Skip to content

Database error handling #3

@elliotblackburn

Description

@elliotblackburn

I can see in services/dbhandler.js that we are just swallowing errors if the query returns one. The whole db querying seems a bit peculiar for node. Generally we'd make a query which would then have a callback and inside that callback we would then format the response and call the entire operations callback. A bit like below

var pgQuery  = require('pg-query'); // nice wrapper library to pg

function getSomeData(id, callback) {
  var query = 'SELECT * FROM myTable WHERE id=$1';
  var params = [ id ];

  pgQuery(query, params, function(err, rows, results) {
    if(err) {
      return callback(err);
    }
    //Format the results somehow...
    callback(null, results);
  });
}

Then somewhere, updateProfile would be called somewhat like this.

function show(req, res, next) {
  var id = req.params.id;
  db.getSomeData(id, function(err, theData) {
    res.status(200).json(theData);
  }
}

I'm curious as to why we're swallowing errors, is this an oversight or is this done for a reason?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions