Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Fix typos in third migration tutorial part #151

Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Fix typos in third migration tutorial part #151

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
339187f
Purge trailing whitespace
paultcochrane Jul 17, 2024
0cf66f3
Fix typos in ThirdMigration tutorial part
paultcochrane Jul 17, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
88 changes: 44 additions & 44 deletions lib/DBIx/Class/Migration/Tutorial/ThirdMigration.pod
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,17 +7,17 @@ DBIx::Class::Migration::Tutorial::ThirdMigration - A more advanced database chan
In the last version we added a new table C<country> and enforced a one to many
relationship between C<country> and C<artist> such that every C<artist> belonged
to a single C<country>. These types of database changes, when you add a table
and some columns, tend to be some the easier types of migrations to perform.
and some columns, tend to be some of the easier types of migrations to perform.

Lets take on a more complicated migration problem, and add some more fixture
Let's take on a more complicated migration problem, and add some more fixture
configurations and seed data. This time we will add a table and change the
nature of an existing relationship. Is L<DBIx::Class::Migration> up to the
job? Let's find out!

=head1 The New Business Requirement

In version 1 and 2 the relationship between Artist and CD is one to many; for
each Artist there are zero to many CDs, and for each Cd there is one and only
each Artist there are zero to many CDs, and for each CD there is one and only
one artist. Now we realize that a CD could have a bunch of Artists working
together as a collaboration. We need to change our database to reflect this
need, and we additionally need to migrate our existing data to this new schema.
Expand All @@ -26,7 +26,7 @@ Let's get to it!
=head1 Change your Schema

Let's make the Version 3 schema. We need to to add a Bridge class between
Artist and Cd so that we can express the idea of 'many to many'.
Artist and CD so that we can express the idea of 'many to many'.

touch lib/MusicBase/Schema/Result/ArtistCd.pm

Expand Down Expand Up @@ -60,9 +60,9 @@ And then open that in your text editor, enter the following code:

1;

So if you are familiar with L<DBIx::Class> you'll already know this is a normal
pattern. We have a bridge table where each row points to one Artist and on CD
row, and no duplications are allowed (the Same Artist can't be linked to the
So if you are familiar with L<DBIx::Class> you'll already know this is a normal
pattern. We have a bridge table where each row points to one Artist and one CD
row, and no duplications are allowed (the Same Artist can't be linked to the
same CD twice, for example).

Now we need to change the Artist and CD Result classes. Open your editor on
Expand Down Expand Up @@ -106,9 +106,9 @@ C<lib/MusicBase/Schema/Result/Artist.pm>:
So we didn't change the columns, but we removed the direct relationship to Cd
and replaced it with a relationship to the new ArtistCd result class. We also
created one pseudo 'many to many' relationship across the bridge so that we can
directly pick up all the Cds for a given Artist.
directly pick up all the CDs for a given Artist.

Now we need to change the Cd. Open C<lib/MusicBase/Schema/Result/Cd.pm>
Now we need to change the CD. Open C<lib/MusicBase/Schema/Result/Cd.pm>
and make the following change:

package MusicBase::Schema::Result::Cd;
Expand Down Expand Up @@ -142,8 +142,8 @@ and make the following change:

1;

Changes here are a bit deeper. First we removed the C<artist_fk> column since
we no longer constrain each Cd to a single Artist. Then we removed the
The changes here are a bit deeper. First we removed the C<artist_fk> column since
we no longer constrain each Cd to a single Artist. Then we removed the
relationship directly to Artist (since we no longer have that) and replaced it
with a new relationship C<artist_cd_rs> which connects us to the C<ArtistCD>
bridge table. Last, we added another pseudo relationship so that you can
Expand All @@ -167,7 +167,7 @@ C<MusicBase::Schema>:

1;

Very good, we've changed the database to reflect our new requirement. Lets
Very good, we've changed the database to reflect our new requirement. Let's
start working on the migration.

=head2 Prepare the migrations
Expand Down Expand Up @@ -237,7 +237,7 @@ design (change indexes, storage engines, etc.)
The real fun begins under C<share/migrations/SQLite/upgrade/2-3/*>. Again you
have the C<001-auto.sql> file, which is the first, best guess on how to upgrade
the database. We need to poke into that and shape it into something that can
work for both our database structure and our data. Let's look at the the
work for both our database structure and our data. Let's look at the
suggestion. Open C<share/migrations/SQLite/upgrade/2-3/001-auto.sql> in your
text editor:

Expand Down Expand Up @@ -276,9 +276,9 @@ text editor:
COMMIT;

This doesn't look too bad. From first review it looks to me like we are just
missing transfering data from the old relationship to the new bridge table. So
missing transferring data from the old relationship to the new bridge table. So
like last time we will bust this up into a few steps, under new file names but
under this directory. Lets handle this in bits.
under this directory. Let's handle this in bits.

touch share/migrations/SQLite/upgrade/2-3/001-create_artist_cd.sql

Expand All @@ -301,7 +301,7 @@ C<001-auto.sql>, having to do with creating the new table and indexes over:
COMMIT;

Feel free to add additional SQL style comments, in order to help anyone down
the road understand what you are doing!
the road understand what you are doing!

Now we need to populate that with existing data. We will make a separate
upgrade step for that:
Expand All @@ -321,7 +321,7 @@ here's the SQL to enter:
Luckily everything we need exists in the current C<cd> table, so this is a
straightup insert. In this case I didn't use a Perl deploy run script since
I felt the performance benefit of a native SQL approach outweighed the
advantage of database portability. If I ever needed to make this work on say
advantage of database portability. If I ever need to make this work on say
MySQL or Pg, I'd need to rewrite it, and there's not a lot of SQL so I am
willing to take that risk.

Expand All @@ -330,7 +330,7 @@ relationship:

touch share/migrations/SQLite/upgrade/2-3/003-alter_cd.sql

Then open that in your text editor, and lets bring over the last part from
Then open that in your text editor, and let's bring over the last part from
C<001-auto.sql>:

BEGIN;
Expand Down Expand Up @@ -365,7 +365,7 @@ upgrade script:

rm share/migrations/SQLite/upgrade/2-3/001-auto.sql

And you'd probably wish to commit these files now if you are using an source
And you'd probably wish to commit these files now if you are using a source
control system (and if you are not, prepare for pain!)

Next step will be to perform the upgrade.
Expand All @@ -376,7 +376,7 @@ Next step will be to perform the upgrade.
Schema is 3
Deployed database is 2

SO the database is in the expected state. If you've been messing with the data
So the database is in the expected state. If you've been messing with the data
you might which to 'clean' things up, with something like:

## optional step to 'clean up' database
Expand All @@ -397,7 +397,7 @@ Let's go ahead with the upgrade:
Schema is 3
Deployed database is 3

Looks good. Lets peek in the database and do a quick sanity check. You should
Looks good. Let's peek in the database and do a quick sanity check. You should
also consider writing some test cases similar to what we did in the previous
section:

Expand All @@ -408,9 +408,9 @@ section:
Enter SQL statements terminated with a ";"

sqlite> .tables
artist country
artist country
artist_cd dbix_class_deploymenthandler_versions
cd track
cd track

sqlite> select * from artist_cd;
1|1
Expand All @@ -435,15 +435,15 @@ good and making sure you made good downgrades (we didn't :) ) or you can force
install the database to an arbitrary older version and start again with the
C<prepare> command. We will take this second option for this tutorial.

So for the purposes of our tutorial, lets say that suddenly we realize our
design for the C<country> table is terrible wrong. We've been putting real
So for the purposes of our tutorial, let's say that suddenly we realize our
design for the C<country> table is terribly wrong. We've been putting real
country names in the table, and in English, but now we want to internationalize
our site. That means we should avoid English words in our seed data, and
instead use normalized codes that our UI layer can use and leverage existing
internationalization and localization tools against. So we need to change that
country table, and do so in a way to make sure we keep our existing country
information correct. Lastly, we want to add a few new countries to the list
as well as one more artist to the system. Thats a bunch of changes, so lets
as well as one more artist to the system. That's a bunch of changes, so let's
get to it!

