Abridged summary of comp.lang.tcl@googlegroups.com - 66 updates in 11 topics

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

Abridged summary of comp.lang.tcl@googlegroups.com - 66 updates in 11 topics

comp.lang.tcl mailing list
Rohan Yadav <[hidden email]>: Jan 23 10:00AM -0800

Hello all,
In my organization we are currently running 2000+ tcl script simultaneously to perform the business function. Currently we are experiencing memory outage each day. We decided to check ...more
Don Porter <[hidden email]>: Jan 23 02:08PM -0500

On 01/23/2018 01:00 PM, Rohan Yadav wrote:
> In my organization we are currently running 2000+ tcl script simultaneously...
 
Does that mean 2000 Tcl_Interp's concurrently running, each with a ...more
[hidden email]: Jan 23 11:18AM -0800

On Tuesday, 23 January 2018 11:08:29 UTC-8, Don Porter wrote:
> | [hidden email] Information Technology Laboratory |
> | http://math.nist.gov/~DPorter/ ...more
Don Porter <[hidden email]>: Jan 23 02:59PM -0500

> We have ... 10 threads inside a process [each with its] own interpreter.
 
>>> we are experiencing memory outage each day.
 
What are the symptoms of a "memory outage" precisely?
...more
[hidden email]: Jan 23 10:53AM -0800

Is there a better way to accomplish the following?
 
> spawn -open [chan push [exp_open] [my_transform new]]
 
In other words, I want to filter the output of a command before operating on it with ...more
Cecil Westerhof <[hidden email]>: Jan 23 02:02AM +0100

I have a function to fetch all fonts. But this can of-course only be
called if Tk is loaded. Is there a way to verify that a package (in
this case Tk) is loaded?
 
I do not want to use: ...more
Rich <[hidden email]>: Jan 23 02:14AM


> I do not want to use:
> package require Tk
 
> Because that creates a window if Tk was not loaded.
 
"man package"
 
Specifically read the part about "present".
 
Then experiment in a Tcl ...more
Cecil Westerhof <[hidden email]>: Jan 23 04:44AM +0100

>> package require Tk
 
>> Because that creates a window if Tk was not loaded.
 
> "man package"
 
Sadly the man pages are not installed on my system. But luckily I
found:
...more
Rich <[hidden email]>: Jan 23 04:01AM


>>> Because that creates a window if Tk was not loaded.
 
>> "man package"
 
> Sadly the man pages are not installed on my system.
 
I would highly recommend you install the man pages. ...more
Cecil Westerhof <[hidden email]>: Jan 23 07:07AM +0100


> I would highly recommend you install the man pages. That is unless
> your 'system' is windows, then installing them likely won't work at
> all.
 
I was planning to do that. I wanted to ask on ...more
Christian Gollwitzer <[hidden email]>: Jan 23 07:34AM +0100

Am 23.01.18 um 04:44 schrieb Cecil Westerhof:
 
> Sadly the man pages are not installed on my system. But luckily I
> found:
> http://wiki.tcl.tk/9943
 
The wiki is a nice place, if you want ...more
Cecil Westerhof <[hidden email]>: Jan 23 09:31AM +0100

> while evaluating {is_Tk_present }
 
> But when I define it in a script and use it in the script, it works as
> expected.
 
If I define (a bit more general) function in a package and use that ...more
Rich <[hidden email]>: Jan 23 11:44AM

> do it. I find it a bit strange because of all other packages the man
> pages are installed. But it was very easy: I just had to install
> tcl-doc.
 
Ah, yes, Debian, with their silly split of X, ...more
Rich <[hidden email]>: Jan 23 11:48AM

> proc ::dcblUtilities::isPackagePresent {package} {
> expr {![catch {package present ${package}}]}
> }
 
Interactive shell:
 
$ rlwrap tclsh ...more
Cecil Westerhof <[hidden email]>: Jan 23 03:19PM +0100

> happens when one forgets to initially define the namespace in the
> REPL. But any difference in defining a proc between the two is almost
> always the result of a typo in the REPL, not because it ...more
Cecil Westerhof <[hidden email]>: Jan 23 03:31PM +0100

>> tcl-doc.
 
> Ah, yes, Debian, with their silly split of X, X-dev, and X-doc
> packages.
 
