Meta characters in PicoLisp

2013-05-07 Thread Thorsten Jolitz

Hi List, 

citing from the Reference

,
| Internal symbol names can consist of any printable (non-whitespace)
| charac-ter, except for the following meta characters:
| 
|  ’ ( ) , [ ] ‘ ~ { }
`

I forgot what the ~ is about (e.g. in ~( ... )), and could not find
the explanation. 

And isn't ^ another meta character (with what meaning in PicoLisp)?

-- 
cheers,
Thorsten


-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: Meta characters in PicoLisp

2013-05-07 Thread Alexander Burger
Hi Thorsten,

 I forgot what the ~ is about (e.g. in ~( ... )), and could not find
 the explanation. 

It is described in doc/ref.html#macro-io. The tilde is similar to the
backquote, in that both evaluate the following expression, and put the
result into the currently read list. While the backquote inserts the
result as it is, the tilde expects that the result is a list which is
then spliced into the read list.

   : (1 2 `(chop abc) 3 4)
   - (1 2 (a b c) 3 4)

   : (1 2 ~(chop abc) 3 4)
   - (1 2 a b c 3 4)

Besides this, the tilde has a special meaning when separating a namespace
and a symbol name, e.g.  myPackage~mySymbol


 And isn't ^ another meta character (with what meaning in PicoLisp)?

Not in internal symbols. In transient symbols (strings), however, it
denotes a control character. e.g. ^I is TAB and ^M is carriage
return . abc^Jdef^Jghi^J is a text of three lines.

♪♫ Alex
-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: Meta characters in PicoLisp

2013-05-07 Thread Thorsten Jolitz
Alexander Burger a...@software-lab.de writes:

Hi Alex,

 I forgot what the ~ is about (e.g. in ~( ... )), and could not find
 the explanation. 

 It is described in doc/ref.html#macro-io. The tilde is similar to the
 backquote, in that both evaluate the following expression, and put the
 result into the currently read list. While the backquote inserts the
 result as it is, the tilde expects that the result is a list which is
 then spliced into the read list.

: (1 2 `(chop abc) 3 4)
- (1 2 (a b c) 3 4)

: (1 2 ~(chop abc) 3 4)
- (1 2 a b c 3 4)

I knew I read about it, but forgot where. Searching in 'PicoLisp Works'
did not help, since the tilde character is not found (on scribd at
least). 

 Besides this, the tilde has a special meaning when separating a namespace
 and a symbol name, e.g.  myPackage~mySymbol


 And isn't ^ another meta character (with what meaning in PicoLisp)?

 Not in internal symbols. In transient symbols (strings), however, it
 denotes a control character. e.g. ^I is TAB and ^M is carriage
 return . abc^Jdef^Jghi^J is a text of three lines.

I see, thanks!

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: epil - PicoLisp with Emacs-like command-line (was: Meta key in

2012-11-02 Thread Alexander Burger
Hi Thorsten,

 Maybe there could be a
 1. global switch '*Vi' that is true by default, and
 2. a function '(de Toggle-Vi...)' that toggles that switch, and
 3. three modes: Command, Insert, EInsert 
 or 
 3. two modes Command and Insert, with much Command functionality
 duplicated in Insert

Perhaps it is better to keep the two editors separate? This would make
changes and extensions more modular.


 maybe one could actually have two additional library files 'eled.l' and
 'eedit.l', not change anything in 'led.l' and 'edit.l', and simply

Yes.

 insert conditional statements in the startscripts and the functions that
 load the libraries? I.e., add 'eled.l' and 'eedit.l' to the PicoLisp

I would use the names vled.l and eled.l, and have a symbolic link
led.l to one of those. In that way, no changes to the rest of the
system are necessary.

Cheers,
- Alex
-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: epil - PicoLisp with Emacs-like command-line (was: Meta key in

2012-11-01 Thread Alexander Burger
Hi Thorsten,

 I managed to produce a first experimental version of 'epil', i.e. a new
 version of led.l (eled.l) and some small changes in libraries that are
 called at PicoLisp startup. If you clone this github repo: 
 
 ,-
 | https://github.com/tj64/epil
 `-

Thanks! I think this might make the PicoLisp REPL more usable for many
people, as most are not used to a VI-style command line, and probably
are not even aware of it.

In the long term, the Emacs-interface should be available without
explicit installation, and possibly even co-exist with the VI version
(switchable at runtime). Would this be difficult?

Cheers,
- Alex
-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: epil - PicoLisp with Emacs-like command-line (was: Meta key in

2012-11-01 Thread Thorsten Jolitz
Alexander Burger a...@software-lab.de writes:

Hi Alex,

 I managed to produce a first experimental version of 'epil', i.e. a new
 version of led.l (eled.l) and some small changes in libraries that are
 called at PicoLisp startup. If you clone this github repo: 
 
 ,-
 | https://github.com/tj64/epil
 `-

 Thanks! I think this might make the PicoLisp REPL more usable for many
 people, as most are not used to a VI-style command line, and probably
 are not even aware of it.

Although it seems very efficient, and there are not too many keybindings
to remember, the VI-style command line requieres a completely different
way a thinking for somebody who uses Emacs all the time. 

 In the long term, the Emacs-interface should be available without
 explicit installation, and possibly even co-exist with the VI version
 (switchable at runtime). Would this be difficult?

Since now I know how to assign (almost) all kinds of Emacs keybindings
in PicoLisp, I can try to better understand the 'commandline'
implementation. What I did for 'epil' is simply eliminating 'Command
mode' and shifting the commands into 'Insert mode' with new keybindings.

Maybe there could be a
1. global switch '*Vi' that is true by default, and
2. a function '(de Toggle-Vi...)' that toggles that switch, and
3. three modes: Command, Insert, EInsert 
or 
3. two modes Command and Insert, with much Command functionality
duplicated in Insert

where Command and Insert are the Vi-Style modes and EInsert is the
Emacs-style mode. Since there are conflicting keybindings, one would
call the functions for many keys conditional on the value of '*Vi*'. 

Does that sound like a reasonable approach, or is there a simpler/more
elegant solution?

As an alternative approach: 

maybe one could actually have two additional library files 'eled.l' and
'eedit.l', not change anything in 'led.l' and 'edit.l', and simply
insert conditional statements in the startscripts and the functions that
load the libraries? I.e., add 'eled.l' and 'eedit.l' to the PicoLisp
distribution, and offer the 'epil' shell-command too?

For development, at the moment I rather prefer to have 'epil' separated
from standard PicoLisp, so I can mess up things ;)

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: Meta key in PicoLisp program

2012-10-26 Thread Thorsten Jolitz
Alexander Burger a...@software-lab.de writes:

Hi Alex,

thanks for your explanations. 

 So Ctrl and Shift are the only 'meta-keys' that can be used (and

 Yes, they cause the generation of different ASCII characters by the
 keyboard when they are pressed. For example, if you press the 'A' key
 you get the character 97 ('a'), if you press shift, you get 65 ('A'),
 and if you press control, you get 1 ('^A').

my idea is to add an Emacs-like command line to PicoLisp. I thought this
would be about:

1. move all commands from command-mode to insertion-mode (the only
   Emacs-mode)
2. eliminate command-mode
3. replace the vi 1-key bindings with Emacs keychords (with Control and
   Meta prefix-keys) for all commands that have equal or similar semantics to
   Emacs commands
4. eliminate/replace the commands that do not fit into the non-modal Emacs
   concept
5. write some new commands to give PicoLisp commandline editing the
   authentic Emacs feeling. 

but there is no way to emulate Emacs without the Meta/Alt-key, and e.g.
C-d (Ctrld) is 'delete-char' in Emacs, just like C-c (Ctrlc) is a
very common prefix-key.

 combined)? What about F1 to F12 or any other keys not used for
 'self-insertion''? 

 The function keys are usually generated as escape sequences (system
 dependent).

 You may take a look into lib/term.l in the PicoLisp distribution. It
 defines some keys from the terminfo database (using 'tput').

this looks easy enough, but I'm still looking for a document that
explains what you are doing there. I found lots of material about
changing colors and positioning text with tput, but nothing yet about
how define keys. Do you have a link or so where this is explained in a
basic way?

Do the definitions in lib/term.l only work in xterm (X11 session) or in
a console session (without X11) too?

