I was in need of a logging mechanism that prints all log messages with a “severity” that equals or is higher of “INFO” to STDOUT (I’m using multilog for actual logging stuff).
As I wasn’t able to google a solution I came up with the following (not quite elaborated) code:
class OwnLogger < Logger
include Severity
SEV_LABEL = %w(DEBUG INFO WARN ERROR FATAL ANY)
def add(severity, message = nil, progname = nil, &block)
time = Time.now
if ENV['LOG_FROM_INFO_TO_STDOUT']
if severity >= INFO
msg = "#{$$}:#{SEV_LABEL[severity]}:" + time.strftime("%Y-%m-%dT%H:%M:%S.") << "%06d " % time.usec + ":"
if message.nil?
if block_given?
message = yield
else
message = progname
progname = @progname
end
msg += message.to_s
puts msg
end
end
super
end
end
So if you put the following lines in your /config/application.rb you’ll get what I got
config.logger = OwnLogger.new(File.dirname(__FILE__) + "/../log/#{ENV['PROCNAME']}#{Rails.env}.log")
AND (most important) you’ll still have the “normal” logfile containing all messages
Schamlose Werbung
Wenn Ihr liebgewordene Schuhe mal nicht wegwerfen sondern lieber reparieren lassen wollt und zufällig in der Nähe von Solingen seid, dann besucht doch mal die Familie Janisch.
Die erledigen das dann für Euch…
Details unter Schuhmacherei-Janisch.de
Klare Empfehlung von mir!
Just took me 2 hours of Googling…
I found:
http://stackoverflow.com/questions/246983/informix-how-to-get-an-id-of-the-last-inserted-record
[...]The value of the last SERIAL insert is stored in the SQLCA record, as the second entry in the sqlerrd array. Brian’s answer is correct for ESQL/C, but you haven’t mentioned what language you’re using.
If you’re writing a stored procedure, the value can be found thus:
LET new_id = DBINFO(‚sqlca.sqlerrd1‘);
It can also be found in $sth->{ix_sqlerrd}[1] if using DBI
There are variants for other languages/interfaces, but I’m sure you’ll get the idea.
[...]
I got the idea and have created the following procedure
CREATE PROCEDURE lastid ()
RETURNING integer;
DEFINE lastid integer;
LET lastid = DBINFO(‘sqlca.sqlerrd1′);
RETURN lastid;
END PROCEDURE
So to get the latest column you just use
select lastid() from table(set{1})
btw.: If you want to make use of this in Rails, I’d some success with:
class InformixConnect < ActiveRecord::Base
self.abstract_class = true
establish_connection :development_informix
def save
super
self.id = self.connection.select_value(“select lastid() from table(set{1})”)
end
end
so a model like
class Customer < InformixConnect
end
would be able to connect to Informix and to get correct values from the DB without the
usage of sequences!
ERROR
Unfortunately the stuff above (overwriting the save method) does not work for some reasons.
I have patched
class InformixAdapter < AbstractAdapter
def prefetch_primary_key?(table_name = nil)
false
end
….
def insert(sql, name= nil, pk= nil, id_value= nil, sequence_name = nil)
execute(sql)
select_value(“select lastid() from table(set{1})”)
end
…..
This works!
So it happened again:
Most of my application icons were broken, i.e. only visible in a very low resolution.
After googling a bit I found the following discussion:
http://discussions.apple.com/thread.jspa?threadID=2142401
mdi’s suggestion did the trick for me: http://discussions.apple.com/message.jspa?messageID=10296267#10296267
What I did:
- open Terminal.app (If you never did… : Try „spotlight-search“ –> Terminal)
- enter cd /private/var/folders
- enter sudo find . -name “com.apple.QuickLook.thumbnailcache”|xargs sudo rm -rf
(you will have to enter your password)
- Restart Finder.app using „Apple-Menu–> „quit immediately (It’s „Sofort Beenden“ in German. Dunno the english term) or press
- command+option+escape“. Chose „Finder restart“.
- Be happy
OK, to be honest I had to restart to be really happy
Argh…
This f…ine error message above has been thrown during a
records = Model.find(:first).
Guess why:
I made use of a legacy table that contained the attribute „type“. I just had to add a
self.inheritance_column=’itype’
into my model and now everything’s OK.
Rails: Do you consider this as friendly error messages?
Hmmm… Kleine Fingerübung in Perl (pre-alpha
)
Ah ja: Brainfuck: http://lmgtfy.com/?q=Brainfuck
#!/usr/bin/env perl
use Term::ReadKey;
ReadMode 'raw';
my $debug = 0;
my $code;
if ($ARGV[0]){
{ local $/ = undef; local *FILE; open FILE, "<".$ARGV[0]; $code = ; close FILE }
}
$code ||=
">+++++++++[<+++++++++++++>-]<-.-----.+++++++
.>+++[>++++++++++<-]
>++.<<----.+++.
--------- . -------.>++++[<+++++>-]<+.>++++++++++.";
$code =~ s/[^\.,+-\[\]<>]//msg;
$| = 1;
my @val;
@val[ 0 .. 29999 ] = (0) x 30000;
my $vpt = 0;
my @ls;
my @le;
my $lpt = 0;
my $cc = 0;
my @c = split( '', $code );
my $nc = @c;
while ( $cc < $nc ) {
$_ = $c[ $cc++ ];
print "vpt:$vpt \$val[\$vpt]"
. $val[$vpt]
. " loop_counter:$loop_counter loop(counter)_start:"
. $ls[$loop_counter]
. " loop(counter)_end:"
. $le[$loop_counter]
. " cc:$cc command:$_\n"
if $debug;
/ and do { $vpt and $vpt--; next };
/>/ and do { $vpt < @val and $vpt++; next };
/\+/ and do { $val[$vpt] = $val[$vpt] + 1; next; };
/\-/ and do { $val[$vpt] = $val[$vpt] - 1; next; };
/\./ and do { print chr( $val[$vpt] ); next; };
/\,/ and do { $val[$vpt] = ord( ReadKey 0 ); next; };
/\[/ and do {
if ( $val[$vpt] ) { $ls[ $lpt++ ] = $cc - 1 }
else { $cc = defined $le[$lpt] ? $le[$lpt] : (index($code,']',$cc)+1) }
next;
};
/\]/ and do {
$le[ --$lpt ] = $cc;
$cc = $ls[$lpt];
next;
};
}
print "\n";
Umpft.
Ich konnte dem aktuellen http://www.macheist.com Nano-Bundle natürlich doch nicht widerstehen. Monkey-Islands war eine Versuchung wert und vor allen Dingen Tweetie 2.0.
Jetzt habe ich auch noch MacJournal dazu bekommen und bin eigentlich ziemlich begeistert.
- Blogging mit einem vernünftigen Client
- Verschlüsselung privater Einträge
- Export von Einträgen als HTML, PDF etc.
- Volltextsuche via Spotlight.
- Synchronisation via Mobile-Me (muss noch TeamDrive ausprobieren)
Und alle so „Yeah“!
P.S. Und die Probleme von Safari mit WordPress brauchen mich jetzt auch nicht mehr zu kümmern
@johnny von Spreeblick hat eine Spendenaktion für die Erdbebenopfer von Haiti in’s Leben gerufen.
Nicht, dass es nicht auch andere Wege des Spendens gäbe, aber das Problem ist ja (und das löst sich durch solche Aktionen halt), dass “man” (ja, auch ich) in der Regel den A… ja nicht hochkriegt.
Also: Wenn Ihr Eure Kontonummer noch kennt, einfach mal auf obigen Link clicken!
Danke sehr!
Problem: Centos did not boot after a failure with an LVM snapshot.
It got stuck at:
Setting up Logical Volume Management: (part of rc.sysinit)
Solution:
Boot:
1) grub “e” (to edit the first line)
2) append “init = /bin/bash” to kernel parameters
3) boot (Enter, “b”)
(append “Single” should work as well but it didn’t as far as I remember)
Console
1) mount / -o remount (enables rw mode)
2) vi /etc/lvm/lvm.conf
3) replace
locking_dir = “/var/lock/lvm”
by
locking_dir = “/tmp”
(this is because “lvremove” does not recognize the –ignorelockingfailure parameter)
4) cd /sbin
5) ./lvm.static lvdisplay –ignorelockingfailure (not sure if this param is necessary after editing lvm.conf)
6) Find the snapshot path: /dev/your_volumegroup/your_logical_volume_snapshot
7) remove: ./lvm.static lvremove /dev/your_volumegroup/your_logical_volume_snapshot
reboot
9) be happy as it works now!
10) ask yourself if you should have re-edited lvm.conf…
I have forgotten to do so but as “/tmp” is more or less recommended in lvm.conf.
Hope that helps either people or me (in future).
Seems that it sometimes only pretends to have deleted the f…ine dustbin.
Try
cd /d C:\$Recycle.bin
dir /AH
rd /s /q “NameOfFolder”
if you need some space
