DiamondDragon Posts - Answer Overflow

•Created by DiamondDragon on 12/17/2024 in #help

Can't upsert with onConflictDoUpdate & generatedAlwaysAsIdentity ID columns?

I've noticed the type system doesn't recognize the ID columns (since they are supposed to be autogenerated?) How would one upsert or update? The ID field is not available

export const folders = pgTable('folders', {
    id: integer().primaryKey().generatedAlwaysAsIdentity({ startWith: 1000 }),
    name: text('name').notNull(),
    createdBy: integer('created_by')
        .notNull()
        .references(() => users.id),
    parentFolderId: integer('parent_folder_id').references((): AnyPgColumn => folders.id),
})

const db = createDb()
// no id is available to pass to values per TS
await db.insert(folders).values({ name: 'Test', createdBy: 1000 }).onConflictDoUpdate({
    // where is ID passed in?
    target: folders.id, // no type error
    set: { name: 'Test2' },
})

export const folders = pgTable('folders', {
    id: integer().primaryKey().generatedAlwaysAsIdentity({ startWith: 1000 }),
    name: text('name').notNull(),
    createdBy: integer('created_by')
        .notNull()
        .references(() => users.id),
    parentFolderId: integer('parent_folder_id').references((): AnyPgColumn => folders.id),
})

const db = createDb()
// no id is available to pass to values per TS
await db.insert(folders).values({ name: 'Test', createdBy: 1000 }).onConflictDoUpdate({
    // where is ID passed in?
    target: folders.id, // no type error
    set: { name: 'Test2' },
})

2 replies

•Created by DiamondDragon on 12/11/2024 in #support-community

Returning data from Tanstack Query useMutation wrapper

3 replies

•Created by DiamondDragon on 12/3/2024 in #support-community

Zod Unions appear to break `Table` helper (maybe zodToConvex related?)

1 replies

•Created by DiamondDragon on 12/1/2024 in #support-community

Handling Upserts with Optional System Fields

Not sure if this is the best practice here but didn't appear there's a way of getting the Id field to be optional so it can be passed in for upserts (if it doesn't exist, insert, if it does patch). Not sure if my types are entirely correct but seems to work but happy to hear if anyone else agrees

function upsertDeal(ctx: MutationCtx, args: MakeIdOptional<Deals>) {
    if (args._id) { // type _id = string & {} | undefined
        return ctx.db.patch(args._id, args)
    }

    return insertDeal(ctx, args)
}


// Make only _id optional and exclude _creationTime for upserts
export type MakeIdOptional<T> = Omit<T, '_id' | '_creationTime'> & {
    _id?: T extends { _id: Id<any> } ? Id<any> : never
}

function upsertDeal(ctx: MutationCtx, args: MakeIdOptional<Deals>) {
    if (args._id) { // type _id = string & {} | undefined
        return ctx.db.patch(args._id, args)
    }

    return insertDeal(ctx, args)
}


// Make only _id optional and exclude _creationTime for upserts
export type MakeIdOptional<T> = Omit<T, '_id' | '_creationTime'> & {
    _id?: T extends { _id: Id<any> } ? Id<any> : never
}

3 replies

•Created by DiamondDragon on 11/26/2024 in #support-community

HTTP Path Suffix Not Working?

7 replies

•Created by DiamondDragon on 10/16/2024 in #support-community

Unable to get Tanstack Trellaux Start working

7 replies

•Created by DiamondDragon on 9/30/2024 in #workers-help

Error on remote worker: APIError: A request to the Cloudflare API

I'm trying to test out a browser deployed on worker in a DO and getting this error

✘ [ERROR] Error on remote worker: APIError: A request to the Cloudflare API (/accounts/5aa6cae4d0260b6aae9d255f3b63242f/workers/scripts/playground-worker/edge-preview) failed.