 Would Emacs-like key-chords be possible? E.g. 'C-x u', i.e. type first
 'Ctrlx' and then 'u', using 'C-x' as a kind of prefix that can be
 combined with e.g. [a-zA-Z0-9].  

 Yes, all this must be handled as events (or sequences of events). Same
 goes for mouse-clicks and -buttons.

Do you have an example of this kind of event-handling in the PicoLisp
sources?

 BTW - where are Ctrl-d and Ctrl-c defined? Could they be redefined to
 something else in one single place?

 No, they are ASCII values. In general, a control character is the
 corresponding uppercase character minus 64. For example, Ctrl-A is 1,
 Ctrl-B is 2, Ctrl-C is 3 and so on.

But would it be possible to make PicoLisp call 'delete-char' when the
Ctrl-d ASCII is send, and use another (less common) ASCII value to exit
the program? 

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: Meta key in PicoLisp program

2012-10-26 Thread Alexander Burger
Hi Thorsten,

 my idea is to add an Emacs-like command line to PicoLisp. I thought this

Good idea :)

 1. move all commands from command-mode to insertion-mode (the only
Emacs-mode)
 2. eliminate command-mode
 3. replace the vi 1-key bindings with Emacs keychords (with Control and
Meta prefix-keys) for all commands that have equal or similar semantics to
Emacs commands
 4. eliminate/replace the commands that do not fit into the non-modal Emacs
concept
 5. write some new commands to give PicoLisp commandline editing the
authentic Emacs feeling. 

Yes. Quite a lot of stuff, though.


 Do the definitions in lib/term.l only work in xterm (X11 session) or in
 a console session (without X11) too?

They should work where 'tput' and 'terminfo' is available.


  Would Emacs-like key-chords be possible? E.g. 'C-x u', i.e. type first
  'Ctrlx' and then 'u', using 'C-x' as a kind of prefix that can be
  combined with e.g. [a-zA-Z0-9].  

Yes, I think so. By handling sequences of key events, collecting them as
necessary, and then deciding what action to perform.


 Do you have an example of this kind of event-handling in the PicoLisp
 sources?

As a central entry point the 'fkey' function from lib/led.l can be
used. An example usage is in

   http://rosettacode.org/wiki/Keyboard_macros#PicoLisp

The flight simulator in rcsim/main.l uses it too, defining some
function keys to control the airplane.

But these examples don't handle sequences of multiple-key events. For
those, a given key like Ctrl-X might then wait for a short timeout
period (e.g. with (key 200)) for further keys to arrive.


  No, they are ASCII values. In general, a control character is the
  corresponding uppercase character minus 64. For example, Ctrl-A is 1,
  Ctrl-B is 2, Ctrl-C is 3 and so on.
 
 But would it be possible to make PicoLisp call 'delete-char' when the
 Ctrl-d ASCII is send, and use another (less common) ASCII value to exit
 the program? 

Yes. Exiting upon Ctrl-D is explicitly coded into the line editor. So
any other key (or none at all) might be used.

Cheers,
- Alex
-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: Meta key in PicoLisp program

2012-10-26 Thread Alexander Burger
On Fri, Oct 26, 2012 at 11:57:34AM +0200, Axel Svensson wrote:
  my idea is to add an Emacs-like command line to PicoLisp. I thought this
 
 Is there anything you want that's not supported by the GNU readline
 library? Since this library is so popular, it's been well tested
 during the years. Maybe it be possible to create picolisp bindings for
 it?

In fact, there is already a simple version in the PicoLisp distribution,
in lib/readline.l. It is for pil32.

In pil64, 'readline' can be directly called via 'native':

   (de *Led (native /lib64/libreadline.so.5 readline 'S NIL))

but this has a memory leak iirc. I haven't investigated closely.


The problem with both solutions is that they don't integrate as well as
lib/led.l with the PicoLisp runtime environment (e.g. tab completion
on the currently existing internal symbols, locking of the history
file), but this can probably be improved upon.

