guyellis / http-status-check Goto Github PK

NYC provides better coverage for ES6 and has become a more active project than Istanbul.

This issue is to swap out Istanbul for NYC:

1. Add nyc dependency
2. Remove istanbul dependency
3. Change any references in package.json or anywhere elsewhere from istanbul to nyc.
4. Add a .nycrc file with similar settings to .istanbul.yml
5. Update .gitignore
6. Remove .istanbul.yml

This commit from the plant project is a good reference for what needs to be done to complete this issue.

In uriCheck.js there's a startsWith function. Replace this with the equivalent function from string.js

In uriCheck.js we set followRedirect to false. This should be configurable per site and default to false if missing which will allow for checking collections of shortUrls and their ultimate destinations.

Add a data collector that keeps the results of each of run.

• Check that project runs on Node 5.1.0
• Add this version to to engines in package.json
• Check for running version of Node in a test and app start against version in package.json and warn if version mismatch?

Add gulp-typescript (or other strategy) to allow compile checking of code as a type of syntax checker and linter.

Add something like node-cron that allows http-status-check to continuously run at set intervals.

This will be particularly useful if you wanted to use http-status-check as a URL monitoring tool.

If the out-adapter was setup to send failure reports to email or phone then you would be alerted quickly if there was a problem.

Version 10 of Node.js (code name Dubnium) has been released! 🎊

To see what happens to your code in Node.js 10, Greenkeeper has created a branch with the following changes:

• Added the new Node.js version to your .travis.yml
• The new Node.js version is in-range for the engines in 1 of your package.json files, so that was left alone

If you’re interested in upgrading this repo to Node.js 10, you can open a PR with these changes. Please note that this issue is just intended as a friendly reminder and the PR as a possible starting point for getting your code running on Node.js 10.

Your Greenkeeper Bot 🌴

Lots of the syntax in this project might be more readable if in ES6 syntax. Make one or more changes to the code to change it from ES5 syntax to ES6. Where can we use ES6 classes for example? This issue relies on #31 being done first.

Create tests for Twilio Out Adapter.

Add a CHANGELOG.md file. Add info about last version change.

There might be value in adding https://github.com/addyosmani/psi/ to this project for more health reporting. E.g. min score drops below x then fail check.

 3 failing

  1) inAdapters/fileAdapter/ getRunData() should return checksites.js data if exists:

      AssertionError: expected 'samplesites' to equal 'checksites'
      + expected - actual

      -samplesites
      +checksites
      
      at Context.<anonymous> (test/lib/inAdapters/test.fileAdapter.js:35:27)

  2) inAdapters/fileAdapter/ getRunData() should return samplesites.js data if exists and if checksites.js is missing:
     Error: Expecting HTTP_STATUS_CHECK_FILE environment or checksites.js or samplesites.js to exist.
      at findFile (lib/inAdapters/fileAdapter.js:19:11)
      at Object.getRunData (lib/inAdapters/fileAdapter.js:24:14)
      at Context.<anonymous> (test/lib/inAdapters/test.fileAdapter.js:56:31)

  3) inAdapters/fileAdapter/ getRunData() should throw an exception if neither samplesites.js or checksites.js exist:

      AssertionError: expected 2 to equal 3
      + expected - actual

      -2
      +3
      
      at Context.<anonymous> (test/lib/inAdapters/test.fileAdapter.js:79:34)

fix http not existing, now is https

$ npm run start

> [email protected] start /opt/pocs/http-status-check
> node index.js

X Comentarismo API (http://www.linksilk.com) failed. Here are the problems:
  - Expected header 'location' to be http://linksilk.com/ but instead it was https://linksilk.com/
A total of 6 URIs were tested.
Failure count:  1
Success count:  4
Disable count:  1

Increase coverage:

=============================== Coverage summary ===============================
Statements   : 78.71% ( 196/249 )
Branches     : 73.27% ( 74/101 )
Functions    : 82.35% ( 28/34 )
Lines        : 78.63% ( 195/248 )
================================================================================

By adding test.emailAdapter.js Now coverage is:

=============================== Coverage summary ===============================
Statements   : 93.57% ( 233/249 )
Branches     : 81.19% ( 82/101 )
Functions    : 91.18% ( 31/34 )
Lines        : 93.55% ( 232/248 )
================================================================================

Try the alexa100.js file in sitelists/ directory to see that many requests end up with ECONNREFUSED. I suspect this is because there are too many simultaneous requests to the same cluster.

Add a flag to the configuration to surpress the output when it's successful. (Useful to reduce the noise if you have a large list of sites and several are failing.)

• Add Babel 6 to project.
• Add appropriate initialization calls/references so that tests and other files so that ES6 is available.
• Add an ES6 function (not in Node*) to test suite to confirm that it works.
• Add an ES6 function (not in Node*) to target code to confirm that it works.
• Node has some ES6 functionality. To test that Babel is doing the transpiling add something that Node doesn't have.

Currently the configuration of how the utility will run is lumped together with the list of sites to check. Split this apart so that the list of sites can be pulled from a different source and the config of how the utility runs is separate.

Allow a config item on each site entry in the input that is called 'control' or something similar. Defaults to false if missing. This tells the script that we expect this site to be always available and use it to test our internet connection. e.g. google.com, yahoo.com, microsoft.com should always return 200 or 301. If they timeout then there's something wrong with our connectivity and we shouldn't assume a problem with the other sites. Run the control sites at the beginning and at the end.

Move the files that have lists of sites to crawl into their own directory and allow the input source of files to be passed on the command line.

Tests for MongoDB in Adapter.

Write tests for the Email Out Adapter. Mock the email sending with Sinon.js so emails aren't sent when the tests are run.

Create an out adapter that sends the output to recipients via Twilio.

Create an adapter that reads the input (the URLs that need to be scanned) from a MongoDB.

Add a web server that provides a UI into the results of the runs and the ability to add, remove and manage sites that are run.

Complete the Email Out Adapter that's been started so that email(s) get sent to receipients at the end of the run.