Paul Sadauskas

We also have this issue when trying to upgrade to the npm module version with webpacker. Similar error, but with a different css node_module: ``` ERROR in ./app/javascript/packs/active_admin.scss Module build...

Thanks, that was helpful, I didn't know about `!`. Is there any way to tell rofi to not try and execute the script on first launch when the input is...

@jgreben Are you able try on Ruby 2.7.x or 3.x, see if you get different results? I wonder if its a 2.6 thing... Also, what version of ActiveSupport is it...

Ah OK. Looks like Rails 5.2 doesn't do the `if block_given?` part. https://github.com/rails/rails/blob/v5.2.6/activesupport/lib/active_support/notifications/instrumenter.rb#L23 Looks like this was broken by a rubocop cleanup? https://github.com/httprb/http/commit/f4fb3369d030520e981eb891fb5dd25f8e0403d2#diff-bf40ca1851dc97b2ef1fd9b85e44d722ad130f7042df0d8c1c12738d2b44b873L32 cc @ixti

Seems pretty great, I like the "fix" to how addressable handles `#join`. I'm often doing something like: ```ruby base = A::URI.parse("https://foo.example/api/v2") http.get(base.join("users")) # 404 on https://foo.example/api/users ``` ... and I...

Looks like the server is returning the `Content-Length` header twice: ``` $ http https://imagery.pragprog.com/products/595/tpp20.jpg HTTP/1.1 200 OK ACCEPT_RANGES: bytes Accept-Ranges: bytes Age: 4603039 CONTENT_LENGTH: 2802494 Cache-Control: public max-age=315360000 Connection: keep-alive...

I seem to recall reading somewhere that some headers allow multiple values separated by some delimiter, and if one of those headers is specified multiple times then the client should...

@ixti I think the issue is that there's http-level errors that do generate a response that the Features can "see", but there's another class of errors below the http level...

Reduced to a simple test case: ```ruby require "thread" require "http" @http = HTTP.headers("X-Test" => "1") 5.times.map do Thread.new { r = @http.post("https://httpbin.org/post"); puts r.status} end.map(&:join) ``` This breaks in...

@tarcieri So do you think the HTTP gem should allow the simple test case in my comment to work or not? (The one where I re-use the same `@http` ivar...