Casing

Durcno lets you write all identifiers in camelCase — the natural TypeScript convention — and automatically converts them to snake_case when generating SQL and migrations.

How it works

Whenever you define a table, column, enum, or sequence, Durcno converts the name you provide to snake_case before using it in any SQL statement or migration file.

export const BlogPosts = table("public", "blogPosts", {
  id: pk(),
  authorId: bigint({ notNull }),
  publishedAt: timestamp({ notNull }),
});

The generated SQL uses snake_case throughout:

CREATE TABLE "public"."blog_posts" (
  "id" bigserial PRIMARY KEY,
  "author_id" bigint NOT NULL,
  "published_at" timestamp NOT NULL
);

You write TypeScript-style names; Durcno takes care of the rest.

What gets converted

Every identifier you pass to a Durcno schema builder is automatically converted:

Builder	Identifier(s) converted
table(schema, name, ...)	schema, table name, column names
enumtype(schema, name, ...)	schema, enum name
sequence(schema, name, ...)	schema, sequence name

Enum values (e.g. ["admin", "user"]) are not converted — they are stored in the database exactly as written.

Conversion rules

The conversion follows standard camelCase-to-snake_case rules:

You write	SQL uses
blogPosts	blog_posts
firstName	first_name
userRole	user_role
publishedAt	published_at
XMLParser	xml_parser
parseHTTPResponse	parse_http_response

Consecutive uppercase sequences (acronyms) are handled correctly: XMLParser becomes xml_parser, not x_m_l_parser.

Column names in queries

When querying, results are returned with the camelCase keys you defined — not the snake_case names from the database. This keeps your application code consistent with your schema definitions.

const posts = await db.from(BlogPosts).select();

// Result keys match your schema definition
posts[0].authorId; // ✅
posts[0].author_id; // ❌ TypeScript error

Naming recommendations

• Table names: camelCase (blogPosts, userProfiles)
• Column names: camelCase (firstName, createdAt)
• Enum names: camelCase (userRole, orderStatus)
• Schema names: camelCase (public, appData)

tip

You never need to write blog_posts or first_name anywhere in your schema. Stick to camelCase everywhere and let Durcno handle the PostgreSQL side.