Re: multi-threaded win32::console

Um, you obtain $CONSOLE at the top of your code and "share" it via closure(*) with the thread routine.

And in your thread routine you use $CONSOLE...once. Then, for no apparent reason, you switch to using $threadCONSOLE which is never declared or initialised.

And despite your post showing use strict; at the top, when I try syntax checking it, I get:

C:\test>perl -c 632320.pl
Global symbol "$i" requires explicit package name at 632320.pl line 17
+.
Global symbol "$i" requires explicit package name at 632320.pl line 17
+.
Global symbol "$i" requires explicit package name at 632320.pl line 17
+.
Global symbol "$i" requires explicit package name at 632320.pl line 18
+.
Global symbol "$i" requires explicit package name at 632320.pl line 22
+.
Global symbol "$i" requires explicit package name at 632320.pl line 22
+.
Global symbol "$i" requires explicit package name at 632320.pl line 22
+.
Global symbol "$i" requires explicit package name at 632320.pl line 23
+.
Global symbol "$i" requires explicit package name at 632320.pl line 24
+.
Global symbol "$i" requires explicit package name at 632320.pl line 24
+.
Global symbol "$i" requires explicit package name at 632320.pl line 25
+.
Global symbol "$i" requires explicit package name at 632320.pl line 34
+.
Global symbol "$i" requires explicit package name at 632320.pl line 34
+.
Global symbol "$i" requires explicit package name at 632320.pl line 34
+.
Global symbol "$i" requires explicit package name at 632320.pl line 35
+.
Global symbol "$inPattQty" requires explicit package name at 632320.pl
+ line 69.
Global symbol "$outPattQty" requires explicit package name at 632320.p
+l line 69.
syntax error at 632320.pl line 73, near ") {"
632320.pl has too many errors.
[download]

Update: hacking my way past all the errors and warnings, your code seems to run:

Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.

"Science is about questioning the status quo. Questioning authority".

In the absence of evidence, opinion is indistinguishable from prejudice.

"Too many [] have been sedated by an oppressive environment of political correctness and risk aversion."

