fix: prevent Lambda from erroring out if Crawler is already running (#11)

Author: Anders

src/index.js

@@ -2,9 +2,16 @@ const AWS = require('aws-sdk');
 
 exports.handler = function(event, context, callback) {
     const glue = new AWS.Glue();
+    
     glue.startCrawler({ Name: process.env.CRAWLER_NAME }, function(err, data) {
         if (err) {
-            throw new Error(err);
+            // Check if Crawler is already running
+            const response = JSON.parse(this.httpResponse.body);
+            if (response['__type'] == 'CrawlerRunningException') {
+                console.log('Crawler already running; ignoring trigger.');
+
+                callback(null, response.Message);
+            }
         }
         else {
             console.log("Successfully triggered crawler");