Problems with multiple domains setup

[MamesPalmero]

I am trying to configure gettext_i18n_rails to use multiple domains, as it says in the configuration. Here is my current configuration:

config/initializers/fast_gettext.rb

FastGettext.add_text_domain 'app', path: 'locale', type: :po
FastGettext.add_text_domain 'errors', path: 'locale', type: :po
FastGettext.default_available_locales = ['en', 'es'']
FastGettext.default_text_domain = 'app'

Object.send(:include, FastGettext::TranslationMultidomain)

lib/task/gettext.rake

Rake::Task["gettext:setup"].clear

namespace :gettext do
  task :setup => [:environment] do
    FastGettext.translation_repositories.each_key do |domain|
      GetText::Tools::Task.define do |task|
        task.package_name = domain
        task.package_version = "1.0.0"
        task.domain = domain
        task.po_base_directory = locale_path
        task.mo_base_directory = locale_path
        task.files = files_to_translate
        task.enable_description = false
        task.msgmerge_options = gettext_msgmerge_options
        task.msgcat_options = gettext_msgcat_options
        task.xgettext_options = gettext_xgettext_options
      end
    end
  end
end

and in any file of my project app/.../*.rb
d_("errors", "Error")

The problem is that after adding the languages and executing rake gettext: find, the .pot and .po files are generated empty. I think the locale/errors.pot andlocale/language/errors.po files should have an entry with msgid 'error" . Is there something wrong with my configuration or something that I did not understand correctly?

[grosser]

idk ... don't remember the details of how this is supposed to work :( ... I'd recommend opening up the gem to drop a few debugs/puts to find out where it's going wrong, please report your findings back here :)

…

On Tue, Aug 7, 2018 at 11:18 PM MamesPalmero ***@***.***> wrote: I am trying to configure gettext_i18n_rails to use multiple domains, as it says in the configuration. Here is my current configuration: *config/initializers/fast_gettext.rb* FastGettext.add_text_domain 'app', path: 'locale', type: :po FastGettext.add_text_domain 'errors', path: 'locale', type: :po FastGettext.default_available_locales = ['en', 'es''] FastGettext.default_text_domain = 'app' Object.send(:include, FastGettext::TranslationMultidomain) *lib/task/gettext.rake* Rake::Task["gettext:setup"].clear namespace :gettext do task :setup => [:environment] do FastGettext.translation_repositories.each_key do |domain| GetText::Tools::Task.define do |task| task.package_name = domain task.package_version = "1.0.0" task.domain = domain task.po_base_directory = locale_path task.mo_base_directory = locale_path task.files = files_to_translate task.enable_description = false task.msgmerge_options = gettext_msgmerge_options task.msgcat_options = gettext_msgcat_options task.xgettext_options = gettext_xgettext_options end end end end *and in any file of my project app/.../*.rb* d_("errors", "Error") The problem is that after adding the languages and executing rake gettext: find, the .pot and .po files are generated empty. I think the locale/errors.pot andlocale/language/errors.po files should have an entry with msgid 'error". Is there something wrong with my configuration or something that I did not understand correctly? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#176>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAAsZ0gZd7YnQoXKn_wrSddBZmoNzKAqks5uOb3rgaJpZM4VyeRt> .

[jstoup111]

I've got the same issue and tried to follow the code as much as I could before I started getting lost. I am assuming that the parser (based on my research) isn't including the multidomain methods that are used to translate. I'm happy to open a PR if I can be pointed in the right direction. I'm not sure it's even this gem that needs to be updated to support it.

[grosser]

sorry no idea ... just erm dig in and add docs / refactor things that are unclear on the way :(

[yorickpeterse]

It seems the problem is in lib/gettext_i18n_rails/ruby_gettext_extractor.rb, specifically in the method process_call:

def process_call exp
  _call = exp.shift
  _recv = process exp.shift
  meth  = exp.shift
                                    
  case meth
  when :_, :p_, :N_, :pgettext, :s_
    gettext_simple_call(exp)
  when :n_
    gettext_plural_call(exp)
  end
                                    
  until exp.empty? do
    process(exp.shift)
  end
                                    
  s()
end

This method seems to specifically ignore calls such as d_(), D_(), etc. Take this file for example:

_('foo')
D_('bar')

The above code will only find the first translation. If you change it to the following:

def process_call exp
  _call = exp.shift
  _recv = process exp.shift
  meth  = exp.shift
                                              
  case meth
  when :_, :p_, :N_, :pgettext, :s_, :d_, :D_
    gettext_simple_call(exp)
  when :n_
    gettext_plural_call(exp)
  end
                                              
  until exp.empty? do
    process(exp.shift)
  end
                                              
  s()
end

It will find both, producing:

[["foo", "/tmp/test.rb:1"], ["bar", "/tmp/test.rb:2"]]

Having said that, I'm not sure if this output is what is expected when using multi domain translation strings.

[grosser]

I don't remember touching / seeing this ... can you make a PR with the change you think is best 🤷‍♂️
(ideally not breaking anything else :D )

[yorickpeterse]

@grosser Unfortunately, I'm not sure if these changes would even be correct as I'm not familiar with what output is expected by the code that makes use of the above code.

[grosser]

can you come up with a test-case that fails now ? ... so we have something concrete that can be fixed by poking in the code