34 replies to this topic

[afinab]

I have to add a number of Vietnamese names to my v6.09 database, and they have some additional accents aside from the usual french ones. I have the fonts installed on my computer, and they are mapped to the numbers. I can get the names properly typed in Word and Wordpad, but not in TMG.

Number 1-4 are mapped to give the accented letters ă, â, ê, ô respectively and work fine, number 5 through 0 are mapped to apply an accent to a previous typed letter, but give me in TMG instead other already accented letters, or the accent as a separate character.

Does anyone have experience with this?

Thanks for any help,
Afina

[John Moran]

I have to add a number of Vietnamese names to my v6.09 database, and they have some additional accents aside from the usual french ones. I have the fonts installed on my computer, and they are mapped to the numbers. I can get the names properly typed in Word and Wordpad, but not in TMG.

Number 1-4 are mapped to give the accented letters ă, â, ê, ô respectively and work fine, number 5 through 0 are mapped to apply an accent to a previous typed letter, but give me in TMG instead other already accented letters, or the accent as a separate character.

Does anyone have experience with this?

Thanks for any help,
Afina

In WinXP, select Start>>Run
type in "Charmap" without the quotes

Repeat to start two instances of Character Map. Use Mouse in blue at top to separate the two Displays. Set one Charmap to Windows Western. Set the other to Windows Viet Nam.

This will let you see what you are up against: Namely two different character sets. Might as well be English vs Cyrillic or Greek. TMG only supports one 8 byte (256 character/font) set at a time. But maybe you can see the location of the characters you need and then fake it in Word by making two different printouts with two different languages selected, and then interleaving the sheets and throwing away the extras.

[afinab]

John,

Thanks for answering.

When I look at the character map in the two settings (̣Western and Vietnam) some letters are indeed missing in Western. (I usually look at the charactermap in Unicode).

However, if I set the language to Vietnamese on my computer (in control panel: regional and language options), I get the proper Vietnamese characters in Word, Wordpad, Excel, Firefox. So why not in TMG?
I'm new at this, but if TMG support a 256 character/font, and all the Alt +#### codes are below 256, why can't I get them in TMG?

Afina

[John Moran]

OK. I'll admit I'm winging it here.

I seldom actually crank up TMG or any other Genealogy program except to experiment. Having said that, I have at one time or another set my Regional Settings and/or languages to Vietnamese to use it with Family Treemaker, and have also selected Hebrew to see the right to left problems problems in TMG memos. I became interested years ago when some guy wanted to use UFT with Cyrillic.

By accident, before I uninstalled the Hebrew support "recently", I found that FTM memo fields seemed to handle the right to left script in Hebrew correctly, although TMG did not - as of a few updates ago.

But it looks like the Vietnamese character layout is just a slight modification of the Windows Western layout, so if you have no other persons in the database requiring German or French characters (and perhaps sorting conventions), you might be able to get by with Vietnamese only.

That's about all I know about the subject. You get 256 different characters to play with (up to 0xFF in hex) in your TMG database. You can think of each character as being whatever you like, as long as you have a subset of a single byte "font" to print out each of the characters you actually use.

As to why you cannot use different languages simultaneously in TMG as you can in M$ Word, I gue$$ it all boil$ down to dollar$ and cent$, and workload, and perceived demographics of the respective customer bases. TMG does not support "unicode" at this time.

John

[afinab]

OK. I'll admit I'm winging it here.

I seldom actually crank up TMG or any other Genealogy program except to experiment. Having said that, I have at one time or another set my Regional Settings and/or languages to Vietnamese to use it with Family Treemaker, and have also selected Hebrew to see the right to left problems problems in TMG memos. I became interested years ago when some guy wanted to use UFT with Cyrillic.

By accident, before I uninstalled the Hebrew support "recently", I found that FTM memo fields seemed to handle the right to left script in Hebrew correctly, although TMG did not - as of a few updates ago.

