When you are asking for help always tell the full story. You will get the best help that way.
No one asked, the problem is not that I did not say what, the problem is that people are marking assumptions.
I'm trying to create a filename whit correct codeset, the orignal filename is UTF8, so it has to be encoded as if I typed it my self.
Quote:
Input prefs sets the keyboard map. This maps keystrokes to codes. At the moment these are single bytes sequences for the most as far as I can recall.
I undertread.
So there is table that maps the rawkeys to ASCII tabel some where. And you need locale.library to convert the ASCII string to Glyphs to be displayed on the screen.
Quote:
There is no distinct input character set.
are you shore I see the one from Input.prefs has “n_#?” prefix, and one from GetEnv does not have “n_#?” prefix
Quote:
Sometimes the name of the character set that the author of the keymap expects you to use is included in the "English description field"
What your saying is that struct InputPrefs -> ip_KeymapName will not always have this value?
Quote:
That is a convenience to enable people to use the right keyboard with the right character set,
As long as they see the characters glyphs they expect, when they type it can't be that bad.
Quote:
but you can't rely on it, you can't rely on the format of it to parse it etc etc. Particularly if it happens to be a custom third party charset.
Is not codeset code 1 to 1 font glyph?
(NutsAboutAmiga)
Basilisk II for AmigaOS4 AmigaInputAnywhere Excalibur and other tools and apps.
No matter how much you ignore it it's still an user error, your settings are broken, you use ISO-8859-1 as charset (english_ISO-8859-1 instead of english_ISO-8859-15 as first language in Prefs/Locale) but a keymap in ISO-8859-15 which obviously can't work with it.
I don't know, it might be a freak accident that I can see øæåØÆÅ symbols, when I type on my keyboard.
Quote:
The "CountryName" your broken program
yes I did not change "CountryName" string, it was just to prove that ISO code was not the same in Input.prefs as in locale.prefs.
Anyway it's same program as one on wiki.amiga.com whit some modifications, whit all it original faults plus possibly some of mine.
Quote:
displays is just the file name of the non-working keymap file (KEYMAPS:n_ISO-8859-15) you are currently using.
Maybe the Keymap is broken, but I don't see anything wrong whit it, I can see and read æ,ø,å,Æ,Ø and Å glyphs, when I write.
(NutsAboutAmiga)
Basilisk II for AmigaOS4 AmigaInputAnywhere Excalibur and other tools and apps.
What your saying is that struct InputPrefs -> ip_KeymapName will not always have this value?
Just look in KEYMAPS:, about 1/3 the files don't include a charset in the name.
Quote:
Maybe the Keymap is broken, but I don't see anything wrong whit it, I can see and read æ,ø,å,Æ,Ø and Å glyphs, when I write
You are trying to use a keymap in ISO-8859-15 with a system using ISO-8859-1 as charset. Most of ISO-8859-1 and ISO-8859-15 are the same, IIRC 8 chars changed. To check that it doesn't work you have to type a char which is different in both charsets, for example the Euro sign (€, ALT-e), which is in your ISO-8859-15 keymap, but can't work on your ISO-8859-1 system.
Maybe the Keymap is broken, but I don't see anything wrong whit it, I can see and read æ,ø,å,Æ,Ø and Å glyphs, when I write.
What about the € symbol?
ISO-8859-1 and ISO-8859-15 are very similar so a keyboard map designed to work with the latter will work with probably 250 or more of characters but not the euro and maybe a couple of others I can't remember which others differ.
May I know why you have dismissed the standard Amiga library Codesets?
It's only required on old AmigaOS versions which didn't have support for charsets yet, AmigaOS 4.x supports different (8 bit and in some parts like catalogs UTF-8) charsets and there is nothing in this library which AmigaOS 4.x doesn't support itself.
It can be used on old AmigaOS versions as replacement for the missing AmigaOS 4.x features, for example for supporting UTF-8 catalogs like YAM does now on AmigaOS 3.x.
Yes of course, it's an abstraction layer but last time I tried to use it, it was easier to use and straight forward (and not the least using Amigaish API).
Just look in KEYMAPS:, about 1/3 the files don't include a charset in the name.
Thanks for explaining, I will do what you suggest before, I just did not expect the ISO-8859-1 tabel to have table that was 1 to 1, 2 to 2 … 255 to 255.
it makes sense now.
(NutsAboutAmiga)
Basilisk II for AmigaOS4 AmigaInputAnywhere Excalibur and other tools and apps.
No matter how much you ignore it it's still an user error, your settings are broken, you use ISO-8859-1 as charset (english_ISO-8859-1 instead of english_ISO-8859-15 as first language in Prefs/Locale) but a keymap in ISO-8859-15 which obviously can't work with it.
