Beefy Boxes and Bandwidth Generously Provided by pair Networks
Keep It Simple, Stupid
 
PerlMonks  

Re: Problem with ReadKey

by liverpole (Monsignor)
on Nov 04, 2010 at 13:04 UTC ( #869456=note: print w/ replies, xml ) Need Help??


in reply to Problem with ReadKey

Hi jimhenry, To try to help you, I thought I would run your program and see how the execution differs from what's expected.

I quickly realized that you omit any of the code in the main program.

Well that shouldn't take me too much more work, I'll just see what you wrote (something about text files), and see if I can cobble together something small that calls slow_print.

Then I looked at slow_print, and saw that it depends on an externally defined array called @paras, an external variable $rewrap_margin, and at least two external subroutines, rewrap and handle_keystroke.  (By "external", I mean external to what you gave us; maybe in the main program or some library?).

My point is, I've now lost interest.  There are plenty of other questions where the poster is giving enough information for me to get to the fun stuff; the actual sleuthing of the problem.  To do that with your code one has to anticipate the definitions of two other data structures and two other subroutines, as well as guess how you might have called this subroutine.  Sorry, but that's work!

Please consider providing a small, standalone program which runs on its own, and illustrates the problem.  Then I'll be glad to come back and take a look, knowing that I can dive right into solving the problem.


s''(q.S:$/9=(T1';s;(..)(..);$..=substr+crypt($1,$2),2,3;eg;print$..$/


Comment on Re: Problem with ReadKey
Select or Download Code
Re^2: Problem with ReadKey
by jimhenry (Novice) on Nov 05, 2010 at 00:37 UTC
    I apologize for the partial, unrunnable code in my post; I shouldn't post when sleep-deprived and frustrated. If I had taken the time to figure out a minimal script that reproduces the problem, I could have asked a much more specific question -- which I'll do now. The sample script provided by jethro worked fine on my system. I then tried to figure out what was the salient difference between his script and mine, and quickly remembered that my script was (because of the huge number of text files it might need to handle) reading a long list of filenames from stdin rather than the command line. It's normally run in a pipe from a find command, like so:
    find $HOME -name \*.txt | textual-slideshow.pl $*
    It slurps the filenames from standard input, then reads random sample paragraphs from some of them, prints random paragraphs a line at a time, and repeats. My simplified version that's partway between jethro's script and mine looks like this:
    #! /usr/bin/perl use Term::ReadKey; use Time::HiRes qw(time); my @slurp_stdin = <>; ReadMode 3; while (1) { my $key; my $wait_until = time + 3; while ( time < $wait_until ) { $key = ReadKey( -1 ); if ( defined $key ) { print STDERR "keystroke $key\t"; } } print "Something\n"; }
    And it has the same problem as my script, not surprisingly. The problem clearly has to do with running in an environment where stdin is redirected. So I tried to fix that by closing STDIN and reopening it after reading the filenames. That didn't help. I tried adding an explicit STDIN argument to the ReadMode and ReadKey calls; that didn't help either. This is what I've got now, and it also has the same problem as my original script or the above modification of jethro's script:
    #! /usr/bin/perl -w use Term::ReadKey; use Time::HiRes qw(time); my @slurp_stdin = <STDIN>; close STDIN; open STDIN, "-"; while (1) { my $key; my $wait_until = time + 3; while ( time < $wait_until ) { ReadMode 3, STDIN; # 'noecho'; $key = ReadKey( -1, STDIN ); if ( defined $key ) { print STDERR "keystroke $key\t"; } ReadMode 0, STDIN; } print "Something\n"; }
    This repeatedly prints "Something"; but when I press keys, I don't get the "keystroke (key value)" message, only the value of the key. My full original script is at http://jimhenry.conlang.org/scripts/textual-slideshow.zip.

      open STDIN, "-"; is just reopening STDIN with the standard input, which is still redirected. I don't think that you can get at the terminal output that easy.

      Just remove the redirection and use the backticks operator `find $HOME -name \*.txt` or File::Find inside your perl script to get at the filenames. No redirection, no problem

        Hi jimhenry,

        I agree with jethro that the best way to do this is to read the list of files in the program.  You shouldn't have to specify STDIN in your Term::ReadKey calls, either.  Also, consider putting "$| = 1" at the beginning of the program to flush STDOUT.

