NotZelda Posts - Answer Overflow

NotZelda

Explore posts from servers

•Created by NotZelda on 3/13/2025 in #❓-help

What are the best practices for keeping TRPC and ts-server fast?

I decided to migrate my entire app to TRPC over a couple days as I identified it was a great solution for my use case. However, now my typescript auto-complete and typing checking appear to be much slower. I think TRPC is still the best solution, I'm just curious if there are good practices to follow to ensure that my ts-server stays fast as I scale even further. ts-go may help with this in the future, but for now I'm stuck waiting a few seconds for intellisense. Anyone have any resources I can review?

6 replies

TtRPC

•Created by NotZelda on 3/3/2025 in #❓-help

How to define per route staleTime on queryClient?

It looks like I am unable to override the default stale time provided to react query client, no matter how I define the the route. Is TRPC overriting these dafult somewhere down the chain?

export const createQueryClient = () => {
  const queryClient = new QueryClient({
    defaultOptions: {
      queries: {
        // With SSR, we usually want to set some default staleTime
        // above 0 to avoid refetching immediately on the client
        staleTime: 1000 * 60 * 30,
      },
      dehydrate: {
        serializeData: SuperJSON.serialize,
        shouldDehydrateQuery: (query) =>
          defaultShouldDehydrateQuery(query) ||
          query.state.status === "pending",
      },
      hydrate: {
        deserializeData: SuperJSON.deserialize,
      },
    },
  });

  queryClient.setQueryDefaults(["performanceMetrics", "getEmployeeMetrics"], {
    staleTime: Infinity,
  });

  return queryClient;
};

export const createQueryClient = () => {
  const queryClient = new QueryClient({
    defaultOptions: {
      queries: {
        // With SSR, we usually want to set some default staleTime
        // above 0 to avoid refetching immediately on the client
        staleTime: 1000 * 60 * 30,
      },
      dehydrate: {
        serializeData: SuperJSON.serialize,
        shouldDehydrateQuery: (query) =>
          defaultShouldDehydrateQuery(query) ||
          query.state.status === "pending",
      },
      hydrate: {
        deserializeData: SuperJSON.deserialize,
      },
    },
  });

  queryClient.setQueryDefaults(["performanceMetrics", "getEmployeeMetrics"], {
    staleTime: Infinity,
  });

  return queryClient;
};

By creating the queryClient in this way, I would expect the api route: api.performanceMetrics.getEmployeeMetrics to have an infinite stale time, but it still coming in with the default 30 seconds. I would like to avoid having to wrap every route with my own definition of staleTime, however I also want to make sure no matter why I call these hooks in my app they all have the same staleTime.

8 replies

DTDrizzle Team

•Created by NotZelda on 12/3/2024 in #help

How to dynamically select all columns from one table, and join a single column from another table?

Its pretty often that I need to grab some details from a relational table and add it in to my original query. However, when I do this, I then need to explicitly call out all columns on the original table, and compose in my values from the join table. This can become tedious when I have many columns or when a column is added on the original table. I was hoping I might be able to spread out my reference to the PgTable and just add in one other key for the other table, but this isn't a valid query. Wrong

    const existingReport = await db
      .select(
        {
          ...reports,
          templateName: fileUpload.originalFileName,
        }
      )
      .from(reports)
      .orderBy(desc(reports.lastUpdated))
      .limit(1)
      .leftJoin(fileUpload, eq(reports.linkedFileID, fileUpload.id));

    const existingReport = await db
      .select(
        {
          ...reports,
          templateName: fileUpload.originalFileName,
        }
      )
      .from(reports)
      .orderBy(desc(reports.lastUpdated))
      .limit(1)
      .leftJoin(fileUpload, eq(reports.linkedFileID, fileUpload.id));

"Correct"

const existingDetails = await db
    .select({
      id: reports.id,
      name: reports.name,
      linkedFileID: reports.linkedFileID,
      effectiveDate: reports.effectiveDate,
      uploadedBy: reports.uploadedBy,
      lastUpdated: reports.lastUpdated,
      createdOn: reports.createdOn,
      nextRelease: reports.nextRelease,
      templateName: fileUpload.originalFileName,
    })
    .from(reportDetails)
    .where(eq(reportDetails.name, reportName))
    .leftJoin(fileUpload, eq(reportDetails.template, fileUpload.id));

const existingDetails = await db
    .select({
      id: reports.id,
      name: reports.name,
      linkedFileID: reports.linkedFileID,
      effectiveDate: reports.effectiveDate,
      uploadedBy: reports.uploadedBy,
      lastUpdated: reports.lastUpdated,
      createdOn: reports.createdOn,
      nextRelease: reports.nextRelease,
      templateName: fileUpload.originalFileName,
    })
    .from(reportDetails)
    .where(eq(reportDetails.name, reportName))
    .leftJoin(fileUpload, eq(reportDetails.template, fileUpload.id));

I am almost certain I am missing something here to be able to generate my select from the table schema itself. Does anyone know what the proper methods are to achieve this?

3 replies

Gaming

Programming