Again its more correct to say I was misled, by the fact that window did not show the “Charset”, so picking the right one, was not that easy in the first place.
as you can see there is virtually no difference between. English and English-British, I do not see what encoding they use.
Then I click the “Clear Language” button, and what is this a extra column, called “Charset”, it was not there before I clicked the button.
Only if you have not picked one, can you see the Charset.
And more the € symbol is not even visible in the first list, only after clicking the button to clear the selection, there is no way of knowing if I have selected the right one, once some thing is selected.
Edited by LiveForIt on 2014/3/6 19:50:50
(NutsAboutAmiga)
Basilisk II for AmigaOS4 AmigaInputAnywhere Excalibur and other tools and apps.
Again its more correct to say I was misled, by the fact that window did not show the “Charset”, so picking the right one, was not that easy in the first place.
We told you at least 2 times in this thread that you have to clear the list to select a language with correct charset:
#20 Quote:
It sets the codepage as well, the one from the first language is used. To change the codepage you have to clear all selected languages, only if the list is empty the codepages are displayed for the languages, and use one with the codepage you want as the first one.
#26 Quote:
So if you want a system which is purely using ISO-8859-15 (like I have), you need to make the correct language selection in Locale prefs: delete all chosen languages, select as your first one a language with Euro sign/ISO-8859-15 extension, and after that select any secondary language choices.
Quote:
Only if you have not picked one, can you see the Charset.
Because there can only be one charset, the one of the first language. Additional languages you add in locale prefs, up to 10 are supported, can't use an own/different charset.
I did not mean that you or anyone misled me, it was just the Input prefs looked more like the correct one because of the File-name and what was listed there.
I did have that configuration long before I was told to change it.
(NutsAboutAmiga)
Basilisk II for AmigaOS4 AmigaInputAnywhere Excalibur and other tools and apps.
I agree with you that the way to select the system charset in Locale prefs is not very intuitive and might benefit from a separate control for the charset selection (which should of course be read-only after selecting the first language) - or some other way to make it clearer that you are setting the charset in addition to your primary language.
This is why I always take great care to describe the process, including clearing the chosen languages before selecting one with the desired charset.
Good to see that you ended up understanding it, even if it took some attempts .
I agree with you that the way to select the system charset in Locale prefs is not very intuitive
It is unintuitive in that it requires a step (clearing all languages) that is not logically related to what the user wants to do (change a charset). The fact that charset has something to do with language is a technical thing, not a logical relation.
You (or nbache) are assuming that the user actually cares (or understands the consequences of) which charset they are using.
This is another area where UTF-8 would really help, as the user wouldn't need to select a character set. As already indicated, the method is unintuitive, however the process shouldn't be needed at all.
Talking of which, I take it listbrowser.gadget has UTF-8 support? As last time I checked, Cyrillic wasn't part of ISO-8859-15 and neither was Hebrew. Or is it a ridiculous "change the 8-bit charset to something appropriate for the listbrowser cell I'm filling in" thing?
Talking of which, I take it listbrowser.gadget has UTF-8 support?
No idea if there is codeset 106 support in some gadgets and/or images already, but it wouldn't help at all since there are no Text() functions in AmigaOS yet which support anything but 8 bit fonts. (Of course the bullet API is there since AmigaOS 2.1 already and much easier to use with the new AmigaOS 4.x diskfont.library functions, but you still have to render each glyph yourself.)
Quote:
As last time I checked, Cyrillic wasn't part of ISO-8859-15 and neither was Hebrew. Or is it a ridiculous "change the 8-bit charset to something appropriate for the listbrowser cell I'm filling in" thing?
Yes, IA_Font in label.image isn't limited to struct TextAttr but supports struct TTextAttr as well (TA_CharSet tag).
You (or nbache) are assuming that the user actually cares (or understands the consequences of) which charset they are using. [...] the process shouldn't be needed at all.
Absolutely agreed. Considering how things are in AOS, I simply didn't dare to think aloud that far
News feed
OS4 Feedback forum
OS4Depot Feedback forum
Amigabounty forum
OpenAmiga forum
AmiCygnix forum
ABC shell forum
AmiKit forum
Cinnamon Writer forum
CodeBench forum
Digital Universe forum
Dopus 5 forum
E-UAE forum
Gnash forum
Ibrowse forum
JAmiga forum
Odyssey forum
OWB forum
Qt forum
SmartFileSystem forum
Timberwolf forum
TouchDevice forum
TuneNet forum
Unsatisfactory Software forum
Donate