        Here's a small subroutine which will let you avoid using File::Find and just get the files ending in ".txt":

        sub find_all_textfiles { my ($dir) = @_; my $fh = new IO::File; opendir($fh, $dir) or die "Failed to read dir '$dir' ($!)\n"; my @files = readdir($fh); closedir $fh; my @found = ( ); foreach my $fname (@files) { next if $fname eq '.' or $fname eq '..'; my $path = "$dir/$fname"; if (-f $path and $path =~ /[.]txt$/) { push @found, $path; } elsif (-d $path) { push @found, find_all_textfiles($path); } } return @found; }

        Of course this will require that you read each file individually, and somehow store their contents.

        If it matters which file the text came from, you could save the lines from each file in a separate array reference, and then return a reference to those array references (note my use of Data::Dumper, which is invaluable for visualizing your data!):

        use strict; use warnings; use Data::Dumper; use IO::File; use Term::ReadKey; use Time::HiRes qw(time); $| = 1; my @files = find_all_textfiles("."); my $a_text = read_files(@files); printf "Results of reading [@files] => %s\n", Dumper($a_text); sub read_files { my (@files) = @_; my $a_text = [ ]; foreach my $fname (@files) { print "Reading '$fname' ...\n"; my $fh = new IO::File($fname) or die "Can't read '$fname' ($!) +\n"; chomp(my @text = <$fh>); push @$a_text, [ @text ]; } return $a_text; }

        If you don't care which text came from which file, just throw it all in one big array:

        use strict; use warnings; use Data::Dumper; use IO::File; use Term::ReadKey; use Time::HiRes qw(time); $| = 1; my @files = find_all_textfiles("."); my @text = read_files(@files); printf "Results of reading [@files] => %s\n", Dumper([ @text ]); sub read_files { my (@files) = @_; my @text = ( ); foreach my $fname (@files) { print "Reading '$fname' ...\n"; my $fh = new IO::File($fname) or die "Can't read '$fname' ($!) +\n"; chomp(my @lines = <$fh>); push @text, @lines; } return @text; }

        In fact, in the last example, you could even combine the finding of files with the saving of each file's text, to create a single subroutine.  I'm using a nifty trick here, which is to pass the array reference $a_text which is the aggregate of all text in each recursive call to read_textfile_lines; only at the top level is it undefined (but in that case, you initialize the array ref with:  $a_text ||= [ ];):

        use strict; use warnings; use Data::Dumper; use IO::File; use Term::ReadKey; use Time::HiRes qw(time); $| = 1; my $a_text = read_textfile_lines("."); printf "Text from ALL textfiles in current dir: %s\n", Dumper($a_text +); sub read_textfile_lines { my ($dir, $a_text) = @_; $a_text ||= [ ]; my $fh = new IO::File; opendir($fh, $dir) or die "Failed to read dir '$dir' ($!)\n"; my @files = readdir($fh); closedir $fh; foreach my $fname (@files) { next if $fname eq '.' or $fname eq '..'; my $path = "$dir/$fname"; if (-f $path and $path =~ /[.]txt$/) { $fh = new IO::File($fname) or die "Can't read '$fname' ($! +)\n"; chomp(my @lines = <$fh>); push @$a_text, @lines; close $fh; } elsif (-d $path) { read_textfile_lines($path, $a_text); } } return $a_text; }

        Now you're ready to call some subtroutine process_text (or whatever), passing the ref to the array of all text $a_text.  It will do something like:

        sub process_text { my ($a_text) = @_; while (1) { my $key; my $wait_until = time + 3; while ( time < $wait_until ) { ReadMode 3; $key = ReadKey( -1 ); if ( defined $key ) { print "keystroke $key\t"; } ReadMode 0; } print_some_random_text($a_text); } } sub print_some_random_text { my ($a_text) = @_; # Replacing the next line with meaningful code is left as # an exercise for the OP. ## print "[Debug]\n"; }

        As you see, I've left for you the fun of deciding what to print out in the subroutine print_some_random_text, as well as adding code for trapping relevant keys from the user in the subroutine process_text.  Good luck!


        s''(q.S:$/9=(T1';s;(..)(..);$..=substr+crypt($1,$2),2,3;eg;print$..$/

Log In?
Username:
Password:

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://869456]
help
Chatterbox?
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others musing on the Monastery: (4)
As of 2014-09-21 00:16 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?

    How do you remember the number of days in each month?











    Results (165 votes), past polls