Cheers,
- Alex
-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: Meta key in PicoLisp program

2012-10-26 Thread Alexander Burger
On Fri, Oct 26, 2012 at 12:27:21PM +0200, Alexander Burger wrote:
 The problem with both solutions is that they don't integrate as well as
 lib/led.l with the PicoLisp runtime environment (e.g. tab completion

BTW, the historic reason for PicoLisp's simple line editor is that the
GNU readline library is bigger than the whole PicoLisp interpreter. So
it didn't feel right to link with it.
-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: Meta key in PicoLisp program

2012-10-26 Thread Axel Svensson
 GNU readline library is bigger than the whole PicoLisp interpreter. So
 it didn't feel right to link with it.

Probably right feeling there. In production, command line interaction
might not even be used at all. But the option is nice, so if someone's
willing to tweak readline.l it would be a welcomed contribution.
-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: Meta key in PicoLisp program

2012-10-26 Thread Thorsten Jolitz
Alexander Burger a...@software-lab.de writes:

Hi Alex,

 my idea is to add an Emacs-like command line to PicoLisp. I thought this

 Yes. Quite a lot of stuff, though.

I already have a basic emacs commandline version running, the most basic
Emacs 'movement, deletion and paste' commands, just by changing the
keybinding of existing functions to Emacs bindings with Ctrlprefix. 

But there are still several things I don't really understand: 

  Would Emacs-like key-chords be possible? E.g. 'C-x u', i.e. type first
  'Ctrlx' and then 'u', using 'C-x' as a kind of prefix that can be
  combined with e.g. [a-zA-Z0-9].  

 Yes, I think so. By handling sequences of key events, collecting them as
 necessary, and then deciding what action to perform.


 Do you have an example of this kind of event-handling in the PicoLisp
 sources?

 As a central entry point the 'fkey' function from lib/led.l can be
 used. An example usage is in

http://rosettacode.org/wiki/Keyboard_macros#PicoLisp


in lib/term.l I find: 

,--
| (setq
|*XtF1(in '(tput kf1) (line T))
|*XtF2(in '(tput kf2) (line T))
`--

but in the web I still did not find a list of all the keys like kf1
and kf2 that 'tput' knows about - where can I look them up? I would
e.g. need to know how to adress the Alt key - kalt? I guess not... 

and what about this notation: 

,
| *XtMenu  ^[[29~#?
`

where do I find the explanation for ^[[29~?


In http://rosettacode.org/wiki/Keyboard_macros#PicoLisp I find: 

,--
| (load @lib/term.l)
|  
| (fkey *XtF1
|(prinl Initialized value to  (setq *Number 1)) )
|  
| (fkey *XtUp
|(prinl Incremented to  (inc '*Number)) )
`--

thats easy enough. 

 The flight simulator in rcsim/main.l uses it too, defining some
 function keys to control the airplane.

same thing here: 

,
| (fkey *XtUp
|(down *Model) )
`

 But these examples don't handle sequences of multiple-key events. For
 those, a given key like Ctrl-X might then wait for a short timeout
 period (e.g. with (key 200)) for further keys to arrive.

could you give a little example how this would look like, e.g. if I want
to define the Emacs binding 'C-x u' ('Ctrl-x u'), i.e. press C-x first
(as a prefix) and then u?

 Yes. Exiting upon Ctrl-D is explicitly coded into the line editor. So
 any other key (or none at all) might be used.

I redefined CtrlD to CtrlQ, that was easy. 
(What about CtrlC - where is that defined?)



-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: Meta key in PicoLisp program

2012-10-26 Thread Thorsten Jolitz
Thorsten Jolitz tjol...@googlemail.com
writes:

 but in the web I still did not find a list of all the keys like kf1
 and kf2 that 'tput' knows about - where can I look them up? I would
 e.g. need to know how to adress the Alt key - kalt? I guess not... 

ok, I found the list of keys in the terminfo man-page (I should have
looked first there), so this part of the question is obsolete...

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: Meta key in PicoLisp program

2012-10-24 Thread Thorsten Jolitz
Alexander Burger a...@software-lab.de writes:

Hi Alex,

 AltV (M-V or M-S-v in Emacs)
 CtrlALTV (C-M-V or C-M-S-v in Emacs)
 
 instead of something like this (CtrlV):

 Unfortunately, this is not directly possible.

