On 18 Dec 2013, at 21:20, Dennis Boone <drb at msu.edu> wrote:
Are these encoded as four separate characters or is the renderer
meant to figure out the ligatures and render the correct form of MEEM
when presented with it from memory?
In both 8859-6 and unicode, there is one code point for the letter, and
the renderer is to figure out which form belongs at that place in the
text.
It appears that xterm isn't really able to do this. It may also screw
up the Unicode bi-directional algorithm, since I see the Arabic
characters in L-R order. Not sure if one could sort this out by
embedding RLE/PDF characters; my unicode entry setup is in a state of
disarray due to upgrades.
I say screw this and we all learn Phoenician and standardise on that!
Who needs vowels anyway? They're for children.
Are these encoded as four separate characters or is the renderer
meant to figure out the ligatures and render the correct form of MEEM
when presented with it from memory?
In both 8859-6 and unicode, there is one code point for the letter, and
the renderer is to figure out which form belongs at that place in the
text.
It appears that xterm isn't really able to do this. It may also screw
up the Unicode bi-directional algorithm, since I see the Arabic
characters in L-R order. Not sure if one could sort this out by
embedding RLE/PDF characters; my unicode entry setup is in a state of
disarray due to upgrades.
De
On 18 Dec 2013, at 18:58, Johnny Billquist <bqt at softjar.se> wrote:
I'd say it's safe to assume that VMS do not do any advanced stuff that you're looking for now, Sampsa. You might have a separate application that can understand all this, but VMS itself do not. In general, each byte is one character, and once output, it stays.
Any rendering of characters differently depending on position will either have to be done inside the terminal, or else by special software in VMS, which is not a part of any normal piece.
What a shame, mind you Microsoft Word couldn't figure this out until about 2007 and the Terminal.app on my Mac still can't join letters.
On 2013-12-18 17:50, Sampsa Laine wrote:
On 18 Dec 2013, at 18:37, Dennis Boone <drb at msu.edu
<mailto:drb at msu.edu>> wrote:
Does VMS support Arabic text in the console (VT-series terminals etc)
or graphical modes (X11)?
In about 1983, I did some work on VMS 3.x systems in Riyadh. The user
terminals were Tandberg 2200 family devices with Arabic language
support. I don't recall there being anything special about the
installed VMS: it had no significant localizations in terms of system
messages, etc. We were able to do mixed English and Arabic text on
these terminals using standard system utilities (EDT), COBOL
applications and the Cincom Total database. The terminals operated on a
set-shift model, and it seems like maybe the | character was the shift.
Everything was single-byte character sets. IIRC the terminals dealt
with the Arabic positional shifting character forms.
I suspect this isn't exactly what you're looking for, though.
De
It might be actually, I'd love to be able to type Arabic on my VMS
boxes, we're getting a retrotech crowd together in Egypt and Lebanon.
The problem with Arabic is that each letter has up to four shapes
depending on its position in the word, i.e. the initial, medial, final
and independent.
So for example the letter MEEM ( - looks like a little circle with a
vertical line hanging off it, sometimes) looks like this when attached
to a SEEN ( - three small vertical bumps):
Independent:
Initial (MEEM-SEEN):
Medial (SEEN-MEEM-SEEN):
Final (SEEN-MEEM):
Are these encoded as four separate characters or is the renderer meant
to figure out the ligatures and render the correct form of MEEM when
presented with it from memory?
I've attached a PNG of the example above in case some of you guys don't
have Arabic fonts :)
I'd say it's safe to assume that VMS do not do any advanced stuff that you're looking for now, Sampsa. You might have a separate application that can understand all this, but VMS itself do not. In general, each byte is one character, and once output, it stays.
Any rendering of characters differently depending on position will either have to be done inside the terminal, or else by special software in VMS, which is not a part of any normal piece.
Johnny
On 18 Dec 2013, at 18:37, Dennis Boone <drb at msu.edu> wrote:
Does VMS support Arabic text in the console (VT-series terminals etc)
or graphical modes (X11)?
In about 1983, I did some work on VMS 3.x systems in Riyadh. The user
terminals were Tandberg 2200 family devices with Arabic language
support. I don't recall there being anything special about the
installed VMS: it had no significant localizations in terms of system
messages, etc. We were able to do mixed English and Arabic text on
these terminals using standard system utilities (EDT), COBOL
applications and the Cincom Total database. The terminals operated on a
set-shift model, and it seems like maybe the | character was the shift.
Everything was single-byte character sets. IIRC the terminals dealt
with the Arabic positional shifting character forms.
I suspect this isn't exactly what you're looking for, though.
De
It might be actually, I'd love to be able to type Arabic on my VMS boxes, we're getting a retrotech crowd together in Egypt and Lebanon.
The problem with Arabic is that each letter has up to four shapes depending on its position in the word, i.e. the initial, medial, final and independent.
So for example the letter MEEM ( - looks like a little circle with a vertical line hanging off it, sometimes) looks like this when attached to a SEEN ( - three small vertical bumps):
Independent:
Initial (MEEM-SEEN):
Medial (SEEN-MEEM-SEEN):
Final (SEEN-MEEM):
Are these encoded as four separate characters or is the renderer meant to figure out the ligatures and render the correct form of MEEM when presented with it from memory?
I've attached a PNG of the example above in case some of you guys don't have Arabic fonts :)
sampsa <sampsa at mac.com>
Does VMS support Arabic text in the console (VT-series terminals etc)
or graphical modes (X11)?
In about 1983, I did some work on VMS 3.x systems in Riyadh. The user
terminals were Tandberg 2200 family devices with Arabic language
support. I don't recall there being anything special about the
installed VMS: it had no significant localizations in terms of system
messages, etc. We were able to do mixed English and Arabic text on
these terminals using standard system utilities (EDT), COBOL
applications and the Cincom Total database. The terminals operated on a
set-shift model, and it seems like maybe the | character was the shift.
Everything was single-byte character sets. IIRC the terminals dealt
with the Arabic positional shifting character forms.
I suspect this isn't exactly what you're looking for, though.
De
Does VMS support Arabic text in the console (VT-series terminals etc) or graphical modes (X11)?
I _THINK_ I saw something in the Hobbyist license PAK related to Hebrew..
sampsa <sampsa at mac.com>
mobile +44 7961 149465
Just thought I'd mention this - I just installed the Alpha-VM product by those Russian guys and have it configured as a 1 CPU ES40 with 512 MB RAM, no JIT CPU acceleration.
However, I decided to stick the system / data (this is a toy system, so only using one 20 GB disk image) on my SSD and the thing FLIES.
Even though VUPS.COM and WHETSTONE (not great benchmarks for Alphas I know, but I was just doing some ballpark numbers) show that the box should be 5x slower than CHIMPY, it feels just as a fast or ever faster..
I think in a few years it'll be possible to run some pretty beefy Alpha boxes on a combo of SSD storage and commodity Intel server hardware (almost considering migrating CHIMPY onto this box to save on power :)
Now if I only had a license for their 4 core, JIT compiled version that supports like 16 GB of RAM..Sigh...
sampsa <sampsa at mac.com>
mobile +44 7961 149465
