|0027 Codepage info|
|0027 NLS formatting|
|Positive number example|
|Negative number example|
|Digits after decimal separator|
|Negative format symbol|
|Number list separator|
|Leading and trailing zeros: 0.700 formatted|
|System of measurements|
|Default time format|
|Time format separator|
|All time formats|
|Default short date format|
|Default long date format|
|All short date formats|
|All long date formats|
NLSWeb originated as a reference tool for testers. When an application is tested to comply to local conventions (defined by user locale), may be handy to have a perfect sample. The etalon can be found by reading the NLS settings from the operating system NLS API. But the most complete information is available under Windows 2000 only - while it may be required when tests are running on Windows 95/98 or Windows Me.
The solution was found: collect the information on a Windows 2000 server and display it to everybody interested through a Web page. No data is collected on the client: perfect NLS information is not related on the platform where the test is run. On the server side, on the other hand, a lot of work has to be done. An ASP page was written.
NLS data fits well into tree structure of XML; this format leaves room for future expansion if NLS API grows; it is easily built and formatted. The drawback of XML format - it is not supported by all browsers, and those supporting it may be limited in formatting capabilities. That's why NLSWeb receives XML data from the ATL control, formats it using XSL filters, and sends HTML data to the client.
Normally, language-specific information is required for one language at a time; ATL control running on our page collects information for all languages. To keep the user from drawning in all that data, while providing an easy way to access any language information, the ASP used two XSL filters. First - the default one - formats the start page, showing a list of locales that server supports. The second, single-language filter , formats a language-specific page; it receives a single - language subset of the XML data that ATL.
The data sent is multi-lingual and encoded in UTF-8. Certain problems may show up when multi-lingual data is displayed on non-multilingual platforms. To resolve some of those problems, a Unicode font was created from Sylfaen font supplied with Microsoft Windows 2000. This font is embedded into the output HTML, so that you see less "default glyphs" (result of system's failure to find a symbol to display text) in the output. However, this solution works only with Microsoft Internet Explorer 4.0 or better. For more information on this technique and for the tool that generates Web fonts, check the link: http://www.microsoft.com/typography/web/embedding/weft3/default.htm.