in config/database.yml.)”

Note that this only makes sense to someone who didn’t follow the advice on page 70 and decided to use MySQL as their development database, and thus, read through the next 4 pages on setting up the config/database.yml. It might be a good idea to point out (in a slightly different way than saying “…we set up…”) that the database.yml file tells the application where to look for the appropriate table for those readers who didn’t setup their db config file earlier, otherwise this statement could cause a bit of confusion for the reader.

flash[:notice] = “User #{user.name} deleted”

should be

flash[:notice] = “User #{@user.name} deleted”

I have a problem with the migrations, I use MySQL and I can’t create the foreign keys with :options parameter…, I can’t use it for nothing… t.integer number, :options => “unsigned” fails…(do the same without :options), I also probe with commas.

sorry for my english, I don’t know if is the site for this problem re-sorry

thanks

(Sam says: the example is only for sqlite)

maybe it should be better to change the code to:
respond_to do |wants|
wants.html {redirect_to_index}
wants.js
end

So people get used to doing that instead of doing request.xhr?

(Sam says: Unfortunately, I have been able to make such code return back js for non xhr requests, but adding “if request.xhr?” to the wants.js addresses that.)

At the end of section 15.3 (String Extensions) it says: “In a Rails application, these changes can go in the file environment.rb in the config directory.”
Wouldn’t it be a bit cleaner to put custom inflections into “config/initializers/inflections.rb”?

Attempting to run rake db:migrate from top level dir the first time gives error:
kant:~/ruby_rails/depot cpaule$ rake db:migrate
(in /Users/cpaule/ruby_rails/depot)
rake aborted!
no such file to load — sqlite3

(See full trace by running task with —trace)
The database.yml file looks like:
$ more config/database.yml

1. SQLite version 3.x
2. gem install sqlite3-ruby (not necessary on OS X Leopard)
   development:
   adapter: sqlite3
   database: db/development.sqlite3
   timeout: 5000

1. Warning: The database defined as “test” will be erased and
2. re-generated from your development database when you run “rake”.
3. Do not set this db to the same as development or production.
   test:
   adapter: sqlite3
   database: db/test.sqlite3
   timeout: 5000

production:
adapter: sqlite3
database: db/production.sqlite3
timeout: 5000
So it appears that Rails wants to see a previously created db file named sqlite3.
But the book on page 70 says "there are no steps required to create a database with Sqlite3.
I am running Rails 2.1.0 and ruby 1.8.6 (2007-03-13 patchlevel 0) [powerpc-darwin8.11.0] on a Mac OS X Version 10.4.11.
My email is cpaule@umn.edu
Thanks!

(Sam says: user apparently didn’t install sqlite3-ruby as described on page 36)

In the section titled “Version Control”, second paragraph, “It’s simple way …” should be “It’s a simple way …”

When trying to create my own rails API documentation as descibed at the top of this page, I got an error when executing the command “rake rails:freeze:gems” of:

uninitialized constant Gem::GemRunner

I found that I needed to edit the file /usr/lib/ruby/gems/1.8/gems/rails-1.2.6/lib/tasks/framework.rake and add the line “require ‘rubygems/gem_runner’” as follows:

deps = %w(actionpack activerecord actionmailer activesupport actionwebservice)
require ‘rubygems’
require ‘rubygems/gem_runner’
Gem.manage_gems

(Dave says: This seems like a problem with your installation. I’d ask on the Rails list)

Its “titled multiple model in form”
The example is regarding new and creating models.

Product.transaction do
details.product = product
product.save! details.save!
redirect_to :action => :show, :id => @product
end

Here in the above code, one cant have details.product = product without saving the product first. The code should be as follows:
Product.transaction do
product.save! details.product = product details.save!
redirect_to :action => :show, :id => @product
end
I was trying it out on one of my project and took me an hour to figure it out. Hope this error will be corrected in the next edition of the e-book.

I am trying this on a linux system that uses postgresql rather than sqlite3. When trying to perform the say/hello action to get the Missing Template screen, I got several errors about database. To get around it, I had to modify the config/database.yml file and issue postgres commands to create the database. Perhaps a Note here would help those who aren’t using exactly the same setup would be useful.

(Sam says: A database should not be required for /say/hello. This is covered in chapter 6.)

I have been following along doing everything on a linux system with postgresql. Everything has worked OK and I have made the appropriate changes for postgresql. Now that I try to use the browser to go to products I get the following Routing Error:

No route matches “/products” with {:method=>:get}

I am not quite sure how to proceed at this point.

(Sam says: Added a suggestion to restart the server)

Paginate has been removed from the Rails 2.1 and as such it is recommended to you will_paginate as plugin. I tried the will_paginate and it was cool.

It would be worth mentioning about it in the book as I toiled with the method paginate not found error on rails 2.1. I had to figure it out that the paginate support has been taken out of core rails 2.1

Hope this is useful.

The last sentence of 19.1 (Validation) under subsection “Validation Error Messages” states: “You’ll probably want to put this in the environment.rb file in your application’s config directory.”
Wouldn’t something like “config/initializers/active_record_errors.rb” be a cleaner location to put reconfigured error messages?

For those out there trying to follow along with the book but use postgresql rather than sqlite3, then the migration file that creates line_itmes needs to look like this (took me several hours to figure this out):

class CreateLineItems < ActiveRecord::Migration
def self.up
create_table :line_items do |t|
t.integer :product_id, :null => false, :options =>
“CONSTRAINT line_items_product_id_fkey FOREIGN KEY (product_id) REFERENCES products(id)”
t.integer :order_id, :null => false, :options =>
“CONSTRAINT line_items_order_id_fkey FOREIGN KEY (order_id) REFERENCES orders(id)”
t.integer :quantity, :null => false
t.decimal :total_price, :null => false, :precision => 8, :scale => 2

t.timestamps
end
end

def self.down
drop_table :line_items
end
end

Please emphasize this code: “def self.find_products_for_sale
find(:all, :order => ”title“)
end” has to go before the validation lines to avoid an error…

The book suggests the line:

The listing of depot/app/views/layouts/store.html.erb has arrows highlighting several lines to be added to the original version of this file.

Unfortunately, two added lines are not highlighted in this way:

<% if @cart %>

and its corresponding end

<% end %>

The rake command should be < rake db:schema_migrations >.

rhtml, rjs and rxml aren’t used anymore.

directory = File.dirname(FILE)
should be
directory = File.join(File.dirname(FILE), “dev_data”)

Having the before_filter applied before the creation of a session can result in being redirected to login twice in a row when the suggested improvements to save the original request URI are made.

class ApplicationController < ActionController::Base
before_filter :authorize, :except => :login
session :session_key => ‘_depot_session_id’

“schema_mirations”

The full code of user.rb starts off with require ‘digest/sha1’ on the first line, however it’s not mentioned in the instructions leading up to this. It might be a little confusing for some.

(Sam says: Even in B1.3, it was mentioned on page 160)

StoreController should be StoresController. Also, store_controller.rb should be stores_controller.rb.

(Same says: correct as is: http://tinyurl.com/6e3pe8)

class name is wrong, says
“class ApplicationController < ActionController::Base”
instead of
“class AdminController < ActionController::Base”

Also, the code sample is repeated. I think it was supposed to be broken up into smaller chunks?

Get an error when installing rake, isn’t it just part of rails now anyway?

cap —apply-to /local/project/path [applicationname]
isn’t that now just:
capify .

There hasn’t been a controller generated for Orders yet.

It might be a good idea to remind readers to add your password to the database.yml file here.

the command: sudo gem update sqlite3-ruby does not seem to do an update (system: ubuntu) if sqlite3-ruby is not already installed?? maybe this should be specified in some way?
thanks,brice

In Section 6.1 we initiate the “depot” application using the default “database.yml” containing sqlite3 adapter.

Afterwards we reset it and use mysql adapter, and we execute some queries in mysql environment to test our configuration.

At the beginning of page 77 (pdf) we are asked to check the existence of table “schema_migrations” in sqlite3 environment.

Is there a flow erratum?

(Sam says: missed the second paragraph on page 70 which says to skip to page 74)

validate :price_must_be_at_least_a_cent
protected
def price_must_be_at_least_a_cent
errors.add(:price, ‘should be at least 0.01’) if price.nil? ||
price < 0.01
end

Now shouldn’t the if-clause if price.nil? || price < 0.01 have :price instead of price?, note the colon.

(Sam says: The syntax is correct as is. There is a difference between the name of the variable (which is used to match the field on the form) and the value of the variable (which is compared against nil and 0.01))

On the very last row of regular text: ‘“po,”’ should read ‘“po”,’

“We saw an example of multiple sessions in the test on page 225.” The example actually starts on 226.

When creating the “person” model, the right command to do the database migration isn’t “rake db:migrate” instead of “ruby db:migrate”?

“to-level” should probably be “top-level” in the following sentence:

Every new Rails application already has one example of this—a declaration such as the following is added to the to-level application controller to set an application-specific cookie name for session data.

Rails 2.1 now uses Mongrel by default instead of WEBrick when running ‘ruby script/server’.

I’ve apparently run into a common issue in learning helpers. I got the idea from the text that I could use a helper function inside my controller by using the helper: method. It took me several more readings to realize that helper functions are only available in views.

I did find this in a rails faq:

Is it possible to call a helper from the controller?

No, but you can move the helper method into the controller and add helper_method :method_name after the definition to make it available to the view again.

So, suggestion: call this out and make it clear - you can’t use helper functions from within controllers.

Figure 9.1 shows the ‘empty cart’ button below the cart in the sidebar, but when you refer to figure 9.1 on p 128, we haven’t changed the code to move the button yet.

(Sam says: Page 127, the button is in the partial)

For reasons I can’t understand at all, with Rails 2.1, my shopping cart in the sidebar was not getting refreshed until I removed the
respond_to { |format| format.js }

line from the end of my StoreController.add_to_cart method. After that, my add_to_cart.js.rjs worked fine. ?

In the StoreController.add_to_cart revision that degrades gracefully in the absence of Javascript, you did not include the arrows showing the new code.

“It always returns true for nil and false, and
it always returns false for numbers and for true.”

Huh?

“However, Rails 1.2 includes a fix for this.”

Should be

“However, Rails 1.2 included a fix for this.”

If the reader has done the ‘playtime’ suggestion that involves making the clicking the book images perform the add_to_cart action then they won’t work when the button is ajaxified. It would be worth mentioning that in index.html.erb the line

<%= link_to image_tag(product.image_url), {:action => :add_to_cart, :id => product} %>

should be replaced by

<% form_remote_tag :url => {:action => :add_to_cart, :id => product } do %>
<%= image_submit_tag(product.image_url, :class => “click_img”) %>
<% end %>

Also the line in depot.css that reads

#store .entry img {

should be changed to

#store .entry .click_img {

(Sam says: Fixed by adding another playtime activity in chapter 9)

On this and several pages before it, you describe the use of <->, instead of <%= %>. It causes server errors.

(Sam says: I don’t see this in my copy)

As you go into the definition of the method password, it would be worth noting that these are public defs, as opposed to the private defs we just did with the salt.

The first argument to the method post_via_redirect is:
“/store/save_order”

However, when I use that arg, the following error occurs:
“Expected response to be a <:success>, but was <500>
<”Missing template store/save_order.html.erb in view path …"

I corrected this by replacing the second argument with:
“/store/checkout”

The first argument to the method post_via_redirect is:
“/store/save_order”

It should be:
“/store/checkout”

I saw you were releasing another version early next week, and thought I should get this in.

