> Hi Jon,
>
> Sorry.  I forget we have a diverse collection of operating platforms.
> Windows XP Pro is what I am running.
>
> Thanks,
> Rick.
>

I still do not have an answer for you but I'm wondering if you can use
something link Putty or Plink to manage the SSH connection and continue to
use FTP assuming that your server provides an FTP service behind SSH.

This requires far more investigation and is outside of my current
knowledge base as all my SSH connections are typically remote X-terms
which work well.

Jon.

>
>
> ----- Original Message ----
> From: "jon@..." <jon@...>
> To: jasspa@yahoogroups.com
> Sent: Tuesday, February 20, 2007 12:51:48 PM
> Subject: Re: [jasspa] ME and Secure FTP
>
>
>> The age of insecurity (pun intended) has arrived inside our corporate
>> firewall.  As of the end of the month, all applications that pass
>> plaintext passwords over the network will be turned off.  I do most of
>> my
>> editing on remote files retrieved using the built-in ftp in ME.  This
>> new
>> policy forces me to use some sort of secure ftp to retrieve and save
>> files.  I don't know much about the secure world (I've lived in
>> insecurity
>> all my life :).
>>
>> I have cygwin installed (included ssh and ssl) if that helps.  How do I
>> configure ME to use a secure protocal.
>>
>> Thanks,
>> Rick.
>
> Nice question!
>
> I was also considering this but currently do not know how to answer this
> because I have not tried it.
>
> What is your ME platform, UNIX, LINUX or Windows?
>
> Jon.
>
>
>
>
> __________________________________________________________________________
>
> This is an unmoderated list. JASSPA is not responsible for the content of
> any material posted to this list.
>
> To unsubscribe, send a mail message to
>
>     mailto:jasspa-unsubscribe@yahoogroups.com
>
> or visit http://groups.yahoo.com/group/jasspa and
> modify your account settings manually.
>
>
>
> Yahoo! Groups Links
>
>
>
>     http://docs.yahoo.com/info/terms/

Is it possible to use actual TAB characters but set the TABs to be 4
spaces?

Thanks

Bob Brown

Bob Brown wrote:
> Is it possible to use actual TAB characters but set the TABs to be 4
> spaces?
>
> Thanks
>
> Bob Brown
>
>

Hi Bob,

I think what you are asking is to set the tab width to 4 displayable
characters.

You can do this globally or per buffer, Note that if you enable globally
then you might still have to adjust the buffer types individually where
the buffer major mode is over-riding the global mode.

I would expect that you will do this on a per buffer basis.

You can show/hide the white spaces using
Menu->View->Whitespaces

Per Buffer Type
================

  From the file type where you want to do this then:

M-x major-mode-setup
or Main Menu->Tools->Major Mode Setup ..

  From the major mode then:

Disable "TAB" buffer mode - enable literal tabs.
Set the "Tab Width" to 4

Once setup whenever you load a file of the specified type then these
setting will be used.

Globally
=========

M-x user-setup

General Tab
Disable "TAB" - always use literal tabs unless over-ridden by buffer.

The global $tab-width does not have a dialogue so if you wanted to make
this 4 by default then do this in your <user>.emf i.e

set-variable $tab-width 4

Remember that in the buffer the variable is called $buffer-tab-width
which has local scope, its initial value is seeded from $tab-width.

Hope that helps.

Regards
Jon.

Follows an extension to the 'Bad 'C' (also C++/Java) highlighting in
hkc.emf in the "; Pick up bogus 'C' constructs extension" block.

; Pick up a bad assignment in a conditional statement. e.g.
; if (a = SOME_CONST)
; while (a = SOME_CONST)
hilight .hilight.c  2 "\\s\\(if\\|while\\)[ \t]*([ \t]*\\w+[ \t]*=[
\t]*[a-zA-Z_0-9]+[ \t]*)"  .scheme.hlred

Seems to pick up bad assignments and not cause any miss highlighting.
Will only pick up a simple expression.

Jon.

I just moved to 2006/08/10 in the process of upgrading my Linux box and
ran into the following:

line 386 of me.emf says
      4 popup-window

which, according to the docs, makes the window fit on the desktop, or
something:

> When n is 4 then both the frame position and frame size on the Desktop
> Window is adjusted, this is the same as 3 except the frame size may
> also be adjusted to fit on the desktop. The name argument is not
> required.

Well, it does awful things to mine.  (NB I'm running it on SuSE 10.1,
from a Win2k box with three monitors, using a pretty up-to-date Cygwin
as the X server.)  It freaks out.  ME sits there spanning two monitors,
from the middle one off to the side and bottom of the right-hand one,
hanging off the right and bottom sides; it looks like it's refreshing
the window like mad; it flickers up (otherwise not unusual) messages
about not finding spelling dictionaries; and I have to crack the whip to
get it back into shape (i.e., hit ^G a few times, resize it as small as
possible and then drag the carcass back over so all of it is at least
being displayed on my screens).

I tried debugging me.emf by doing 'set-variable $debug 1' but that
didn't work at all.  (I found out later: 2 is the new 1, in this
regard.)  I had to clear out the file and paste it back bit by bit.
Eventually I figured out that it was this popup-window thing causing the
problem.

So.  I thought you'd want to know: it looks like this is a Bad Idea.  I
don't know what it's doing, but it's not happy.  I commented it out, for
now.

I remember getting around the problem of "make the whole window appear
on the screen" by resizing it one column and one row bigger than I
really wanted it, and then resizing it one smaller.  I think I took that
out of my startup script years ago because it's been working okay.

p.s. I did figure out the spelling dictionaries issue: the spell
directory wasn't in my MEPATH.

Cheers

-Th


--
Thomas Hundt <tom@...> +1-415-867-6698

Was perusing the release notes for 2006/08/10 and noticed

      * insert-password(3) password generation utility for system
      administrators who tire of thinking of new passwords.

Apparently this generates random sequences of characters, e.g.,

1XuJg2w6 (One - X-RAY - uniform - JULIET - golf - Two - whisky - Six)

This kills me.  This is the same algorithm my boss uses to generate
passwords.  ("Hit a bunch of random keys on the keyboard.")  It
generates extremely INSECURE passwords because the first thing you have
to do is write them down somewhere -- or stick them in some text file
(hopefully encrypted -- just how secure is 'me -k'?).  I hate these
types of passwords, because I'm constantly opening this file to look for
them.

Lately, for passwords, I've been using the system once popularized by
CompuServe, which amounts to <some-word> <some-punctuation-char>
<some-other-word>, for example, barge%water.  Which is very similar to
the algorithm used by dotcoms to generate unique company (and domain)
names:  things like RedGorilla and BlueToenail.  So, hey, let's hear it
for the automatic company name generator feature... ;-)


--
Thomas Hundt <tom@...> +1-415-867-6698

Another problem I had with my recent re-build was this: the ME build
process expects libtermcap to be in one of the usual places.  SuSE, as
well as (apparently) some others, now park libtermcap.a and
libtermcap.so under /usr/lib/termcap.  (See below.)  I ended up
symlinking them up a directory, but I really don't like doing stuff like
that, since for sure I'm not going to remember it the next time I build
a box.  It would be nice if the ME build would be smart and look for it
in the new place.  Apparently this new location is going to become more
and more common:

> The termcap database is an obsolete facility
> and many distributions have moved libtermcap.so and libtermcap.a
> from /usr/lib/ to /usr/lib/termcap/.  The short-term solution
> is to add "-L/usr/lib/termcap" to the CFLAGS.  The long-term
> solution is to remove the dependency on termcap.

http://bugs.mysql.com/bug.php?id=3663


--
Thomas Hundt <tom@...> +1-415-867-6698

Thomas Hundt wrote:
> Another problem I had with my recent re-build was this: the ME build
> process expects libtermcap to be in one of the usual places.  SuSE, as
> well as (apparently) some others, now park libtermcap.a and
> libtermcap.so under /usr/lib/termcap.  (See below.)  I ended up
> symlinking them up a directory, but I really don't like doing stuff like
> that, since for sure I'm not going to remember it the next time I build
> a box.  It would be nice if the ME build would be smart and look for it
> in the new place.  Apparently this new location is going to become more
> and more common:
>
>> The termcap database is an obsolete facility
>> and many distributions have moved libtermcap.so and libtermcap.a
>> from /usr/lib/ to /usr/lib/termcap/.  The short-term solution
>> is to add "-L/usr/lib/termcap" to the CFLAGS.  The long-term
>> solution is to remove the dependency on termcap.
>
> http://bugs.mysql.com/bug.php?id=3663
>
>

As far has we have been able to ascertain the newer versions of Linux
use libncurses rather than termcap.

The 2.6 Linux make should use ncurses when termcap is not found. All of
the 2.6 versions we are building are ncurses.

This should be in the latest Linux make files.

Jon.

Thomas Hundt wrote:
> I just moved to 2006/08/10 in the process of upgrading my Linux box and
> ran into the following:
>
> line 386 of me.emf says
>      4 popup-window
>
> which, according to the docs, makes the window fit on the desktop, or
> something:
>
>> When n is 4 then both the frame position and frame size on the Desktop
>> Window is adjusted, this is the same as 3 except the frame size may
>> also be adjusted to fit on the desktop. The name argument is not
>> required.
>
> Well, it does awful things to mine.  (NB I'm running it on SuSE 10.1,
> from a Win2k box with three monitors, using a pretty up-to-date Cygwin
> as the X server.)  It freaks out.  ME sits there spanning two monitors,
> from the middle one off to the side and bottom of the right-hand one,
> hanging off the right and bottom sides; it looks like it's refreshing
> the window like mad; it flickers up (otherwise not unusual) messages
> about not finding spelling dictionaries; and I have to crack the whip to
> get it back into shape (i.e., hit ^G a few times, resize it as small as
> possible and then drag the carcass back over so all of it is at least
> being displayed on my screens).
>
> I tried debugging me.emf by doing 'set-variable $debug 1' but that
> didn't work at all.  (I found out later: 2 is the new 1, in this
> regard.)  I had to clear out the file and paste it back bit by bit.
> Eventually I figured out that it was this popup-window thing causing the
> problem.
>
> So.  I thought you'd want to know: it looks like this is a Bad Idea.  I
> don't know what it's doing, but it's not happy.  I commented it out, for
> now.

OK Thanks. I certainly have never run with multi-headed configuration -
I do not have that many Graphics cards and monitors. I believe that the
Windows multi-head works so not sure why the Linux one does not.

>
> I remember getting around the problem of "make the whole window appear
> on the screen" by resizing it one column and one row bigger than I
> really wanted it, and then resizing it one smaller.  I think I took that
> out of my startup script years ago because it's been working okay.
>
> p.s. I did figure out the spelling dictionaries issue: the spell
> directory wasn't in my MEPATH.

You still using a MEPATH? Not used a MEPATH for years normally put the
jasspa directory somewhere where ME automatically finds it and then dump
the spelling directories in the spell sub-directory.

Jon.

>
> Cheers
>
> -Th
>
>

Thomas Hundt wrote:
> Was perusing the release notes for 2006/08/10 and noticed
>
>      * insert-password(3) password generation utility for system
>      administrators who tire of thinking of new passwords.
>
> Apparently this generates random sequences of characters, e.g.,
>
> 1XuJg2w6 (One - X-RAY - uniform - JULIET - golf - Two - whisky - Six)
>
> This kills me.  This is the same algorithm my boss uses to generate
> passwords.  ("Hit a bunch of random keys on the keyboard.")  It
> generates extremely INSECURE passwords because the first thing you have
> to do is write them down somewhere -- or stick them in some text file
> (hopefully encrypted -- just how secure is 'me -k'?).  I hate these
> types of passwords, because I'm constantly opening this file to look for
> them.
>
> Lately, for passwords, I've been using the system once popularized by
> CompuServe, which amounts to <some-word> <some-punctuation-char>
> <some-other-word>, for example, barge%water.  Which is very similar to
> the algorithm used by dotcoms to generate unique company (and domain)
> names:  things like RedGorilla and BlueToenail.  So, hey, let's hear it
> for the automatic company name generator feature... ;-)
>
>

We will leave that as an exercise for the reader. You should be able to
hook some words out of the spelling dictionary.

