T.R | Title | User | Personal Name | Date | Lines |
---|
592.1 | UWM location | VINO::WITHROW | Robert Withrow | Thu Apr 13 1989 11:42 | 9 |
| A version of UWM is now available at:
VINO::BORDEAUX:[WITHROW.PUBLIC]UWM.SAVZ
This is a compressed backup saveset. It contains the sources and executable
for the UWM window manager. It has been partially ported to VMS, but is
lacking subprocess support. This has been added and is being tested, and
the above will shortly be replaced with the newer version.
|
592.2 | What is a standard X11 client? | SDSVAX::SWEENEY | Wall Street is my beat | Thu Apr 13 1989 11:54 | 11 |
| Could someone help me out and define what is meant by a "standard X11
client"?
Is this some convention represented by the files' positions on the MIT/
X Consortium release tape, if so what is it? Has the X Consortium
declared what the "standard X11 clients" are?
I would hate to see this usage become common. If it does it will
create the impression to customers that somehow, the VMS release of
DECwindows is incomplete because it omits the "standard X11 clients".
|
592.3 | The stuff on the CORE tape is what I mean... | VINO::WITHROW | Robert Withrow | Thu Apr 13 1989 12:37 | 10 |
| > Could someone help me out and define what is meant by a "standard X11
> client"?
I am referring to the things that show up on the CORE distribution tape.
I don't know nuttin' about consortium policy, but I think that I have
seen somewhere that there is no REQUIREMENT for them to be provided, but
I think there is an expectation that they WILL be provided, or that
equivalent means will be provided. For many things equivalents ARE
provided, but not all. XRDB is an example where (as far as I know) no
equivalent is provided by VMS DECWindows.
|
592.4 | Some of it has been done | 25519::SCHOELLER | Who's on first? | Thu Apr 13 1989 13:15 | 15 |
| Take a look in ELKTRA::DW_EXAMPLES. You will find xsetroot and xlsfonts.
I think some of the hacks over there may cover xwininfo and I think I have
seen xmodmap.
In .1 you mention uwm. If you had bothered to look around in this conference,
you would have seen that there already is a complete VMS port of uwm floating
around.
If there is one that has not been done, I would expect xterm to be it.
That is because (IMHO) DECterm is a much more complete terminal emulator.
BTW what does xrdb do?
Dick
|
592.5 | xrdb...documentation | FUEL::graham | if ya want home cookin, stay home | Thu Apr 13 1989 15:02 | 27 |
|
RE.4
The following is only a scracth of the xrdb man page from an Ultrix
system....you can read about the rest on your nearest Ultrix Workstation.
Kris..
xrdb(1X)
NAME
xrdb - X Window System server resource database utility
SYNTAX
xrdb [ options ] [ file ]
ARGUMENTS
file Specifies the defaults file to be loaded into
the specified server. The filename, or the
standard input if no filename is specified,
is optionally passed through the C preproces-
sor. The following symbols, which depend on
the server being used, are defined:
...
.....
|
592.6 | UWM confusion? | POOL::BALLOU | It's not slow, it's careful! | Thu Apr 13 1989 16:18 | 25 |
| Re .4:
I think I'm a bit confused by this note. In note 275.1, I answered an inquiry
as to whether UWM was running under VMS. I did this port some time back,
having decided that our window manager, ahem, left something to be desired ...
However, I knew it was an "unsupported hack," and I wasn't completely
confident that UWM was reliable. Well, I was finally fairly sure that it was
stable, but because of security measures, I could not give it a home. Since
then, Robert Withrow asked me for a copy, tried it out, decided it was
acceptable, and generously gave it a home. This is what is said in .1.
(Also, let me add that Robert worked on the code some more. I just tried to
get UWM to stop getting ACCVIOs. That meant I ignored the subprocess code.
Robert put that in.)