In the fifth paragraph on the page:
“files woulc match” should be “files would match”

Last paragraph prior to “Options for Creating Tables” section.
“creates the both the” should be “creates both the”

Our demonstration of routes did not specify starting in any particular rails application directory.

The second thing we are asked to do - “puts rs.routes” - comes up with only two actions, which indicates that we must be in a default rails application. However, I tried using a default rails app, and then trying
rs.recognize_path “/store”
resulted in an error:
ActionController::RoutingError: No route matches “/store” with {}

Trying the same thing in the depot application worked fine, but then I got far more output to the “puts rs.routes”.

I suggest giving a little more guidance - tell folks to use the depot application, and if you do, tell them the rs.routes output they will get is much more than what you’ve shown.

On this page (and page 26) we discuss “the model” and “the view”. It seems more natural here to say “a model” and “a view” (as it does in at least one place here).

3rd paragraph: “This triumvirate—the model, view, and controller—together form an architecture known as MVC.”

How about: “This triumvirate—the model, view, and controller—together FORMS an architecture known as MVC.”

Bottom: “That directory contains the template files for the controller’s views.”

How about: “That directory WILL CONTAIN the template files for the controller’s views.”

(after all, at this point the directory is empty)

Footnote 5: “While you can continue to use rhtml extension…”
How about: “While you can continue to use the .rhtml extension…”

Footnote 5: “Information on how this can be accomplished can be found in on page 243.”

How about something like: “Information on how this can be accomplished can be found in Chapter 14 on page 243.”

(Sam says: Link removed; generated applications now do this by default)

2nd-to-last paragraph: “In our case, we’ll also use a new action method to handle the page (although that isn’t always the case, as we’ll see later in the book).”

How about: “In our case, we’ll also use a new action method to handle the page (although that NEEDN’T always BE the case, as we’ll see later in the book).”

3rd paragraph: “In a real application we might want to make these proper buttons, but for now we’ll just use hyperlinks.”

How about removing this sentence? (What’s wrong with hyperlinks anyway?)

Paragraph beginning with “One choice would be to stores session …”; I believe “stores” should be “store”.

I had to do a “mongrel_rails restart” after adding admin controller. Was getting “No controller responded to.. etc..” when trying to view /admin. Guessing there’s some caching problem with mongrel — might want to add a note that a restart may be required.

(Sam says: this is a non-deterministic bug in Rails. DHH is aware)

As far as I can discern from the examples, when passing a cart collection to _cart_item.html.erb the table is not created if the collection is empty. Therefore, when the add_to_cart.js.rjs is triggered it references the id ‘current_item’, but with a new empty cart this table is not created before the rjs is triggered and ends up with a ‘null value’ rjs error. How I see to fix the problem is in the rjs template:

page[:current_item].visual_effect :highlight, :startcolor => “#88ff88”, :endcolor => ‘#114411’ unless @cart.total_items == 1

This still solves the problem since the blind_down effect is done on the creation of a new cart and still catches the eye, and any subsequent additions to the cart causes the flash effect

(Sam says: The add to cart response should only be produced after a product has been added to the cart. This would imply that the collection should never be empty.)

I wonder why there is not an <h1….> tag for the Show action. After applying the css styling this possible inconsistency is evident because the dashed blue line is visible for Index and Edit actions. Anyway, it is not a big deal but it did alert me for a few seconds and I started getting confused (I’m a little slow!).

Perhaps should be made clear that accessor methods should not be private (and why) or this will cause some confusion shortly when attempting to use the model

The examples given for the number_to_currency helper method show corresponding example outputs. However, the chosen indicator, “=>”, which directs readers attention to the ouput, is the same as one of the operands in the code. Essentially there are too many “=>” symbols.

2nd bullet item: “Check that the user name and password in database.yml match the one you created on page 70.”

How about replacing “one” with something like “user”, “account”, or “ones”.

I’m finding the migration numbers in the depot example somewhat confusing. With endings like “000001”, “000002”, etc, they appear to be sequence numbers appended to a date, until we get to the footnote on page 79 that says they are ‘clearly fictitious’.

Also, from the footnote on this page (76) it appears we can still refer to them by version number (‘VERSION=0’) when moving backward or forward to a specific migration.

And on top of all that page 89 says: “Don’t be concerned if the timestamp of the file you downloaded is before others that you have already migrated as Rails knows which migrations have been completed and which ones have yet to be done.”

This has me wondering, how does Rails order the migrations? And shouldn’t that ordering be obvious to any developer who looks at the project? If the ordering of the timestamps can’t be relied upon, how do we keep it straight?

(Sam says: Moved footnote to first occurrence. At the place where the footnote was, the text indicates that the timestamps are used for ordering.)

Top: “The speed with which you can turn words into a working application with Rails means that you’re never far from being able to let the client play with their application. These short feedback cycles mean that both you and the client get to understand the real application sooner, and you waste far less time reworking your application.”

That last “application” at the end sounds like one too many uses of the word after everything that precedes it!

Bottom: “…the same reloading that caused Rails to notice the earlier change to our schema also means it will always use the latest version of our code, too.”

The “too” at the end feels redundant after the “also”.

The superscript “11” seems out of place, since the footnote has nothing to do with the price being greater than zero. It might go better with one of the earlier paragraphs.

Footnote 12: “Checking the number is at least one cent ensures…”
How about: “Checking THAT the number is at least one cent ensures…”

Url validation part makes me wonder if there is a built-in validator in Rails that checks whether a string is a legal url…

(Sam says: There is a plugin http://github.com/RISCfuture/validates_url/tree/master)

Top half: the “data-only migration” in the tutorial has me wondering, why mix test data with the schema definitions? This seems kindof ugly, and surely we don’t want this test data inserted into our production database. What is the benefit of setting up our test data this way?

Line numbers might be helpful in the code sample, since the text afterward refers to the ‘eighth line’.

After rake:db:sessions:clear when using MySQL the app breaks.

Mysql::Error: Lost connection to MySQL server during query: SELECT * FROM `sessions`

It’s fixed by restarting the app server.

(Sam says: this is a non-deterministic bug in Rails. DHH is aware)

it currently says: “:allow_blank boolean If true, nil attributes are considered valid ”

but it should be
“:allow_blank boolean If true, blank attributes are considered valid”

the hidden_div_if helper method causes a bunch of problems if you simply switch the default layout to “store”. for one, we placed the helper in StoreController so this raises a NoMethodError when attempting to access the Users controller. a fix is placing the helper method in the application helper. second, we still have an issue with this since the way that it is written asks the object @cart if it is empty. this will give a nil object error since if there is no cart and the user is coming directly to /users the cart is never created. the fix is to modify the logic a bit to check for a cart and then check to see if the cart is empty in the store layout:

<% if cart %> <% hidden_div_if(cart.items.empty?, :id => “cart”) do %>
\t<%= render(:partial => “cart”, :object => @cart) %>
<% end %>
<% end %>

(Sam says: That’s exactly what pages 173-174 describe, complete with triangles pointing to the lines you need to add.)

apologies, the next page (174) fixes the hidden_div_if problem as I described, and checking for @cart ‘should’ make sure that the hidden_div_if is never called but it’s still safer to move it into the application_helper. also, the book suggests navigating to the admin controller right after making the change to the ‘store’ layout anyway which will cause errors immediately

we didn’t create an ‘orders’ controller, so you may not want to have that in the html currently…

Top: “The fact that ProductTest is a subclass of the Test::Unit::TestCase class tells us that Rails generates tests based on the Test::Unit framework…”

In the example right before this text, ProuctTest is obviously a subclass of ActiveSupport::TestCase, but I’m not clear on what the relationship is between ActiveSupport::TestCase and Test::Unit::TestCase…

4th paragraph: “there’s a short cut.”

Should “short cut” be one word?

3rd paragraph: “Next, we’ll test that we’re validating the image URL ends with one of .git, .jpg, or .png.”

How about: “Next, we’ll test that we’re validating THAT the image URL ends with one of .git, .jpg, or .png.”

End of 2nd paragraph of “David Says…” sidebar: “…then paying his valid_credit_card…”

How about: “…then paying WITH his valid_credit_card…”

Top: “And, in fact, Rails is already doing this, (convention over configuration for the win!), but you can control…”

How about: “And, in fact, Rails is already doing this (convention over configuration for the win!), but you can control…” (remove 3rd comma)

Right after “Using Fixture Data” heading: “Now we know how to get fixture data into the database, we need to find ways of using it in our tests.”

How about: “Now THAT we know how to get fixture data into the database, we need to find ways of using it in our tests.”

Footnote 2: “If you are using a database other my SQLite3…”
How about: “If you are using a database other THAN SQLite3…”

Also: “If so, you will need to create such a database.”
How about changing “such a” to “a” or “your”

2nd-to-last paragraph: “If I want to run all my tests, I use this rake task.” (referring to ‘rake test:units’)
How about: “If I want to run all my UNIT tests…”

The test we write manually (CartTest) is inheriting from Test::Unit::TestCase while the generated ProductTest we saw earlier inherited from Test::Unit::TestCase. Wondering what the distinction is, and if there’s a reason to choose one over the other. (I see this was submitted for build 1.0 for this page number, but I think my question is a little different).

2nd-to-last paragraph: “Notice that we’ve included the existing products fixture into this test.”

Should this be, “…into this test CASE.”? Or is it correct either way?

othermost should probably have been outermost

“Now we’re ready to test the index action again. We have to remember to add
the fixtures directive to the login controller test class.”
- there should be admin controller mentioned here.

The method “url_for” can be used on the console. Simply call it on the “app” object, which the console creates for you. Eg:

?> app.url_for :controller => :store, :action => :display, :id => 123

The phrase “as described Section 20.2” should be “as described in Section 20.2”.

Using Rails v2.1.0 and running the “generate scaffold” command listed on the previous page, the last line of output is not “route map.resources :articles”. That line is located near the middle of the output.

In the test_two_people_buying a hash MIKES_DETAILS is used but not previously mentioned in the text. Of course the appropriate listing can be found in the depot_r/test/integration/dsl_user_stories_test.rb but it would be better to show it explicitly.

The phrase “the one that ends blog/add_comment” should be “the one that ends in blog/add_comment”.

The text box labelled “edit.html.rb” should actually be “edit.html.erb” (IE: “.erb”).

The phrase “a name ending _tag” should be “a name ending with _tag”.

To be consistent, “we say how on page 494” should be “we saw how …”.

“Your terminal man not display the Unicode character correctly, but you can verify that the length is correct.”
I believe that: “Your terminal may not…” would be better.

The definition of the self.encrypted_password method contains this line:

string_to_hash = password + “wibble” + salt

No explanation of what purpose “wibble” serves here is given. It would typically not be a big deal, except the password generation code is already fairly circuitous for a beginner, so it’s confusing to use both both a salt and this odd extra tacked on.

Paragraph on @response: “Every time we send a request back to a browser, Rails is populating a response object behind the scenes.”

Seems like this would be clearer as: “Every time we send a RESPONSE back to a browser, Rails is populating a response object behind the scenes.”

3rd paragraph: “The get method, a convenience method loaded by the test helper, simulates a web request (think HTTP GET) to the index action of the AdminController and captures the response.”

It seems like we’re not exercising (or otherwise following) the RESTful routes here. After all, the index action is implicitly the result of a GET (a POST to the same url would be routed to the create action, right?)

Page 222 talks about calling get with a url parameter for integration tests… I’m wondering if this can also be done for controller tests.