Jon.

---

From: jasspa@yahoogroups.com [mailto:jasspa@yahoogroups.com] On Behalf Of Jon Green
Sent: Sunday, February 25, 2007 3:41 PM
To: jasspa@yahoogroups.com
Subject: [jasspa] Addition to bad 'C' hilighting

Follows an extension to the 'Bad 'C' (also C++/Java) highlighting in
hkc.emf in the "; Pick up bogus 'C' constructs extension" block.

; Pick up a bad assignment in a conditional statement. e.g.
; if (a = SOME_CONST)
; while (a = SOME_CONST)
hilight .hilight.c 2 "\\s\\(if\\|while\\)[ \t]*([ \t]*\\w+[ \t]*=[
\t]*[a-zA-Z_0-9]+[ \t]*)" .scheme.hlred

Seems to pick up bad assignments and not cause any miss highlighting.
Will only pick up a simple expression.

Jon.

>> p.s. I did figure out the spelling dictionaries issue: the spell
>> directory wasn't in my MEPATH.
>
> You still using a MEPATH? Not used a MEPATH for years normally put the
> jasspa directory somewhere where ME automatically finds it and then dump
> the spelling directories in the spell sub-directory.
>
> Jon.

Hmmm.  Yes, I see: /opt/jasspa or /usr/share/jasspa or /usr/local/jasspa
are the wired-in locations.

I guess I should be using MEINSTALLPATH to point to the top of my
install tree, instead of setting MEPATH.  It would be a little less
typing ;-)

BTW I use the following lame wrapper script to run the thing.  It puts
it in the background, so as not to monopolize the xterm (a la Windows),
or runs the console version if X11 isn't available.

#!/bin/sh
#------------------------------------------------------------
# me.sh  Thundt  rev 3  03-Aug-04
# Wrapper around MicroEmacs; doesn't monopolize xterm
# Copy into your ~/bin
#------------------------------------------------------------
# Rename 'me' to 'me32'. Put this wrapper in ~/bin/me.

##set -v  # uncomment for debug
##set -x

#-----
# Make dirs and set paths as needed
#-----
if [ ! -d ${HOME}/.jasspa ]; then
      mkdir ${HOME}/.jasspa
fi

MEBACKUPPATH="${HOME}/Backup/Jasspa_backups/"
if [ ! -d ${MEBACKUPPATH} ]; then
      mkdir -p ${MEBACKUPPATH}
fi

MEPATH=${HOME}/.jasspa:/home/thundt/App/Jasspa/company:/home/thundt/App/Jasspa/m\
acros:/home/thundt/App/Jasspa/spelling
export MEBACKUPPATH MEPATH

#-----
# Run it in X or console version, as appropriate
#-----
if [ "${DISPLAY}x" = "x" ]; then
      /home/thundt/bin/me32 -n $*  # no DISPLAY -> run termcap version
(in fg)
else
      if [ "$1" = "-n" -o  "$2" = "-n" -o  "$3" = "-n" -o  "$4" = "-n" \
          -o  "$5" = "-n" -o "$5" = "-n" -o  "$6" = "-n" -o  "$7" = "-n" \
          -o  "$8" = "-n" -o  "$9" = "-n" ]; then
          /home/thundt/bin/me32 $*  # user requested non-X; run in fg
      else
          /home/thundt/bin/me32 $* &  # X version; run in bg
      fi
fi



--
Thomas Hundt <tom@...> +1-415-867-6698

Phillips, Steven wrote:
> Im not sure how this works because it shouldn't, the hilighting does not
> support words in groups only single letters - following is a correction:
>
>         hilight .hilight.c 2 "\\s\\{if <file://\\s\\{if>[ \t]*([
> \t]*\\w+[ \t]*=[ \t]*\\w+[ \t]*)" .scheme.hlred
>         hilight .hilight.c 2 "\\s\\{while <file://\\s\\{while>[ \t]*([
> \t]*\\w+[ \t]*=[ \t]*\\w+[ \t]*)" .scheme.hlred
> Note I have changed the [a-zA-Z_0-9] to a \w as they are equivalent,
> however I wonder whether this should be simply testing for  'if ( a =
> S', this would spot things like 'if(a=b-1)' and any assignment in an
> if/while can avoid this hilighting and compile warning by using double
> brackets, i.e. 'if((a=b-1))'
>
> Steve
>

Yep - darned must have not executed it again when I added the while
(final trimming that killed the turkey).

I like the \S will give this a try.

Sorry.
Jon.

Thomas Hundt wrote:
>>> p.s. I did figure out the spelling dictionaries issue: the spell
>>> directory wasn't in my MEPATH.
>> You still using a MEPATH? Not used a MEPATH for years normally put the
>> jasspa directory somewhere where ME automatically finds it and then dump
>> the spelling directories in the spell sub-directory.
>>
>> Jon.
>
> Hmmm.  Yes, I see: /opt/jasspa or /usr/share/jasspa or /usr/local/jasspa
> are the wired-in locations.
>
> I guess I should be using MEINSTALLPATH to point to the top of my
> install tree, instead of setting MEPATH.  It would be a little less
> typing ;-)
>
> BTW I use the following lame wrapper script to run the thing.  It puts
> it in the background, so as not to monopolize the xterm (a la Windows),
> or runs the console version if X11 isn't available.
>
> #!/bin/sh
> #------------------------------------------------------------
> # me.sh  Thundt  rev 3  03-Aug-04
> # Wrapper around MicroEmacs; doesn't monopolize xterm
> # Copy into your ~/bin
> #------------------------------------------------------------
> # Rename 'me' to 'me32'. Put this wrapper in ~/bin/me.
>
> ##set -v  # uncomment for debug
> ##set -x
>
> #-----
> # Make dirs and set paths as needed
> #-----
> if [ ! -d ${HOME}/.jasspa ]; then
>      mkdir ${HOME}/.jasspa
> fi
>

You should not need this as when you run me for the first time then it
will create this if it does not exist.