      at throwFetchError
  (/Users/am/Coding/playground/node_modules/wrangler/wrangler-dist/cli.js:159817:18)
      at fetchResult
  (/Users/am/Coding/playground/node_modules/wrangler/wrangler-dist/cli.js:159730:5)
      at process.processTicksAndRejections
  (node:internal/process/task_queues:95:5)
      at async createPreviewToken
  (/Users/am/Coding/playground/node_modules/wrangler/wrangler-dist/cli.js:174177:29)
      at async createWorkerPreview
  (/Users/am/Coding/playground/node_modules/wrangler/wrangler-dist/cli.js:174198:17)
      at async start
  (/Users/am/Coding/playground/node_modules/wrangler/wrangler-dist/cli.js:174924:34)
  {
    text: 'A request to the Cloudflare API
  (/accounts/5aa6cae4d0260b6aae9d255f3b63242f/workers/scripts/playground-worker/edge-preview)
  failed.',
    notes: [
      {
        text: 'An unknown error has occured. If this
  error persists, please file a report in the
  workers-sdk GitHub repository
  (https://github.com/cloudflare/workers-sdk/issues) or
  reach out to your account team. [code: 10002]'
      }
    ],
    location: undefined,
    kind: 'error',
    code: 10002,
    accountTag: '5aa6cae4d0260b6aae9d255f3b63242f'
  }

✘ [ERROR] Error on remote worker: APIError: A request to the Cloudflare API (/accounts/5aa6cae4d0260b6aae9d255f3b63242f/workers/scripts/playground-worker/edge-preview) failed.

      at throwFetchError
  (/Users/am/Coding/playground/node_modules/wrangler/wrangler-dist/cli.js:159817:18)
      at fetchResult
  (/Users/am/Coding/playground/node_modules/wrangler/wrangler-dist/cli.js:159730:5)
      at process.processTicksAndRejections
  (node:internal/process/task_queues:95:5)
      at async createPreviewToken
  (/Users/am/Coding/playground/node_modules/wrangler/wrangler-dist/cli.js:174177:29)
      at async createWorkerPreview
  (/Users/am/Coding/playground/node_modules/wrangler/wrangler-dist/cli.js:174198:17)
      at async start
  (/Users/am/Coding/playground/node_modules/wrangler/wrangler-dist/cli.js:174924:34)
  {
    text: 'A request to the Cloudflare API
  (/accounts/5aa6cae4d0260b6aae9d255f3b63242f/workers/scripts/playground-worker/edge-preview)
  failed.',
    notes: [
      {
        text: 'An unknown error has occured. If this
  error persists, please file a report in the
  workers-sdk GitHub repository
  (https://github.com/cloudflare/workers-sdk/issues) or
  reach out to your account team. [code: 10002]'
      }
    ],
    location: undefined,
    kind: 'error',
    code: 10002,
    accountTag: '5aa6cae4d0260b6aae9d255f3b63242f'
  }

My toml

name = "playground-worker"
main = "src/index.ts"
compatibility_date = "2024-09-23"
compatibility_flags = [ "nodejs_compat" ]

observability = { enabled = true }

workers_dev = false
send_metrics = false
browser = { binding = "MYBROWSER" }

[durable_objects]
bindings = [
    { name = "BROWSER_DO", class_name = "BrowserDurableObject" }
]

[[migrations]]
tag = "v1" # Should be unique for each entry
new_classes = ["BrowserDurableObject"]

[triggers]
# every 5 minutes
crons = ["*/5 * * * *"]

name = "playground-worker"
main = "src/index.ts"
compatibility_date = "2024-09-23"
compatibility_flags = [ "nodejs_compat" ]

observability = { enabled = true }

workers_dev = false
send_metrics = false
browser = { binding = "MYBROWSER" }

[durable_objects]
bindings = [
    { name = "BROWSER_DO", class_name = "BrowserDurableObject" }
]

[[migrations]]
tag = "v1" # Should be unique for each entry
new_classes = ["BrowserDurableObject"]

[triggers]
# every 5 minutes
crons = ["*/5 * * * *"]

6 replies

•Created by DiamondDragon on 9/6/2024 in #help

Drizzle Zod refining not working?

