Find strings that have not been wrapped for localization

Can you help me find and wrap strings for localization?

Problem

Programs typically have many source code files, sometimes hundreds. Some are written by the developer. Some are obtained online and incorporated into the app. Every source code file can contain strings that might potentially need to be wrapped with a localization function. Finding all those strings and wrapping them is tedious work and frequently strings that should be localized are missed. There has to something better than opening and scanning through every line of every file, and there is.

Solution

This is a task that can be improved through automation. Over time we have found many ways to automatically identify strings to present to the developer so that they can make a choice; localize, don’t localize, undecided. Over time we’ve seen many ways developers accidentally obfuscate strings. Each new pattern is added to the cumulative data by which apps are analyzed, continually expanding our ability to determine which strings should potentially be translated. The resulting dataset is an actively-evolving trade secret.

Impact

By isolating and displaying the unwrapped strings, the developer can quickly go through all their source code files and decide what to localize. Saves the developer a great deal of time compared to manually reviewing every line of code, in every source code file. Also, this process frequently finds old files that they thought they had been examined but still contain strings that need to be reviewed.

Status

Trade Secret