Last paragraph: “…that’s the way we designed the AdminController. It uses a before filter to intercept calls to actions that aren’t available to users without an administrative login.”

The example immediately below shows this before filter in the ApplicationController, not the AdminController (which makes sense of course, otherwise it wouldn’t be used application-wide).

Bottom:
We’ll call the user “dave.”

I think the period should be outside the quotes…

(Dave says: in this context, it’s correct where it is.)

Last paragraph: “Parameter two is an empty hash—this represents the HTTP headers to be passed to the action.”

Headers? Page 210 says it is ‘parameters’, not headers. I’m assuming ‘parameters’ means either urlencoded key-value pairs such as “?key1=value1&key2=value2…” or the type that can go into the request body (in the case of POST or PUT), and that ‘headers’ means HTTP headers such as ‘Content-Type:’ and ’Accept-Encoding:".

Of course, this makes me wonder: shouldn’t we be able to specify both? If this argument is for parameters, how do we specify headers (and vice versa)?

Also, this says ‘parameters’ in my copy of the 2nd edition. Has something in Rails changed here?

Also, headers are in the integration testing versions of these request methods! On page 229 they are described with arguments shown for both params and headers.

3rd-to-last paragraph: “Before we go much further, let’s look at some of the Rails-specific conveniences for teting controllers.”

Seems like we’ve been doing this already. Should “let’s look at” be “let’s take a closer look at”?

“You’ll often want to post form data within a function test.”
How about:
“You’ll often want to post form data within a FUNCTIONAL test.”

“To do this, you’ll need to know that the data is returned as a hash nested inside the params hash.”
How about:
“To do this, you’ll need to know that the data is SUBMITTED as a hash nested inside the params hash.”

“So, to post a form to the edit action containing User model data…”

Wouldn’t a post go to a create or update action (not edit)?

“You can simulate an xml_http_request using…”

How about a little mention about what is different about an ‘xml’ http request from the server’s perspective (is it simply a value in one of the HTTP headers?)

Also, in the description of xml_http_request immediately below, it says: “The first parameter will be :post or :get.” Is this a restriction on the xml_http_request method being described here? Why have such a restriction? (I think browsers are moving away from this…)

On page 229, there’s a description of a seemingly more permissive version of xml_http_request for integration tests, which accepts HEAD, GET, POST, PUT or DELETE.

Under “Assertions” heading: “In addition to the standard assertions we listed back on page 203, additional function test assertions…”

The use of both “In addition” and “additional” sounds redundant.

Description for assert_dom_(not_?)equal:
“Because the assertion compares a normalized version of both strings it is fragile in the face of application changes.”
This seems a little fragile, logically. How about:
“Because the assertion compares the complete (normalized) contents of both strings it is fragile in the face of application changes.”
It’s fragile because it’s comparing the complete contents, not because it’s comparing the normalized contents.

assert_redirected_to description:
“Asserts that the redirection options passed in match those of the redirect called in the last action.”
How about:
“Asserts that the redirection options passed in match those of the redirect RETURNED BY the last action.”

(I’m assuming that at this point, we’ve gotten back one of the 3xx redirect responses, and haven’t actually followed it yet…)

follow_redirect description, at bottom:
“If the preceding action generated a redirect, this method follows it by issuing a get request.”

How about adding, “otherwise it does nothing” (if that’s the case)

example in with_routing description:
Looks like the assert_equal call is indented too much (and doesn’t match its closing ‘end’)

Bottom paragraph describing HTML::Selector library:
It looks like this library is described as working only for HTML and not XML in general. I’m curious why that is, since everything about it seems applicable to xml.

Every item in the “Chained Selectors” section shows “sel_1” (with no ‘s’) but ‘sel_2s’ (with an ‘s’). The ‘s’ seems out of place in each case.

Description for “sel_1 + sel_2s” says “immediately follow”. The comment right under it just says “follow”. How about changing that to “immediately follow”.

Last item on page, “sel1, sel_2s” seems a little out of place lumped in with “chained slectors”, which are described as describing “relationships between elements”. There’s no relationship between elements that is described by this selector.

Every one of the pseudoclasses describe on this page and the next (219), except the first two, describe the selector as if it were a strict parent-child relationship and not a looser ancestor-descendant relationship, despite the fact that the two simple selectors in each case are separated by a space (which pg 217 says means ‘ancestor’) and not a ‘>’ (which pg 217 says means ‘parent’).

Do we need ’>’s in all these examples to make the descriptions match? Or do pseudoclasses operate a little differently? There’s no mention here either way.

If the ‘>’ is needed, that would also affect the example on pages 215 and 221.

:nth-child description:
“Contrast this with nth-of-type, described later.”

This is a bit confusing - what are we supposed to be contrasting here? Since this comes right after the mention that “n counts from 1”, it could be suggesting that in nth-of-type, n doesn’t count from 1. Consider removing this, and/or adding “counting from 1” to the description for nth-of-type.

only-of-type description (bottom):
“Selects all sel_2 nodes that are the only children of sel_1 nodes.”
Add: “of type sel_2”

Near bottom: “Finally, you can invert the sense of any selector.”
How about “invert the meaning” or simply “invert”

Bottom line:
div :not(p) # all the non-paragraph nodes of all divs

How about “all the non-paragraph child nodes” or “all the non-paragraph descendant nodes” (depending on which is correct)

Top: “Now we know how to select nodes in the response, let’s see…”
How about: “Now THAT we know how to select nodes in the response, let’s see…”

About 3/4 down page:
:count => n Exactly n nodes must have been selected.

I’m wondering here: Can a range be passed to count as with the non-hash example of calling assert_select above:

assert_select “div#main div.entry”, 1..10 # one to 10 entries on a page

Near top:
assert_select “div#cart”

This seems unnecessary, since this would be covered by the other two asserts following this one.

Near bottom:
“When you create a model or controller, Rails creates the corresponding unit and functional tests.”

Wondering if this should be:
“When you create a model or controller, Rails creates the corresponding unit OR functional test.”

since a model would get a unit test and a controller would get a functional test, but neither would get both. (Not sure here)

Near bottom:
“Integration tests are not automatically created, however, so you’ll need to use a generator to create one.”
How about:
“Integration tests are not automatically created, however, but you can use a generator to create one.”

(since you don’t need to use a generator, and it doesn’t seem to be doing much anyway in this instance)

End of 3rd paragraph:
“The test helper method post_via_redirect generates the post request and then follows any redirects returned until a regular 200 response is returned.”

How about changing the ending to something more like:
“…follows any redirects returned until a NON_REDIRECT response is returned.”

In the regular_user code example, in the user.checks_out method, it takes the items in the ‘details’ hash that was passed in, and then uses them to rebuild an identical hash to pass as the :order parameter.

This could be replaced with “:order => details” as is done in the performance test example at the top of page 232.

Bottom of page: “For example, here’s a test that verifies that there’s no interaction between two users buying products at the same time.”

A little strong perhaps? I doubt this “verifies” that, but it certainly could uncover a problem along those lines…

Big paragraph near bottom:
“However, you can also create explicit sessions (using the open_session method) and invoke these methods on it directly.”

How about:
“However, you can also create explicit sessions (using the open_session method) and invoke these methods on THEM directly.”

Bottom:
“Integration test sessions have the following attributes.”
How about something like:
“Integration test sessions have a collection of attributes.”

(they don’t ‘follow’ quite yet - not til after an example and another mention of ‘the list that follows’)

cookies description:
“A hash of the cookies.”
How about:
“A hash of the cookies kept by the simulated client.”

description for ‘host’:
“Set this value to the host name to be associated with the next request.”
How about:
“Set this value to the SERVER host name to be associated with the next request.” (as opposed to the client host name)

Same for the description for ‘remote_addr’ farther down:
“The IP address to be associated with the next request”
How about:
“The SERVER IP address to be associated with the next request”

(Sam says: Actually, remote_addr refers to the client)

description for ‘request’ and ‘response’ attributes:
Both of these are said to be “used by the last request.” How about “associated with the last request.”?

descriptions for ‘status’ and ‘status_message’ attributes:
Both say “status code of the last request”.

How about changing ‘request’ to ‘response’ in both of these, since it’s the response that has the status code.

Interesting that the follow_redirect!() function has a ‘!’ in its name, while the (.+)_via_redirect functions farther down, which do the same thing (I think*) don’t have a ‘!’ in their names.

*The description for follow_redirect!() has no mention of following possible subsequent redirects, as the other methods I mentioned do. Does this method follow subsequent redirects or not?

xml_http_request description:
“Performs a HEAD, GET, POST, or PUT, DELETE, or XML_HTTP request with the given parameters.”

How about:
“Performs a HEAD, GET, POST, PUT or DELETE, for an XmlHttpRequest using the given HTTP method.”

As it stands, the first ‘or’ seems out of place, the second I’m guessing should be ‘for’, and the ‘XML_HTTP’ seems to belong in there as one more HTTP method name.

It would also be nice to have a brief mention of what makes an ‘xml’ http request different (unless I’m mistaken, it’s just an extra header that browsers happen to set when they’re acting on a javascript call).

description for (.+)_via_redirect(path, args={}) functions:
What’s the ‘args’ parameter here? Is it the same thing as the ‘params’ parameter in the methods listed right above these? Also, why can you pass headers to that set of functions but not to this set?

Description for ‘open_session’:
“Creates a new session object. If a block is given, pass the session to the block; otherwise return it.”

Isn’t the session returned either way (not ‘otherwise’)? The example on page 226 seems to depend on this…

‘reset!()’ description (at bottom):
I’m wondering here how this resets the session. For example, does it clear the session variable used by controllers?

Start of 6th paragraph: “In the old days (yes, that was last year)”

I assume this is from the 2nd (or first!) edition? What year are we talking about?

3rd paragraph: “Now we need to write a performance test. Again, we want to keep them separate”

How about: “…keey THESE separate”

Example showing order_speed_test.rb:

In the first two ‘require’ lines, do we need all this rather than “require ‘test_helper’”? as on page 205? In other words, does the new convention described back there not apply to performance tests?

And these next two lines…

1. Reraise errors caught by the controller.
   class StoreController; def rescue_action(e) raise e end; end

I’m a little lost there - is this bit of code changing anything? It looks like anything it catches and reraises would have bubbled down in the same way if this code weren’t there. I don’t see any description of what purpose this code is serving.

Also, is there a reason the OrderSpeedTest class is inheriting from Test::Unit::TestCase and not ActionController::TestCase?

In the first paragraph, we have ‘script/profiler’ and ‘script/benchmarker’ but in the examples that follow we have ‘script/performance/profiler’ and ‘script/performance/benchmarker’

Near bottom: “This subdirectory holds all the stub files that are used in the test environment. If we wanted to stub out files while in the development environment, we’d have put our stubs in the directory test/mocks/development.”

I’m wondering at this point: does this mean the contents of test/mocks/development are checked before other locations while in development mode? And the contents of test/mocks/production will take precedence over other loacations while in production mode?

Speaking of how the ‘mocks’ we’re using are really stubs, how about including mocks into the narrative rather than just mentioning them as a couple footnotes.

Listing of foreign_key helper is partly coloured green (after hash signs) what is a bit misleading because these are not comments in fact.

(Dave says: Our syntax highlighter can get confused—there’s no much we can do about it without including a full Ruby parser, which just isn’t practical)

It would be useful to change the schema version in the listing showing Schema Information added by “rake annotate_models” because currently it looks a bit different (i.e. 20080816214522).

