[BUG] Method Finder DNU with array literal as receiver

I had a Method Finder open. I searched for:

{ 1 . 2 . 3 . 4 } . 2 .  2

Then I clicked on the answer that read:

{1. 2. 3. 4} at: 2 --> 2

And I got this walkback.

30 April 2002 3:24:11 pm

VM: unix - Squeak3.2gamma of 15 January 2002 [latest update: #4843]
Image: Squeak3.2gamma [latest update: #4843]

SmallInteger(Object)>>doesNotUnderstand:
	Receiver: 1
	Arguments and temporary variables: 
		aMessage: 	a Message with selector: #includes: and arguments: #($:)
	Receiver's instance variables: 
1

SelectorBrowser>>selectedMessageName
	Receiver: a SelectorBrowser
	Arguments and temporary variables: 
		example: 	'{1. 2. 3. 4} at: 2 --> 2'
		tokens: 	#(#{ 1 #. 2 #. 3 #. 4 #} #at: 2 #- #-> 2)
		aSymbol: 	nil
	Receiver's instance variables: 
		dependents: 	#(a SystemWindow(1) a PluggableTextMorph(3349) a

Re: rounded corners (was: Re: straw-man 3.2 default preferences)

At 1:59 AM -0700 4/30/02, Scott Wallace wrote:
>I suppose I agree, reluctantly, with Henrik's position about rounded 
>corners, and would be inclined to change the out-of-box look of 3.2 
>accordingly.
>
>Opinions?

An opinion, but not a strong one:

I have not noticed a performance problem with rounded corners (and I 
haven't timed how much time they consume), but I have been running 
with square corners for a while. I find the square corners a bit 
nicer-looking, and I find it easier to visually distinguish windows 
from buttons and flaps when some are square and some are rounded than 
when all are rounded.

-Martin

Re: [e-lang] Java Serialization Re: Alan Kay in the News

Shell scripts for squeak on UNIX (was: Please use MIME attachments)

(CC'd back to the list because others may like the scripts)

On Wed, 1 May 2002, Richard A. O'Keefe wrote:

> 	I'm the same way.... I do all my email on this machine (over ssh) and all
> 	my squeak work on my machine. I use Pine to send, receive, and read
> 	emails... I SCP over any files I need, in both directions. And so far, I
> 	haven't had a problem.
>
> So how do you send change sets to this mailing list?

I scp them from my machine to the host that I do email on, then attach
them. (Its a lot nicer if you use SSH RSA authenticaation to auto-log-onto
the mail host from your squeak machine. :)

I've just made and tested an alias that does this for me.. ('sqmail'. Note
that it will probably fail if there's whitespace (or strange characters)
in filenames.)

> Pine can do attachments, but do you have to manually gzip the change sets?
>

I can... I usually don't.

If you want to gzip them, add something like:

   ssh mailhost.foo.bar gzip ~/csSend/*.cs

to 'sqmail'... But you'll have to edit the part that makes the cut&paste
command if you want the output to still be cut&pasteable. (See below)

Then with one line:

     sqmail foo.cs bar.cs Fulltext*.cs  /foo/bar/*.cs

I automatically copy those changesets into a folder on the mail host, then
automatically echo a line I can cut&paste into Pine's 'Attachments' header
that will attach them from that directory. :)

  ``csSend/FullText-DemoAdaptors.1.cs,  csSend/FullText-DemoAdaptors.2.cs,
csSend/FullText-DemoAdaptors.3.cs,  csSend/FullText-Engines.1.cs,
csSend/FullText-Engines.2.cs,  csSend/FullText-Engines.3.cs,
csSend/FullText-Engines.4.cs,  csSend/FullText-Engines.5.cs,
csSend/FullText-Engines.6.cs''  (all one line, and ready to cut&paste)

--

I've also got a few other quick bash scripts.. For diff'ing changesets (to
see what changed, which is GREAT), and for viewing them with something
normal like 'less'.

