@graphile-contrib/pg-simplify-inflector
This plugin simplifies field names in the PostGraphile schema; e.g. allUsers
becomes simply users
, User.postsByAuthorId
becomes simply User.posts
, and
Post.userByAuthorId
becomes simply Post.author
.
Adding this plugin to your schema is almost certainly a breaking change, so do
it before you ship anything! This is the primary reason this isn't enabled by
default in PostGraphile.
This plugin is recommended for all PostGraphile users.
Customising
This plugin is implemented as a single JS file that does not need to be compiled
at all - you can simply copy it into your project and customise it as you see
fit.
Alternatively, you can
write your own inflection plugin.
Changes:
Given these tables:
create table companies (
id serial primary key,
name text not null
);
create table beverages (
id serial primary key,
company_id int not null references companies,
distributor_id int references companies,
name text not null
);
Query.allCompanies
👉 Query.companies
(disable via
pgSimplifyAllRows = false
)
Query.allBeverages
👉 Query.beverages
Beverage.companyByCompanyId
👉 Beverage.company
Beverage.companyByDistributorId
👉 Beverage.distributor
Company.beveragesByCompanyId
👉 Company.beverages
(because the
company_id
column follows the [table_name]_id
naming convention)
- All update mutations now accept
patch
instead of companyPatch
/
beveragePatch
(disable via pgSimplifyPatch = false
)
- If you are using
pgSimpleCollections = "only"
then you can set
pgOmitListSuffix = true
to omit the List
suffix
- Fields where the singular and plural are the same and a distinct plural is
required are force-pluralised ("fishes") to avoid conflicts (e.g.
singularize("fish") === pluralize("fish")
).
Note: Company.beveragesByDistributorId
will remain, because distributor_id
does not follow the [table_name]_id
naming convention, but you could rename
this yourself with a smart comment:
comment on constraint "beverages_distributor_id_fkey" on "beverages" is
E'@foreignFieldName distributedBeverages';
or with a custom inflector:
module.exports = makeAddInflectorsPlugin(
{
getOppositeBaseName(baseName) {
return (
{
// These are the default opposites
parent: "child",
child: "parent",
author: "authored",
editor: "edited",
reviewer: "reviewed",
// 👇 Add/customise this line:
distributor: "distributed",
}[baseName] || null
);
},
},
true
);
Installation:
yarn add @graphile-contrib/pg-simplify-inflector
or
npm install --save @graphile-contrib/pg-simplify-inflector
Usage:
CLI:
postgraphile --append-plugins @graphile-contrib/pg-simplify-inflector
Library:
const PgSimplifyInflectorPlugin = require("@graphile-contrib/pg-simplify-inflector");
// ...
app.use(
postgraphile(process.env.AUTH_DATABASE_URL, "app_public", {
appendPlugins: [PgSimplifyInflectorPlugin],
// Optional customisation
graphileBuildOptions: {
/*
* Uncomment if you want simple collections to lose the 'List' suffix
* (and connections to gain a 'Connection' suffix).
*/
//pgOmitListSuffix: true,
/*
* Uncomment if you want 'userPatch' instead of 'patch' in update
* mutations.
*/
//pgSimplifyPatch: false,
/*
* Uncomment if you want 'allUsers' instead of 'users' at root level.
*/
//pgSimplifyAllRows: false,
/*
* Uncomment if you want primary key queries and mutations to have
* `ById` (or similar) suffix; and the `nodeId` queries/mutations
* to lose their `ByNodeId` suffix.
*/
// pgShortPk: true,
},
// ... other settings ...
})
);
Naming your foreign key fields
By naming your foreign key along the lines of author_id
or author_fk
, e.g.:
CREATE TABLE posts (
id serial primary key,
author_id int not null references users,
...
);
We can automatically extract the field prefix: author
and call the relation
author
rather than the default: user
. This allows for a post to have an
author
, editor
, reviewer
, etc. all which point to users
.
The reverse, however, is not so easy. On the User type, we can't call the
reverse of all these different relations posts
. The default inflector refers
to these as postsByAuthorId
, postsByEditorId
, etc. However we'd rather use
shorter names, so we introduce a new inflector: getOppositeBaseName
. This
inflector is passed a baseName (the part without the _id
/_fk
suffix, e.g.
author
, editor
, reviewer
above) and should return the opposite of that
base name which will be prepended to the target type to produce, e.g.
authoredPosts
, editedPosts
, reviewedPosts
. Failing this, we just fall back
to the default (verbose) inflector; it will be up to you to add smart comments
or a custom inflector to override these.
Handling field conflicts:
In most cases, the conflict errors will guide you on how to fix these issues
using smart comments.
Smart Tags
@foreignSimpleFieldName
@foreignSimpleFieldName
was added to override the naming of the foreign-side
of a one-to-many relationship's simple collections field (if you're using simple
collections). By default we'll take the @foreignFieldName
and add the "list
suffix" ("List" by default, but "" if pgOmitListSuffix
is set), but if you
prefer you can override it entirely with @foreignSimpleFieldName
. If you set
@foreignSimpleFieldName
and you're using simpleCollections 'both'
then you
should also set @foreignFieldName
explicitly or unexpected things may occur.
Applies to:
@listSuffix
@listSuffix
allows you to override the default naming on a per-entity basis,
overriding pgOmitListSuffix
. For example, with pgOmitListSuffix: true
, you
can apply @listSuffix include
to have the -List
suffix appended to the
simple collection generated for that table, and remove the -Connection
suffix
from the Relay connection. When pgOmitListSuffix
is not true
, you can use
@listSuffix omit
to selectively omit the -List
suffix on simple collections
and append -Connection
to the Relay connection instead.
If @listSuffix
is set, the only valid values are "omit"
and "include"
. Any
other value will cause an error.
| @listSuffix omit | @listSuffix include |
---|
Relay Connection | companiesConnection | companies |
Simple Collection | companies | companiesList |
NOTE: @listSuffix
will have no effect when using @foreignSimpleFieldName
.
Applies to:
- tables
- foreign key constraints
- computed column functions returning
SETOF <record type>