> MEBACKUPPATH="${HOME}/Backup/Jasspa_backups/"
> if [ ! -d ${MEBACKUPPATH} ]; then
>      mkdir -p ${MEBACKUPPATH}
> fi

You do need this, although you could add this to your me.emf (including
the creating the directory).

>
>
MEPATH=${HOME}/.jasspa:/home/thundt/App/Jasspa/company:/home/thundt/App/Jasspa/m\
acros:/home/thundt/App/Jasspa/spelling
> export MEBACKUPPATH MEPATH
>

I still don't think you need MEPATH if you do a non-system installation.

If you move the directories "spelling", "macros" and "company" into the
.jasspa directory then they will be found automatically without $MEPATH.
i.e. this is the same as unpacking metree in ${HOME} and renaming jasspa
to .jasspa

If you do want to retain the directories in their current location then
you could use $MEINSTALLPATH="/home/thundt/App/Jasspa" the content of
the subdirectories will be found.

Having said all of that $MEPATH is fastest as it does not search any
other location.

> #-----
> # Run it in X or console version, as appropriate
> #-----
> if [ "${DISPLAY}x" = "x" ]; then
>      /home/thundt/bin/me32 -n $*  # no DISPLAY -> run termcap version
> (in fg)
> else
>      if [ "$1" = "-n" -o  "$2" = "-n" -o  "$3" = "-n" -o  "$4" = "-n" \
>          -o  "$5" = "-n" -o "$5" = "-n" -o  "$6" = "-n" -o  "$7" = "-n" \
>          -o  "$8" = "-n" -o  "$9" = "-n" ]; then
>          /home/thundt/bin/me32 $*  # user requested non-X; run in fg
>      else
>          /home/thundt/bin/me32 $* &  # X version; run in bg
>      fi
> fi
>
>
>

Modified:
         ;a leading : or math operator indicates a continued line
(except * because of javadoc)
         indent #l5 s "^\\s*[\\.\\+-/%:][^\\*/]" t

Added:
         ; a trailing = starts a continuation
         indent #l5 c "=\\s*$" t


Example:
        SomeRidiculouslyLongClassName myRediculousThing =
                 ThingFactory.createSomething();
        assert myRediculousThing.isReallySomething()
                 : "omg these classes and methods have long names";


Cheers
-- bryan

I swear this used to work... (I'm using 2006/09/09 now.)
Hitting a period no longer quits query-replace-mode.
It seems to have been replaced by an 'L' for "last" command.
(What I want is a way to quit out of it elegantly, rather than by brute
force with C-g.)

On the plus side, I see we now have an 'a' for "replace all remaining"
now, which I've been wanting for a long time.  (I would have bound this
to '!'.)

Is there a way to rebind these keys, so I can fix this myself?

(Whenever possible, I make all software use the same bindings, for
obvious reasons.  And I try to standardize on the standard GNU Emacs
bindings over Jasspa's, because when faced with an unfamiliar machine,
the odds are about 10,000 to 1 that GNU will be installed, and not
Jasspa.  It's nice to just be able to get to work.)

Thanks


FYI here's the GNU help for this function (query-replace):
> Type Space or `y' to replace one match, Delete or `n' to skip to next,
> RET or `q' to exit, Period to replace one match and exit,
> Comma to replace but not move point immediately,
> C-r to enter recursive edit (C-M-c to get out again),
> C-w to delete match and recursive edit,
> C-l to clear the screen, redisplay, and offer same replacement again,
> ! to replace all remaining matches with no more questions,
> ^ to move point back to previous match,
> E to edit the replacement string



--
Thomas Hundt <tom@...> +1-415-867-6698

! works for "replace all remaining", I think it's an undocumented feature.

On 3/22/07, Thomas Hundt <thundt@...> wrote:
>
>
>
>
>
>
> I swear this used to work... (I'm using 2006/09/09 now.)
>  Hitting a period no longer quits query-replace-mode.
>  It seems to have been replaced by an 'L' for "last" command.
>  (What I want is a way to quit out of it elegantly, rather than by brute
>  force with C-g.)
>
>  On the plus side, I see we now have an 'a' for "replace all remaining"
>  now, which I've been wanting for a long time.  (I would have bound this
>  to '!'.)
>
>  Is there a way to rebind these keys, so I can fix this myself?
>
>  (Whenever possible, I make all software use the same bindings, for
>  obvious reasons.  And I try to standardize on the standard GNU Emacs
>  bindings over Jasspa's, because when faced with an unfamiliar machine,
>  the odds are about 10,000 to 1 that GNU will be installed, and not
>  Jasspa.  It's nice to just be able to get to work.)
>
>  Thanks
>
>  FYI here's the GNU help for this function (query-replace):
>  > Type Space or `y' to replace one match, Delete or `n' to skip to next,
>  > RET or `q' to exit, Period to replace one match and exit,
>  > Comma to replace but not move point immediately,
>  > C-r to enter recursive edit (C-M-c to get out again),
>  > C-w to delete match and recursive edit,
>  > C-l to clear the screen, redisplay, and offer same replacement again,
>  > ! to replace all remaining matches with no more questions,
>  > ^ to move point back to previous match,
>  > E to edit the replacement string
>
>  --
>  Thomas Hundt <tom@...> +1-415-867-6698
>

The '.' command used to exit immediately moving the cursor back to the
starting point (regardless of whether you made any substitutions), I
failed to see the use of this and given it was also incompatible to GNU
emacs I removed it in sept 2006.

The 'l' command is equivalent to GNU Emacs '.', we do not have an
equivalent to 'q' or  RET which simply exits at the current location, I
think this is an omission that should be fixed, 'return' seems in
keeping with isearch.

