Show usage on missing or wrong command.

[johannhof]

It's pretty annoying that you only get a stack trace when simply running web-ext without any arguments for the first time, so I added help messages.

[coveralls]

Coverage increased (+0.0007%) to 99.818% when pulling 2a54587 on johannhof:master into 03b1b7a on mozilla:master.

[kumar303]

I think it's better to do this with yargs.strict(). Could you try it? It was started here but abandoned: #169 Also, it needs to work for incorrect sub-command arguments. Example: web-ext build --not-an-option should show an error.

[kumar303]

Were you trying on web-ext 1.0.1 or one built against master? When I try against master it's pretty good but it needs this patch to handle sub-command options:

diff --git a/src/program.js b/src/program.js
index 7b21b16..2f03ae4 100644
--- a/src/program.js
+++ b/src/program.js
@@ -37,7 +37,7 @@ export class Program {
       }
       // Calling env will be unnecessary after
       // https://github.com/yargs/yargs/issues/486 is fixed
-      return yargs.env(envPrefix).options(commandOptions);
+      return yargs.strict().env(envPrefix).options(commandOptions);
     });
     this.commands[name] = executor;
     return this;

[kumar303]

^ that causes the test process to exit though because yargs exits the process. My idea for dealing with that was: #169 (comment)

[johannhof]

@kumar303 pushed another approach which works for me, anything wrong with that one? Tests seem to be passing.

[coveralls]

Coverage remained the same at 99.827% when pulling f6a2a26 on johannhof:master into fd2c4d6 on mozilla:master.

[johannhof]

The thing that looks ugly on master for me is simply typing web-ext without any sub-commands, but now it should look good in all cases.

[kumar303]

Thanks! Looks good and the patch works for me. It doesn't address #169 but we can do that in a separate patch (it's a related but separate issue).