The :dependent option tells Active Record what to do to child rows when you
destroy a row in the parent table. It has five possible values.

should be

The :dependent option tells Active Record what to do to with child rows when you
destroy a row in the parent table. It has five possible values.

Regarding this: “You’ll need to research on
your own how to make a migration run in production—I’d rather not say here.”

It would be helpful to point to specific places to research this, as it seems to be critical piece to understand for deployment. One reference is the “Deploying Rails Applications” book, but I’d like to see other references as well.

Adding “:xml => @product.to_xml(:include => :orders)” to format.xml will generate more verbose output (basically, a record dump) than is indicated in the book. The books shows what the output would be if the “:xml” argument was left out.

The phrase “which we’ll discuss Section 23.1” should be “… discuss in Section 23.1” .

depot> ls -p app/helpers
application_helper.rb products_helper.rb store_helper.rb

The Rails generators automatically created a helper file for each of our controllers (produts and store).

NOTE: “(PRODUTS and store)”

(Sam says: fixed in Beta 1.4)

7 lines from the bottom, “overide” should be “override”

Bottom: “the add_to_cart doesn’t exist yet”
How about: “the add_to_cart ACTION doesn’t exist yet”

List item 5: “Create a simple stylesheet.”

We created the stylesheet in the previous chapter. In this chapter, we just updated it.

List item 5: “Create a simple stylesheet.”

We created the stylesheet in the previous chapter. In this chapter, we just updated it. (ps - and that was part of step 7 (add a button…))

“So, to post a form to the edit action containing User model data…”

but the example (post :create) shows a post to the create action, not the edit action

You can fix this by deleting a the h that appears

The question is regarding the line in the index.html.erb code:

<% for product in @products -%>

First, you might want to explain with the minus is doing in there (removing whitespace apparently). And second, isn’t there supposed to be one at the opening of the line, e.g.:

<- for product in @products ->

I’m new to all of this, so I had to google what the erb tags are, and what they do.

(Sam says: Covered in section 4.2 on page 54)

The paragraph that begins: “The extras parameter again contains…” pertains not to the previous example but the previous-1 example. It seems a bit out of place.

In explaining the process of finding a method for the action this line is the first “or-else” part. At best it is awkward. But I think it is also out of place. “If not, but the controller implements method_missing, that method
is called, passing in the action name as the first parameter and an empty
argument list as the second.”
The logic would be that a “method” is sought to match the “action”, if not “If no method can be called, the controller looks
for a template named after the current controller and action. If found, this
template is rendered directly.”
Now the text says that “an “Unknown
Action” error is generated“. My suspicion is that the ”method_missing" method is called with the action which does some logging and then raises the error.
Oh one strange note. My PDF copy does not have any footer on the bottom of page 439???

“hide_action” is only discussed on this page in a chapter on how the controller processes incoming action requests. The text mentions that it hides a public method that you don’t want to be private. The text indicates that this might occur for a method that is put in a library. There is sparse information about library modules in this text. The index points to “/lib” directory for a little assistance. But the above “hide_action” is not mentioned there.
IMHO “hide_action” should be discussed with “private” and “public” sections of a module/file as well as the “/lib” area.
And a couple of questions that all this raises: Does the controller parsing function try to find the “action” in a list of “public” actions that are not in the “/lib” directory? Do all methods in libraries need to be put after “hide_action”??

“See the documentation of ActionController::AbstractRequest for full details.” appears near the top of the page and after a code block (“class BlogController”). This code block seems quite unnecessary as it only references 2 of the 20 odd request attributes that have been mentioned in the previous 1.5 pages.

In the paragraph: “By convention, the template for action action of controller control will be in the
file app/views/control/action.type.xxx (where xxx is one of erb, builder, or rjs). The
app/views part of the name is the default. It may be overridden for an entire
application by setting
ActionController::Base.template_root =dir_path” —- control, action and xxx are described but nothing is said about “type”

Sorry about the previous post on method_missing. On 444 text explains that method_missing is only called if in development mode. The wording here is still awkward.

At the bottom ‘Note’ part, in item 3 at the first line ‘but it simply is a a method call which is passing’, there are 2 letters of ‘a’.

I use NetBeans 6.1 to try this depot application. The browser I used is Firefox 2.0. After I disable the browser’s javascript (in Firefox, Tools -> Options -> Uncheck ‘Enable JavaScript’), I didn’t see an error message even the cart is empty. But the highlight visual effect doesn’t work. Once javascript is enabled in the browser, everything is working.

(Sam says: The highlight visual effect requires JavaScript. I’m unclear what error message you were expecting to see when the cart is empty.)

It shows:
work> cd depot
depot> ls -p
I’m using the Ruby console window opened from Instant rails.
I get, “ls is not recognized ,,,” dir works ’so it makes me wonder whether I’m doing things correctly (which I find desperately important when I’m feeling my way around!).
Should it show, “.. or dir in Windows”?

closing div is missing

“Next, we have to tell Rails to use database storage for our application”
How about:
“Next, we have to tell Rails to use database storage for our application SESSIONS”

Maybe this isn’t the place to mention this, since here the book is quoting directly from Rails’ environment.rb file, but in the section that reads, “Use the database for sessions instead of the cookie-based default”, shouldn’t “cookie-based” be “filesystem-based”?

What we’re changing by uncommenting this setting is where the server is keeping its sessions. Either way, we’re using cookies to store session ids right?

2nd-to-last paragraph:
“cookies are stored on the browser by name.”
How about something like:
“cookies are stored on the browser by server hostname.”

On pages 116-118, exceptions are said several (I noticed seven) times to be ‘thrown’ rather than ‘raised’. Other languages use this terminology all the time, but the Pickaxe book makes a distinction between ‘raising’ exceptions (handled with ‘rescue’) and ‘throwing’ symbols (handled with ‘catch’).

I’m not sure if the two are interchangeable (eg you can rescue something that you throw, or catch something that you raise) but it would seem more consistent with the pickaxe book if these uses of ‘throw’ (or ‘throws’, ‘thrown’ etc) were changed to ‘raise’.

Footnote 6: “It’s often a good idea to keep a window open showing new lines as they are added to this file.”

I’ve noticed that when keeping a window open with a log as described here and another with the output from script/server, much of the same information shows up in both. I’m not clear on what the difference is between the two, or if one has a superset of the information in the other.

“We’ll also need a new CSS styling for the notice box.”
Should this be, “We’ll also need a new CSS styling RULE for the notice box.”?

In the “What We Just Did” bullet list:
“Using the flash to pass errors between actions”
How about:
“Using the flash to pass errors and other messages between actions”

2nd paragraph:
“We see whether our list of items already includes the product we’re adding; if it does, we bump the quantity, and otherwise we add a new CartItem.”

How about:
“WE’LL see whether our list of items already includes the product we’re adding; if it does, WE’LL bump the quantity, otherwise WE’LL add a new CartItem.”
(removed extra ‘and’)

Be careful as you add methods to this con-
troller as we work further on the product

should be “model”

Start of last paragraph:
“Now if you hit Refresh, and you’ll see a different error”
How about:
“Now if you hit Refresh, you’ll see a different error”

In the first “playtime” item:
“Add a new variable to the session to record how many times the user has accessed the index action.”
How about:
“Add a new variable to the session to record how many times the user has accessed the store controller’s index action.”

Footnote: “And if you’ve updated your CSS appropriately… See the listing on page 692 for our CSS.”

I think the CSS needed for this change (adding the cart to the sidebar) is already in the CSS file I downloaded earlier. There’s already a section in it called “Styles for the cart in the sidebar”.

Footnote 3: “it simply is a a”
How about: “it is simply a”

Also: “We will cover respond_to method”
How about: “We will cover THE respond_to method”

End of 1st paragraph: "the content of

Top: “what Microsoft call quirks mode”
How about: “what Microsoft CALLS quirks mode”

Adding a user did not work for me until I restarted my server. I don’t know why. Is RESTful routing not correctly generated until the server restarts? This is a very confusing issue.

(Sam says: this is a non-deterministic bug in Rails. DHH is aware)

3rd paragraph:
“Because the add_to_cart template is invoked only when we add something to the cart, THEN we know that we have to…”
(emphasis on THEN added - I don’t think it belongs)

Last paragraph: “So, on the server, we can tell that we’re talking to a JavaScript-enabled browser by testing to see whether the incoming request was generated by an xhr object. And the Rails request object, available inside controllers and views, makes it easy to test for this condition: it provides an xhr? method.”

It would be nice to de-mystify xhr requests a bit and say what is actually different about these types of requests. I think it’s one of the HTTP headers?

“deterministically ordered far outways the miniscule risk of this occuring.” should read “deterministically ordered far outweighs the miniscule risk of this occuring.”

Outweighs is misspelled.

There’s an example of the Order class with a “#…” seemingly suggesting that other stuff follows the line “has many :line_items”. But at this point, there’s nothing else in that file.

2nd paragraph: “belongs_to tells Rails that rows in the line_items table are children of rows in the orders and products tables: the line item cannot exist unless the corresponding order and product rows exist.”

Should “the line item cannot exist” be “a line item cannot exist”?

Last paragraph: “Now we have our tables and our models, we can…”
How about: “Now THAT we have our tables and our models, we can…”

Footnote 6: “You can save yourself some keystrokes on commands like these by creating a file named .sqlite3rc and put it in your home directory.”

“put” should be “putting”
Also, should this file be in your home directory? Or the project’s root directory? A relative path like db/development.sqlite3 makes no sense if referenced in a file in the developer’s home directory…

I haven’t tried this yet, but, surely instead of using number_to_currency() twice in add_to_cart.html.erb and once in index.html.erb, this would be better placed in the models, as per your suggestions earlier in the book.

I only spotted this because way back I had set a currency other than dollars in my index.html.erb, so my catalog was showing dollars and my cart pounds.

By the way the book rocks!

end statement from previous code block has become separated by figure 9.1

(Dave says: we’ll fix this during layout)

3rd paragraph: “Add a little CSS magic (see the listing in the appendix)…”

Doesn’t seem necessary. The CSS downloaded earlier on gives me the same thing shown in Figure 10.2.

There is no download link for the code associated with Iteration D4:Hide an Empty Cart. Also there are no triangle shaped pointers showing which lines have been modified.

(Sam says: The next paragraph says “So let’s not use this code”)

Code block that starts def empty_cart should have a triangular pointer next to line redirect_to_index showing this line has changed

In section 9.6 you talk about using FF and Safari, one with JS on and one with JS off. I’m not favouring (yeah, I know British spelling) one over the other but FF has a massively popular add-on called NoScript that makes checking JS/no JS a one-click breeze.

Curious why error_message_for uses ‘order’ (a string) but form_for uses :order (a symbol)

2nd playtime item:
“Hint: variables set in the controller are available in the layout as well as in the directly rendered template.”
How about:
“Hint: variables set in the controller are available in LAYOUTS AND PARTIALS as well as in the directly rendered template.”

(The simplest way I came up with to solve this one was accessing a variable from a partial, not a layout.)

The code shown for the new user form “app/views/users/new.html.erb” looks significantly different than what I see after generating the file with the scaffold script. I am suspicious that the code shown in the book is from an older version of rails. I’m using rails 2.1.0. and scaffold creates code that looks like this:

New user

<% form_for(@user) do |f| %>
<%= f.error_messages %>

<%= f.label :name %>

<%= f.text_field :name %>

<%= f.label :hashed_password %>

<%= f.text_field :hashed_password %>

<%= f.label :salt %>

<%= f.text_field :salt %>

<%= f.submit “Create” %>

<% end %>

<%= link_to ‘Back’, users_path %>

I think it would be less confusing if the code in the book was changed to more closely follow what is currently generated by scaffold.