Changes to be made:
Expand Down Expand Up @@ -522,10 +522,10 @@ blowing away anything important. You'd expect some output like so:
Overwriting existing DDL file - ...
Your Database version must be lower than than your schema version
in order to prepare upgrades / downgrades
Copying Fixture Confs from .../MusicBase/share/fixtures/2/conf to
Copying Fixture Confs from .../MusicBase/share/fixtures/2/conf to
.../MusicBase/share/fixtures/3/conf

Output above has been abbreviated a bit to highlight the important information.
The output above has been abbreviated a bit to highlight the important information.
Don't worry about that "Copying Fixture Confs from ..." overwriting any of your
custom changes, if there is a file in the target directory matching we just
skip the copy (we always assume if the file is there that you may have made
Expand Down Expand Up @@ -554,7 +554,7 @@ Now, let's look at the new C<share/migrations/SQLite/upgrade/2-3/001-auto.sql>
DROP TABLE country_temp_alter;

Again, to be brief I've only included above the new statements related to our
changes to the C<country> table. This actually seems pretty good. Lets
changes to the C<country> table. This actually seems pretty good. Let's
break that out into a separate file, and add some statements to move data
from the old name to the new code columns:

Expand Down Expand Up @@ -604,7 +604,7 @@ Don't forget to delete the C<001-auto.sql> file:

rm share/migrations/SQLite/upgrade/2-3/001-auto.sql

In order to complete our new requirements, lets create some Perl run files to
In order to complete our new requirements, let's create some Perl run files to
add some new country codes, and one new Artist:

touch share/migrations/SQLite/upgrade/2-3/005-new_countries.pl
Expand Down Expand Up @@ -651,13 +651,13 @@ your editor and add the following:

You might notice that the relationship names in C<006-new_artist.pl> don't
exactly match those in our schema. As I mentioned before, this is because the
C<$schema> that is passed as the first (and only) argument to your anonymous
C<$schema> that is passed as the first (and only) argument to your anonymous
subroutinues is NOT the schema that comes from L<MusicBase::Schema> but instead
it is generated directly from the database using L<DBIx::Class::Schema::Loader>
it is generated directly from the database using L<DBIx::Class::Schema::Loader>.
This is because your schema is going to change a lot, we can't rely on it always
being backwardly compatible with every version of the database.

If you every get confused about what the auto generated schema looks like, you
If you ever get confused about what the auto generated schema looks like, you
can always use the C<make_schema> command:

## example command, don't need to run as part of the tutorial
Expand All @@ -670,7 +670,7 @@ dump to STDOUT the generated classes:
## example command, don't need to run as part of the tutorial
export DBIC_MIGRATION_DEBUG=1

Everything is ready to go. Lets run the upgrade:
Everything is ready to go. Let's run the upgrade:

dbic-migration -Ilib upgrade

Expand All @@ -682,7 +682,7 @@ As before, let's peek inside the database for a quick sanity check:
Enter SQL statements terminated with a ";"

sqlite> .tables
artist country
artist country
artist_cd dbix_class_deploymenthandler_versions
cd track

Expand Down Expand Up @@ -723,7 +723,7 @@ Then open that in you text editor, add the following:
fixtures_ok ['all_tables'];

is Country->count, 6,
'Correct Number of Tests';
'Correct Number of Countries';

ok my $artist = Artist->first,
'Got one artist';
Expand All @@ -737,18 +737,18 @@ Then open that in you text editor, add the following:
done_testing;

So this is just a basic test to see that all the new countries exist and that
the new many to many between Artist and Cd works. Lets run the test suite:
the new many to many between Artist and Cd works. Let's run the test suite:

prove -l t

This gives you:

t/more-than-1.t ... 1/? DBIx::Class::ResultSet::count(): No such relationship cd_rs on Artist
t/more-than-1.t ... 1/? DBIx::Class::ResultSet::count(): No such relationship cd_rs on Artist
t/more-than-1.t ... Dubious, test returned 9 (wstat 2304, 0x900)
All 1 subtests passed
All 1 subtests passed
t/upgrade-1to2.t .. skipped: not correct schema version
t/upgrade-2to3.t .. ok
t/use.t ........... ok
t/upgrade-2to3.t .. ok
t/use.t ........... ok

Test Summary Report
-------------------
Expand Down Expand Up @@ -790,10 +790,10 @@ Open C<lib/MusicBase/Schema/ResultSet/Artist.pm> and make the following change:
We just change the join condition to match the new relationship, and try again:

$prove -l t
t/more-than-1.t ... ok
t/more-than-1.t ... ok
t/upgrade-1to2.t .. skipped: not correct schema version
t/upgrade-2to3.t .. ok
t/use.t ........... ok
t/upgrade-2to3.t .. ok
t/use.t ........... ok
All tests successful.
Files=4, Tests=7, 2 wallclock secs ( ... )
Result: PASS
Expand Down