Here are all 4:

sqless () {
  tr \\015 \\n <"$1" | less
}

sqcat () {
  tr \\015 \\n <"$1"
}

sqmail () {
  scp -C -v $* qwe3.math.cmu.edu:csSend/
  ls $* | sed '-es%^\(.*/\)%%' | sed '-es%^%csSend/%'     \
         | tr \\n \~ | sed -es/~/,\ \ /g
}

sqdiff () {
  tr \\015 \\n <"$1" > /tmp/sq1
  tr \\015 \\n <"$2" > /tmp/sq2
  shift ; shift ;
  diff $* /tmp/sq1 /tmp/sq2
}

These are all bash scripts and work fine for me.

--

The last can be used to pass extra options to diff:

  sqdiff foo.1.cs foo.2.cs -y    # for side-by-side comparison. :)
  sqdiff foo.1.cs foo.2.cs -u    # for context-aware diff to see what
                                 #    changed and the surrounding context.
  sqdiff foo.1.cs foo.2.cs -U 8  #  (same, but more context)

These are *really nice* when trying to compare changeset versions.  Check
your local 'diff' manpage for the other goodies. :)

Scott

[ENH] Timestamp

Hannes Hirzel asked (in the DateAndTime thread) what my Timestamp class does
and if it works with 3.2.  Well, it files into my 3.2 alpha image just fine
(updated to 4664).  I'm sorry I don't have a 3.2 gamma image handy.  The
class is fairly straightforward, Hannes, if you want to take a quick look at
it.

As was noted in the preamble, this class does not implement time zones, but
I think it is useful even without them.  If time zones are also needed, I
suppose someone could look at the work that was done by Lex Spoon and
others.  In any case, my class is very lightweight (no refactoring of the
base classes is necessary to use Timestamp in your image). 

I implemented Timestamp when I needed to port some of my code from VSE to
Squeak.  At the time, I banged on this code pretty hard, so I expect that it
should work as you would expect.

I'm sorry I haven't had time to write SUnit tests for this.  But before
sending this note to the list, I inspected each of the following statements
(see below), exercising a good chunk of the protocol -- everything seemed
fine.

I hope you find this class as useful as I do.  I imagine that it could be
used quite handily in such applications as Celeste, the PDAMorph, etc.
Squeak really deserves to have a Timestamp class; lugging dates and times
around in arrays is too painful.

Cheers,

---==> Chris

PS>  For history buffs & people who are already using my Timestamp class: in
the original changeset (posted to the list in October of 2000), I had named
the class TimeStamp.  Since that time, I've discovered that timestamp is a
word in its own right, so I changed the name of my class to Timestamp.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Here are some protocol tests:

Timestamp current
Timestamp date: (Date fromString: '1/1/2002') time: Time now
Timestamp fromSeconds: 1000
Timestamp fromString: '1/2/1002 4:44:00pm'
Timestamp midnight
Timestamp midnightOn: (Date today)
Timestamp noon
Timestamp noonOn: (Date today)
Timestamp current minusDays: 4
Timestamp current minusSeconds: 500000
Timestamp current plusDays: 4
Timestamp current plusSeconds: 500000
Timestamp current asDate
Timestamp current asSeconds
Timestamp current asTime
Timestamp current asTimestamp
Timestamp current dateAndTime
Date today asTimestamp
Time fromString: '11:55:23 am'
Time midnight
Time noon

Re: Remote access to Celeste db?

Aaron J Reichow <reic0024 <at> d.umn.edu> wrote:
> I'd like to switch to Celeste.  What has kept me back from doing so is
> that there's no lo-tek way to access my old email remotely.  Which is why
> I continue to use pine. By lo-tek, I mean some way that's accessible by
> just telnet/ssh.

Just by the way, if you have access to a graphical terminal and a
cable-modem or better network connection, you can use Nebraska.  That's
what I do to access my mail remotely.  There is a changeset floating
around to add authentication to Nebraska.

Squeak is nice.  Trying to dumb it down to a text interface or a WWW
interface seems like a doomed project  Granted, email in particular
should be doable....  Alternatively, someone could write an IMAP-based
email reader, and you could simply never download your email for real. 
Or, someone could write an IMAP server in Squeak that reads from Squeak
MailDB's.  So many possibilities, none of which are implemented.  :)

Lex

RE: How to force AcceptableCleanTextMorph to accept for each character typed?

Hi, Ken,

The way you would do this is to wait until 4846iBeamInNullText 
appears in the external update streams.