Unfortunately there is no reasonable way to changing these binding
(might be posible to wrap the command around calls to translate-key but
I think this would be too dangerous to be worth it),

Steve

Thomas Hundt wrote:
>
> I swear this used to work... (I'm using 2006/09/09 now.)
> Hitting a period no longer quits query-replace-mode.
> It seems to have been replaced by an 'L' for "last" command.
> (What I want is a way to quit out of it elegantly, rather than by brute
> force with C-g.)
>
> On the plus side, I see we now have an 'a' for "replace all remaining"
> now, which I've been wanting for a long time. (I would have bound this
> to '!'.)
>
> Is there a way to rebind these keys, so I can fix this myself?
>
> (Whenever possible, I make all software use the same bindings, for
> obvious reasons. And I try to standardize on the standard GNU Emacs
> bindings over Jasspa's, because when faced with an unfamiliar machine,
> the odds are about 10,000 to 1 that GNU will be installed, and not
> Jasspa. It's nice to just be able to get to work.)
>
> Thanks
>
> FYI here's the GNU help for this function (query-replace):
> > Type Space or `y' to replace one match, Delete or `n' to skip to next,
> > RET or `q' to exit, Period to replace one match and exit,
> > Comma to replace but not move point immediately,
> > C-r to enter recursive edit (C-M-c to get out again),
> > C-w to delete match and recursive edit,
> > C-l to clear the screen, redisplay, and offer same replacement again,
> > ! to replace all remaining matches with no more questions,
> > ^ to move point back to previous match,
> > E to edit the replacement string
>
> --
> Thomas Hundt <tom@... <mailto:tom%40hundt.com>> +1-415-867-6698
>
>

Attached (this time) some more refinements for Doxygen with 'C' dealing
with [in,out] labelling of parameters.

Similar will apply to cpp.

Jon.

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;
; Author : $Author$
; Created By : Jon Green
; Created : Wed Nov 8 21:32:44 2006
; Last Modified : <070324.1608>
;
; Description : Override the C template.
; - Change the margin indent to 40 for comment align.
; - Extend Doxygen to highlight the parameter name not just the keyword.
;
; Notes : Copied and extended from hkc.emf
;
; Copyright (c) 2006 Jon Green.
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

; Extend the indentation to allow for margin indentation.
!if &sin "d" .fhook-c.setup
     indent .hilight.c "m" 40
!endif

; Doxygen highlighting extensions. Highlights keywords in Doxygen
!if &and &sin "h" .fhook-c.setup &band .hilight.flags 0x02
     !if &sin "l" .fhook-c.setup
         ; Create a new doxygen parameter hilighting scheme
         !if ¬ &exi .hilight.pdoxygen
             set-variable .hilight.pdoxygen &pinc .hilight.next 1
             set-variable .hilight.pdoxygencpp &pinc .hilight.next 1
         !endif
         ;
         ; Special hilighting for a 'C' parameter - branch from the doxygen
         ; space into scheme for the parameter. Note that @param is defined
         ; after the generic '@' definitions.
         ;
         ; Special for @param, @see, @retval
         hilight .hilight.cdoxygen 0x80 "[\\\\@]file"   .hilight.pdoxygen
.scheme.keyword
         hilight .hilight.cdoxygen 0x80 "[\\\\@]struct" .hilight.pdoxygen
.scheme.keyword
         hilight .hilight.cdoxygen 0x80 "[\\\\@]param"  .hilight.pdoxygen
.scheme.keyword
         hilight .hilight.cdoxygen 0x80 "[\\\\@]see"    .hilight.pdoxygen
.scheme.keyword
         hilight .hilight.cdoxygen 0x80 "[\\\\@]enum"   .hilight.pdoxygen
.scheme.keyword
         hilight .hilight.cdoxygen 0x80 "[\\\\@]retval" .hilight.pdoxygen
.scheme.keyword
         hilight .hilight.cdoxygen 0x80 "[\\\\@]a"      .hilight.pdoxygen
.scheme.keyword
         ; Doxygen parameter hilighting space.
         0 hilight .hilight.pdoxygen 2 50 .scheme.error
         ; Hilight the next word and finish.
         hilight .hilight.pdoxygen 0x80 "\\s+[a-zA-Z0-9.()!_\\.]+"
.hilight.cdoxygen .scheme.prepro
         ; Hilight the optional parameter definitions.
         hilight .hilight.pdoxygen 0 "\\s*\\[\\s*in\\s*\\]" .scheme.variable
         hilight .hilight.pdoxygen 0 "\\s*\\[\\s*out\\s*\\]" .scheme.variable
         hilight .hilight.pdoxygen 0 "\\s*\\[\\s*in\\s*,\\s*out\\s*\\]"
.scheme.variable
         hilight .hilight.pdoxygen 0 "\\s*\\[\\s*out\\s*,\\s*in\\s*\\]"
.scheme.variable
         ; We must return to C when we see the close of comment "*/"
         hilight .hilight.pdoxygen 0x80 "\\*/" .hilight.c .scheme.error
         ;
         ; Special hilighting for a 'C++' parameter - branch from the doxygen
         ; space into scheme for the parameter. Note that @param is defined
         ; after
         ; the generic '@' definitions.
         ;
         ; Special for @param, @see, @retval
         hilight .hilight.cdoxygencpp 0x80 "[\\\\@]a"      .hilight.pdoxygencpp
.scheme.keyword
         hilight .hilight.cdoxygencpp 0x80 "[\\\\@]file"   .hilight.pdoxygencpp
.scheme.keyword
         hilight .hilight.cdoxygencpp 0x80 "[\\\\@]struct" .hilight.pdoxygencpp
.scheme.keyword
         hilight .hilight.cdoxygencpp 0x80 "[\\\\@]param"  .hilight.pdoxygencpp
.scheme.keyword
         hilight .hilight.cdoxygencpp 0x80 "[\\\\@]see"    .hilight.pdoxygencpp
.scheme.keyword
         hilight .hilight.cdoxygencpp 0x80 "[\\\\@]enum"   .hilight.pdoxygencpp
.scheme.keyword
         hilight .hilight.cdoxygencpp 0x80 "[\\\\@]retval" .hilight.pdoxygencpp
.scheme.keyword
         ; Doxygen parameter hilighting space.
         0 hilight .hilight.pdoxygencpp 2 50 .scheme.error
         ; Hilight the optional parameter definitions.
         hilight .hilight.pdoxygencpp 0 "\\s*\\[\\s*in\\s*\\]" .scheme.variable
         hilight .hilight.pdoxygencpp 0 "\\s*\\[\\s*out\\s*\\]" .scheme.variable
         hilight .hilight.pdoxygencpp 0 "\\s*\\[\\s*in\\s*,\\s*out\\s*\\]"
.scheme.variable
         hilight .hilight.pdoxygencpp 0 "\\s*\\[\\s*out\\s*,\\s*in\\s*\\]"
.scheme.variable
         ; Hilight the next word and finish.
         hilight .hilight.pdoxygencpp 0x80 "\\S+[a-zA-Z0-9.()!_\\.]+"
.hilight.cdoxygencpp .scheme.prepro
         ; We must return to C when we see the close of comment "*/"
         hilight .hilight.pdoxygencpp 0x80 "$" .hilight.c .scheme.error
     !endif
     ; Pick up bogus 'C' constructs extension
     !if &sin "v" .fhook-c.setup
         ; Detect a for loop with no body - put semi-colon on next line, i.e.
         ; for (i = 1; i < 10; i++);
         ; { ... }
         hilight .hilight.c  2 "\\s\\{for[ \t]*([^()]*)[ \t]*;" .scheme.hlred
         ; Detect a bad 'if' statement that is terminated with ';' e.g.
         ; if (i < x);
         ; { ... }
         hilight .hilight.c  2 "\\s\\{if[ \t]*([^()]*)[ \t]*;"  .scheme.hlred
         ; Detect a spurious '\' at the end of the line that is
         ; not a string or #define. i.e.
         ; char *d = "This is a " \
         ;           "Duff string continuation";
         hilight .hilight.c  2 "\\\\$"                   .scheme.hlred
         ; Pick up a bad assignment in a conditional statement. e.g.
         ; if (a = SOME_CONST)
         ; while (a = SOME_CONST)
         hilight .hilight.c 2 "\\s\\{if[ \t]*([ \t]*\\w+[ \t]*=[ \t]*\\w+[ \t]*)"
.scheme.hlred
         hilight .hilight.c 2 "\\s\\{while[ \t]*([ \t]*\\w+[ \t]*=[ \t]*\\w+[
\t]*)" .scheme.hlred
     !endif
!endif