Shouldn't this code for zod make an optional field firstName required? https://drizzle.run/t2zowcjw5f1xbb5ny7zq2j3n

3 replies

•Created by DiamondDragon on 8/31/2024 in #help

Dynamic function get use typed column names using getTableConfig

I have both UUIDs and NanoIDs on every one of my tables. I am considering adding utils to map from uuid -> nanoID & vice versa. how would i make a dynamic function like this? basically under the assumption every table has a id and shortId column which is the nanoid

export async function getUuidFromShortId(table: PgTable, shortId: string) {
    const { columns, name, schema } = getTableConfig(table)
    const [result] = await db
        .select({ id: columns.id })
        .from(name)
        .where(eq(columns.shortId, shortId))
        .limit(1)
    return result[0]?.id ?? null
}

export async function getUuidFromShortId(table: PgTable, shortId: string) {
    const { columns, name, schema } = getTableConfig(table)
    const [result] = await db
        .select({ id: columns.id })
        .from(name)
        .where(eq(columns.shortId, shortId))
        .limit(1)
    return result[0]?.id ?? null
}

1 replies

•Created by DiamondDragon on 8/25/2024 in #pages-help

Error using API Create PagesProject with pulumi - Err 'You must set a production branch.'

Calling this from a monorepo. I want to use pulumi to create a pages project/update it

import { PagesProject, type PagesProjectArgs } from '@pulumi/cloudflare'

const projectName = 'spa-cf'
const CLOUDFLARE_ACCOUNT_ID = process.env.CLOUDFLARE_ACCOUNT_ID
const CLOUDFLARE_API_TOKEN = process.env.CLOUDFLARE_API_TOKEN

const pagesProjectConfig: PagesProjectArgs = {
    accountId: CLOUDFLARE_ACCOUNT_ID,
    name: projectName,
    productionBranch: 'main',
    buildConfig: {
        buildCaching: true,
        buildCommand: 'bun run build',
        destinationDir: '../apps/spa/dist',
        rootDir: '../apps/spa',
    },
}

export const pagesProject = new PagesProject(projectName, pagesProjectConfig)

🚀 ~ file: pages.ts:58 ~ createProject ~ url: https://api.cloudflare.com/client/v4/accounts/xxxx/pages/projects
🚀 ~ file: pages.ts:58 ~ createProject ~ request: {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    Authorization: 'Bearer XXX'
  },
  body: '{"accountId":"XXXX","name":"spa-cf","productionBranch":"main","buildConfig":{"buildCaching":true,"buildCommand":"bun run build","destinationDir":"../apps/spa/dist","rootDir":"../apps/spa"}}'

import { PagesProject, type PagesProjectArgs } from '@pulumi/cloudflare'

const projectName = 'spa-cf'
const CLOUDFLARE_ACCOUNT_ID = process.env.CLOUDFLARE_ACCOUNT_ID
const CLOUDFLARE_API_TOKEN = process.env.CLOUDFLARE_API_TOKEN

const pagesProjectConfig: PagesProjectArgs = {
    accountId: CLOUDFLARE_ACCOUNT_ID,
    name: projectName,
    productionBranch: 'main',
    buildConfig: {
        buildCaching: true,
        buildCommand: 'bun run build',
        destinationDir: '../apps/spa/dist',
        rootDir: '../apps/spa',
    },
}

export const pagesProject = new PagesProject(projectName, pagesProjectConfig)

🚀 ~ file: pages.ts:58 ~ createProject ~ url: https://api.cloudflare.com/client/v4/accounts/xxxx/pages/projects
🚀 ~ file: pages.ts:58 ~ createProject ~ request: {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    Authorization: 'Bearer XXX'
  },
  body: '{"accountId":"XXXX","name":"spa-cf","productionBranch":"main","buildConfig":{"buildCaching":true,"buildCommand":"bun run build","destinationDir":"../apps/spa/dist","rootDir":"../apps/spa"}}'

Initial deployment triggered: {
  result: null,
  success: false,
  errors: [ { code: 8000033, message: 'You must set a production branch.' } ],
  messages: []
}

Initial deployment triggered: {
  result: null,
  success: false,
  errors: [ { code: 8000033, message: 'You must set a production branch.' } ],
  messages: []
}

4 replies

•Created by DiamondDragon on 6/20/2024 in #help

Use results of a column to pass where conditions elsewhere in the query?

I have a field that stores filters that I parse into the SQL conditions for a where clause. Or at least, that's the goal. That field is in my views table. Is it possible to pass the result of that field in views into a nested where condition?

export async function getTableData(opts: { db: Db; tableShortId: string; viewShortId: string }) {

    const { db, tableShortId, viewShortId } = opts

    const tableData = await db.query.views.findFirst({
        where: and(eq(views.shortId, viewShortId), eq(views.tableId, tableShortId)),
        columns: {
            shortId: true,
        },
        with: {
            // Get the view for the table
            table: {
                columns: { shortId: true },
                with: {
                    fields: {
                        columns: { shortId: true },
                    },
                    records: {
                                                //! is it possible to filter records by view.jsonColumn after being parsed
                        columns: { shortId: true },
                        with: {

export async function getTableData(opts: { db: Db; tableShortId: string; viewShortId: string }) {

    const { db, tableShortId, viewShortId } = opts

    const tableData = await db.query.views.findFirst({
        where: and(eq(views.shortId, viewShortId), eq(views.tableId, tableShortId)),
        columns: {
            shortId: true,
        },
        with: {
            // Get the view for the table
            table: {
                columns: { shortId: true },
                with: {
                    fields: {
                        columns: { shortId: true },
                    },
                    records: {
                                                //! is it possible to filter records by view.jsonColumn after being parsed
                        columns: { shortId: true },
                        with: {

1 replies

•Created by DiamondDragon on 6/10/2024 in #help

construction a dynamic query builder generically ?

Ok i'm not really sure how to ask this but I'm trying to allow a user the ability in UI to build out a query builder where they can add many filters with potentially nested filters with AND/OR conjuctions. Like airtable. Anyone have any patterns how to do this? I had attempted to try to construct a mapping on the SQL operators and the drizzle functions but I can't get the types to work . I'm playing aroung just with the sql operator to try making ti work with more raw SQL. trying to wrap my head around this and its a pattern i've seen few apps implement and never seen any code online to use as a guide Ideally a user can do a query in the UI like below and the queries can be constructed dynamically.

Filter employes where 
  name = john AND
  salary >= 50,000 AND
  relatedDepartment contains 'billing' AND
  ( department_city does not contain NYC OR
    whatever... equals blah
  )

Filter employes where 
  name = john AND
  salary >= 50,000 AND
  relatedDepartment contains 'billing' AND
  ( department_city does not contain NYC OR
    whatever... equals blah
  )

const operators: Record<OperatorNames, Operator> = {
    contains: {
        label: 'Contains',
        name: 'contains',
        sql: 'LIKE',
        drizzle: (column: Column, value: string | SQLWrapper): typeof like => {
            return like(column, value)
        },
    },
        // other operators...
}

const operators: Record<OperatorNames, Operator> = {
    contains: {
        label: 'Contains',
        name: 'contains',
        sql: 'LIKE',
        drizzle: (column: Column, value: string | SQLWrapper): typeof like => {
            return like(column, value)
        },
    },
        // other operators...
}

1 replies

HHono

•Created by DiamondDragon on 6/8/2024 in #help

Serialization of Date?