Cheers,

   -- Scott

PS: Sorry for the long delay in responding.

At 5:33 PM -0700 4/1/02, Ken G. Brown wrote:
>How would you get the text insertion bar to show up when you first click in
>this little pane? It appears when you start typing but it would be nice to
>be there beforehand so you have an idea the pane has keyboard focus.
>Thx.
>Ken
>
>>  Ken,
>>
>>  The type-in pane of the "Objects Tool" (seen when its "find" tab is
>>  selected) had a similar need.  Check out ObjectsTool.newSearchPane to
>  > see how it does it.

Re: Remote access to Celeste db?

On Tue, 30 Apr 2002, Lex Spoon wrote:

> Just by the way, if you have access to a graphical terminal and a
> cable-modem or better network connection, you can use Nebraska.  That's
> what I do to access my mail remotely.  There is a changeset floating
> around to add authentication to Nebraska.

I considered that.  But my cable modem at home is capped at like 7K/s
going out, which is painful for Nebraska.  Also, even if it were zippier,
going through all the work of installing Squeak (and then hunting down and
filing in the authentication CS) just to find sdome old email.  Even so,
that isn't possible for a lot of the machines on campus.

> Squeak is nice.  Trying to dumb it down to a text interface or a WWW
> interface seems like a doomed project.

Pffft.  There's a lot of that kind of attitude on this list, and I
understand it.  Squeak is awesome.  I wish all the computers into
with I came in contact ran SqueakNOS, and there were super fast connexions
for running Nebraska.  But it is not so.  Having my information accessible
via some text or www interface allows me to get what I need on any
computer, not just computers fast enough, on a fast enough link, and open
to disk-quota-less software configuration.  I mean nothing personal
against you or anyone else one here, so please don't take it as such. :)

While a WWW interface may be more generically useful for others, I think
the first thing I'll create is an IRC bot for interacting with MailDB.  I
know what you're thinking.  Should be easier to hack together than a
Comanche module (I wrote a bot framework for SqIRC a while back), but
still achieves my goal.

Regards,
Aaron

  Aaron Reichow  ::  UMD ACM Pres  ::  http://www.d.umn.edu/~reic0024/
"the profit system follows the path of least resistance and following the path
    of least resistance is what makes a river crooked."   ::  u. utah phillips

Re: Remote access to Celeste db?

Aaron J Reichow wrote:
> 
> I'd like to switch to Celeste.  What has kept me back from doing so is
> that there's no lo-tek way to access my old email remotely.  Which is why
> I continue to use pine. By lo-tek, I mean some way that's accessible by
> just telnet/ssh.
Hmm, how about some sort of web access to Celeste as well? Craig & I
have considered trying a telnet server in Flow for another project. Or
would making Celeste work with IMAP be useful? - or even make an IMAP server?

The main thing I'd like to see sorted out in Celeste is not fetching
messages twice when leaving messages on thre server. This would make it
easy to use at home and on the road.

tim

Re: Is BC worth the image format switch?

As Anthony pointed out the key factor is that he has implemented proper
Closure behaviour. In the process he sucked in my new CompiledMethod
format changes, fixed a few method header infelicities and reimplemented
the callstack mechanism. The latter is responsible for most of the
performance improvement; for some explanation of what happens you might
find it useful to read Eliot Miranda's OOPSLA 97 paper on BrouHaHa.
Anthony implemented a variant of that stack mechanism but exposed it to
the Smalltalk code rather than hiding it. I'm not entirely sure I like
that part, but I haven't had time to examine it properly yet. The worst
problem of which I'm aware is the lack of support for swapSender: and
callCC - important for coroutining etc. 

The VI4 project is trying to gather together all the plausible changes
that peolpe want and which would require an image format change or would
other wise break back compatibility. For example, renaming the plugins
more sensibly ( change Klatt to KlattPlugin, etc) would save some
confusion (ever got lost trying to work out where the code for ZipPlugin
is? - in InflatePlugin & DeflatePlugin classes!) but means changing
assorted image code.

We could even convert all bitmaps to little-endian, since only Mac uses
big-endian and we know that nobody important uses Macs. 

tim