Attached some more refinements for Doxygen with 'C' dealing with
[in,out] labelling of parameters.

Similar will apply to cpp.

Jon.

Hi all,

I would like to enlarge the system wide font size for ME on Solaris.
The new 19 inch TFTs in our computer lab make the default font size (15
pt) look very tiny.
According to change-font(2) I could set thing up in my personal
.Xdefaults file, but I see no way to do that in a system wide ME global
setup file (I installed ME there) for all users. Otherwise most people
will refuse to consider ME a useful tool.

Gruß / Regards
		    Christof

--
http://badvista.fsf.org - it's time to leave!

Christof Boeckler wrote:
> Hi all,
>
> I would like to enlarge the system wide font size for ME on Solaris.
> The new 19 inch TFTs in our computer lab make the default font size (15
> pt) look very tiny.
> According to change-font(2) I could set thing up in my personal
> .Xdefaults file, but I see no way to do that in a system wide ME global
> setup file (I installed ME there) for all users. Otherwise most people
> will refuse to consider ME a useful tool.
>
> Gruß / Regards
> 		   Christof
>

Sounds like you are sysadmin on this - why not edit me.emf and add a
little start-up macro that:

a) Sets the font to new default.
b) Checks the users .Xdefaults file for a "MicroEmacs.font" entry, if it
does not exist then adds one (or creates the file if it does not exist).

I like option b) better than a) because it will not cause any re-sizze
anomalies and allows the user to change it themselves and not get your
default as part of the start-up.

Now I'm sure that writing  a macro for b) is going to be fairly easy -
but a little late for me at the moment.

i.e.
find-file ~/.Xdefaults
search-forward "^MicroEmacs.font"
if it does not exist
     end-of-buffer
     insert-string "MicroEmacs.font ..."
     save the file
endif
kill-buffer ".Xdefault"


Regards
Jon.

Jon Green wrote:
> Christof Boeckler wrote:
>> Hi all,
>>
>> I would like to enlarge the system wide font size for ME on Solaris.
>> The new 19 inch TFTs in our computer lab make the default font size (15
>> pt) look very tiny.
>> According to change-font(2) I could set thing up in my personal
>> .Xdefaults file, but I see no way to do that in a system wide ME global
>> setup file (I installed ME there) for all users. Otherwise most people
>> will refuse to consider ME a useful tool.
>>
>> Gruß / Regards
>> 		   Christof
>>
>
> Sounds like you are sysadmin on this - why not edit me.emf and add a
> little start-up macro that:
>
> a) Sets the font to new default.
> b) Checks the users .Xdefaults file for a "MicroEmacs.font" entry, if it
> does not exist then adds one (or creates the file if it does not exist).
>
> I like option b) better than a) because it will not cause any re-sizze
> anomalies and allows the user to change it themselves and not get your
> default as part of the start-up.
>
> Now I'm sure that writing  a macro for b) is going to be fairly easy -
> but a little late for me at the moment.
>
> i.e.
> find-file ~/.Xdefaults
> search-forward "^MicroEmacs.font"
> if it does not exist
>     end-of-buffer
>     insert-string "MicroEmacs.font ..."
>     save the file
> endif
> kill-buffer ".Xdefault"
>
>
> Regards
> Jon.
>

A good place to start is newuser.emf this is the start up that
configures the environment for a new user. You can probably just change
the font and force it into the session - probably easier than above.

Jon.

Hi,
while using 20060909 for 10.3 MacOS....

... accidentially managed to unbind the RETURN terminator, then had to
force-quit the shell entirely, as no amount of fiddling allowed me to
recover it.

How do I prevent that from happening in the future?

(in other words: what is the prescribed way of exiting the potentially
destructive global-unbind-key op ["esc C-k"] ?)

Also, sorry if that's an RTFM question: is there an equivalent to
ME3.9 emacs.rc file in the distribution, one executed by default upon
launch?

Thanks in advance.