At the top of page 150, the book claims there’s only one tricky thing about the form code on the previous page (149). However, I was perplexed by another issue which is why was the object so important to a label. Why did a label even care? Now that I look at the underlying code, I realize that the label uses the object as a namespace to prefix onto the names of the fields, presumably so one form can use multiple object with similar fields. I believe page 150 would be a great place to mention that potential stumbling block for curious but web-inexperienced readers.

Thanks. Great book by the way. :-)

Opening line (grammar): Shouldn’t it be “If it weren’t for…” as subjunctive (instead of “If it wasn’t for…”)?

Addition to “What We Just Did”?

We unified our layouts into a single application-wide layout in the ApplicationController class in the file application.rb in app/controllers.
[i.e., on page 174.]

Code sample appears truncated

<%= link_to ‘Destroy’, user, :confirm => ‘Are you sure?’, :method => :delet…TRUNC

Near bottom: “This time, we’ll use a little rails magic when we create the model. We can give a list of columns and types, and Rails will fill in the migration for us.”
(script/generate scaffold example folllows)

We actually started out the depot application with one of these, way back on page 77

string_to_hash = password + “wibble” _ salt

What useful purpose could the “wibble” in the middle possibly be serving?

Paragraph in middle: “Because the incoming password is in plain text, we have to read the user record using the name as a key then use the salt value in that record to construct the hashed password again.”

How about: “…using the name as a key AND then use the salt value…”

2nd paragraph: “It defines the standard methods to index, show, create new and edit users.”

Do we want update and destroy in this list?

2nd paragraph: “So lets avoid the redirect…”

How about “let’s” instead of “lets”

Last paragraph: “After clicking Add User, the page is redisplayed with a cheery flash notice.”

It looks like the code handling this (on pg 165) redirects to the index (rather than redisplaying the form).

It looks like the login method shown here (and again on page 173) will log the user out with that first line, “session[:user_id] = nil” after the simple act of viewing the login page. It doesn’t check for a POST request until afterward, meaning a GET will have the side effect of logging the user out (I read RESTful Web Services - GETs shouldn’t have side effects!)

How about moving that statement inside the “if request.post?”

In the listing for application.rb, the first line (“before_filter…”) has a colored arrow next to it showing that it’s a new addition, but the authorized method at the bottom (also new) doesn’t.

“A Friendlier Login System” sidebar, 3rd paragraph:
“We also need to clear down that stored URI once used.”

Was this meant to be “clear out” rather than “clear down”?

Not sure why “Enter User Details” shows up twice in the figure - the view code seems to use it only once, as the ‘legend’ tag.

typo: “Within an builder template”… -> “Within a builder template”.

Footnote: “Later, in Section 16.4, Data Migrations, on page 282, we’ll look at options for populating database tables as part of a migration.”

These were first introduced in iteration A4, but this note makes it sound like it hasn’t been mentioned yet

(Sam says: Not quite. There is a subtle distinction between populating the database schema and populating the tables themselves.)

“Navigate to http…/admin/. The filter method intercepts us on the way to the product listing”

Shouldn’t be “http…/products” instead ?

Shouldn’t it be “<%= link_to ‘Orders’, :controller => ‘admin’ %>” rather than “<%= link_to ‘Orders’, :controller => ‘orders’ %>” ?

(Same says: Next beta will generate the orders controller)

In the citation of the depot.css seems to be a partial duplication of text:

#store .entry form, #store .entry form div {
display: inline;
}

(Sam says: That’s actually correct CSS)

“The second, lib/tasks/capistrano.rake, adds some tasks to your application so that you can directly call Capistrano tasks using Rake.”
This second file is not created on my system (Using Debian-Etch, Rails 2.1, rubygems 1.2). Maybe it is not added everytime for some reason?

Updating the default fixture for products.yml because with rails 2.1.0 I have this:

one:
title: MyString
description: MyText
image_url: MyString

two:
title: MyString
description: MyText
image_url: MyString

Would it be possible to show how making the stubbing method gateway.collect returning true in a first test case and false in another one ? Is there a dynamic solution rather than “return true” ?

Thanks.

2nd paragraph: “What it [REST] really means is that you use HTTP verbs (GET, POST, DELETE, and so on) to send requests and responses between applications.”

Don’t think “and responses” belongs, since responses don’t use HTTP verbs. I suppose the response equivalent would be status codes…

Near top: “We’ll explain why this works on page 434- - -for now, just take it on faith.”

This is rendered as three short dashes. I’m guessing it’s meant to be one long dash?

The after_destroy method only seems to do its job if it’s public. I got briefly caught by that one, might be worth a mention.

After generating the store controller but before trying localhost:3000/store, config/routes.rb needs to be updated to map.resources :products, :store

(Sam says: Not true, store is not a resource, instead it depends on the default routes)

“The app/controllers directory and its subdirectories.
• All directories whose names start with an underscore or a lowercase letter under app/models.
• The directories app, app/models, app/controllers, app/helpers, app/services, config, lib, and vendor.”

Seems like you say “app/controllers” and “app/models” directories are included twice…

The last paragraph on the page, before the list of special column names starts; A number of column names that have special significance to Active Record.

Should it read; There are a number of column names that have special significance to Active Record.

If you’re using Firefox browser you should give a try to an add-on called SQLite Manager; great for administering our db, tables etc.

On bottom of page, inside the following method:

bad.each do |name|
product = Product.new(:title => “My Book Title”, :description => “yyy”, :price …TRUNC
:image_url => name)
assert !product.valid?, “saving #{name}”
end

…TRUNC should be replace with => 1

The file depot_f/app/controllers/store_controller.rb already contains the add_to_cart action, even though that is yet to be added to the store controller.

(Sam says: oreshadowing :-) More seriously: a lot of changes to a small number of files are happening in a short period of time here, and while more snapshots could be taken, it wouldn’t help in understanding.)

It would help if you provide some explanation of how @items.find works in this case, or just provide the relevant page number in the Ruby documentation at the end of the book. Thanks.

paginate words just like find–it just
doesn’t fetch all of the records.

I believe you want paginate works just like find — it just …

“until you deside it is time to upgrade.”

The footnote on this page flows over to the next page’s footnotes.

(Dave says: We fix stuff like this during layout (although footnotes do indeed sometimes break across pages)

You might want to add a bullet for the Rails helper method image_tag as well.

The comment “# get the limit amount from the form” should be “# get the name from the form”

“That’s because the render method in the main template arranges to set a variable with the same name as the partial template to the current item each time around the loop.”

It’s not entirely clear that the ‘loop’ in this case is implicit, and not explicitly coded.

Second bullet point says the book uses ‘Rails version 2.0 (specifically the 2.1 Rails RubyGem)’

I had to restart my server after adding the admin controller, I didn’t notice you mention that.

(Sam says: this is a non-deterministic bug in Rails. DHH is aware)

You might want to update footnote 1 to indicate that Capistrano also supports Git.

I think it would be nice to have one of the triangles showing which line is new for the code at the very bottom of this page. (code for: depot_r/app/controllers/info_controller.rb)

(Sam says: The whole body of the method is new)

The first paragraph after the “rake routes” listing (with a nested articles->comments routing):

The book says “Note that the named route for /articles/:article_id/comments/:id is article_comments, not simply comments.” when in fact the name of the route should be “article_comment”, in singular (as it is in the listing). Also note that the following subordinate clause should also have the singular form “not simply comment”. Minor errors indeed. :)

Bump the class def code snippet to the next page after the “Prices, Dollars, and Cents” sidebar — it’ll be a lot clearer to the reader what’s going on.

:dependent => false (or nil) to prevent child row from being updated or deleted when parent is destroyed does not seem to be supported in Rails 2.1 (nor edge at the moment). The same goes for the has_many association.

Normally when using a database schema with synthetic ids this might not be a big loss, but when using a legacy database with natural keys it often is…

I found the placement of the code fragment (fig 13.1) at the top of this page (@@default_error_messages = …) confusing, especially when it follows the sentence “The complete file now looks like this.” I think the book would be clearer if it was moved closer to where it is relevant, or better separated from other pieces of code.

(Dave says: we handle this during layout when the book is finalized)

the part :message=> ‘must be a URL for GIF, JPG ’ + ‘or PNG image.(gif|jpg|png)’ would be a lot more tidy without the last part (gif|jpg|png), which is confusing for non-geek end users of our app.

Maybe you should precise that the name of the file is important when you create a migration and that it should have the same name as the class, isn’t it?
And you should precise that in this case the views and layout won’t be upgraded (it should be done manually).
thank you very much for your book, it’s very good at all.

Best Regards
Maxime

Sorry for my english but I’m french, you can email me at helfer.m@free.fr if you have some questions.

The “this line has changed” triangle appears here and seems out of place since this is the listings appendix.

The “this line has changed” triangles appears here and seem out of place since this is the listings appendix.

The “this line has changed” triangles appear here and seem out of place since this is the listings appendix.

per previous comment, add the triangle to the

format.html

line that was added for the change that was just made.

Seems like the following line should be bold, italics, underlined, flashing, and otherwise drawing attention to itself:

Although it isn’t obvious in print, you must use spaces, not tabs, at the start of each of the data lines, and all the lines for a row must have the same indentation

Top:
“Once you have files in the lib directory, you use them in the rest of your application.”

How about:
“Once you have files in the lib directory, you CAN use them in the rest of your application.”

2nd paragraph:
“If the file is in the lib directory itself, you require it directly by name.”

How about:
“If the file is in the lib directory itself, you CAN require it directly by name.”

In the code sample near the bottom of this page, the ‘puts’ statement isn’t indented inside the do..end block

Middle paragraph: “Early ones may only be compatible only with earlier versions of Rails…”

Using “only” twice here sounds redundant

“If you want to go back to the system-wide versino of Rails, you can either delete the vendor/rails directory or run the command ‘rake rails:unfreeze’”

This has me wondering if these two are equivalent. My guess based on this passage is that they are.

3rd paragraph:
“This is less risky than having your project dynamically update as the core team make changes each day…”

How about:
“This is less risky than having your project dynamically update as the core team MAKES changes each day…”

“rake rails:freeze:edge”

I assume from the earlier discussion that this command will put a copy of the latest version of edge rails in vendor/rails?

Top: “Running on the Edge means getting all the latest improvements and techniques as they emerge from extraction.”

I’m wondering what is meant here by “extraction”

Near top:
“…it arranges to load itself from there, rather from the shared library code.”

“…it arranges to load itself from there, rather THAN from the shared library code.”

Outdated references to .rhtml, .rxml in Figure 14.3

Footnote 3: “With gems, usegem install…” appears to lack a space between “use” and “gem”

“Rails extends all Ruby objects with the blank? method. It always returns false for nil and false, and it always returns true for numbers and for true.”

This description looks backward to me.

now = Time.now
…
puts now.at_midnight # => Thu May 18 00:00:00 CDT 2006
…
puts now.midnight # => Thu May 18 00:00:00 CDT 2006

I’m wondering what the difference between these is…

Last sentence:
“Your terminal man not display…”

should be:
“Your terminal MAY not display…”

(Sam says: Fixed in B1.5)

The last paragraph on this page talks about the “apply” and “undo” parts of a migration. Should this be “up” and “down” instead?