But it looks like the Vietnamese character layout is just a slight modification of the Windows Western layout, so if you have no other persons in the database requiring German or French characters (and perhaps sorting conventions), you might be able to get by with Vietnamese only.

That's about all I know about the subject. You get 256 different characters to play with (up to 0xFF in hex) in your TMG database. You can think of each character as being whatever you like, as long as you have a subset of a single byte "font" to print out each of the characters you actually use.

As to why you cannot use different languages simultaneously in TMG as you can in M$ Word, I gue$$ it all boil$ down to dollar$ and cent$, and workload, and perceived demographics of the respective customer bases. TMG does not support "unicode" at this time.

Thanks John, I guess I'll have to improvise :-)
Afina

[Roy_Delos_Reyes]

As to why you cannot use different languages simultaneously in TMG as you can in M$ Word, I gue$$ it all boil$ down to dollar$ and cent$, and workload, and perceived demographics of the respective customer bases. TMG does not support "unicode" at this time.

John

More like TMG *can't* support unicode at this time because, last I heard, Microsoft hasn't seen fit to have Foxpro support unicode.

[afinab]

More like TMG *can't* support unicode at this time because, last I heard, Microsoft hasn't seen fit to have Foxpro support unicode.

Let's hope Bill gets with it some time this century :-)

[David Reno]

I am intererested in adding Farsi names, so I have a similar problem.

Here's the real question:
Can we get someone from TMG development to weigh in on a roadmap for Unicode support? Is this something that is not on the roadmap at all due to FoxPro limitiations? Is this something that is a 7.x or 8.x feature?

[Karla Huebner]

I hope that Wholly Genes will be able to add Unicode support in the near future, as a huge number of users (current but especially potential) need to be able to deal with Vietnamese, Czech, Polish, Romanian, Hungarian, and other languages that use the Roman alphabet with variations. That's not even getting into non-roman alphabets, where at least one typically transliterates for readers using the Roman alphabet. But it would be great to have support for Cyrillic, Hebrew, Arabic, and other alphabets.

Karla

[mosh]

The lack of bidirectional Hebrew support in TMG is for me a serious deficiency. As far as I know, both FTM and Legacy have had such support for awhile already. The OS supports such languages. To continue to develop the program as is when its FoxPro foundation is so outdated seems to be a dead-end at best.

Mosh

[sam.m]

Yes unicode is possible in Visual FoxPro 9 but programming it may be very difficult.

For the more technical you can read about it here:


http://www.west-wind.com/presentations/fox.../foxunicode.asp

"Using Unicode in Visual FoxPro Web and Desktop Applications"
...
Summary

Using Unicode in Visual FoxPro is definitely not for the faint of heart. If you are lucky and you can get away with using a single locale specific Unicode language, Visual FoxPro’s code page translation mechanism can get you reasonable language capabilities in your application without too much extra work. Problems don’t arise until you have more than one language involved.
....
It’s a shame VFP doesn’t have better support for Unicode.
....


Whollygenes may have to change to a new database backend, which probably means starting over! Or building a bridge to a new database?
Lets hope.

[sam.m]

Not sure but the following seems like a possible solution for making TMG unicode compliant without having to rewrite tmg from scratch.

Steven Black's INTL Toolkit for Visual FoxPro is designed to flexibly localize your Visual FoxPro applications. INTL encapsulates strategies to localize the following:

* Interface strings
* Fonts
* Images
* Data sources
* Currencies
* Right-To-Left writing systems
* Dynamic dialogs

http://www.stevenblack.com/intl%205.asp

I am not a programmer, but I do realise that this is just one possible part of converting tmg to unicode.
Heres hoping.

[retsof]

Late news from Microsoft is that they say that VF9 will be the last one and there will be no VF10. The last update for VF9 does not include Unicode either. Somebody, VF10 may appear with Unicode due to "popular demand", meaning y'all.

If VF10 does NOT appear, TMG may have a decision to make several years down the road, depending on whether something better comes along.