Ian
this post <http://tech.groups.yahoo.com/group/jasspa/message/2247>

Ian wrote:
> Hi,
> while using 20060909 for 10.3 MacOS....
>
> ... accidentially managed to unbind the RETURN terminator, then had to
> force-quit the shell entirely, as no amount of fiddling allowed me to
> recover it.
>
> How do I prevent that from happening in the future?
>
> (in other words: what is the prescribed way of exiting the potentially
> destructive global-unbind-key op ["esc C-k"] ?)
>
> Also, sorry if that's an RTFM question: is there an equivalent to
> ME3.9 emacs.rc file in the distribution, one executed by default upon
> launch?
>
> Thanks in advance.
>
> Ian
> this post <http://tech.groups.yahoo.com/group/jasspa/message/2247>
>

Hi Ian,

I know you are transitioning across (I saw your subscribe message) - it
will take a little while to get familiar, but not long.

Hopefully you have found <user>.emf file - not sure where this lives on
a Mac, but it is in your home directory. This is equivalent to your old
ue emacs.rc (You should find it here $home/.jasspa/<user>.emf). esc-x
user-setup tells you where it is.

In the distribution macros i.e. /opt/jasspa/contrib or
/usr/share/jasspa/contrib or c:\program files\jasspa\microemacs\contrib
etc. then there is a file called "user.emf". This has lots of stuff in
it that may be useful for new users. You can copy this into your
<user>.emf in full or take bits from it and add.

Right at the bottom of this file you will see how to disable the unbind
key which is as follows:

; If you loose the C-k kill-line key then you are miss-typing and unbinding.
; Remove the global-unbind-key binding so this cannot happen.
!force global-unbind-key "esc C-k"

So add this to your <user>.emf and then restart (or do esc-x
execute-buffer to load it in).

Hope that helps.

Regards
Jon.

Below is an initial implementation of a language template for the
Groovy programming language.

cheers
-- bryan


;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;
;  Created By    : Bryan Schofield
;  Created       : Mon Apr 2 09:38:19 2007
;  Last Modified : <070402.0940>
;
;  Description
;   Groovy Language Template
;   http://groovy.codehaus.org
;
;  Notes
;
;  History
;
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;


define-macro fhook-groovy
     set-variable $buffer-mask "luh1"
     @# buffer-init "groovy"
     buffer-init-hooks
!emacro

; buffer-init variables
set-variable .fhook-groovy.setup ® "/history/fhook/groovy" "bdfghinopx"
set-variable .fhook-groovy.setup-mask "abdefghikmnoptux"
set-variable .fhook-groovy.comment "|/*| */|*| * | * |fr|"
set-variable .fhook-groovy.comment-1 "|/**| */|*| * | * |f|"

set-variable .fhook-groovy.collapse-open  "^class"
set-variable .fhook-groovy.collapse-close "^}"
set-variable .fhook-groovy.collapse-mclose "1"
set-variable .fhook-groovy.collapse-mnext "-1"

set-variable .fhook-groovy.item-list-s1 "^[\t ]*class[\t
]+\\(\\w[1-9A-Za-z_]*\\)[\t ]*<?<?[\t ]*[A-Za-z0-9_]*"
set-variable .fhook-groovy.item-list-r1 "class \ecB\\1\ecA"
set-variable .fhook-groovy.item-list-s2 "^[\t ]*def[\t
]+\\(\\w[1-9A-Za-z_.]*\\)[\t ]*.*$"
set-variable .fhook-groovy.item-list-r2 "def \ecB\\1\ecA"

!if ¬ &exist .hilight.groovy
     set-variable .hilight.groovy  &pinc .hilight.next 1
!endif

!if &and &sin "h" .fhook-groovy.setup &band .hilight.flags 0x02
     0 hilight .hilight.groovy 0x00 $global-scheme
     ; Comments
     hilight .hilight.groovy 20 "/\\*" "*/" "" .scheme.comment
     hilight .hilight.groovy  2 "//"           .scheme.comment
     ; Comment TODO's
     hilight .hilight.groovy 20 "/\\*\\s+[Tt][Oo][Dd][Oo]" "*/" "" .scheme.error
     hilight .hilight.groovy 18 "//\\s*[tT][oO][dD][oO]"           .scheme.error

     ; Strings
     hilight .hilight.groovy  4 "\"" "\"" "\\"  .scheme.string
     hilight .hilight.groovy  4 "\'" "\'" "\\"  .scheme.string
     hilight .hilight.groovy  4 "\"\"\"" "\"\"\"" "\\"  .scheme.quote
     ; groovy string literal /.foo[0-9]\s*bar/ hiligh gets confused
with simple division.
     ;hilight .hilight.groovy  4 "/" "/" "" .scheme.quote

     ; constants: A_CONSTANT_VARAIBLE
     hilight .hilight.groovy 1    "[A-Z_]+"          .scheme.constant
     ; class names: AJavaClassName
     hilight .hilight.groovy 1    "[A-Z]+[a-z]+\\w*" .scheme.type

     ; Keywords
     hilight .hilight.groovy 0x12 "^\\s*import"       .scheme.prepro
     hilight .hilight.groovy 0x12 "^\\s*package"      .scheme.prepro
     hilight .hilight.groovy 1    "abstract"     .scheme.keyword
     hilight .hilight.groovy 1    "as"           .scheme.keyword
     hilight .hilight.groovy 1    "assert"       .scheme.keyword
     hilight .hilight.groovy 1    "boolean"      .scheme.type
     hilight .hilight.groovy 1    "break"        .scheme.operator
     hilight .hilight.groovy 1    "byte"         .scheme.type
     hilight .hilight.groovy 1    "case"         .scheme.operator
     hilight .hilight.groovy 1    "catch"        .scheme.operator
     hilight .hilight.groovy 1    "char"         .scheme.type
     hilight .hilight.groovy 1    "class"        .scheme.keyword
     hilight .hilight.groovy 1    "continue"     .scheme.operator
     hilight .hilight.groovy 1    "def"          .scheme.keyword
     hilight .hilight.groovy 1    "default"      .scheme.operator
     hilight .hilight.groovy 1    "do"           .scheme.operator
     hilight .hilight.groovy 1    "double"       .scheme.type
     hilight .hilight.groovy 1    "else"         .scheme.operator
     hilight .hilight.groovy 1    "enum"         .scheme.keyword
     hilight .hilight.groovy 1    "extends"      .scheme.keyword
     hilight .hilight.groovy 1    "false"        .scheme.constant
     hilight .hilight.groovy 1    "final"        .scheme.keyword
     hilight .hilight.groovy 1    "finally"      .scheme.operator
     hilight .hilight.groovy 1    "float"        .scheme.type
     hilight .hilight.groovy 1    "for"          .scheme.operator
     hilight .hilight.groovy 1    "if"           .scheme.operator
     hilight .hilight.groovy 1    "implements"   .scheme.keyword
     hilight .hilight.groovy 1    "import"       .scheme.keyword
     hilight .hilight.groovy 1    "in"           .scheme.operator
     hilight .hilight.groovy 1    "instanceof"   .scheme.keyword
     hilight .hilight.groovy 1    "int"          .scheme.type
     hilight .hilight.groovy 1    "interface"    .scheme.keyword
     hilight .hilight.groovy 1    "long"         .scheme.type
     hilight .hilight.groovy 1    "native"       .scheme.keyword
     hilight .hilight.groovy 1    "new"          .scheme.keyword
     hilight .hilight.groovy 1    "null"         .scheme.constant
     hilight .hilight.groovy 1    "package"      .scheme.keyword
     hilight .hilight.groovy 1    "print"        .scheme.keyword
     hilight .hilight.groovy 1    "println"      .scheme.keyword
     hilight .hilight.groovy 1    "private"      .scheme.keyword
     hilight .hilight.groovy 1    "property"     .scheme.keyword
     hilight .hilight.groovy 1    "protected"    .scheme.keyword
     hilight .hilight.groovy 1    "public"       .scheme.keyword
     hilight .hilight.groovy 1    "return"       .scheme.operator
     hilight .hilight.groovy 1    "short"        .scheme.type
     hilight .hilight.groovy 1    "static"       .scheme.keyword
     hilight .hilight.groovy 1    "strictfp"     .scheme.keyword
     hilight .hilight.groovy 1    "super"        .scheme.keyword
     hilight .hilight.groovy 1    "switch"       .scheme.operator
     hilight .hilight.groovy 1    "synchronized" .scheme.keyword
     hilight .hilight.groovy 1    "this"         .scheme.keyword
     hilight .hilight.groovy 1    "throw"        .scheme.operator
     hilight .hilight.groovy 1    "throws"       .scheme.keyword
     hilight .hilight.groovy 1    "transient"    .scheme.keyword
     hilight .hilight.groovy 1    "true"         .scheme.constant
     hilight .hilight.groovy 1    "try"          .scheme.operator
     hilight .hilight.groovy 1    "void"         .scheme.keyword
     hilight .hilight.groovy 1    "volatile"     .scheme.keyword
     hilight .hilight.groovy 1    "while"        .scheme.operator

     hilight .hilight.groovy 1    "; *$"       .scheme.error

!endif

!if &sin "d" .fhook-groovy.setup
     !if ¬ &exi .indent.groovy-comm
         set-variable .indent.groovy-comm &pinc .hilight.next 1
         set-variable .indent.groovyT &pinc .hilight.next 1
     !endif


     0 indent .indent.groovy-comm 1 20
     0 indent .indent.groovyT 1 500
     0 indent .hilight.groovy 2 20 .indent.groovyT

     ; switch to comment indentation mode
     indent .indent.groovyT t "/\\*"   .indent.groovy-comm
     indent .hilight.groovy  x "/\\*" 1 .indent.groovy-comm
     indent .indent.groovy-comm x "\\*/" -1 0
     indent .indent.groovyT t "\\*/" 0
     indent .hilight.groovy o "\\*/" -1


     ; strings
     indent .hilight.groovy e "\"" "\"" "\\"
     indent .hilight.groovy e "\'" "\'" "\\"
     ; standard braces
     indent .hilight.groovy n "{"  t
     indent .hilight.groovy o "}" -t
     indent .hilight.groovy n "(" 3/2t
     indent .hilight.groovy o ")" -3/2t

!endif

buffer-init-fhook "groovy"

Is there any short cut to insert the current buffer's file name in to
a shell command, using shell-command, pipe-shell-command, or
ipipe-shell-command? Something like vi's %?

for example,

    someScript -opt %f

There seems to be ("system") variables of

$buffer-fname

	 and

$file-names

which may be what you're looking for... see

"M-x list-variables" and/or "M-x command-apropos"

	 $buffer

Ian
this post <http://tech.groups.yahoo.com/group/jasspa/message/2251>

On 4/2/07, Bryan Schofield <schofield.bryan@...> wrote:
> Is there any short cut to insert the current buffer's file name in to
> a shell command, using shell-command, pipe-shell-command, or
> ipipe-shell-command? Something like vi's %?
>
> for example,
>
>    someScript -opt %f

I don't think there's a shortcut.

I think you're meant to construct the command string yourself.
As Ian just pointed out, list-variables shows various variables that
hold various forms of the filename.

If you want not to have to worry about the filename, maybe try
pipe-shell-command (and supply your own result buffer name to catch the
results).

-Th


Bryan Schofield wrote:
> Is there any short cut to insert the current buffer's file name in to
> a shell command, using shell-command, pipe-shell-command, or
> ipipe-shell-command? Something like vi's %?
>
> for example,
>
>    someScript -opt %f
>

---

From: jasspa@yahoogroups.com [mailto:jasspa@yahoogroups.com] On Behalf Of Thomas Hundt
Sent: Tuesday, April 03, 2007 1:35 AM
To: jasspa@yahoogroups.com
Subject: Re: [jasspa] shell commands

I don't think there's a shortcut.

I think you're meant to construct the command string yourself.
As Ian just pointed out, list-variables shows various variables that
hold various forms of the filename.

If you want not to have to worry about the filename, maybe try
pipe-shell-command (and supply your own result buffer name to catch the
results).

-Th

Bryan Schofield wrote:
> Is there any short cut to insert the current buffer's file name in to
> a shell command, using shell-command, pipe-shell-command, or
> ipipe-shell-command? Something like vi's %?
>
> for example,
>
> someScript -opt %f
>