bundle / cri/option_parser / foreman interaction error #1243
Labels
Comments
|
I’ve experienced
My preference goes out to the former. Can you let me know whether or not either of these approaches work? |
|
Oh interesting. I didn't catch the double-bundle, that's a good point. I tried it both ways and still no luck, tho: with I also tried deleting my whole bundle directory and re-bundling in case I had a bad old gem around or something, but no difference. Does |
|
Can you share the crash.log for that error? I don't recognize it and would like to investigate. Can you also run "gem pristine --all" and see whether or not the error persists? |
|
this is after running |
|
Can you confirm that you’ve copy-pasted the entire I’m wondering whether this problem is related to the use of |
|
Can you change your Gemfile to load cri from GitHub ( gem 'cri', github: 'ddfreyne/cri'… and check whether the problem still exist? |
|
|
Some questions:
This error has me stumped. |
|
@jm3 I’m still at a loss w.r.t this error — can you answer the questions above? |
|
I deleted the whole bundle and re-bundled to make sure that it was clean, using your git branch source for I tried the Procfile with bundle exec and without bundle exec, and pasted the errors above. |
|
Can you try removing
|
result: no change; removing foreman from the bundle yields the same error, guard 2.14.1 => pry 0.11.2 => Error: wrong argument type Nanoc::CLI::CleaningStream (expected File) |
|
There are two separate issues here:
The You can get around that, I believe, by configuring stdout/stdin for pry: Pry.config.input = STDIN
Pry.config.output = STDOUTThat should get rid of the The |
|
@jm3 What error do you get when you run without pry? (Remove pry from the Gemfile) |
|
Nevermind that question — pry is a dependency so you can’t run without it. However, changing the Procfile to read rather than (thus, add I still cannot reproduce the |
|
@jm3 Do you have a reproduction case for me? I fail to find out what’s wrong. |
|
Oh wow, ugh. It looks like it was the Thanks for your help isolating that. Starts fine now, with and without |
|
Glad to hear! By the way, I’m working on making the live-recompilation part of Nanoc more integrated (easier to set up and use, as well as faster). If you don’t mind living on the edge, you can:
This will be quite similar to your setup with Guard. Some things will still change before this feature is properly released:
|
|
Very cool, I'm trying it out now. I followed the instructions (github gem source, new gem deps, re-bundle, my Guardfile looks like this: |
|
Whoops, that’s meant to be You won’t need a Guardfile anymore after this, but please keep it for now, as this feature is still experimental! |
|
Also, this commit will make it not look like you’ll get the police after you — “illegal option” is confusing and jargon-y. |
|
OK cool, that helped. FYI, if the directory "output" doesn't exist, this error is thrown: I created output and it fixed it. Nanoc loads and reloads now, but the actual site isn't reloading in the browser when something changes — it requires a manual refresh in Chrome. Is that the expected behavior? |
|
Oops, thanks for reporting the Chrome should automatically reload. Does the console show anything that’s out of the ordinary? |
|
Reported as #1254.
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
bundle exec foreman startused to correctly runnanoc viewwithout complaint; now nanoc errors with acri/option_parsererror. Running the two individual commands in foreman's Procfile works fine:Procfile:
% bundle exec guard
% bundle exec nanoc view
but
% bundle exec foreman start
which runs the same two commands, now bombs with a
cri/option_parseerror:Ruby version 2.4.0, Gemfile:
crash.log
The text was updated successfully, but these errors were encountered: