The Unicode Consortium Discussion Forum (CLOSED)

Right way of coding internationalized documents
Page 1 of 1

Author:  msta [ Fri Jan 04, 2013 12:08 am ]
Post subject:  Right way of coding internationalized documents


We are now trying to develop simple program, which is supposed to parse some texts into earlier prepared document templates. These teplates are prepared in RTF format but I believe it is not so important there and also the same problem would occur to HTML templates.

Let say a teplate is supposed to work as an invoice. People may see the standard texts and tables and may feel the empty places in the template with some texts which are written in a desktop software.

The program is written using of course some local software tools with local fonts, characters etc.
Our goal is to assure that the program will work properly for anybody (using any language, operating system etc).

I believe there is no problem with preparing a template, because the templete is supposed to be perepared by user (uning MS Word or WordPad). So standard (fixed) texts in templetes should be no problem. They even do not need to be written using unicode characters as we may be sure that users operating system will take care that he is preparing template in using right language codes. Anyway we alco cannot exect that use will write a template using unicode.

But what about texts, which are supposed to be parsed into the template? We will supply as a default some english templates but user should be able to feel these templates with texts in his original language. besides still he may writte his own templates.

I am not expecting any detailed guidance what to do. We only need some hints, which may help us to see it from right perspective. We will not be able to test how our program will work with some unknown for us languages (arablic, chineese etc.)



Author:  asmus [ Sun Jan 13, 2013 5:21 pm ]
Post subject:  Re: Right way of coding internationalized documents

For testing purposes, you can prepare strings in your own language and insert the letter "a" or "x" or whatever, at every other location. That way, testers can tell whether they were corrupted, or whether they were replaced by non-localized strings, without the need to learn a new language.

If you like, you can find sets of 26 Chinese characters even, which look like Latin letters (you may need to squint a bit). If you substitute those, you can still "read" the text in English, but you would make sure that the software correctly handles "Chinese" characters.

Page 1 of 1 All times are UTC - 6 hours [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group