muta...@gmail.com
2021-06-16 04:44:17 UTC
First imagine (in some cases, it is a reality), that the
following things exist as physical objects that may
have been there since 1990:
ASCII ANSI terminal
EBCDIC ANSI terminal
ASCII 3270 terminal
EBCDIC 3270 terminal
ASCII HTML terminal
EBCDIC HTML terminal
ASCII Telnet+ terminal
EBCDIC Telnet+ terminal
And I have emulators for all of the above, called
termansi.exe, term3270.exe, termhtml.exe and
termtel.exe. The terminal emulators are mainly
focused on running on PDOS/386.
The name Telnet+ may be confusing.
What I'm looking for is a magical terminal, that, when
given an appropriate control sequence, will switch
between any of the "big 3" terminal types above, ie
ANSI, 3270, HTML.
This way I can have one C program on PDOS/3X0 that
emits ANSI control characters, and another that emits
3270 control characters, and at the command prompt
I can issue some sort of command (tset?) that switches
between the terminal types dynamically.
But let's focus on EBCDIC ANSI first. That will be the
standard control sequence that any of my applications
emit.
Rather than burden the OS with the need to support
multiple application-generated control sequences, I
was thinking that what I need is a "terminal controller".
PDOS/3X0 will not speak directly (via CCW) to some
terminal calling via the internet, but will instead speak
to a termcont.exe running on my Windows machine,
which accepts e.g. a term3270 caller, and will translate
the application EBCDIC ANSI stream into an EBCDIC
3270 stream.
Even though my terminal controller will only accept
(somehow) one connection at a time, it will accept
different terminal types by if they use port 80 it is
assumed that they are ASCII HTML. Port 64000 is
assumed to be EBCDIC ANSI.
Most interesting to me at the moment is the EBCDIC
ANSI to ASCII HTML conversion.
If my PDOS/3X0 application does 3 printfs with newline,
I am expecting (my requirements at the moment)
fresh HTML pages to be generated. The first page will
have the <HTML> marker with just one line (with <br>
in it). The second page will be the same HTML marker
with 2 <br> etc. Now the bit that is outside my knowledge.
What is the simplest way, after having finished outputting
all HTML pages, to then turn around and request a single
character to be entered? I am happy to put out another
HTML page, perhaps this time with the "javascript"
keyword to solicit a function to input a key. I want all
keys to be able to be entered by the user, e.g. 'A' or
ctrl-a or cursor up. I don't care what HTML response
that generates, so long as it is standard. That HTML
response will be converted into EBCDIC ANSI by the
terminal controller, then fed back to PDOS/3X0, and
finally into some PDOS/3X0 application, and my eye
is on micro-emacs. In other words, I want to drive my
existing micro-emacs executable that produces
EBCDIC ANSI sequences, from a browser.
Is that possible or am I missing something?
Thanks. Paul.
following things exist as physical objects that may
have been there since 1990:
ASCII ANSI terminal
EBCDIC ANSI terminal
ASCII 3270 terminal
EBCDIC 3270 terminal
ASCII HTML terminal
EBCDIC HTML terminal
ASCII Telnet+ terminal
EBCDIC Telnet+ terminal
And I have emulators for all of the above, called
termansi.exe, term3270.exe, termhtml.exe and
termtel.exe. The terminal emulators are mainly
focused on running on PDOS/386.
The name Telnet+ may be confusing.
What I'm looking for is a magical terminal, that, when
given an appropriate control sequence, will switch
between any of the "big 3" terminal types above, ie
ANSI, 3270, HTML.
This way I can have one C program on PDOS/3X0 that
emits ANSI control characters, and another that emits
3270 control characters, and at the command prompt
I can issue some sort of command (tset?) that switches
between the terminal types dynamically.
But let's focus on EBCDIC ANSI first. That will be the
standard control sequence that any of my applications
emit.
Rather than burden the OS with the need to support
multiple application-generated control sequences, I
was thinking that what I need is a "terminal controller".
PDOS/3X0 will not speak directly (via CCW) to some
terminal calling via the internet, but will instead speak
to a termcont.exe running on my Windows machine,
which accepts e.g. a term3270 caller, and will translate
the application EBCDIC ANSI stream into an EBCDIC
3270 stream.
Even though my terminal controller will only accept
(somehow) one connection at a time, it will accept
different terminal types by if they use port 80 it is
assumed that they are ASCII HTML. Port 64000 is
assumed to be EBCDIC ANSI.
Most interesting to me at the moment is the EBCDIC
ANSI to ASCII HTML conversion.
If my PDOS/3X0 application does 3 printfs with newline,
I am expecting (my requirements at the moment)
fresh HTML pages to be generated. The first page will
have the <HTML> marker with just one line (with <br>
in it). The second page will be the same HTML marker
with 2 <br> etc. Now the bit that is outside my knowledge.
What is the simplest way, after having finished outputting
all HTML pages, to then turn around and request a single
character to be entered? I am happy to put out another
HTML page, perhaps this time with the "javascript"
keyword to solicit a function to input a key. I want all
keys to be able to be entered by the user, e.g. 'A' or
ctrl-a or cursor up. I don't care what HTML response
that generates, so long as it is standard. That HTML
response will be converted into EBCDIC ANSI by the
terminal controller, then fed back to PDOS/3X0, and
finally into some PDOS/3X0 application, and my eye
is on micro-emacs. In other words, I want to drive my
existing micro-emacs executable that produces
EBCDIC ANSI sequences, from a browser.
Is that possible or am I missing something?
Thanks. Paul.