Hi,

I am trying to enter japanese fonts (kanji) on my Amigaone, using Aozora Mincho .ttf

I can enter (insert) individual characters, but I am trying to locate individual kanji
and PGS won’t print the font specimen, presumably because it is too large.
Everytime I insert 1 character the insert window closes, and I have to re-open
it scroll down 8 times and select the next character.

All the font characters show a Ctrl-D value, and I thought that would probably be a better way.

i have looked at the pagestream documentation web site, and it states that shift ctrl-D should
allow me to enter individual characters quicker.
But I cannot get shift Ctrl-D to work. I have tried several key combinations, but all I get is
alternative’s of character D, or the date prefs, or script window, or preferences.

Does anyone have any idea what I am doing wrong?

Do I need any particular settings, or other programs installed to get this to work.


I am running PGS 4.1.5.6 on a Amigaone G3/800, using OS4.1 update 6.



thanks for any help.


Peter Swallow (aka Swoop)

Received from peter.ams@virgin.net on 04/04/2014

Hi Peter.

> All the font characters show a Ctrl-D value, and I thought that would probably be a better way.
>
> i have looked at the pagestream documentation web site, and it states that shift ctrl-D should
> allow me to enter individual characters quicker.
> But I cannot get shift Ctrl-D to work. I have tried several key combinations, but all I get is
> alternative's of character D, or the date prefs, or script window, or preferences.
>
> Does anyone have any idea what I am doing wrong?

I think you may be misunderstanding how the Ctrl-d function works. For a start, Shift-Ctrl-d opens the insert date requester, as it should, so don't do that.

So to insert any character, you first need to find its number. e.g. if I wanted to insert a bullet point as part of the Triumvirate-Normal font:
1. Open insert character box (ctrl-z)
2. Find the bullet character and click on it
3. The three fields in the bottom right corner show:

Key: [blank]
Ctrl-C: bu
Ctrl-D: 8226

Key = blank means you cannot insert this character using a single key press, while with some characters you obviously can, e.g. just pressing the A key types an A chr

Remember the bu and 8226 codes which are specific to this character in this font. Once you know these codes, you don't need to open the insert window box again.

4. Close the insert window
5. Put the text cursor in the position to insert the bullet point, ensure you have the correct font selected
6. Press Ctrl+c then type bu
7. The bullet mark will appear
8. Press Ctrl+d then type 08226 (I think you always need to enter 5 digits with Ctrl-d, hence the leading 0, or you could do Ctrl-d 8226 Enter)
9. The bullet mark will appear

This works fine for me with OS3.9 and PGS v4.1.5.6.

This part of the documentation explains it fairly well:
http://pagestream.org/?action=Documents&id=127

Another way to possibly go about the entire thing is instead of entering the individual Kanji characters, just type the English neumonic for them while writing the document. Then when done, search and replace the specific neumonic with the single Kanji character. The find and replace box will let you select any single character (or a sequence of them) for replacement if you press the >> button. A word of warning, I think the find and replace thing probably won't work if you're not using the same font everywhere. i.e. the neumonic font and the kanji replacement font need to be the same. And the find and replace thing does seem to have a few bugs around it, so be sure to save your work first.

Or you could do as above, but use an Arexx script to globally go through and replace all of the different English neumonics with their respective Kanji character instead of manually replacing them one at a time.

Hope that helps,
Anthony.

Hi,

I am trying to enter japanese fonts (kanji) on my Amigaone, using Aozora Mincho .ttf

I can enter (insert) individual characters, but I am trying to locate individual kanji
and PGS won’t print the font specimen, presumably because it is too large.
Everytime I insert 1 character the insert window closes, and I have to re-open
it scroll down 8 times and select the next character.

All the font characters show a Ctrl-D value, and I thought that would probably be a better way.

i have looked at the pagestream documentation web site, and it states that shift ctrl-D should
allow me to enter individual characters quicker.
But I cannot get shift Ctrl-D to work. I have tried several key combinations, but all I get is
alternative’s of character D, or the date prefs, or script window, or preferences.

Does anyone have any idea what I am doing wrong?

Do I need any particular settings, or other programs installed to get this to work.


I am running PGS 4.1.5.6 on a Amigaone G3/800, using OS4.1 update 6.



thanks for any help.


Peter Swallow (aka Swoop)

A good, informative reply. But I can add just a little:

ctrl-d: you do not have to enter all five digits *but* there has to be a way for PgS to know that you are done. For example, if you wanted to insert 8226 followed by a one ctrl-d 8, 2, 2, 6, 1 would not insert the bullet followed by a one. But if you type ctrl-d 8, 2, 2, 6, space — that will insert a bullet character followed by a space.

find/replace and fonts: PgS doesn’t care what the font is, the substitution will work just fine no matter your font. However, if the font doesn’t have a glyph for the character then it won’t display the way you want. But if you were to use mnemonics in Times Roman you could use the Find requester to replace things just fine. They wouldn’t display right because Times Roman doesn’t have kanji characters — but all you would need to do is switch to a font that does have the characters. Not a big deal, really.

