OWB 2.7

	Bottom Previous Topic Next Topic
Register To Post

« 1 2 (3)

xeron

Re: OWB 2.7

Posted on: 2008/9/3 18:52 #41

Not too shy to talk

@ZeroG

Well, there you go. Amiga clipboard already supports charsets, then.

ZeroG

Re: OWB 2.7

Posted on: 2008/9/3 20:16 #42

Not too shy to talk

@xeron

Not really...

All software that reads or writes the clipboard have to support the CSET-Chunk and it has to convert the text to the charset that it uses.

Chris

Re: OWB 2.7

Posted on: 2008/9/3 20:43 #43

Amigans Defender

@ZeroG

also... the spec is a bit vague.

The FTXT spec doesn't mention CSET at all, however it does say all characters in the CHRS chunk must be in the core character set of ISO/DIS 6429.2 and ANSI X3.64-1979 standards.

the CSET spec is very minimal, it's a LONG to specify the charset (I'd imagine we are talking MIBenum here as used elsewhere in OS4), followed by seven reserved LONGs. It relates to text in any IFF FORM, and I would agree this includes CHRS chunks in FTXT.

I haven't seen Detlef's updated CSET spec, as the public SDK is typically out of date and doesn't contain it, or a number of other functions I'd quite like to use.

Chris

ZeroG

Re: OWB 2.7

Posted on: 2008/9/3 21:29 #44

Not too shy to talk

@Chris
I dont know if Detlef has put the updated CSET spec in the SDK, but he has send me a copy of it.
Quote:

chunk for specifying the character set of a text

Registered by Martin Taillefer.
Updated by Detlef W?rkner.

A chunk for use in any FORM, to specify the character set used for
text in FORM.

struct CSet
{
uint32 CodeSet;
uint32 Reserved[7];
}

Explanation of possible CodeSet values:
---------------------------------------
0 = undefined, has to be interpreted as ECMA-94 Latin 1.
ECMA-94 Latin 1 was the only standard Amiga charset before
AmigaOS4, it is identical to ISO-8859-1.

See the autodoc for diskfont.library/ObtainCharsetInfo()
how to interpret non-zero values.

See Documentation/Charsets.doc on the OS4 CD for an explanation
of character sets.

Here is an incomplete(!) list of character sets which are supported
by some parts of AmigaOS4.

3 - US-ASCII
4 - ISO-8859-1
5 - ISO-8859-2
6 - ISO-8859-3
7 - ISO-8859-4
8 - ISO-8859-5
9 - ISO-8859-6 Caution, ISO-8859-6-E has value 81, ISO-8859-6-I has 82.
10 - ISO-8859-7
11 - ISO-8859-8 Caution, ISO-8859-8-E has value 84, ISO-8859-8-I has 85.
12 - ISO-8859-9
13 - ISO-8859-10
ISO-8859-11 is not an official IANA charset. Use TIS-620 instead.
109 - ISO-8859-13
110 - ISO-8859-14
111 - ISO-8859-15
112 - ISO-8859-16
106 - UTF-8 Can already be used in catalogs and keymaps.
1012 - UTF-7 Supported by the CharsetConvert command.
1013 - UTF-16BE Supported by the CharsetConvert command.
1014 - UTF-16LE Supported by the CharsetConvert command.
1018 - UTF-32BE Supported by the CharsetConvert command.
1019 - UTF-32LE Supported by the CharsetConvert command.
2042 - IBM CodePage 437
2084 - KOI8-R
2104 - Amiga-1251 See SDK:Documentation/Localization/Charsets/Amiga-1251/
2250 - windows-1250
2251 - windows-1251
2252 - windows-1252
2253 - windows-1253
2254 - windows-1254
2255 - windows-1255
2256 - windows-1256
2257 - windows-1257
2258 - windows-1258
2259 - TIS-620

AmigaOS may support additional character sets in future.
IANA may define additional character sets in future.

Chris

Re: OWB 2.7

Posted on: 2008/9/3 21:39 #45

Amigans Defender

@ZeroG

Ah, that spec is the same as the one I have then, except with the extra explanation and clarification of valid values for CodeSet (and change of LONG to uint32, which seems sensible)

Chris

xeron

Re: OWB 2.7

Posted on: 2008/9/3 22:08 #46

Not too shy to talk

@ZeroG

Quote:

ZeroG wrote:
@xeron

Not really...

All software that reads or writes the clipboard have to support the CSET-Chunk and it has to convert the text to the charset that it uses.

I know that, my point is that the clipboard itself is not a problem, and right now OWB can start putting a CSET chunk on the clipboard.

The sooner apps start doing that, the better.

nbache

Re: OWB 2.7

Posted on: 2008/9/3 23:20 #47

Just can't stay away

@xeron

Quote:

xeron wrote:

[...]

I know that, my point is that the clipboard itself is not a problem, and right now OWB can start putting a CSET chunk on the clipboard.

The sooner apps start doing that, the better.

But Shinkuro's original problem was with text copied into OWB. Solving that would still require that the other program knew about the CSET chunk and set it correctly.

Best regards,

Niels

Rudei

Re: OWB 2.7

Posted on: 2008/9/4 3:39 #48

Not too shy to talk

I've seen so much about OWB, I've barely used it, about time it was installed on my A1XE.

Time for a change from Aweb perhaps?

Rude!

Nissan GTR '08

TMTisFree

Re: OWB 2.7

Posted on: 2008/9/4 19:39 #49

Just popping in

@Rudei

There is a new version coming soon.

Bye,
TMTisFree

"Never ascribe to malice that which is adequately explained by incompetence." (Napol?on Bonaparte)
"I would love to change the world, but they won?t give me the source code." (Unknown)

redfox

Re: OWB 2.7

Posted on: 2008/9/6 17:19 #50

Quite a regular

The problem I was having at amigaworld is fixed now with OWB 2.8.

Thanks.

redfox

Chris

Re: OWB 2.7

Posted on: 2008/9/7 18:25 #51

Amigans Defender

@xeron

I see another problem here - Datatypes textclass doesn't have any charset information, and (I'm guessing) the FTXT datatype doesn't look at the CSET chunk.

This isn't a problem for most applications that use the clipboard, as they will be reading and writing clips directly or through iffparse rather than using Datatypes.

However, if the clipboard is viewed through Multiview using Datatypes, or an FTXT file not on the clipboard is read through Datatypes, the Charset will be ignored. This probably isn't relevant to OWB but it does need fixing (the addition of a tn_CharSet value to the Text structure and a flag to indicate that this is available would be a good start)

Register To Post	« 1 2 (3)
	Top Previous Topic Next Topic

Currently Active Users Viewing This Thread: 1 ( 0 members and 1 Anonymous Users )