But as far as I know Debian does this not with other packages. Or
maybe it installs the manpages only ...more
Cecil Westerhof <[hidden email]>: Jan 23 03:52PM +0100

> happens when one forgets to initially define the namespace in the
> REPL. But any difference in defining a proc between the two is almost
> always the result of a typo in the REPL, not because it ...more
Ralf Fassel <[hidden email]>: Jan 23 05:14PM +0100

* Cecil Westerhof <[hidden email]>
| > Interactive shell:
| >
| > $ rlwrap tclsh
--<snip-snip>--
| I found the problem. In my ~/.tclshrc I have:
| ::tclreadline::Loop
...more
Rich <[hidden email]>: Jan 23 05:20PM


> But as far as I know Debian does this not with other packages. Or
> maybe it installs the manpages only for the default packages. I do not
> use man a lot.
 
What little I know of Debian (which ...more
Cecil Westerhof <[hidden email]>: Jan 23 06:29PM +0100

> | ::tclreadline::Loop
 
> Just guessing, assuming package tclreadline adds readline-like editing
> to tclsh: either use tclreadline OR rlwrap, not both.
 
Oops, that happens when you do ...more
Cecil Westerhof <[hidden email]>: Jan 23 07:17PM +0100

> readline).
 
> As I don't use tclreadline and instead just let rlwrap handle it, I
> don't know why this might be happening.
 
Ted Nolan provided the solution: '!' is the history interpolation ...more
Cecil Westerhof <[hidden email]>: Jan 23 04:04PM +0100

When I have the following ~/.tclshrc:
if {${tcl_interactive}} {
package require tclreadline
 
::tclreadline::Loop
}
 
I get a very strange behaviour. When I enter: ...more
[hidden email] (Ted Nolan <tednolan>): Jan 23 05:38PM

In article <[hidden email]>,
> %
 
>So am I doing something wrong, or did I find a bug in
>::tclreadline::Loop?
 
'!' is the history interpolation character for the readline ...more
Cecil Westerhof <[hidden email]>: Jan 23 07:11PM +0100


> '!' is the history interpolation character for the readline package.
 
> Add a space after it so it gets passed through to Tcl rather than the
> loop trying to interpret it.
...more
[hidden email]: Jan 23 07:01AM -0800

Need some help understanding one of the examples given in the "The Tcl Programming Language"
 
 
I am new to TCL and have been trying to learn TCL using "The Tcl Programming Language: A Comprehensive ...more
Gerald Lester <[hidden email]>: Jan 23 09:17AM -0600


> What I am confused is with the command " set str [string toupper [lindex $args end]] ". I am wondering whats the significance of using [lindex $args end]] here. if [lindex $args end]] is printed ...more
Ralf Fassel <[hidden email]>: Jan 23 05:06PM +0100

* [hidden email]
| proc puts args {
| set str [string toupper [lindex $args end]]
| builtin_puts {*}[lreplace $args end end $str]
| }
| ------------------------
...more
[hidden email]: Jan 23 08:33AM -0800

> to the real puts, now called builtin_puts. The uppercase string is
> substituted for the real string via the lreplace on the next line.
 
> --
 
 
 
Many Thanks for the quick reply Gerald ...more
[hidden email]: Jan 23 08:52AM -0800

On Tuesday, January 23, 2018 at 4:06:08 PM UTC, Ralf Fassel wrote:
> element of the list.
 
> HTH
> R'
 
 
Thanks Raif. I think I am confused with the behaviour of lindex.
From tclsh if I do ...more
Rich <[hidden email]>: Jan 23 05:32PM

> This is from tclsh
 
> May be I should go deep in to the other chapters to get a deeper
> understanding.
 
Yes, you likely have not reached that part of the book yet. The "args" ...more
Robert Heller <[hidden email]>: Jan 23 12:06PM -0600


> --RUN--
> % testIndex "This is from tclsh"
> This is from tclsh
 
In the case of testIndex, the argument args is a *list* of its arguments.
Specificly, testIndex is being passed one ...more
Alexandru <[hidden email]>: Jan 22 07:34PM -0800

Hi,
 
What would be the best, simplest, most general way to block double-clicks on buttons? Per default, a double-click triggers two times the widget command, which is actually not wanted.
...more
Rich <[hidden email]>: Jan 23 04:04AM


> What would be the best, simplest, most general way to block
> double-clicks on buttons? Per default, a double-click triggers two
> times the widget command, which is actually not wanted.
...more
Alexandru <[hidden email]>: Jan 22 08:13PM -0800

Am Dienstag, 23. Januar 2018 05:04:15 UTC+1 schrieb Rich:
> performed in the -command proc, should be ".b configure -state normal".
 