If I was doing this I would do one of two things:

1. use a script to provide an input mechanism. Basically, it would open a dialog, enter the mnemonic and it would insert the desired character. This would be okay for inputting small amounts of text and is what I did for Sumerian cuneiform.

2. use a script to automate the replacing of mnemonics with characters. This would be handy because you could make it somewhat readable by using dashes between mnemonics that should appear without white space between them and the script could strip out the dashes as it encountered them.

I guess there is, technically, a third option. A bit more technical and requiring access to a modern font editor. The short of it is that you can define — in the font — character sequences that are replaced by an arbitrary character. Think ligatures and you get the idea. PgS supports this, though there are some issues (reported to Deron, not sure when an update to address them will occur). I did this for a special font that has unusual ligatures (aa, bb, cc, …, in addition to ch, ph, th, sh and some others). In short, you have the *font* do the translation from mnemonic to glyph. Importantly, the typed characters are still there and if you switch the font to something else they are what will be displayed. Its amazing stuff and allows for some really cool things. (Not to mention security implications if you could get a browser to display a URL using a special font…)

Tim Doty

On Apr 4, 2014, at 4:15 PM, Anthony Hoffman <ahoffman@clear.net.nz> wrote:

> Received from peter.ams@virgin.net on 04/04/2014
>
> Hi Peter.
>
> > All the font characters show a Ctrl-D value, and I thought that would probably be a better way.
> >
> > i have looked at the pagestream documentation web site, and it states that shift ctrl-D should
> > allow me to enter individual characters quicker.
> > But I cannot get shift Ctrl-D to work. I have tried several key combinations, but all I get is
> > alternative's of character D, or the date prefs, or script window, or preferences.
> >
> > Does anyone have any idea what I am doing wrong?
>
> I think you may be misunderstanding how the Ctrl-d function works. For a start, Shift-Ctrl-d opens the insert date requester, as it should, so don't do that.
>
> So to insert any character, you first need to find its number. e.g. if I wanted to insert a bullet point as part of the Triumvirate-Normal font:
> 1. Open insert character box (ctrl-z)
> 2. Find the bullet character and click on it
> 3. The three fields in the bottom right corner show:
>
> Key: [blank]
> Ctrl-C: bu
> Ctrl-D: 8226
>
> Key = blank means you cannot insert this character using a single key press, while with some characters you obviously can, e.g. just pressing the A key types an A chr
>
> Remember the bu and 8226 codes which are specific to this character in this font. Once you know these codes, you don't need to open the insert window box again.
>
> 4. Close the insert window
> 5. Put the text cursor in the position to insert the bullet point, ensure you have the correct font selected
> 6. Press Ctrl+c then type bu
> 7. The bullet mark will appear
> 8. Press Ctrl+d then type 08226 (I think you always need to enter 5 digits with Ctrl-d, hence the leading 0, or you could do Ctrl-d 8226 Enter)
> 9. The bullet mark will appear
>
> This works fine for me with OS3.9 and PGS v4.1.5.6.
>
> This part of the documentation explains it fairly well:
>http://pagestream.org/?action=Documents&id=127
>
> Another way to possibly go about the entire thing is instead of entering the individual Kanji characters, just type the English neumonic for them while writing the document. Then when done, search and replace the specific neumonic with the single Kanji character. The find and replace box will let you select any single character (or a sequence of them) for replacement if you press the >> button. A word of warning, I think the find and replace thing probably won't work if you're not using the same font everywhere. i.e. the neumonic font and the kanji replacement font need to be the same. And the find and replace thing does seem to have a few bugs around it, so be sure to save your work first.
>
> Or you could do as above, but use an Arexx script to globally go through and replace all of the different English neumonics with their respective Kanji character instead of manually replacing them one at a time.
>
> Hope that helps,
> Anthony.
>
>
>

On 04/05/2014, Anthony wrote:

> [...]
>> i have looked at the pagestream documentation web site, and it
>> states that shift ctrl-D should allow me to enter individual
>> characters quicker. But I cannot get shift Ctrl-D to work. I
>> have tried several key combinations, but all I get is alternative's
>> of character D, or the date prefs, or script window, or
>> preferences.
>[...]
>
> I think you may be misunderstanding how the Ctrl-d function
> works. For a start, Shift-Ctrl-d opens the insert date
> requester, as it should, so don't do that.
>
> So to insert any character, you first need to find its number.
> e.g. if I wanted to insert a bullet point as part of the
> Triumvirate-Normal font: 1. Open insert character box (ctrl-z)
> 2. Find the bullet character and click on it
> 3. The three fields in the bottom right corner show:
>
> Key: [blank]
> Ctrl-C: bu
> Ctrl-D: 8226
>[...]
> 4. Close the insert window
> 5. Put the text cursor in the position to insert the bullet
> point, ensure you have the correct font selected 6. Press Ctrl+c
> then type bu 7. The bullet mark will appear
> 8. Press Ctrl+d then type 08226 (I think you always need to enter
> 5 digits with Ctrl-d, hence the leading 0, or you could do Ctrl-d
> 8226 Enter) 9. The bullet mark will appear
>
> This works fine for me with OS3.9 and PGS v4.1.5.6.
> [...]