 Alt/Meta key combinations have no representation in ASCII or UTF-8.
 Instead, they are generated as events by the runtime system, where key
 events may have certain bits set for modifier keys.

 As such, they are very system-dependent. External libraries like
 'termcap/terminfo' or 'curses' can be used to handle them.

 BTW, this is one of the reasons why the PicoLisp line editor uses a
 'vi'-style user interface. It needs only ASCII-keys.

So Ctrl and Shift are the only 'meta-keys' that can be used (and
combined)? What about F1 to F12 or any other keys not used for
'self-insertion''? 

Would Emacs-like key-chords be possible? E.g. 'C-x u', i.e. type first
'Ctrlx' and then 'u', using 'C-x' as a kind of prefix that can be
combined with e.g. [a-zA-Z0-9].  

PS
BTW - where are Ctrl-d and Ctrl-c defined? Could they be redefined to
something else in one single place?

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Meta key in PicoLisp program

2012-10-23 Thread Thorsten Jolitz


Hi List, 

how would I write a 
- Meta/Alt key
- AltGr key

in a PicoLisp program?

E.g. if I want to write 
AltV (M-V or M-S-v in Emacs)
CtrlALTV (C-M-V or C-M-S-v in Emacs)

instead of something like this (CtrlV):

,---
| (^V (insChar (key)))
`---

-- 
cheers,
Thorsten


-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: Meta key in PicoLisp program

2012-10-23 Thread Alexander Burger
Hi Thorsten,

 AltV (M-V or M-S-v in Emacs)
 CtrlALTV (C-M-V or C-M-S-v in Emacs)
 
 instead of something like this (CtrlV):

Unfortunately, this is not directly possible.

Alt/Meta key combinations have no representation in ASCII or UTF-8.
Instead, they are generated as events by the runtime system, where key
events may have certain bits set for modifier keys.

As such, they are very system-dependent. External libraries like
'termcap/terminfo' or 'curses' can be used to handle them.

BTW, this is one of the reasons why the PicoLisp line editor uses a
'vi'-style user interface. It needs only ASCII-keys.

Cheers,
- Alex
-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: meta!

2011-03-14 Thread Alexander Burger
Hi Dexen,

 could we please have the ``Content-Transfer-Encoding: quoted-printable''=20
 handling fixed? :)
 
 I.e., the stuff that currently creates the endless =3D20 etc.

Hmm, OK. But how? The source is in the PicoLisp distribution in
misc/mailing, it doesn't do anything with the encoding.

Cheers,
- Alex
-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: meta!

2011-03-14 Thread dexen deVries
Hi Alex,


On Monday 14 of March 2011 19:13:54 you wrote:
  could we please have the ``Content-Transfer-Encoding:
  quoted-printable''=3D20 handling fixed? :)
 =20
  I.e., the stuff that currently creates the endless =3D3D20 etc.
=20
 Hmm, OK. But how? The source is in the PicoLisp distribution in
 misc/mailing, it doesn't do anything with the encoding.

Let's see :-)


Cheers,
=2D-=20
dexen deVries

[[[=E2=86=93][=E2=86=92]]]

``In other news, STFU and hack.''
mahmud, in response to Erann Gat's ``How I lost my faith in Lisp''
http://news.ycombinator.com/item?id=3D2308816
-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: meta!

2011-03-14 Thread dexen deVries
On Monday 14 of March 2011 19:13:54 you wrote:
  could we please have the ``Content-Transfer-Encoding:
  quoted-printable''=20 handling fixed? :)
  
  I.e., the stuff that currently creates the endless =3D20 etc.
 
 Hmm, OK. But how? The source is in the PicoLisp distribution in
 misc/mailing, it doesn't do anything with the encoding.


And it didn't work.

In short, the messages I receive don't have the ``Content-Transfer-Encoding'' 
header (which is a companion to ``Content-Type''), even the ones I sent with 
that header for sure.

Without this header, the mail client doesn't know to interpret the escape 
codes starting with the ``equals'' sign.

Now I dont' understand the code too well, but it seems only fields indicated by 
a shortlist are copied from incomming message to the outgoing message. Perhaps 
if you added the ``Content-Transfer-Encoding'' field to this list (to be just 
copied), it would work?