> The above assumes the button is named .b, adjust accordingly ...more
Gerald Lester <[hidden email]>: Jan 23 12:06AM -0600

On 01/22/2018 10:13 PM, Alexandru wrote:
 
>> The above assumes the button is named .b, adjust accordingly for your
>> scripts.
 
> Thanks. I was aware if this possiblity, which I avoided until now ...more
Alexandru <[hidden email]>: Jan 22 10:34PM -0800

Am Dienstag, 23. Januar 2018 07:06:59 UTC+1 schrieb Gerald Lester:
> | Gerald W. Lester, President, KNG Consulting LLC |
> | Email: [hidden email] ...more
Rich <[hidden email]>: Jan 23 11:54AM

> because of possible errors in the callback script that would block
> setting the button state back to normal. I regard this method as
> "unsafe".
 
Any other way you utilize to 'block' the ...more
Rich <[hidden email]>: Jan 23 11:55AM


> Oh, that sounds good. But is there a way to automate this? Instead of
> putting the same text into the widgets -command property, make sure
> all buttons have this behavior?
...more
Alexandru <[hidden email]>: Jan 23 06:28AM -0800

Am Dienstag, 23. Januar 2018 12:55:58 UTC+1 schrieb Rich:
> > all buttons have this behavior?
 
> Build a proc that builds procs. I.e., what the java camp might refer
> to as a "Factory".
...more
[hidden email]: Jan 23 06:43AM -0800

On Monday, January 22, 2018 at 10:34:23 PM UTC-5, Alexandru wrote:
 
> What would be the best, simplest, most general way to block double-clicks on buttons? Per default, a double-click triggers two ...more
Alexandru <[hidden email]>: Jan 23 06:50AM -0800

> Implement two bindings: single click and double click. The binding on the regular press will match on the first press of the 'double'. Then the command bound to the double will match on the second ...more
Andreas Leitgeb <[hidden email]>: Jan 23 04:40PM

>> bind . <1> {desired script}
>> bind . <Double-1> {do nothing}
> Thaks, I will check this.
 
That doesn't seem to work. It will call the "do nothing" and
still call the -command a second time ...more
Alexandru <[hidden email]>: Jan 23 08:48AM -0800

> Implement two bindings: single click and double click. The binding on the regular press will match on the first press of the 'double'. Then the command bound to the double will match on the second ...more
Alexandru <[hidden email]>: Jan 23 08:50AM -0800

Am Dienstag, 23. Januar 2018 17:40:43 UTC+1 schrieb Andreas Leitgeb:
> if {$::running} { return }
> set ::running 1; after 1000 { set ::running 0 }
> ...
 
It does work. ...more
Alexandru <[hidden email]>: Jan 23 08:53AM -0800

> Implement two bindings: single click and double click. The binding on the regular press will match on the first press of the 'double'. Then the command bound to the double will match on the second ...more
Harald Oehlmann <[hidden email]>: Jan 23 05:54PM +0100

Am 23.01.2018 um 17:50 schrieb Alexandru:
 
> It does work. See previous post.
 
> Cheers
> Alexandru
 
I would always end bindings with a "break" to avoid the other bindings
in the binding ...more
Alexandru <[hidden email]>: Jan 23 08:57AM -0800

Am Dienstag, 23. Januar 2018 17:54:05 UTC+1 schrieb Harald Oehlmann:
 
> So:
> bind $w <Double-1> break
 
> Harald
 
Hi Harald,
 
what is the explanation for bind $w <Double-1> {return 1} still ...more
Rich <[hidden email]>: Jan 23 05:13PM

>> to as a "Factory".
 
> Could you explain what exactly do you mean? I have no idea about Java
> and Google didn't help.
 
A "Factory" in Java is an object that creates other objects.
...more
Rich <[hidden email]>: Jan 23 05:15PM


>> Harald
 
> Hi Harald,
 
> what is the explanation for bind $w <Double-1> {return 1} still working. What is the difference between return and break in this context of bindings?
 
man n bind:
...more
Alexandru <[hidden email]>: Jan 23 09:18AM -0800

Am Dienstag, 23. Januar 2018 18:15:09 UTC+1 schrieb Rich:
> processing binding scripts associated with other tag's. If the break
> command is invoked within a binding script, then that script ...more
Harald Oehlmann <[hidden email]>: Jan 23 06:21PM +0100


>> Harald
 
> Hi Harald,
 
> what is the explanation for bind $w <Double-1> {return 1} still working. What is the difference between return and break in this context of bindings?
 
