FHEM Forum

I have several device names longer then 20 characters, which messes up the padding when using cmd: list devspec value. Increasing this to 30 seems to work well in my case. This could be hardcoded also.

Before patch:

Ntf_EnO_DimToStoredValue     active
Ntf_EnO_Occupancy_Low_Battery     active
Ntf_EnO_Smoke_Alarm      active
Ntf_EnO_Smoke_Dead_Sensor     active
Ntf_EnO_Smoke_Low_Battery     active
Ntf_EnO_Sync_FUD61       active

With patch:

attr global listNamePadding 30

Ntf_EnO_DimToStoredValue           active
Ntf_EnO_Occupancy_Low_Battery      active
Ntf_EnO_Smoke_Alarm                active
Ntf_EnO_Smoke_Dead_Sensor          active
Ntf_EnO_Smoke_Low_Battery          active
Ntf_EnO_Sync_FUD61                 active

--- - 2019-10-05 12:25:57.074228921 +0200
+++ /srv/fhem/fhem.pl 2019-10-05 12:24:38.000000000 +0200
@@ -342,6 +342,7 @@
   language:EN,DE
   lastinclude
   latitude
+  listNamePadding
   logdir
   logfile
   longitude
@@ -2460,6 +2461,7 @@
CommandList($$)
{
   my ($cl, $param) = @_;
+  my $namefmt = '%-'.AttrVal('global', 'listNamePadding', 20).'s';
   my $str = "";

   if($param =~ m/^-r *(.*)$/i) {
@@ -2498,7 +2500,7 @@
       next if(IsIgnored($d) || ($cl && !Authorized($cl, "devicename", $d, 1)));
       my $t = $defs{$d}{TYPE};
       $str .= "\n$t:\n" if($t ne $lt);
-      $str .= sprintf("  %-20s (%s)\n", $d, $defs{$d}{STATE});
+      $str .= sprintf("  $namefmt (%s)\n", $d, $defs{$d}{STATE});
       $lt = $t;
     }

@@ -2525,13 +2527,13 @@
                 $val = ($val->{NAME} ? $val->{NAME} : # ???
                         join(" ", map { "$_=$val->{$_}" } sort keys %{$val}));
               }
-              $str .= sprintf("%-20s %*s   %*s %s\n", $first?$sdev:'',
+              $str .= sprintf("$namefmt %*s   %*s %s\n", $first?$sdev:'',
                         $arg[2]?19:0, '', $arg[2]?-15:0, $arg[2]?$n:'', $val);

             } elsif($defs{$sdev}{READINGS} &&
                     defined($defs{$sdev}{READINGS}{$n})
                     && (!$fType || $fType eq "r:")) {
-              $str .= sprintf("%-20s %s   %*s %s\n", $first?$sdev:'',
+              $str .= sprintf("$namefmt %s   %*s %s\n", $first?$sdev:'',
                       $defs{$sdev}{READINGS}{$n}{TIME},
                       $arg[2]?-15:0, $arg[2]?$n:'',
                       $defs{$sdev}{READINGS}{$n}{VAL});
@@ -2539,7 +2541,7 @@
             } elsif($attr{$sdev} &&
                     defined($attr{$sdev}{$n})
                     && (!$fType || $fType eq "a:")) {
-              $str .= sprintf("%-20s %*s   %*s %s\n", $first?$sdev:'',
+              $str .= sprintf("$namefmt %*s   %*s %s\n", $first?$sdev:'',
                         $arg[2]?19:0, '', $arg[2]?-15:0, $arg[2]?$n:'',
                         $attr{$sdev}{$n});



Anybody else interested in this patch?
Objections?

Zitat von: rudolfkoenig am 05 Oktober 2019, 19:44:30
Anybody else interested in this patch?
Objections?

Maybe this value can be passed as a parameter to the command 'list' instead.

list .+ NR REGEXP padding=30

Zitat von: Nestor am 05 Oktober 2019, 20:14:02
There is also a bug when multiple values are requested and the first value is not defined but the second is, the name will not be printed.

Patch with fix:
--- - 2019-10-06 07:42:47.000000000 +0200
+++ fhem.pl 2019-10-06 07:40:48.000000000 +0200
@@ -2527,6 +2527,7 @@
               }
               $str .= sprintf("%-20s %*s   %*s %s\n", $first?$sdev:'',
                         $arg[2]?19:0, '', $arg[2]?-15:0, $arg[2]?$n:'', $val);
+              $first = 0;

             } elsif($defs{$sdev}{READINGS} &&
                     defined($defs{$sdev}{READINGS}{$n})
@@ -2535,6 +2536,7 @@
                       $defs{$sdev}{READINGS}{$n}{TIME},
                       $arg[2]?-15:0, $arg[2]?$n:'',
                       $defs{$sdev}{READINGS}{$n}{VAL});
+              $first = 0;

             } elsif($attr{$sdev} &&
                     defined($attr{$sdev}{$n})
@@ -2542,10 +2544,10 @@
               $str .= sprintf("%-20s %*s   %*s %s\n", $first?$sdev:'',
                         $arg[2]?19:0, '', $arg[2]?-15:0, $arg[2]?$n:'',
                         $attr{$sdev}{$n});
+              $first = 0;

             }
           }
-          $first = 0;
         }
       }


And yet another patch which fixes unescaped entities in the list cmd documentation and adds undocumented prefixes for value argument.

--- - 2019-10-06 11:58:26.000000000 +0200
+++ docs/commandref_frame.html 2019-10-06 11:55:02.000000000 +0200
@@ -941,7 +941,7 @@
<a name="list"></a>
<h3>list</h3>
<ul>
-  <code>list [devspec] [value]</code><br>
+  <code>list [devspec] [value ...]</code><br>
   or<br>
   <code>list {-r|-R} devspec</code><br>
   <br><br>
@@ -951,13 +951,13 @@
   See the <a href="#devspec">Device specification</a> section for details on
   &lt;devspec&gt;.
   <br>
-  If value is specified, then output this property (like DEF, TYPE, etc) or
-  reading (actuator, measured-temp) for all devices from the devspec.
+  If <tt>&lt;value&gt;</tt> is specified, then output this property (internal, reading or attribute) for all devices from the devspec.
+  <tt>&lt;Value&gt;</tt> can be restricted with prefix <tt>i:</tt> for internals, <tt>r:</tt> for readings and <tt>a:</tt> for attributes.
   <br><br>
   Example:
-  <pre><code>  fhem> list
+  <pre><code>  fhem&gt; list

-  Type list <name> for detailed info.
+  Type list &lt;name&gt; for detailed info.

   Internal:
     global               (Internal)
@@ -988,7 +988,7 @@
   </code></pre>
   If specifying <code>name</code>, then a detailed status for <code>name</code>
   will be displayed, e.g.:
-  <pre><code>  fhem> list fl
+  <pre><code>  fhem&gt; list fl

   Internals:
     CODE       5102

Thanks for the hint, I fixed the bug mentioned in the 3rd post, and the documentation (both english and german).
For the first patch: if there are at least 2 user voting for it, I will integrate it.