What you need for the help question you are asking is to make sure that the catalog files (and by the way, the HTML help pages, if you decide to have it localised too) are displayed using the same codepage in which they were composed. E.g. my translation of FreeCOM is best suited to CP 850/858.
As far as I know, what usually happens with MS-DOS is that the CONFIG.SYS/AUTOEXEC.BAT automatically generates code to support certain codepage. E.g., in my Win98 it says
=== CONFIG.SYS
DEVICE=DISPLAY.SYS CON=(EGA,437,1)
=== AUTOEXEC.BAT
MODE CON CODEPAGE PREPARE=((850) C:\WINDOWS\COMMAND\EGA.CPI)
MODE CON CODEPAGE SELECT=850
===
(as a side note, Matthias noticed long time ago that this is quite not correct, and last line in autoexec should be replaced with
NLSFUNC
CHCP 850
because as it is, both kernel and PRINTER.SYS are living in 437, so when you create files or print files you risk to lose data; note however that in FreeDOS we do not have a NLSFUNC, so we have to do similar to my Win98 shown above).
Then in practice Microsoft ensures that the help pages, the program messages and the text files have been composed using the codepage 850. But you may still have problems if you get a, say, Portuguese text file to be read on a Spanish MS-DOS distro, just because Portuguese use 860 (most western Europe use 850, but for Portuguese you need some extra characters, same for scandinavian langs, slavic langs).
Thus, you never know. You cross your fingers and hope that user's AUTOEXEC/CONFIG select the appropriate codepage. For Germans (as for Spanish, Italian, French and many more) that is 850, although I always recommend to use 858, which is EXACTLY equal to 850, except that it contains the EURO sign (and drops the Turkish dotless i, useless for us, and 850 is useless for Turkish anyway).
I currently ignore wether codepages are selected in the post-install CONFIG/AUTOEXEC script. I seem to recall that Henrique Peron made a list of what AUTOEXEC/CONFIG should have for each country, so it could be there already.
And second just ask the people that write the strings for you wether they use the same codepage that Henrique adscribed in the table (the most appropriate for each country).
Aitor
Robert Platt escribi�:
Hi folks,
I could do with advice and hints on internationalisation, to make sure I do it correctly.
HTML Help has no international support at present, and I was planning to tackle this next. And now I have had an enquiry into creating a German translation, so I now really want to get a move on.
My plan is to add NLS using kitten, so that help messages can have translated versions.
But my main query is that I want to support html character entities. These will allow translations to use accents and non-English character sets. But the ASCII extended character sets vary with language, they are not standard. So how can I make sure html help matches a given character (for example scharfes s) to the correct ASCII code?
Regards Rob
------------------------------------------------------- This SF.Net email is sponsored by: IBM Linux Tutorials Free Linux tutorial presented by Daniel Robbins, President and CEO of GenToo technologies. Learn everything from fundamentals to system administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click _______________________________________________ Freedos-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freedos-devel