Meanwhile, a later note (302.0) refers to a different port of UWM. My
impression from replies to that note is that this port also works, although
there was at least one mention of an undocumented feature :-) (This feature
does not seem to be present in the port on VINO::. Maybe I just got lucky :-)
Meanwhile, Robert and I have expressed interest in hearing from anyone using
our port of UWM who finds any undocumented features.
- Ken
|
592.7 | Explaining my concerns... | CALL::SWEENEY | Wall Street is my beat | Fri Apr 14 1989 00:49 | 20 |
| I'm concerned that unless customer access to these "hacks" is provided
with the "right" level of documentation, support, etc., there's
a near certainty of serious customer satisfaction problems down the
road.
As a contributor of VMS DECwindows software to DW_EXAMPLES, I don't
think I'd appeciate urgent phone calls to me and my boss that a large
order depends on some modification or fix to such software, whether I
wrote it or not.
A Software Services program called ASSETS offers such software to
customers for what I think are appropriate pricing and conditions of
sale.
Having said that, I think freely sharing source inside Digital has both
helped me immensely and given me an opportunity to help others. I just
urge caution in distributing this stuff, especially when you can't
maintain it and the porting author hasn't explicitly told you that you
can distribute it to customers.
|
592.8 | Would profer SPECIFIC references | VINO::WITHROW | Robert Withrow | Fri Apr 14 1989 13:07 | 14 |
| > Take a look in ELKTRA::DW_EXAMPLES. You will find xsetroot and xlsfonts.
> I think some of the hacks over there may cover xwininfo and I think I have
> seen xmodmap.
I was hoping to make this note a "1 stop shopping" type of thing, where each
utility is pointed to as in reply .1, rather than a method of sending the
reader on a notes safari. Rather than referring the reader nonspecifically to
another notes conference, I would prefer that either the actual location of the
tool be posted, or, at least, the note number in the target notes conference.
> BTW what does xrdb do?
Allows the user to set and query properties used by clients as custimization
info.
|
592.9 | R4 tape an option | KOBAL::VANNOY | Jake VanNoy | Sun Apr 16 1989 14:51 | 12 |
| An alternate option for distribution of MIT clients is to put them
back on the MIT R4 distribution with the appropriate "#ifdef VMS"'s
in them. I talked to Bob Schiefler about this friday and he was
very open to this idea, as he gets asked about VMS support in these
clients frequently.
I plan to pursue this. My first step is to see if I could get real
resources assigned. Failing that, I will appeal to the hackers so
well represented in this notesfile.
jake
|
592.10 | | CVG::PETTENGILL | mulp | Mon Apr 17 1989 13:39 | 13 |
| I've found that for a number of the clients (for example those I've posted
in DW_EXAMPLES), the changes required are very minor and if the logicals sys
and x11 are defined before the compile, no changes are required other than
bug fixes.
For anyone trying to convert X11R3 clients to VMS, I make this suggestion:
check each call to `free' very carefully. Most of these should be calls to
XFree. The distinction is become increasingly important as Xlib implementors
improve the performance by changing the internal operation of XFree to take
advantages of the Xlib memory use patterns.
By the way, the symptoms of incorrect use of free on VMS are strange ACCVIOs.
|
592.11 | Yes, but.... | VINO::WITHROW | Robert Withrow | Tue Apr 18 1989 11:36 | 12 |
| > An alternate option for distribution of MIT clients is to put them
> back on the MIT R4 distribution with the appropriate "#ifdef VMS"'s
> in them.
I like that idea, but from the point of view of an internal DEC user,
there is one major drawback. Extracting these clients from the core
distribution tape using the tools generally available to VMS users
(namely LZDCMP and TARREAD) takes nigh onto 100,000 blocks and large
amounts of time. Perhaps the CORE tape should be broken into three
tar sets: Servers and fonts, Library and toolkits and widgets, Clients.
Or perhaps someone will develop the nerve to port PDTAR to VMS?
|