Last paragraph:
“If the current schema_migrations table has a row with a higher version numbers than this target number, Rails takes the migration with the highest version number…”

How about:
“If the current schema_migrations table has a row with a higher version NUMBER than this target number, Rails takes the migration FOR THE ROW with the highest version number…”

The misspelled “outways” mentioned in the errata for pg 271 is actually on pg 270.

(Sam says: It’s on page 271 in B1.5)

End of 6th paragraph: “This is true even if the version number was before one or more of the already applied migrations. […] If we then run migrations this new migration file—and only this migration file—will be executed.”

Really? Wouldn’t this lead to migrations applied out of order? Is this done because it is assumed that out-of-order migrations created this way will be independent of one another? Some mention of this issue might help…

3rd paragraph:
“Rails migrations automatically add a primary key called id to all tables it creates.”

How about:
“Rails migrations automatically add a primary key called id to all tables THEY CREATE.”

The “create table tickets (…” DDL shown on this page seems to be missing the “updated_at” field.

3rd paragraph: “Note that we’re talking here about creating data that’s a convenience for the developer when they play with the application and for creating ”fixed" data such as lookup tables."

…but page 284 says not to do this. It seems like we’re getting conflicting advice here. It seems like what we’d really want is something that does this in development-mode only…

The first code sample on this page creates the directory “db/migrate/dev_data”. I’m wondering if this is meant to imply a restriction to the development environment, as opposed to test or production. While that seems to be what we would want, it doesn’t appear to be what we’re doing in this example…

2nd paragraph: “Be warned: the only data you should load in migrations is data that you’ll also want to see in production: lookup tables, predefined users, and the like. Do not load test data into your application this way.”

As mentioned in my comments on the last couple pages, this seems to put us back on square one, without a good way of loading test data for the development environment.

2nd paragraph:
“By default, Active Record assumes that the name of the table is the plural form of the name of the class.”

How about:
“By default, Active Record assumes that the name of the table is the LOWERCASE plural form of the name of the class.”

(Sam says: That varies by db. In DB2, for example, it is uppercase.)

2nd paragraph: “if the schema specifies columns with no decimal places, they are mapped to integers; otherwise they are mapped to Ruby BigDecimal objects…”

Should “integers” be changed to “Fixnums” here? (since we’re mentioning the exact class in the decimal case)

Top: “If we fetched this row into an Active Record object, that object would have five attributes.”

Looking at the example immediately above this, it looks like seven rather than five.

5th paragraph:
“For example, the following call creates a connection to a SQLite3 database called railsdb on the server dbserver.com using the given user name and password.”

This description looks like it hasn’t been updated in the move from MySQL to SQLite3, and doesn’t match the code example below it.

Last paragraph: “Note that in this case a RecordNotFound exception is returned if any of the ids cannot be found…”

I assume this should say “raised” rather than “returned”?

“David Says” sidebar, first paragraph: “A call to Person.find(5) is based on our knowledge of the persons table.”

This should be “people table” rather than “persons table” right?

2nd paragraph after sidebar:
“So far we’ve just scratched the surface of find, using it to return one or more rows based on ids that we pass in as a parameter.”

How about:
“So far we’ve just scratched the surface of find, using it to return one or more rows based on ids that we pass in as PARAMETERS.”

Description for “:conditions”:
“…whenever we talk about an SQL parameter, assume the method can accept either an array or a string.”

How about adding “or a hash” to this statement?

2nd paragraph: “…have the question include the author name in the result set.”

Why “question” rather than “query”?

In the first code sample on the third line, how about changing “order” to “orders” since it is doing a find_all

4th paragraph:
“Both read the corresponding rows in the database table into model objects and call the instance-level destroy method of that object.”

How about:
“Both read the corresponding rows in the database table into model objects and call the instance-level destroy method of THOSE OBEJCTS.”

“xxx_id
Default name of a foreign key reference to table named with the plural form of xxx”

add an “a” or “the” before “table”

Bottom: “If you used some other name, you’ll need to add a declaration so Rails can find it.”

How about saying how to do this, or pointing to a book or web site to find out more.

Under “One-to-One Relationships” heading:
“…implemented using a foreign key in one row in one table to reference at most a single row in another table.”

This seems like it would be more correct to say:
“…implemented using a foreign key in AT MOST one row in one table to reference a single row in another table.”

(moved “at most” to an earlier point in the sentence)

3rd paragraph:
“…and the foreign key field is the singular form of the parent class name with _id appended”

How about:
“…and the foreign key field is the LOWERCASE singular form of the parent class name with _id appended”

“orders(force_reload=false)
Returns an array of orders associated with this customer (which may be empty if there is none). The result is cached, and the database will not be queried again if orders had previously been fetched unless true is passed as a parameter.”

How about:
“orders(force_reload=false)
Returns an array of orders associated with this customer (which may be empty if there ARE none). The result is cached, and the database will not be queried again if orders HAVE previously been fetched unless true is passed as a parameter.”

(2 changes shown in caps)

I just realized that all the errata submissions I made today up until now are for B1.4, whereas I submitted them as B1.5. Sorry, my bad.

3rd paragraph:
“What do we do if we wanted…”

How about:
“What WOULD we do if we wanted…”

4th paragraph: “…the join table is not an Active Record object…”

How about “model” rather than “object”

Example shows a find method on the Person model to find objects of classes that inherit from Person. I’m wondering if we can find managers in this table using Manager.find for example…

Bottom: “…access it via the model object’s indexing interface…”

I’m wondering what the “indexing interface” is

2nd paragraph: “There’s an important point in this DDL…”

but we’re looking at a migration here, not a DDL

Footnote 7 is unchanged from the 2nd ed of the book and is now out of date.

In section “Friendlier Login System”, perhaps we can augment and secure the statement “session[:original_uri] = request.request_uri” by writing “session[:original_uri] = request.request_uri if request.get?”. Thus, we make sure that only GET-method actions will be executed, and noone can “back-button” or bookmark or POST by hard-code or request later an inconsistent POST action.

Need to change the color of the “IDENTIFIED BY” line to indicate that the user typed it, and fix “somepass” and “some_pass” so they are the same.

Also, eliminate the semicolon at the end of the “GRANT ALL PRIVILEGES” line — there should only be one semicolon for this line, after the “IDENTIFIED BY” line. Indeed, if you use a semicolon after “GRANT ALL PRVILEGES”, you won’t be prompted for the rest of the line with the password.

The database.yml listing is not quite right. This is what Rails 2.1.1 generates when you have the —database=mysql option set:

development:
adapter: mysql
encoding: utf8
database: depot_development
username: root
password:
socket: /tmp/mysql.sock

(Sam says: The generator produces minor variations based on the platform, hence the “something like” on the page and “may or may not appear” on the next.)

I think this explanation might be a bit over-simplified: “(it’s called adapter because Rails uses this information to adapt to the peculiarities of the database)”. Technically, I think it’s called adapter because it’s named after the design pattern. Perhaps you should take out the parenthetical statement, or add a footnote that mentions the adapter design pattern and where to find more information about design patterns.

Here’s another somewhat likely problem that happened to me that can happen when users try rake db:migrate: an error happens that says you need to update RubyGems. I was running Rails 2.0.2 and when I upgraded to Rails 2.1.1 it told me that I needed to upgrade my RubyGems to at least RubyGems 1.1.1. A fair number of users may upgrade their Rails version before starting the depot application, so they would encounter this error.

(Sam says: Covered in chapter 3)

It would be better style to change this sentence: “So, let’s go ahead and create the model and views and controller and the migration for our products table.” to be “So, let’s go ahead and create the model, views, controller and migration for our products table.”

In the first code listing on this page, it is difficult to distinguish between the user’s command and the resulting output. Use of color and bolding, such as in the listing on the bottom of the page, should help this.

It’s probably worth it to go into more detail on what Rake is (maybe put it in one of those gray text boxes you use when you take a little tangent so that you don’t interrupt the flow of the main idea).

Did you know that “whether or not” is redundant? You should just say “whether”. (Yes, this is picky ;)

The example uses WEBrick, which won’t be used if people type script/server with recent versions of Rails.

(Sam says: Depends on what you have installed. See footnote 2 on page 46.)

The first paragraph under Run the Maintenance Application is too wordy. Consider replacing the whole thing with “First, we’ll start a local web server.”

“Be careful as you add methods to this controller” should be “be careful as you add methods to this model.” The text is talking about modification of the product model, not the controller.

I believe this statement is incorrect “the default is to use the filesystem”. As I understand it, in Rails 2 the default is to use cookies to store all of the session data, not just the session id.

You may want to mention that if you may not be able to see the new flash CSS styles until you hit refresh on your browser.

There is a great deal of troubleshooting advice mixed in with the Depot tutorial. It might be worth considering making a special sidebar for troubleshooting advice.

on an image I can see a beautiful output of the documentation - I wish there would be some instruction how to achieve this output (probably it’s all about providing a correct URI)

(Sam says: Point your browser at the index.html file in the doc/app directory.)

After adding cart I get “ActionController::InvalidAuthenticityToken in Store#index ”

from suggestion in #32678: I have tried to add:

protect_from_forgery :only => [:create, :update, :destroy] after the class StoreController…. statement however this does not change the error message.

I have downloaded and repalced my handwritten code with the examples provided.

assistance appreciated … thank you

Hugh Tucker

(Sam says: If you restarted your browser, you need to back up to the form, hit refresh on that, and then proceed. This is covered on page 114.)

link to depot_r/test/performance/order_speed_test.rb lacks “setup” block of code.

(Sam says: This specific test doesn’t require any particular setup.)

in code ~1/3 of page is line “{ :order => DAVES_DETAILS },” which takes details from hash which isn’t defined in this file (but is defined in downloadable one)

(Sam says: From the text: “we’ll use the same hash of values we used in the integration test”. This hash was originally shown on page 227.)

in ~2/3 of page: “we’ll base the
test on a standard functional test (and we’ll cheat by copying in the boilerplate
from store_controller_test.rb)” - probably it’s not store_controller_test.rb (which seems untouched), but rather integration test named dsl_user_stories_test.rb

“Let’s run the performance test.” at the bottom of the page - for me it run with error (test takes almost 6 seconds) - probably you’d like to prepare users with older machines for such error

At the top of the page you have … “we’ll tart it up in a minute”.

The would ‘start’ is missing the leading ‘s’.

I was getting a ActionController::InvalidAuthenticityToken error when I was trying to add items to my cart on page 112.

I went back to the code on page 109 and added:

protect_from_forgery :only => [:create, :update, :destroy]

It seemed to get it working again.

Cheers,

Brett

Below “David says…” - “blank? method. It always returns false for nil and false, and
it always returns true for numbers and for true.” - apparently words “false” and “true” should be exchanged.

It may be me since I’m new to Rail, which is why I bought the book, but is there a way to debug your tests? That is, to verify that your test is written correctly and the app isn’t working rather than the other way around.

For example, I kept getting an error running the integration test, UserStoriesTest on the ‘assert_template “index”’ line after the ‘post_via_redirect…’ call, stating that it was working with “checkout” when expecting “index”. It appeared to be a timing issue (the order was saved properly, as verified by the preceding assert statement) and was resolved by moving this call to the end of the test (after verifying the order details). Since my app is stock and the test case was copied from the download file, I would assume that it should complete successfully. However, since it failed, I didn’t know if the test case was bad or whether there was a glitch in the application. Since the application functions just fine interactively, it led me to believe that there was an error in the test case.

