# Globalize1 migration path, scopes and namespaces

**URL:** <https://www.ruby-forum.com/t/globalize1-migration-path-scopes-and-namespaces/184690>\
**Category:** Rails I18n\
**Created:** [March 6, 2010, 7:45pm UTC](https://www.ruby-forum.com/t/globalize1-migration-path-scopes-and-namespaces/184690 "2010-03-06T19:45:04Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![JSSS\_rgen\_FeSSSSlme](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/e56c9b/32.png) [@JSSS\_rgen\_FeSSSSlme](https://www.ruby-forum.com/u/JSSS_rgen_FeSSSSlme)\
**Post date:** [March 6, 2010, 7:45pm UTC](https://www.ruby-forum.com/t/globalize1-migration-path-scopes-and-namespaces/184690/1 "2010-03-06T19:45:04Z")

</div>

Hi,

I am working on this plugin I am calling globalize\_bridge. It is meant  
to provide a seamless migration path for rails apps using Globalize  
(1) and utilizing the built in rails I18n mechanism. I am pretty much  
done covering the entire scope of mapping the Globalize API, however,  
only Globalize namespaces still make me scratch my head.

The problem is: with a translation string “Foo. And Bar.”.t,  
containing periods, the “.” is mistaken for the scope separator.  
Luckily, later versions of the I18n gem have the scope\_separator  
directive, to support gettext, etc. So this come handy for Globalize1  
usage, however, Globalize also had the concept of namespaces (~ I18n  
scopes), and it would be a shame if we could port the namespaces over.

Here is an example:

“Foo. And Bar.”.tn(:my\_namespace) # Globalize1 syntax

# translates into this

I18n.t “Foo. And Bar.”, :separator =\> “\001”, :scope =\> :my\_namespace

So this does not pick up the correct translation and ignores the scope  
altogether.

I was looking for a seamless solution. However, the only workaround I  
can think of is to define a secondary separator e.g. :separator =\> “|”  
tweak my translations yml/db.

Anyone having a better idea?

Greetings,  
Juergen

---

<div class="post-metadata">

**Author:** ![JSSS\_rgen\_FeSSSSlme](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/e56c9b/32.png) [@JSSS\_rgen\_FeSSSSlme](https://www.ruby-forum.com/u/JSSS_rgen_FeSSSSlme)\
**Post date:** [March 6, 2010, 9:06pm UTC](https://www.ruby-forum.com/t/globalize1-migration-path-scopes-and-namespaces/184690/2 "2010-03-06T21:06:26Z")

</div>

On Sat, Mar 6, 2010 at 19:40, Jürgen Feßlmeier  
[removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote:

> “Foo. And Bar.”.tn(:my\_namespace) # Globalize1 syntax
> 
> # translates into this
> 
> I18n.t “Foo. And Bar.”, :separator =\> “\001”, :scope =\> :my\_namespace
> 
> So this does not pick up the correct translation and ignores the scope  
> altogether.

One solution would be to replace the translation keys, e.g. with a  
slugalized version of the string. So you’d use “foo\_and\_bar” in this  
case. ActiveSupport has String#parameterize for that.

Then you could stick the original “Foo. And Bar.” string into a yml  
file instead (YAMLator may be helpful in this:  
[Helps you update Rails i18n YAML files programmatically. · GitHub](http://gist.github.com/293581)).

Sounds like a useful plugin.

---

<div class="post-metadata">

**Author:** ![JSSS\_rgen\_FeSSSSlme](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/e56c9b/32.png) [@JSSS\_rgen\_FeSSSSlme](https://www.ruby-forum.com/u/JSSS_rgen_FeSSSSlme)\
**Post date:** [March 6, 2010, 9:41pm UTC](https://www.ruby-forum.com/t/globalize1-migration-path-scopes-and-namespaces/184690/3 "2010-03-06T21:41:32Z")

</div>

Hi Henrik,

that’s a pretty smart idea!

The plugin utilizes I18n::Backend::Database plugin as the  
ViewTranslation replacement by default, so the string is automatically  
added. However, YAMLator could be an interesting tool if one decides  
to add a special backend and add translations automatically to the YML  
files.

Just need to make sure that the user understands that the string in  
all or some cases is slugalized if not automatically written and how  
to deal with it (what it is converted to, etc.).

Thanks for your input.

Greetings,  
Juergen