Hi Alexandru, ...more
Alexandru <[hidden email]>: Jan 23 09:23AM -0800

Am Dienstag, 23. Januar 2018 18:21:04 UTC+1 schrieb Harald Oehlmann:
> same as continue and thus should not work. I don't know...
 
> Sorry,
> Harald
 
That's the point: return 1 *blocks* the ...more
Rich <[hidden email]>: Jan 23 05:42PM

>> Harald
 
> That's the point: return 1 *blocks* the execution of other
> bindings... This is was not expected by me...
 
The "return" command is not documented for use inside binding scripts. ...more
Robert Heller <[hidden email]>: Jan 23 12:06PM -0600

> if {$::running} { return }
> set ::running 1; after 1000 { set ::running 0 }
> ...
 
In some cases, it isn't just the one button being "double clicked" on but some
case where once *A* ...more
Robert Heller <[hidden email]>: Jan 23 12:06PM -0600


> I was actually aware about the meaning of break and continue. My quastion w=
> as about return. Observed behavior: Returning any value has the same effect=
> as break. But why?
...more
[hidden email]: Jan 23 07:02AM -0800

> Windows binaries are easy to build. The problem is that they embed the
> path to the rivet library.
> Perhaps a configuration directive to specify the library location is needed.
...more
Juge <[hidden email]>: Jan 23 03:32AM -0800

I am implementing an edit feature to my table where I select multiple rows and
edit a cell on one of them. Afterwards I want to have the edited values on all of the same cells (column) of the other ...more
Juge <[hidden email]>: Jan 23 03:58AM -0800

Ok, I was able to achieve something with binds.
I made cells -editable 0
 
and then:
set InputbodyTag [$mytbl bodytag]
bind $InputbodyTag <Double-B1-ButtonRelease> [list StartEdit %W %x %y]
...more
[hidden email]: Jan 22 09:01PM -0800

Le dimanche 21 janvier 2018 23:50:33 UTC+1, briang a écrit :
> If you are counting sys calls, you should be writing in C. You could write your own extension open command.
 
> Eliminating those ...more
briang <[hidden email]>: Jan 22 10:06PM -0800

At any time a virtual file system can be mounted at any point along the path. For each open the path must be checked just in case it changed. The lstat is actually the last check.
 
-Brian
...more
Rich <[hidden email]>: Jan 23 11:31AM

> * the "ioctl" call: it does not seem appropriate on a regular file
 
The tty devices are just "regular files" from the perspective of an
open() call. The ioctl is to see if the file, that was ...more
[hidden email]: Jan 22 08:44PM -0800

> if you are seeking women for sex? find here local girls online looking for near me men for sex, romance, flirt, dating, nsa relationship. choose here best dating profile for fuck tonight. ...more
Matthew Hiles <[hidden email]>: Jan 22 01:00PM -0800

On Monday, January 22, 2018 at 1:28:06 PM UTC-5, Cecil Westerhof wrote:
> Cecil Westerhof
> Senior Software Engineer
> LinkedIn: http://www.linkedin.com/in/cecilwesterhof
 
Between all the [] and ...more
[hidden email]: Jan 22 01:27PM -0800

On Monday, January 22, 2018 at 1:28:06 PM UTC-5, Cecil Westerhof wrote:
> Cecil Westerhof
> Senior Software Engineer
> LinkedIn: http://www.linkedin.com/in/cecilwesterhof
 
Hold on gang...not so ...more
Cecil Westerhof <[hidden email]>: Jan 22 11:57PM +0100

> set firstname rick
> set fullname $firstnameHorwitz
> set fullname ${firstname}Horwiz
 
That was one of the reasons I used it in bash. If the variable
firstnameHorwitz existed, you got another ...more
[hidden email]: Jan 22 05:06PM -0800

Indeed, using braces around variable names is embraced by the tcl world...
 
I have a bit more nuanced set of 'rules' for braces then simply eliminating them whenever possible...
...more
You received this digest because you're subscribed to updates for this group. You can change your settings on the group membership page.
To unsubscribe from this group and stop receiving emails from it send an email to [hidden email].