酷!學園

有一個 table 有 1仟萬筆的資料，想把某欄裡 是 null 跟空白的抓出來。有什麼有效率的方式嗎？
目前我用的是
select * from table where isnull(date,'')=''

公司要買 sql 當 web 的資料庫，可不可以只買 5 u 的授權就可以了。
如果只買 5 u 會不會有限制，iis 只能有5個使用者連到sql server。

ps：個人覺得 不管 iis 有多少的使用者連到 sql server  應只算1個使用權。

一般在 gcc 後面加上這些參數-mcpu=i386, -mcpu=i486, -mcpu=pentium, ...
真得程式會跑的比較快嗎？有沒有什麼程式可以有數據測試？

假設目前系統有一個 table test 而 test 只有一個 int 的欄位
當我下
create table if not exists test select 1 from dual

會發生什麼情況？


============
公佈答案，每執行這個sql一次 test 的資料就會加一筆 1 的資料在裡頭
if not exists test  只影響 create table 不影響 後面的 select

第一次寫class，大部份改至hinet的公用程式。
目前程式已可以用，想問一下 sms_hinet.pm 還有什麼要改的嗎？


EXAMPLE：
#!c:/Apache2/Perl/bin/Perl.exe
use sms_hinet;
my $new_sms=sms_hinet->new();;
$new_sms->user_account('8888');
$new_sms->user_password('9999');
$new_sms->mobile_number('0806449');
$new_sms->message("簡訊內容");
if ($new_sms->check_hinet_account()==0) {
$new_sms->sendmsg();
}




====將以下存成 sms_hinet.pm ====
package sms_hinet;
use Socket;
#作者：我本善良   ：asakous(at)yahoo.com.tw
#版權：請自由取用，但請不要拿掉作者是誰。
sub new {
        my $self  = {};
        $self->{ACCOUNT}   = undef;
        $self->{PASSWORD}    = undef;
      $self->{msg}    = undef;
      $self->{interface_type}    = 0;
      $self->{msg_type}    = 0; # 0:檢查帳號密碼 1:傳送簡訊
      $self->{send_type}    = 100; # 100即時傳送
      $self->{mobile_number}    = undef;
        bless($self);            
        return $self;
    }

   sub user_account {
        my $self = shift;
        if (@_) { $self->{ACCOUNT} = shift }
        return $self->{ACCOUNT};
    }
    sub user_password {
        my $self = shift;
        if (@_) { $self->{PASSWORD} = shift }
        return $self->{PASSWORD};
    }
   
       sub interface_type {
        my $self = shift;
        if (@_) { $self->{interface_type} = shift }
        return $self->{interface_type};
    }
   
       sub msg_type {
        my $self = shift;
        if (@_) { $self->{msg_type} = shift }
        return $self->{msg_type};
    }

       sub send_type {
        my $self = shift;
        if (@_) { $self->{send_type} = shift }
        return $self->{send_type};
    }   
   
       sub message {
        my $self = shift;
        if (@_) { $self->{msg} = shift }
        return $self->{msg};
    }   


   sub mobile_number {
        my $self = shift;
        if (@_) { $self->{mobile_number} = shift }
        return $self->{mobile_number};
    }   
   
sub check_hinet_account {

 my $self = shift;
  $proto = getprotobyname('tcp');
  $i_addr = inet_aton('203.66.172.131');
  $paddr = sockaddr_in('8000', $i_addr);
  socket(SOCK, PF_INET, SOCK_STREAM, $proto);
  connect(SOCK, $paddr) or die "\n### Error: Couldnot using socket to connect to remote_host(203\.66\.172\.131:8000) : $! ";

  #檢查帳號密碼
  $dd = $self->{ACCOUNT} . "\0" . $self->{PASSWORD} . "\0";
  $tmp = pack("C",$self->{interface_type}) . pack("C",$self->{msg_type}) . pack("C",length($dd)) . $dd ;  

  my $send_length = send( SOCK, $tmp, 0 );

  my $scalar="" ;
  my $ret_length = recv( SOCK, $scalar, 129, 0 );


  my $return_c = substr $scalar,0,1 ;
  my $return_code = unpack("c", $return_c);
 
  my $return_message_length = length $scalar;
  $return_message_length--;
  my $return_message = substr $scalar, 1, $return_message_length;

  return $return_code;

