2

I need to check if an application can process files in different languages.

What I've thought so far is making a file for each language with the characters on their alphabets and a small fragment of a book and see if the application can process it.

However this approach leaves out punctuation symbols and other properties those languages might have.

Do you have any experiences or advices on that?

Thanks for your responses.

Some of the languages are:

  • English
  • German
  • Japanese
  • Mandarin (Chinese)
  • Vietnamese
  • Russian
  • Arabic
  • Farsi
1

Asian languages such as Japanese, Chinese and Koreans do have alphabets, but their characters are not written in alphabets.

So your alphabet approach will not work as well as you expected.

What I suggest you do is:

  • Based on the background of your application, E.G. (is it a military themed game? is it a to do list application? is it an online ticketing system? is it a word processing application?) Google mostly common-used characters in each language for that background.
  • If it is a generic application, Google "most common Chinese characters" or "most common phrase in Russian" and use the results in your files.
  • Punctuation wise, test comma, finish, quote, double quotes, for most languages, those punctuation should be the same. You can extend your tests if you have enough time and resources.

You have not told us what "processing a file" does for your application, if you can add it, I may come up with a bit more suggestions.

| improve this answer | |
  • Thanks for your response. By processing a file I mean: Reading from a plain or encoded text file, parsing it to an object, and then store the object and the plain or encoded text file contents to a database – Edwin Jaime Sep 7 '16 at 17:17
2

You could make a file that has all characters possible for a given language. Also you may need to check languages that can be read top to bottom or right to left are handled as expected.

Mandarin is a spoken language so you are probably verifying either simplified or traditional Chinese.

| improve this answer | |
  • That depends on what the app does with the data. Imagine an app that loads a resume, breaks it up into words, stems the words, and then adds the words to an index. The logic not only needs to handle characters from different parts of the Unicode table but also needs to employ locale-specific logic for work-breaking and stemming. – user246 Sep 7 '16 at 11:56
  • Sure but "process a file" is pretty vague. I can only with with the info I have. – David Cain Sep 7 '16 at 16:26
  • Thanks for your response. By processing a file I mean: Reading from a plain or encoded text file, parsing it to an object, and then store the object and the plain or encoded text file contents to a database – Edwin Jaime Sep 7 '16 at 17:18
1

Most of not all language signs are part of the Unicode characters sets. If the application processes UTF-8 and UTF-16 sets you will be fine in nearly all cases.

The development team should only use UTF save functions so you as a tester do not have to worry about this. Let them read: http://utf8everywhere.org/

If you have a legacy application where parts are not yet UTF compatible, then I would create a single test file which contains a good spread of languages and symbols to test this.

Some sources for test-data:

Related questions:

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.