Attached patch attempts to fir it~


Cheers,
-- 
dexen deVries

[[[↓][→]]]

``In other news, STFU and hack.''
mahmud, in response to Erann Gat's ``How I lost my faith in Lisp''
http://news.ycombinator.com/item?id=2308816

--Boundary-00=_xLmfN70LoP+89EH
Content-Type: text/x-patch;
  charset=UTF-8;
  name=0001-attempt-to-fix-Content-Transfer-Encoding-header-not-.patch
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;

filename=0001-attempt-to-fix-Content-Transfer-Encoding-header-not-.patch

From e17e05b34cbbbe342d4350cdce8a16fa3b180ea8 Mon Sep 17 00:00:00 2001
From: dexen deVries dexen.devr...@gmail.com
Date: Mon, 14 Mar 2011 19:47:16 +0100
Subject: [PATCH] attempt to fix Content-Transfer-Encoding header not being 
forwarded
X-Face: 
1TH%]0KlR_DQ/V9+m_neC|Yj3$MU@B/vVI\}^j`G/[9JKe=5`p[$l|z^\MRO%k:9xXL[_!jEZo8$[]yo{7O-]t_@Qj`v5H@L[YyU#j7(H)P\{pMs))9$@Vww1ni-{+{5olk=XYG,}|UHQMC08506\0=+p|75ucH!S!1UcXoC1vgpN_{lOMt_54_N5;)ngh6=D^uZN,}ZI9-k9v!}`jpCUMM,//`H6w{KgeM^bS]jwD8'lAEX9F[ScIHig1.r%65;

---
 misc/mailing |4 +++-
 1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/misc/mailing b/misc/mailing
index 1c7087a..aae2dc5 100755
--- a/misc/mailing
+++ b/misc/mailing
@@ -22,7 +22,7 @@
 (while (setq *From (lowc (till   T)))
(off
   *Name *Subject *Date *MessageID *InReplyTo *MimeVersion
-  *ContentType *ContentDisposition *UserAgent )
+  *ContentType *ContentTransferEncoding *ContentDisposition 
*UserAgent )
(while (split (line)  )
   (setq *Line (glue   (cdr @)))
   (case (pack (car @))
@@ -33,6 +33,7 @@
  (In-Reply-To: (setq *InReplyTo *Line))
  (MIME-Version: (setq *MimeVersion *Line))
  (Content-Type: (setq *ContentType *Line))
+ (Content-Transfer-Encoding: (setq 
*ContentTransferEncoding *Line))
  (Content-Disposition: (setq *ContentDisposition *Line))
  (User-Agent: (setq *UserAgent *Line)) ) )
(if (nor (member *From *Mailings) (= subscribe (lowc 
*Subject)))
@@ -66,6 +67,7 @@
 (and *InReplyTo (prinl In-Reply-To:  @ ^M))
 (and *MimeVersion (prinl MIME-Version:  @ ^M))
 (and *ContentType (prinl Content-Type:  @ ^M))
+(and *ContentTransferEncoding (prinl 
Content-Transfer-Encoding:  @ ^M))
 (and *ContentDisposition (prinl Content-Disposition: 
 @ ^M))
 (and *UserAgent (prinl User-Agent:  @ ^M))
 (prinl ^M)
-- 
1.7.4.1


--Boundary-00=_xLmfN70LoP+89EH--
--
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: meta!

2011-03-14 Thread Alexander Burger
Hi Dexen,

 In short, the messages I receive don't have the ``Content-Transfer-Encoding=
 header (which is a companion to ``Content-Type''), even the ones I sent wit=
 that header for sure.
 
 Without this header, the mail client doesn't know to interpret the escape=20
 codes starting with the ``equals'' sign.

This sounds very reasonable! I wasn't aware of that.


 Attached patch attempts to fir it~

Thanks a lot! I installed a patched version. 
This mail is already sent via the new version :)

Let's see.

Cheers,
- Alex
-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe


Re: meta!

2011-03-14 Thread Alexander Burger
Great!

On Mon, Mar 14, 2011 at 10:05:28PM +0100, dexen deVries wrote:
 testin-testin' :D
 
 foo == bar?

BINGO! :D

Thanks again!
-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe