Knex js How to apply union on 2 tables which has different number of columns

See also questions close to this topic

• Why aren't knex migration not showing in psql database?

  I created knexfile.js

   module.exports = {
  
    development: {
      client: 'pg',
      connection: {
        filename: 'postgres://localhost/myDB'
      }
    }, 
  };
  

  Then I create migration

  npx knex migrate:make items
  

  And have this data

   exports.up = function (knex) {
      return knex.schema.createTable("items", (table) => {
        table.increments("id").primary();
        table.string("title").notNullable();
        table.string("description");
        table.decimal("price").notNullable();
        table.decimal("quantity").unsigned().notNullable();
        table.string("image");
      });
    }; 
    exports.down = function (knex) {
      return knex.schema.dropTableIfExists("items");
    };
  

  Then I run npx knex migrate:latest

  And it indicates that one was migrated

  Then I go into psql myDB, run \dt and nothing comes up!
• What is a set analogue from squel.js in knex.js?

  How can I rewrite it using Knex.js?

  type === 'insert' ?
          sql = squel.insert()
              .set('id', userId)
              .set('type', 'open')
              .set('server', server)
              .set('created_at', timestamp)
              .set('updated_at', timestamp)
              .into('saved_users') :
          sql = squel.update()
              .table('saved_users')
              .set('updated_at', timestamp)
              .where(`id = "${userId}"`)
  
      for (const param in player) {
          if (param === 'full_response') continue;
          const value = player[param];
          if (param === 'kill') {
              sql.set(`\`${param}\``, value ?? 0)
          } else {
              sql.set(param, value ?? 0)
          }
      }
  

  I'm using .set(..) to append the insert or update query. How to do it using Knex?
• Having trouble connecting apollo sql datasource with mssql; getting "Login failed for user <username>"

  I'm trying to send GraphQL queries to a mssql database utilizing the Apollo SQL datasource. I believe the error is in the configuration object, since it's recognizing that I'm trying to access the database. I tried looking through the knex documentation (since that's the library that the Apollo SQL datasource uses) but it doesn't have that much information about MSSQL implementation. I also tried installing tedious like it says to in the documentation, but I'm not sure how I would utilize it with the configuration object. Has anybody encountered something similar when trying to send GraphQL queries to a mssql database through apollo-server-express?

  **index.js**
  const db = require('./datasources/db');
  const dbConfig = {
    client: 'mssql',
    connection: {
      domain: <domain>,
      server: <server>,
      driver: <driver>,
      database: <dbname>,
      port: <port>,
      user: <user>,
      password: <pw>,
      requestTimeout: <timeout>,
      options: {
        encrypt: false,
        enableArithAbort: false
      }
    }
  }
  const database = new db(dbConfig);
  
  ...
  
  async function startApolloServer() {
        const httpServer = http.createServer(app);
        const server = new ApolloServer({
          typeDefs,
          resolvers,
          plugins: [ApolloServerPluginDrainHttpServer({ httpServer })],
          dataSources: () => ({ database })
        });
      
        await server.start();
        
        server.applyMiddleware({ app });
      
        await new Promise(resolve => httpServer.listen({ port: httpListenerPort }, resolve));
        console.log(`🚀 Server ready at http://localhost:${port}${server.graphqlPath}`);
  
        return { server, app };
      }
      
      startApolloServer();
  
  

  error message: Login failed for user '<user>'.
• C++ convert a union into an integer

  I have 2 type aliases:

  struct A { uint64_t i; };
  struct B { uint64_t j; };
  

  A and B are not the same type, and are not uint64_t primarily for readability reasons. They represent IDs of program resources that are inherently different (e.g. A represents the ID of an image, and B represents the ID of a raw buffer).

  For most of the program's lifetime, they are kept separate and do their own thing, however at one point they need to be serialized. To prevent writing identical logic for both, and to prevent using templates (long story), I made a union:

  union ResourceHandle {
      A a;
      B b;
  }
  

  Say I have this struct:

  struct MetaData
  {
      ResourceHandle handle;
      /* other data */
  }
  

  I want to write void Serialize(const MetaData& data);

  I know that the handle is a uint64_t, so I just want to cast the union into this type by adding:

  union ResourceHandle
  {
      A a;
      B b;
          
      operator uint64_t() const { return a; }
  };
  

  I suspect this is undefined behaviour, but I think it will generally work in most systems. Is there a way I can reliably cast from the union into a uint64_t without using additional memory to check which of the two is actually stored?
• Union data type In matlab/Simulink

  I am trying to port embedded driver files to Simulink/Matlab, written in C, using C caller block I was able to port functions that do not use a global union data type.

  My final aim is to generate a C code using the Embedded coder.

  How do I go about porting C functions which use Union data type within its functions?

  One idea is to change all unions into structure datatype. However, that would increase the memory size used.

  Any suggestions are welcome.