Is there any suggested guidance on how to handle js/json serialization? I ran into an issue using the hono client where the types break due to a Date being typed as a string after being returned from a hono route I see there are no data transformers provided by hono, so wondering what suggestions there are. https://discord.com/channels/1011308539819597844/1248521918554963979

26 replies

•Created by DiamondDragon on 5/19/2024 in #help

Losing custom $types in pg jsonb using drizzle-zod

Not sure if this is an expected limitation of zod here

export const records = pgTable('records', {
    id: uuid('id')
        .primaryKey()
        .$defaultFn(() => uuidv7()),
    tableId: uuid('table_id')
        .references(() => tables.id, { onDelete: 'cascade' })
        .notNull(),
    organizationId: uuid('organization_id')
        .references(() => organizations.id, { onDelete: 'cascade' })
        .notNull(),
    recordByFieldId: jsonb('record_by_field_id').$type<RecordData>(),

})

export const records = pgTable('records', {
    id: uuid('id')
        .primaryKey()
        .$defaultFn(() => uuidv7()),
    tableId: uuid('table_id')
        .references(() => tables.id, { onDelete: 'cascade' })
        .notNull(),
    organizationId: uuid('organization_id')
        .references(() => organizations.id, { onDelete: 'cascade' })
        .notNull(),
    recordByFieldId: jsonb('record_by_field_id').$type<RecordData>(),

})

produces using $inferInsert

type Record = {
    organizationId: string;
    tableId: string;
    id?: string;
    recordByFieldId?: RecordData | null;
    createdAt?: Date;
    updatedAt?: Date | null;
}

type Record = {
    organizationId: string;
    tableId: string;
    id?: string;
    recordByFieldId?: RecordData | null;
    createdAt?: Date;
    updatedAt?: Date | null;
}

But when using drizzle zod, i lose RecordData

export const insertRecordSchema = createInsertSchema(records)
export type InsertRecord = z.infer<typeof insertRecordSchema>

// results in
type InsertRecord = {
    organizationId: string;
    tableId: string;
    id?: string;
    recordByFieldId?: any;
    createdAt?: Date;
    updatedAt?: Date | null;
}

export const insertRecordSchema = createInsertSchema(records)
export type InsertRecord = z.infer<typeof insertRecordSchema>

// results in
type InsertRecord = {
    organizationId: string;
    tableId: string;
    id?: string;
    recordByFieldId?: any;
    createdAt?: Date;
    updatedAt?: Date | null;
}

1 replies

•Created by DiamondDragon on 4/23/2024 in #pages-help

Can't deploy a nextjs pages site in subdirectory

1 replies

•Created by DiamondDragon on 4/17/2024 in #help

Generated SQL error / type error with operations on unique index

1 replies

•Created by DiamondDragon on 4/16/2024 in #help

Drizzle studio shows as blank page on Arc browser

1 replies

•Created by DiamondDragon on 4/13/2024 in #workers-help

cron not running locally on the trigger specified

I'm using hono here but when i go to http://localhost:8787/__scheduled?cron=***** i see the response log

Hello
[wrangler:inf] GET /__scheduled 200 OK (11ms)
╭─────────────────────────────────────────────────────

Hello
[wrangler:inf] GET /__scheduled 200 OK (11ms)
╭─────────────────────────────────────────────────────

but I dont see anymore subsequent logs of the trigger again. am i doing anything wrong?

async function scheduledHandler(
    event: ScheduledEvent,
    env: Env,
    ctx: ExecutionContext,
) {
    const testMessages = [
        'Hello',
        'Goodbye',
        'Welcome',
        'Farewell',
        'Greetings',
        'I am leaving',
    ]
    ctx.waitUntil(
        (async () => {
            const randomIndex = Math.floor(Math.random() * testMessages.length)
            const randomMessage = testMessages[randomIndex]
            console.log(randomMessage)
        })(),
    )
}

export default { fetch: app.fetch.bind(app), scheduled: scheduledHandler }

async function scheduledHandler(
    event: ScheduledEvent,
    env: Env,
    ctx: ExecutionContext,
) {
    const testMessages = [
        'Hello',
        'Goodbye',
        'Welcome',
        'Farewell',
        'Greetings',
        'I am leaving',
    ]
    ctx.waitUntil(
        (async () => {
            const randomIndex = Math.floor(Math.random() * testMessages.length)
            const randomMessage = testMessages[randomIndex]
            console.log(randomMessage)
        })(),
    )
}

export default { fetch: app.fetch.bind(app), scheduled: scheduledHandler }

4 replies