jamesshore / lets_code_javascript Goto Github PK

please cancel my subscription

My Ubuntu 15.04 automatically upgraded Firefox to v38.0, which breaks the selenium smoke tests.
The selenium-bug is already reported to Selenium, see SeleniumHQ/selenium#437.

This issue isn't with selenium though, but with the crappy error messages in the smoke tests, and I know how you feel about crappy error messages :).

First time I run the smoke tests I get:

mattias@maas-linux-1:~/git/lets_code_javascript$ ./jake.sh loose=true smoketest
Collating client files: .
Bundling client files with Browserify: ..
Smoke testing app:


  ․․

  1 passing (2s)
  1 failing

  1) Smoke test "before all" hook:
     Uncaught Error: Failed to install profile; firefox terminated with Result(code=null, signal=SIGSEGV)
      at Error (native)
      at Array.forEach (native)
  From: Task: WebDriver.createSession()
      at src/_smoke_test.js:26:14
      at Socket.<anonymous> (src/_run_server.js:18:55)
      at readableAddChunk (_stream_readable.js:163:16)
      at Socket.Readable.push (_stream_readable.js:126:10)
      at Pipe.onread (net.js:538:20)

This message isn't the worst I've seen. At least it mentions both firefox and WebDriver. If I re-run the command though (which I often tend to do when I'm unfocused), the server on port 5000 hasn't shutdown correctly and I instead get:

mattias@maas-linux-1:~/git/lets_code_javascript$ ./jake.sh loose=true smoketest
Collating client files: .
Bundling client files with Browserify: ..
Smoke testing app:


  ․․

  1 passing (11s)
  1 failing

  1) Smoke test "before all" hook:
     Error: timeout of 10000ms exceeded. Ensure the done() callback is being called in this test.

Now it takes 10 seconds to time out and the error message is really vague. It wasn't obvious to me I suddenly had an abandoned server running on port 5000.

Hi James,

I just tried to make use of jake but found really weird issue that jake is unable to load jakefile. Do you have any idea how to resolve it.

Here is trace log:
./jake.sh --trace
jake aborted.
Error: No Jakefile. Specify a valid path with -f/--jakefile, or place one in the current directory.
at api.fail ((/opt/lampp/htdocs/jake_test/node_modules/jake/lib/api.js:221:18)
at _loadJakefile ((/opt/lampp/htdocs/jake_test/node_modules/jake/lib/loader.js:56:11)
at load ((/opt/lampp/htdocs/jake_test/node_modules/jake/lib/loader.js:93:5)
at Object. ((/opt/lampp/htdocs/jake_test/node_modules/jake/bin/cli.js:67:10)
at Module._compile (module.js:456:26)
at Object.Module._extensions..js (module.js:474:10)
at Module.load (module.js:356:32)
at Function.Module._load (module.js:312:12)
at Function.Module.runMain (module.js:497:10)
at startup (node.js:119:16)

Help really appriciated.

I recently updated npm to version 3 and noticed that it now puts additional meta data into the package.json files beneath node_modules/**/. This meta data now includes absolute paths to the project directory, which makes it impossible to check those files into Git without causing merge conflicts.

I started using npm shrinkwrap in my own projects. This creates an additional file that locks down all package versions to the currently installed ones. It doesn't solve the issue of being independent of third party services (AFAIK you'd need to set up a proxy for that) but at least it prevents different package versions caused by different checkout times.