I have essentially the same problem as Peter (though I'm not sure jsut
what he means by "PGS won't print the font specimen"), but with
5.0.5.8 on an X1000.

I just tried your method, more or less. Ctrl-z does nothing, so I go
Type > Insert > Character... (no indication of any shortcut for this
in the menu), choose KochiGothic as font and pick the character for
"what (nani/ka)." It appears in the window to the right; below it a
stack of

Glyph: uni4F55
Key:
Shift-Ctrl-D:
Ctrl-D: 20309

I go ahead at this point and click the Insert button, lower left,
knowing that the insert-character window will close. "Nani" appears
in the cursor postion of my New document. I try to repeat it by
typing "Ctrl-D 20309." Nothing happens.

Typing a string of Japanese writing in PgS is torture. Unfortunately,
copy and paste doesn't seem to work either (though I keep hoping I
might figure out a way); it just turns into "??????."

--
Bart Mathias
AmigaOne X1000 OS4.1.6

On Apr 4, 2014, at 8:17 PM, Bart Mathias <mathias@hawaii.edu> wrote:

> I just tried your method, more or less. Ctrl-z does nothing, so I go

ctrl-z? The discussion has been about ctrl-c (on some platforms a copy command, though in context in PgS it can be used with mnemonics and unicode entry. Ctrl-d is the other contender for entering characters.

> I go ahead at this point and click the Insert button, lower left,
> knowing that the insert-character window will close. "Nani" appears
> in the cursor postion of my New document. I try to repeat it by
> typing "Ctrl-D 20309." Nothing happens.

That’s strange. When I do the same thing I get a y dieresis because although I may select an appropriate font in the dialog it isn’t set in the article and PgS just inserts the character, it doesn’t change the font to match the choice in the dialog. Not sure what the correct behavior there is supposed to be.

> Typing a string of Japanese writing in PgS is torture. Unfortunately,
> copy and paste doesn't seem to work either (though I keep hoping I
> might figure out a way); it just turns into "??????.”

I have no idea what Amiga OS does for clipboard and text encoding, but that suggests that the text is getting mangled going through the clipboard. Although in principle a clipboard can be simple, on other platforms it certainly isn’t. There is usually a scheme for having multiple representations of whatever has been copied and the app has to do the right thing to get the right representation. If you get one to four question marks per character then it may well be a problem going from utf-8 to an ascii interpretation. I happened to see recently that Indic uses three bytes per character in UTF-8, I’d have to look to see what happens with Japanese (and it likely depends on whether or not it is kanji or katakana or hiragana).

Hello Tim

On 04/04/2014, you wrote:

>
> On Apr 4, 2014, at 8:17 PM, Bart Mathias <mathias@hawaii.edu>
> wrote:
>
>> I just tried your method, more or less. Ctrl-z does nothing, so
>> I go
>
> ctrl-z? The discussion has been about ctrl-c (on some platforms a
> copy command, though in context in PgS it can be used with
> mnemonics and unicode entry. Ctrl-d is the other contender for
> entering characters.

Anthony Hoffman listed it as step 1: Open insert character box
(ctrl-z)

>> I go ahead at this point and click the Insert button, lower
>> left, knowing that the insert-character window will close.
>> "Nani" appears in the cursor postion of my New document. I try
>> to repeat it by typing "Ctrl-D 20309." Nothing happens.
>
> That's strange. When I do the same thing I get a y dieresis
> because although I may select an appropriate font in the dialog
> it isn't set in the article and PgS just inserts the character,
> it doesn't change the font to match the choice in the dialog. Not
> sure what the correct behavior there is supposed to be.

If you select the y dieresis and change to an appropriate font, you'll
get what you inserted.

>> Typing a string of Japanese writing in PgS is torture.
>> Unfortunately, copy and paste doesn't seem to work either
>> (though I keep hoping I might figure out a way); it just turns
>> into "??????."
>
> I have no idea what Amiga OS does for clipboard and text
> encoding, but that suggests that the text is getting mangled
> going through the clipboard. Although in principle a clipboard
> can be simple, on other platforms it certainly isn't. There is
> usually a scheme for having multiple representations of whatever
> has been copied and the app has to do the right thing to get the
> right representation. If you get one to four question marks per
> character then it may well be a problem going from utf-8 to an
> ascii interpretation. I happened to see recently that Indic uses
> three bytes per character in UTF-8, I'd have to look to see what
> happens with Japanese (and it likely depends on whether or not it
> is kanji or katakana or hiragana).

I'm pretty sure kanji is all two-byte code like the 4F55 I mentioned
for "nani." Ditto kana and rômaji (see
http://www.rikai.com/library/kanjitables/kanji_codes.unicode.shtml).
One gets one "?" per character anyway.
--
Bart Mathias
AmigaOne X1000 OS4.1.6