At bottom of the page you suggest installing low-level C binding to MySql with “gem install mysql” - I tried that and I got an error from gem: “Could not find mysql (> 0) in any repository”. Thus you should update this part or change the command.

(Sam says: Configuration error? Works for me.)

In the Joe Asks… Box there is a line that reads:
Some folks press the point: “just do these variables get set?”

with an inserted ‘how’ that only displays ho with a few extra spaces after it.
Reading it with Preview Version 4.1 (469.2.1)

(Dave says: I bet it goes away if you reboot. Apple’s PDF viewing has some rough edges)

depot_p/app/views/admin/login.html.erb

Bug: When attempting to log in without correct user-id/password, no errors are displayed.

Solution: Add line containing flash[:notice], or some better formatted error messages.
This solution is from a newbie. It would be nice to be nicely formatted like earlier in the book.

Parts of the chapter “Task T: Testing” doesn’t match the style of the rest of the book. This is especially glaring from the second half of page 211 through the first half of 216. This section reads a lot like The Rails Way.

That said, there’s nothing strictly wrong with the section. It’s just a look-and-feel nitpick.

line 7.
States:
… five directories and a helper file
Should state:
… four directories and a helper file

Context “Index: For Admins Only”

When I checked out the functional tests for my Depot app, I found that some of the controller tests had already been written. You might want to consider how you present this section.

The inference is that the test has to be written: it’s there already! Whilst not complete I’m sure, I think it would be misleading not to mention that Rails is doing a bunch of stuff to help.

For example, the following call creates a connection to a SQLite3 database called railsdb on the server dbserver.com using the given user name and password. It will be the default connection used by all model classes.

ActiveRecord::Base.establish_connection(
:adapter => “sqlite3”,
:database => “db/development.sqlite3”
)

The template file in the PDF is correctly named store.html.erb. In the downloadable zip-archive it is named in plural, stores.html.erb. So when you´re simply copy and paste the whole file into your own project, rails doesn´t find, and therefor doesn´t use the template file.

(Sam says: I can’t find “stores.html.erb” in the downloadable zip)

YYYYAAAAAAAARRRRRRGGGGGGHHHHHHH!!!!!!

Debugger

Ok, so it is my fault. I ass-u-me-d it would work.

In the text above (Step 1 under “To use breakpoints” you say:

Insert a call to the method debugger at the point in your code where you
want your application to first stop. You can pass this method a string if
you’d like—this becomes an identifying message later.

What tossed me off the rails and into the weeds was the second sentence “You can pass this method a string if you’d like”. Well, I did.

I inserted “debugger ”Checking why cart is acting strange"

With that string, the server doesn’t stop. Without that string, the server stops and enters irb. Of even MORE enjoyment is the fact that the community now seems enamored with ruby-debug and I was unable to find any docs on debugger anywhere.

I suggest you include an example of the “pass this method a string” in the text, or simply leave out that sentence.

:) Thanks, I feel so much better getting that rant off my chest now.

- Dave

before listing the code listing you say, “don’t worry too much about the details at the moment; we’ll revisit the user interface shortly,” but then the code is not revisited. Specifically, a link to pg. 418 would have been nice to talk about “XXXX_path” and “XXXX_url” being functions (I did a lot of grepping looking for this function :) and also why index.html.erb uses products (with an s) while new.html.erb, etc uses product (no s)… this is because these are passed data with the link_to variable, right?
Anyhow, I know this is just the introduction but on my fifth reading when I realized I didn’t know what edit_product_path what doing (or why products was different than product) it would have been nice to have a tip.
Thanks.

(Sam says: Revisited in A4 on page 90; index lists a number of products (plural), new creates one product (singular))

Several concepts of version numbers are introduced: how they are generated, how Rails remembers which migrations have been run, and how to revert a schema to a previous version.

You’ve gone into several concepts without any examples, and this will confuse first-time readers. Give examples!

Secondly, when you talk about “how do we revert a schema to a previous version?” the word “version” competes with the term “version number” that we’re talking about, which can confuse readers.

“creating a model called discount also creates
a migration called ddd_create_discounts.rb”

This filename reflects the old migration names. Perhaps the filename should be dddddddddddddd_create_discounts.rb

Components is mis-spelled in the third bullet point: “compoenents”

Missing a comma at the end of this line:

:select => “sku, sum(amount) as amount”

What’s left unexplained is a good way to set up your migration(s) for a habtm relationship.

“We have two controllers, a admin controller….” an?

In sentence starting with “Other than requiring a…” the word “words” is used when “works” is the correct .

I know it can be silly, but after “Useyourfavoriteeditortochange methods
the file say_controller.rbintheapp/controllers directory, adding the hello method
as shown.” it would be useful adding “save the file”.

Just before listing at the bottom is phrase that follows: “To do that, we’ll add it [encrypion helper] to the ActiveRecord::Base class.” - I wish you defined where this piece of code should go - I have no idea whether I should put it in basic Active Record directory where it was installed or in app/models in my rails app.

(Sam says: Download e1/ar/encrypt.rb to see how it could be used)

Below the “Playtime” heading, second bullet, the first sentence is missing a right paren after “vice versa”:

Experiment with adding and removing the minus sign at the end of the
ERb <%= %> sequence (i.e., changing %> into -%>, and vice versa.

at the bottom you have one listing marked twice by “Download e1/ar/transactions.rb”

as before - dupliacate “Download e1/ar/transactions.rb”, but here we see no difference between code and it’s output (output looks as if it was a part of the code) - there should be at least a word of introduction as it appeared before

the first listing on this page:
>> rs.generate({:action => “edit” , :id => 123}, last_request)
=> “/blog/blog/edit/123”.
I suggest stressing that blog controller (which at first appeared out of thin air to me) comes from ‘last_request’ param (in which it is default, and therefore hidden)

just after the output: “Have a look at the last line of the output of this command.”, but later you refer to routes which comes at the middle of the output.

in restful2/app/controllers/comments_controller.rb listed in this page in action create (and update also) we don’t save the parent after doing << operation (or update_attributes in update)

just an FYI, reading the PDF at 100% magnification in Adobe Acrobat on a Mac, the equals sign is indistinguishable from a dash.

The word ‘whitespace’ is improperly hyphenated and broken across lines 4-5 as ‘whites-pace’.

In footnote #9, the word ‘and’ at the beginning of line 2 should be ‘at’: “9. If you still see blank lines in the output check to make sure that there aren’t any blank spaces
and the end of the lines.”

according to the description of render(:file…) method in the middle of the page, I should look for help according changing default template base path at page 245, but either it’s not yet described there or I should look for this help somewhere else (probably in page 443 where you show how to use ActionController::Base.template_root=dir_path or in one of appendices)

The second paragraph starts with “In the old days (up until a year or two ago), browsers were treated as really dumb devices.” This is exactly the same sentence used in the second edition. Surely, the tongue-in-cheek comment in the parentheses is no longer necessary.

Another common error is having only install mongrel not mongrel_cluster. Doing:

sudo gem install mongrel_cluster

fixes the “ERROR RUNNING ‘cluster::configure’: Plugin /cluster::configure does not exist in category /commands” error.

This may be a suggestion, who knows. During the example of adding the total price to the Depot cart the example image shows a formatted Total cell along with a formatted individual item price cell. The used css tags are item-price and total-cell neither of which are defined in the depot.css currently available. Copy & Pasted from the URL provided in the eBook on September 23, 2008. In all other cases where new CSS was introduced a code block was included with the additional css but it appears to be missing in this example.

Hope I didn’t miss something.

4th paragraph, last sentence: has_may s/b has_many

last bullet point: while I’m flattered that you would name a whole format after me, I think JON is supposed to be JSON

2nd to last paragraph, the command to clear database sessions is truncated (appropriately enough, with …TRUNC). If this was intentional, then this is a suggestion to include the rest of the command.

Rails 2.2 Warning box, 2nd to last sentence: onlin s/b online

I recognize that this may be moot if the warning box is not included in the released edition, but though it should be mentioned just in case…

“This gives the translators the opportunity to provice the
correct pluralization of nouns and to match the verbs with the nouns.” provice s/b provide

6th paragraph, last sentence: “once we are completed” s/b “once we are finished” or “once we finish”

also pages 697, 699, listings are truncated with …TRUNC, should probably be line-wrapped instead

hay_may should be has_many

The last sentence, in which you state that “making our application work regardless of whether JavaScript is enabled takes just a two line of code change in the add_to_cart action,” is awkward. Perhaps the following would be better: “. . . enabled requires just two lines of code in the add_to_cart action.”

Hi, You have the following code snippets at the top of this page.
<%= button_to “Checkout” , :action => :checkout %>
<%= button_to “Empty cart” , :action => :empty_cart %>

The second is the line of code to be added but the first reflects the code before the ajax changes have been applied and actually looks like this

<% form_remote_tag :url => { :action => :empty_cart } do %>
<%= submit_tag “Empty cart” %>

This raises questions over which format should be used at this point for the check out button.

Hope this helps.

(Same says: depot_p/app/views/store/_cart.html.erb shows button_to has been used consistently)

James

“Maybe we want to edit an existing model”

Should that read

“Maybe we want to edit an existing order”

byes should be bytes

“…login form, the will see” => “…login form, they will see”

Maybe it’s late, but… Perhaps add a definition for ActiveResource in the opening paragraph of the chapter?

Agile Web Development with Rails 3nd Edition (Beta 1.3 - 20080604).pdf

Building an Application - Task F: Administration - Iteration F2: Logging In - Page 168
———————————————————————————————————————————
Create controller “admin”, actions “login”, “logout” and “index”, by command

\truby script/generate controller admin login logout index

However,

Building an Application - Task T: Testing - Functional Testing of Controllers - Login - Page 206
————————————————————————————————————————————————
“Because the LoginController was created with the generate controller script, Rails
has a test stub waiting for us in the test/functional directory.” (Download depot_r/test/functional/login_controller_test.rb)

I think,
the Controller should be AdminController rather than LoginController,
and the functional test file should be “admin_controller_test.rb” rather than
“login_controller_test.rb”

The demo example doesn`t work, since there is no database table!
Okay, i did not exactly follow the guide… i am using Aptana… but this mistake should occure anyway.

greeting, Chris

(Sam says: Works from the command line, and sqlite3 databases are created as needed.)

In section 14.4
“They’re surprised that they call a model class Person and
Rails somehow knows to go looking for a database table called people”

people table? or person table?

Missing word: “with”
It generally is more popular WITH a number of scripting languages (a notable exception being JavaScript) than with statically typed languages, …

At the beginning of this chapter (Active Resources), before the presentation of alternatives to the chapter’s subject, I would really appreciate at least one short paragraph that outlines what it (the chapter’s subject) actually is.

“As it stands, that’s not particularly interesting: because it’s a new model, all the fields will be empty. However, consider the general case. Maybe we want to edit an existing model.”

I think you’re using the word “model” where the word “order” (an instance of the order model) would be more clear. (Similarly, in the next sentence you say “tried to enter an order”, and I think that’s right.) Suggest:

“As it stands, that’s not particularly interesting: because it’s a new order, all the fields in the model will be empty. However, consider the general case. Maybe we want to edit an existing order.”

I think the list of files from “ls test/functional” may be out of date? In my depot directory at this point, I also have line_items_controller_test.rb and orders_controller_test.rb

At the end of the section on unit tests, “rake test:units” runs cleanly, and the same with the section on integration tests. But at the end of this section on functional tests, “rake test:functionals” has lots of failures. (This is true even if you try it directly on the unmodified “depot_r” directory distributed on the Prag. Prog. web site.)