Comment on Re: multi-threaded win32::console Select or Download Code

Replies are listed 'Best First'.
Re^2: multi-threaded win32::console by goibhniu (Hermit) on Aug 14, 2007 at 00:18 UTC
First variant of my full code. `my $CONSOLE` is in main and `my $threadCONSOLE` is in the thread. Read more... (12 kB) This is to help out in a cryptography hobby. To see it really work, you'll need a copy of Knuth's word list (words.knu). I got it from the ACA web site. Put it in the same folder and run this. The first word pattern will have (on my word list) 177 matches. After that, the output stops and looks like this: `C:\chas_sandbox>test.pl 1/177 . 3/8 4/4 3/16 C:\chas_sandbox>` [download] This show the first thread taking one solution off it's input queue, putting 177 on it's output queue and stopping (the period). The second thread only got 3 Solutions out of those 177 before all output stopped (after the first thread ended). If I remove all references to Win32::Console, it proceeds to completion and shows a list of possible solutions for this four word cryptogram.	[reply] [d/l] [select]
Re^3: multi-threaded win32::console by BrowserUk (Patriarch) on Aug 14, 2007 at 01:21 UTC
When I run the above, I get reams and reams of Use of uninitialized value in exists at C:\test\632320.pl line 161. Use of uninitialized value in null operation at C:\test\632320.pl line + 161. Use of uninitialized value in hash element at C:\test\632320.pl line 1 +67. Use of uninitialized value in hash element at C:\test\632320.pl line 1 +67. Use of uninitialized value in hash element at C:\test\632320.pl line 1 +67. Use of uninitialized value in hash element at C:\test\632320.pl line 1 +67. Use of uninitialized value in exists at C:\test\632320.pl line 161. ... [download] It's not at all clear to me what a 'null operation' is, (I don't think I ever encountered that one before :), but the main problem appears to be that your code is still displaying signs of its `no strict;` heritage. Specifically, within your `compare` sub, you have the loop iterator declared ahead of the first loop: `my $k; foreach $k ( keys %{ $hrefA } ) {` [download] which is rarely better than declaring it in line, but is sometimes necessary if you wish to retain its value beyond the end of the loop. However, in the next loop, which is only entered if you found whatever you are looking for in the first, you have no loop iterator variable, but do not appear to use `$_` anywhere within it? And the first thing you do within that second loop is test if the value of `$k` exists as a key within `$hrefA`. `if ($result) { foreach (keys(%{$hrefB})) { ## No loop variable if ( exists $hrefA->{$k} ) { ## line 161` [download] But, as the value of `$k` is being retained from the first loop where it is explicitly looping over the keys of `$hrefA` (as shown above), then that test would appear to be redundant. Except, the undefined warning for line 161 relates to the fact that somehow you are reaching this line with `$k` undefined. My best guess, without having tried to work my way through and understand everything you are doing, is that your second loop is meant to be iterating the keys of `$hrefB` and checking whether they exist in `$hrefA`, and that if you replaced those two snippets with; `## my $k; foreach my $k ( keys %{ $hrefA } ) { ... if ($result) { foreach my $k (keys(%{$hrefB})) { ## No loop variable if ( exists $hrefA->{$k} ) { ## line 161` [download] That might eliminate a large number of the warnings, and might also get you closer to making your code do what you are expecting. Until you eliminate all the warnings your code is producing, it is impossible (or at least, asking a lot) for us to see symptoms of the problem you describe and so attempt to help you solve that problem. I don't insist on `use strict; use warnings;` although I always use them myself, but in this case, it is very obvious that your code is not doing what you think it is doing, and if you had them enabled, you would have realised it yourself. Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error. "Science is about questioning the status quo. Questioning authority". In the absence of evidence, opinion is indistinguishable from prejudice. "Too many [] have been sedated by an oppressive environment of political correctness and risk aversion."	[reply] [d/l] [select]
Re^4: multi-threaded win32::console by goibhniu (Hermit) on Aug 14, 2007 at 02:26 UTC
Perhaps I'm more of an Initiate (not even a Novice) than I thought. Am I not using strict in all examples (even the admittedly horrible original post) above? In Re^4: multi-threaded win32::console it behaves (produces results without errors or warnings) as I expect. The problem is only when I introduce Win32::console. Which of the above versions of the code produced the line numbers in your warnings / errors? Is there a platform or Perl version difference that is causing the problems? I'm on: C:\chas_sandbox>perl -version This is perl, v5.8.8 built for MSWin32-x86-multi-thread (with 33 registered patches, see perl -V for more detail) Copyright 1987-2006, Larry Wall Binary build 819 [267479] provided by ActiveState http://www.ActiveSta +te.com Built Aug 29 2006 12:42:41 Perl may be copied only under the terms of either the Artistic License + or the GNU General Public License, which may be found in the Perl 5 source ki +t. Complete documentation for Perl, including FAQ lists, should be found +on this system using "man perl" or "perldoc perl". If you have access to + the Internet, point your browser at http://www.perl.org/, the Perl Home Pa +ge. C:\chas_sandbox>ver Microsoft Windows XP [Version 5.1.2600] C:\chas_sandbox> [download]	[reply] [d/l]
Re^5: multi-threaded win32::console by BrowserUk (Patriarch) on Aug 14, 2007 at 04:17 UTC
Re^6: multi-threaded win32::console by goibhniu (Hermit) on Aug 14, 2007 at 07:30 UTC
Some notes below your chosen depth have not been shown here
Re^3: multi-threaded win32::console by goibhniu (Hermit) on Aug 14, 2007 at 00:26 UTC
Second variant of my full code. I've gotten rid of all `my $threadCONSOLE` refernces and am just using `my $CONSOLE` from main. Read more... (13 kB) I get the same results (give or take the non-determinacy in mutliple threads running slightly differently each time). `C:\chas_sandbox>test.pl 1/177 . 3/8 4/4 4/20 C:\chas_sandbox>` [download]	[reply] [d/l] [select]
Re^4: multi-threaded win32::console by goibhniu (Hermit) on Aug 14, 2007 at 00:34 UTC
Third Variant of my full code. There is no Win32::Console; just normal print statements. Read more... (11 kB) with the following output: Read more... (8 kB) (this time the output gets readmore tags)	[reply] [d/l] [select]
Re^2: multi-threaded win32::console by goibhniu (Hermit) on Aug 13, 2007 at 22:49 UTC
Very sorry. I wondered a bit at how detailed to paste code and decided to err on the side of brevity. I did not post actual full code, but tried to boil it down to just what was troubling me from a much larger program. I did not try to re-run it after hacking. In the future I'll try to err on the side of completeness. On the up side, in the interest of more complete communications, I created a user! I'm trying to read about the scratchpad, etc and figure out how things work around here. I humbly seek patience in the mean time. I'll post complete code separately and redescribe the problem.	[reply]