We keep improving our software on a daily basis. We use Continuous Deployment in our development process (maybe I’ll write another blog entry on that topic sometime soon) to deliver new functionality as soon as it is developed and tested without any delays. Once in a while, we’ll combine some of the interesting features we’ve launched recently and blog about them.
Our Translation Delivery Network allows anyone to translate and localize any site in weeks or days compared to months with the regular approach of extracting and translating strings. The integration is really lightweight. All you need to do is add special CSS classes around what to not to translate. For example, you may have a personalized greeting that changes depending on who is currently logged in such as:
You don’t want to translate every single combination of this string that may potentially appear. There can be thousands of different names after all!
The typical integration looks like this:
When translators translate this they will see this string with a non-editable placeholder in place of a name (with a usage example inside of a placeholder):
However in some cases our customers can’t make this change in the code base right away. The engineering team could be busy working on other projects, a code freeze is in place, etc.
With the new Pattern Matching functionality, the localization manager can do it from the Smartling dashboard. To get there: load Content tab to get string approval queue, click on “multiple edit” => you will see “Pattern Match” option. Select a few strings that have similar content to be pattern matched:
Pattern matching is a powerful functionality packed into a very easy-to-use interface. It allows you to create patterns for common phrases with variable placeholders, such as “Welcome, Andrey”, “2 comments”, or “Page 2 of 10”. The static part of the string is only translated once (“Welcome” or “comment”), excluding the variable part, while providing the translator the flexibility of changing sentence structure if needed. It minimizes the amount of content that must be translated, reducing costs and time to market.