Vietnamese in FM7

[4everJang]

Hello all,

I received a set of translated files in FM7 format, but I cannot figure out how to get the Vietnamese character set activated for non-Unicode stuff on Windows XP. Can anyone point me in the right direction ?

Thanks

Jang

[Bob_Niland]

What was the platform (I'd guess Windows) and what was the font (I'd guess a codepage font, where the code points for the Viet glyphs map into Roman space)?

Do you have or did the document provider supply the font?

I've not dealt with this migration problem yet, but I wouldn't be surprised if Frame only automatically changes old code points to Unicode for its known special characters (Dingbats, Symbol and perhaps some of the Roman extensions).

For anything else, you may need to retain the legacy font.

I did a little test in FM7.0/WindowsXP, setting up a para fmt using a circa 1993 Cyrillic font I had on hand.

<PgfTag `Body-cyr'>

   <FPlatformName `W.WP CyrillicA.R.400'>

   <FFamily `WP CyrillicA'>

This is presumably a code page 1251 font, but possibly an older 855 or 866.

For a later Frame version to be able to remap the codepoints to Cyrillic space in Unicode, it would need to get a hint somewhere, such as:

• encoded in the document file, or
• look in the font data structure itself, which would require that the font be installed on the system

Looking at the MIF, in the structure for my trial <PgfTag `Body-cyr'>, I was surprised to see:

  <FEncoding `FrameRoman'>

  <FLanguage NoLanguage>

No sign of a hint about code page 855, 866 or 1251 anywhere.

[Wiedenmaier]

Hi Jang,

if a virtual font, like Arial Vietnamese (Windows Codepage 1258), is used you can activate this virtual font via your registry:

HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\FontSubstitutes

Arial Vietnamese,163 = Arial,163

perhaps you have to restart FM and/or your OS

If you use another font, which supports codepage Windows-1258 (Vietname), replace Arial with the other font name.

Hope this helps

Markus

[4everJang · Accepted solution]

Update on the font problem:

The translators used a special Vietnamese font and told us to purchase that font to make the translation come out correct on our computer (don't you just love those agencies that tell you it is never their problem ?). In the end, I found free VNI fonts that resemble the Arial and Times fonts enough. Check out www.vnisoft.com - VNI-Helve looks enough like Arial and there is a VNI-Times as well.

I had to then 'manually' change the fonts in the paragrah formats to pick up the new font (another thing the translation agency should have done in their FM files before sending them back to me, so that the missing fonts message signals there is something still to be done before sending the stuff out to the customers), which is easy enough by a global search/replace in the MIF files. I also found out that the translation agency had managed to change the paths to all the images (i.e. not changed the paths back to what they were before), so I had to also change all references in the MIFs.

In the end the document really looked Vietnamese, so the customer will be happy when receiving the finished product, and a little less happy when receiving the invoice for the extra work that was involved.

Wouldn't we all be much better off (but maybe jobless) if there had not been a tower of Bable long ago... ?

Jang

[Bob_Niland]

> Wouldn't we all be much better off (but maybe jobless) if there had not been a tower of Bable long ago... ?

If we all still spoke only Sumerian, we'd still need documentation - only the profession of translation would not exist.

Given that we didn't manage to avoid the Bable babble, the font codepage/overlay connundrum was pretty much unavoidable, due to the high cost of RAM and ROM in the first few decades of computing (not to mention the high cost of keyboards, strikers and print balls in the teletype age before that - the first code sets were five bits or less - often couldn't even handle case in a single language).

[4everJang]

Yes, I remember working in 4 KBytes of RAM for my first programs - that was all the memory you could rely on, as it was built into the parallel RISC processor I happened to start my computer career with. In our little startup techie company, building small programs in assembly language turned out to be my favourite challenge: I managed to build a fully functional program that was no more than 5 bytes long (clearing the memory of the processor). Yes, bytes, not MB, not KB, but bytes. Those were the (challenging) days. Now we are facing different challenges. The translation agency challenge is not one of my favourites, though.

Jang