}   

sub sendmsg {
      my $self = shift;
     $self->{msg_type} =1;   
     my $dd = $self->{mobile_number} . "\0" . $self->{msg} . "\0";
     my $tmp = pack("C",$self->{interface_type}) . pack("C",$self->{msg_type}) . pack("C",length($dd)) . $dd . pack("C",$self->{send_type});  
     my $send_length = send( SOCK, $tmp, 0 );
     my $scalar="" ;
     my $ret_length = recv( SOCK, $scalar, 129, 0 );
     #print "\n rec data:" . $scalar;
     my $return_c = substr $scalar,0,1 ;
     my $return_code = unpack("c", $return_c);
     my $return_message_length = length $scalar;
     $return_message_length--;
     my $return_message = substr $scalar, 1, $return_message_length;
   return $return_code;

     
}
   
sub close_session {

my $self = shift;
close(SOCK);
}

1;


====sms_hinet.pm完=========

use strict;
use WWW::Mechanize;

#設定每期樂透公佈的網頁
my $url = "http://www.roclotto.com.tw/topnews/biglotto.asp";
my $mech = WWW::Mechanize->new();

$mech->get( $url );
#抓完存檔
$mech->save_content( '1.html');


my @special_number;
my %lotto_number;
my $line_number=1;

#讀檔
open(output,"< 1.html ") or die;

while (<output>) {

if ($_ =~ /(\d+)、(\d+)、(\d+)、(\d+)、(\d+)、(\d+)/ )

 {  
 #把到的一般號碼抓進hash裡
 %lotto_number=($1=>1,$2=>1,$3=>1,$4=>1,$5=>1,$6 );
  }

#抓特別號
 if ($_ =~ /<font color=\"\#FF0000\">(\d+)<\/font>/ )


 { push @special_number ,$1;  }
 

}
close(output);

#開檔，你的樂透號碼檔
#每一行的格式為(記得把#號去掉，最後一個號碼記得斷行(\n))
#2 6 11 12 27 45
#4 12 17 20 24 47

open(log_file,"< log.txt ") or die;

while (<log_file>) {
my $hit_number=0;
my @line_number=split(/\s/);

foreach my $temp_number (@line_number) {

#看讀進的號碼有沒有中
if (exists $lotto_number{$temp_number}){

$hit_number++;
}


}

#印出行號及中獎的號碼數
print '行號 ',$line_number," 中 ",$hit_number," 號碼\n";
$line_number++;
}

close(log_file);

#
#
#備註：程式中用所使用的正規可能不是很好，也許有發生錯誤的可能
#但目前可用
#希望有其它的人能想出更好的方法。
#忘記寫特別號的判斷，等寫完再補上  
#
#

#-------------以下是有特別號判斷的--------------------

use strict;
use WWW::Mechanize;

my $url = "http://www.roclotto.com.tw/topnews/biglotto.asp";
my $mech = WWW::Mechanize->new();
$mech->get( $url );
$mech->save_content( '1.html');

my (%lotto_number,%special_number);
my $line_number=1;

open(output,"< 1.html ") or die;

while (<output>) {

if ($_ =~ /(\d+)、(\d+)、(\d+)、(\d+)、(\d+)、(\d+)/ )

 {  
 
 %lotto_number=($1=>1,$2=>1,$3=>1,$4=>1,$5=>1,$6 );
  }

 if ($_ =~ /<font color=\"\#FF0000\">(\d+)<\/font>/ )

 { %special_number=($1=>1);  }
 

}
close(output);

open(log_file,"< log.txt ") or die;

while (<log_file>) {
my ($hit_number,$hit_special)=(0,'沒中');
my @line_number=split(/\s/);

foreach my $temp_number (@line_number) {

if (exists $lotto_number{$temp_number}){

$hit_number++;
}

if (exists $special_number{$temp_number}){

$hit_special="有中";
}



}


print '行號 ',$line_number," 中 ",$hit_number," 號碼 "," 特別號 $hit_special\n";
$line_number++;

}

close(log_file);

簽這麼久也沒中超過400元。
閒來無事用perl寫個『電腦選號』
祝大家中大獎。

#!c:/Apache2/Perl/bin/Perl.exe
use Array::Unique;
use strict;
use warnings;
tie my @a, 'Array::Unique';

until ( $#a == 5){
my $num1 = int rand 49 ;
if ($num1 == 0){next;}
push @a,$num1;
}
@a = sort {$a <=> $b } @a;

print $a[0]." ".$a[1]." ".$a[2]." ".$a[3]." ".$a[4]." ".$a[5]."\n";

javascript 能跑 sql query 嗎？

寄件者：Matthew England (mengland@chi.mti.com)
主旨：RE: How automatically to store emails into MySQL?
View: Complete Thread （共 7 條留言）  
Original Format
網上論壇：mailing.database.mysql
日期：2000/03/31
 

The Perl and PHP examples provided by Indrek and Michael are great!  These
examples are just the things I was looking for to help get me started.