That would also mean that UTF-8 GEDCOMS could be imported directly. For the moment, the workaround is to bring them into wordpad and save them as ASCII.

Edited by retsof, 04 January 2008 - 08:10 AM.

[sam.m]

Another Article on the subject..

VFP can do Double Byte Character Sets
http://weblogs.foxite.com/boudewijnlutgeri...03/12/3439.aspx

[John Cardinal]

Another Article on the subject..

VFP can do Double Byte Character Sets
http://weblogs.foxite.com/boudewijnlutgeri...03/12/3439.aspx

IMO, modifying an application the size of TMG using the technique described on that page is impractical.

[sam.m]

IMO, modifying an application the size of TMG using the technique described on that page is impractical.

Hi John,

I am not a programmer.

Can you say why it is impractical from a programming point of view?
Also what would you do as a programmer to have tmg handle Unicode?


What would be a trade off or a possible solution that could be programmed and handled by tmg.
How about just the memo fields using this method to at least preserve the original data from UTF-8 GEDCOMS?

[John Cardinal]

I am not a programmer.

Can you say why it is impractical from a programming point of view?
Also what would you do as a programmer to have tmg handle Unicode?
What would be a trade off or a possible solution that could be programmed and handled by tmg.
How about just the memo fields using this method to at least preserve the original data from UTF-8 GEDCOMS?

The article describes a technique for displaying Unicode text on a VFP-generated form, but specifically says that the data must be stored outside of VFP, so this is a user interface solution only. To leverage this so-called solution, TMG would have to store all text data outside of VFP. That would mean extensive changes to all of the major facilities in TMG including reports, filters, and dozens of other facilities. The end result would be a fragile mess (in software terms). No sane software designer would recommend that solution for a major application like TMG.

In addition, as if any more disincentives were needed, it would be a big job to rewrite the many (hundreds?) of TMG forms to use the controls the article advocates.

The article describes a little hack that someone did to implement a tiny application and the solution simply doesn't apply to a large application like TMG. The article title is misleading: the data is not stored in VFP, and using VFP to create a form or two of a user interface is an uncommon use of the tool.

I do not have a solution to the problem of how to add unicode support to TMG other than to rewrite TMG to use a different database engine than VFP. If there was a practical solution that didn't involve a complete rewrite, Wholly Genes would have done it.

[sam.m]

John,

Thanks for detailed answer.

I appreciate and believe I can see what you mean.

[Mr.Kim.Sanders]

A quick look at http://en.wikipedia.org/wiki/Visual_FoxPro shows us the real problem. Visual FoxPro is COM-based and Microsoft has stated that they do not intend to create a Microsoft .NET version. Unless… maybe we helped them?
We must not hide our head in the sand here. So the "handwriting is on the wall" for both Visual FoxPro and its extensions like TMG. The terminal diagnosis has already been given, even if its vegetative state will be ensured for a few years of life support. What kind of life and hope is that?!
Surely there is an up-and-coming programmer (or several) who is/are also interested in genealogy. Let them come on board so the ship we're on doesn't sink. There clearly needs to be a shift from COM-based to .NET based.
And the sooner the better.

[Mr.Kim.Sanders]

I have to add a number of Vietnamese names to my v6.09 database, and they have some additional accents aside from the usual french ones. I have the fonts installed on my computer, and they are mapped to the numbers. I can get the names properly typed in Word and Wordpad, but not in TMG.

Number 1-4 are mapped to give the accented letters ă, â, ê, ô respectively and work fine, number 5 through 0 are mapped to apply an accent to a previous typed letter, but give me in TMG instead other already accented letters, or the accent as a separate character.

Does anyone have experience with this?

Thanks for any help,
Afina

You'll find some ideas you may find useable at http://www.whollygenes.com/forums201/index...?showtopic=9809. I wish you and everyone else all the best as we do our best to put things accurately into our Tree for ourselves and those we may teach.