“If we were to run migrations again at this point, nothing much would happen.
Each of the version numbers of the migration files would match with a row the
database, so there’d be no migrations to apply.” Shouldn’t this be “with a row IN the database”?

Then we provide a template, which makes use not only of the generic XML
functionality that builder proviles,

There are index entries that say see Component, but there is no index entry for Component.

I found the order by example a little confusing, simply because it was in the context of product orders. Here’s the code:
orders = Order.find(:all,
:conditions => “name = ‘Dave’” ,
:order => “pay_type, shipped_at DESC” )

There are four different uses of the word “order”, and it took me a moment to think it through. This might mean it’s a good example because it forces the reader to think, but I thought I’d point it out just in case it wasn’t intentional.

“rake db:schema_migrations” instead of “rake db:schema_version”,
since “task :schema_migrations”

“This behave the same as validation does
in ActiveRecord.” behaveS

If you used some other name,
you’ll need to add a :foreign_key declaration so Rails Rails can find it.

The :dependent option tells Active Record what to do with child rows when you
destroy a row in the parent table. It has five possible values.

Five or three?

On the Second paragraph on the first line there is the sentence: “Normally, this doesn’t matter, because HTML…”

The end of the line is ends up with “white-” and the next line have “pace.”, it should be “white-space” and not “white-pace”.

Reference/link “Postgres hxxp://ruby.scripting.ca/postgres/” is incorrect. Link is broken (yields 404 page; looks like domain wasn’t renewed and got picked up by domain squatter.)

From what I’ve been able to gather, the (most) official source for a Ruby-based interface to PostgreSQL specifically (as opposed to Ruby/DBI) is via the Ruby-PG project:
hxxp://rubyforge.org/projects/ruby-pg

[Had to munge the protocol since the errata form doesn’t allow URL posting in errata reports.]

Hasn’t Rails 2.1 been RESTful for awhile? I don’t understand “…as Rails moves in a RESTful direction” with a footnote to 7:
7. As this book is being finalized, a plugin called Simply Helpful is being worked on. This plugin
makes it even easier to integrate models, REST, and form_for. The plugin might end up in core Rails.

The styles shown in that page are not for the task 7.2 (Add a Page layout). I believe it’s for the Cart layout.

The stylesheet modifications at the end of section 7.2 don’t have anything to do with the output of the main “store” page. All the needed styles are already in the previously referenced stylesheet.

Minor point: The styling of the page is such that the price of the last entry in the catalog (Pragmatic Version Control) doesn’t clear the picture when the page is very wide (I am using 1680x1050). Adding a ‘br clear=“left”’ tag fixes this, but adds unneeded space…

“show her the result of our morning’s work.” - according to the story you are spinning, it is after lunch and you have made these changes in the afternoon. Perhaps “day’s work”?

Some explanation of this syntax would be helpful:

current_item = @items.find {|item| item.product == product}

Since later in the book (p118) you are swapping the order the calls to find_cart and Product.find (lines 2 and 3 of store_controller.rb), it is a little clearer to swap their order here, so that it is clearer on p118 what changes

The method name of “redirect_to_index” is perhaps a little misleading; it doesn’t take into account the flash. Perhaps call is “flash_and_redirect_to_index”. A little long, I admit.

Inconsistent use of parentheses:

<%= h(cart_item.title) %>

Line 7 - “Others finds it useful” should be “Others find it useful”

the 2nd line of the 2nd loop for the test_image_url method is truncated. It’s easy to figure out but I wonder why the previous loop had each parameter of the Product.new() call on a new line but the second loop does not. Previous pages / examples make no elusion to the idea that space is a concern in this book and I find it strange that it appears to be now.

—Matt Stuart

Typesetting software is producing ligatures in the text - it’s fine for the most part, but when I copy+pasted “find_products_for_sale” in the second line, the initial fi ligature caused some “issues” with the Ruby interpreter until I took a closer look at the code…relatively easy to spot if you’re using a monospaced font (doesn’t everyone?)

It’s not a huge deal, but perhaps something that might be looked at?

(Dave says: we provide all the source from the examples as downloads (normally by simply clicking in the PDF).)

Doesn’t this command work for all operating systems, not just Mac OS X?
When would this step be necessary?
Couldn’t one just do
rake db:migrate
to see if the database is configured?

The book states:
The following step is rarely necessary, but can be helpful if things continue to
go wrong. You can verify which version of sqlite3 your sqlite3-ruby inter face
is bound to by running the following either as a standalone program, or from
within irb, or from within ruby script/console.
require ‘rubygems’
require ‘sqlite3’
tempname = “test.sqlite#{3+rand}”
db = SQLite3::Database.new(tempname)
puts db.execute(‘select sqlite_version()’)
db.close
File.unlink(tempname)

Instead of saying 10.4 and 10.5, and then latter referring to Tiger and Leopard, but not telling the reader that 10.4 is Tiger, and 10.5 is Leopard,

why not say 10.4.x and 10.5.x and for versions greater than 10.5.x look on your website for instructions?

That way, 3 mac releases from now, people won’t be pondering what version was Tiger and what version was Leopard.

Also do these instructions imply equally well for the client version as well as the server version of 10.4.x and 10.5.x?

Was:
How to change the version of Rails that your
application is using is described in the sidebar on page 246.

To: How to change the version of Rails that your application is using is described on pages 245 to 247.

Comment:
freeze is mentioned on pages 41, 245, 246, and 247

Since there are multiple authors, does it make sense to say

I write my Rails programs using a programmer’s editor

wiihout identifying who I is?

Perhaps one could say that Mac OS X users prefer TextMate, Xcode, Netbeans, vi, vim, and emacs.

Linux users use:

Window users use:

What does James Duncan Davidson use and why?

To view the Rails API documentation, open the location

doc/api/index.html ( or rails_apps/demmy_app/doc/api/index.html)

with a browser.

I had omitted the “require ‘digest/sha1’” from my User model and got all the way up to successfully login in before I noticed. Does it really need to be there?

Possible error (I’m just learning)—my cart would not update and it would only add one item. I’d return to the store, add another and only the newly added item would be there. The session was not updated. This was corrected with the line

session[:cart] = @cart

added before the end of the add_to_cart method. (I am using mySQL instead of SQLite3, so maybe in SQLite this updates automatically? Where is the session actually set to our cart? The other way happens in the find_cart method called in add_to_cart, but don’t we need to update the session’s cart?)

Re: “We created a virtual attribute representing the plain-text password and coded it to create the hashed version whenever the plain-text version is updated”

The word “updated” is incorrect - we never addressed the ‘edit’ action upon users, so it is currently impossible to “update” a user. The sentence needs to be re-written to be more correct.

(Sam says: The model does exactly this; updating the view is a playtime activity)

Re second bullet point about GET/POST differences. I’m sorry, I went over it twice and can’t see a thing relating to this. We updated the ‘update’ action to redirect to the index, but other than that…

(Sam says: iteration F2, method login)

Page 69 states
A few “magic” commands worth knowing, explained in detail later: rake
db:sessions:clear and rake db:migrate VERSION=0.
Page 77 states
And that’s it. Rake looks for all the migrations not yet applied to the database

Page 78 states
and applies them. In our case, the products table is added to the database
defined by the development: section of the database.yml file.7

7. If you’re feeling frisky, you can experiment with rolling back the migration. Just type
depot>rake db:migrate VERSION=0

Shouldn’t there be a caveat here about the order of the timestamps does matter, if one
does a db:migrate VERSION=0
followed by a
db:migrate

if I use

1) my own 20090601000002 AddPriceToProduct: which says 2009 rather than 2008
(the db:migrate will work as you have stated)

2) Your book file db/migrate/20080601000003_add_test_data.rb
(the db:migrate will work as you have stated)

However,
when I then do a
db:migrate VERSION=0
followed by a db:migrate
then
add_test_data will be run before addPriceToProduct,
so add_test_data will fail because the price column isn’t there.

(Sam says: Migrations is covered in chapter 16; at this early stage what is being described is only enough to get you started)

The link to Tobias Lütke’s ActiveMerchant library is not correct. This seems to be better: www.activemerchant.org

Payment module∗seems to be dead -> discussion seemst to indicate that people should use ActiveMerchant instead. Suggest to kill reference here.

In the explanation of why we use ‘self.salt = ….’ it says: “… Without the self., Ruby would
have thought we were assigning to a local variable, and our code would have no effect. ”
That is pretty clear, what is NOT so clear is, why is the writer not using ‘@salt = …’? After looking at the example several times I am still not sure whether those two are equivalant in this case.

“Typos happen to the best of us and side explorations
are not only are possible, they are positively encouraged.” Should read “side explorations are not only possible, ”?

at the beginning of 11.4 we add layout “store” and are told to visit ‘admin’, and only later the text mentions the need to remove references to cart.
That should be done before attempting to visit the said urls

in “<% for cart_item in cart.items %>" cart_item is different from the previously used "<% for item in cart.items %>”
It would avoid confusion to continue to use the previous “item” version instead of “cart_item”. It is also more logic.

we never added product id 1, the example should use 2 (or 3 or 4)

hxxp://www.rubyveil.com/ is no longer valid (domain not renewed by original registrant)

hxxp://www.rubygarden.org/ - “RubyGarden is offline temporarily due to excessive spam.” - listing a ‘resource’ that’s offline due to a spam DoS is questionable.

The total price should be calculated like this:

def total_price
@items.sum { |item| item.price * item.quantity}
end

(Sam says: CartItem.price is defined to be product.price * quantity)

Having completed Iteration D3 everything works fine as long as there are already products in the cart. But if I empty the cart and add a book to the cart, I get the following error:

RJS error: TypeError: $(“current_item”) is null

I next get a longer error message:

Element.update(“cart”, "

\

\
\

\
\

\

\

\

\
\
<tr class \\“total-line\\”>\

\

\

\
\

\
\

“);
$(”current_item“).visualEffect(”highlight“, {”endcolor“: ”#114411“, ”startcolor“: ”#88ff88"});

The cart then updates, but without highlighting the change. If I add another copy of the same item, everything work as expected again.

“script/plugin install git://github.com/rails/auto_complete.git”

The above assumes that git is installed on your machine.
If git is not installed, then one needs to install it.

.. such as AJAX and RESTful
inter faces into their code: support is built in. (And if you’re not familar with
AJAX and REST inter faces, never fear —we’ll explain them later on.)

AJAX is not defined until page 125.
Restful is never defined. A little googling showed that Sam Ruby has written “Restful Web Services” and DHH wrote the forward for the book, so perhaps you know too much about Restful.

All I was looking for is a simple definition of Restful which I found at wikipedia (so maybe it isn’t even right)

http en.wikipedia.org /wiki/ Representational_State_Transfer
Systems which follow Fielding’s REST principles are often referred to as “RESTful”.

(Sam says: REST is described in section 20.3)

hxxp://pragprog.wikidot.com/rails-play-time - doesn’t contain reference/discussion for first bullet point in “Playtime” for Task D.

[Yes I know it’s a wiki and I could get an account…meh.]

The letter “e” is missing from the word “online” in the phrase: “but they are onlin and can be downloaded” (which is located closer to the bottom “Rails 2.2 Warning” section).

In footnote 5, the text “because an application can’t assume” should be “because an application shouldn’t assume”.

We all know of applications that do assume, so “can’t” isn’t correct: “shouldn’t” is more appropriate.