Any other email-header parsing utilities besides MIME::Parser?  Anything I
can access via PHP?

I answer Indrek's question below:

At 3/30/00 05:28 PM, indrek siitan wrote:
>um. care to explain, what you're planning to do with the e-mail
>afterwards?

I like to describe my goal as an application that mimicks an email-based
"trouble-tracking" system (eg, GNATS, a GNU email-based bug-tracking).  I
want to collect messages/info about customers, products, and events and be
able to retrieve these messages accordingly.  We've chosen to distribute
messages via email in our initial product implementation.

Some of the emails will contain specific, company-proprietary
information.  Other emails will be free form.

I absolutely need to store each message/email in it's entirety--no headers
chopped off or anything else.  This is very important for my
message-archiving requirements.

I will also want to parse the email headers so that I can store things like
date, time, "case ID", originator, etc.  I'm not yet concerned about data
redudancy (from the header info being copied into record
fields/columns).  Should I be concerned about this?

-Matt

>one thing is for sure - you can't use mysql client for that. the
>easiest way is to forward mail to a perl script, what does the job.
>the easiest script should look something like this (i'm writing
>it directly to the e-mail, so i'm not responsible for this being
>absolutely bugfree):
>
>--- 8< ---
>
>#!/usr/bin/perl
>use DBI;
>
>$dbh=DBI->connect("dbi:mysql:database", 'user', 'password');
>
>my $mailbody=''; while (<STDIN>) { $mailbody.=$_; }
>my $qmailbody=$dbh->quote($mailbody);
>
>$sth=dbh->prepare("INSERT INTO mail (body) VALUES($qmailbody)");
>$sth->execute;
>
>$dbh->disconnect;
>
>
>--- 8< ---
>
>if you want to store e-mail headers into separate fields, i
>recommend you taking a look at MIME::Parser module. this does
>a really nice job taking the mail into pieces.
>
>i have a script that does that, but it's not complete yet
>(craps out on some types of attachments)
>
>
>Rgds,
>   Tfr
>
>   --==< tfr@cafe.ee >==< http://tfr.cafe.ee/ >==< +1-504-4467425 >==--


--
---------------------------------------------------------------------
Please check "http://www.mysql.com/Manual_chapter/manual_toc.html" before
posting. To request this thread, e-mail mysql-thread32776@lists.mysql.com

To unsubscribe, send a message to:
    <mysql-unsubscribe-mysql=freebsd.csie.nctu.edu.tw@lists.mysql.com>

> We have three WAN links each of capacity 2 Mbps.
>
> There is a one linux box per link which is used for NAT and firewall. So
> I have three linux boxes and three WAN links to talk to the internet.
>
> We want to do bandwidth shaping over these three links. I know
> individually we can do bandwidth shaping for a particular WAN link using
> linux tc tool. I can repeat the same bandwidth shaping commands for
> other two links also. However it would be better if I can treat three
> links as one big WAN pipe and do the bandwidth shaping for all of them
> at one place. Is it possible?
>
> I am just thinking aloud and not sure if that is possible or not.
It also depends if you want to shape download or upload.  If you want to shape
upload, you have to realize that you can only use 1 link.  For download, you
can use 3 links but only if you NAT and you have a lot of connections so you
can load balance the different connections (like http traffic).  An ftp
download, will only use 1 link.
This all to say that's not so easy to determine what will be the maximum
bandiwdth a class can get.  And that's one of the parameters you need to
create a good tc setup.  But you can create a tc setup that can shape the 3
links in both directions.  Even if this shaping is not perfect, it still
beter then not shaping at all.

Stef

--

stef.coene@docum.org
 "Using Linux as bandwidth manager"
     http://www.docum.org/
     #lartc @ irc.oftc.net