| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Comments:
<0> but is it possible that each time a new entry is entered to have it increment the ID column of it by one so it is unique?
<1> sure
<0> how?
<1> you will most likely have to run 2 queries.
<0> i was afraid of that
<1> or one, if you're clever
<1> col1 = col1 + 1
<1> tada.
<0> i can do it on a column level?
<1> yes.
<0> cool
<0> and it will take the largest value in the column then?
<1> it will take what you feed it
<1> that's up to you to figure out how to get the right value.
<1> a subquery is required.
<0> so lets say i have a col with 3 rows, 1, 2, and 3 so i don't want it to add to 1 or 2 but 3 and next time it should add 1 to 4
<0> whats the best query to do that?
<0> i know this might be a bad query but i'm learning. can i have it evaluate against itself
<1> sure
<0> select * from table where col > col
<0> or is that completely illegal?
<1> from table_name tbl1 inne rjoin table_name tbl2 on tbl1.index = tbl2.index
<1> then use the aliases to control the values.
<0> i only have one table though
<2> draven, on mysql 5 you could do a trigger, i guess
<1> draven: that was using only one table too
<0> !trigger
<1> !m draven aliases
<3> draven: (Database, Table, Index, Column, and Alias Names) : http://dev.mysql.com/doc/mysql/en/Legal_names.html
<0> !m draven trigger
<3> draven: (CREATE TRIGGER Syntax) : http://dev.mysql.com/doc/mysql/en/CREATE_TRIGGER.html
<0> thanks for the info
<4> mysql]# mysqladmin -u dba_user -p create drupal
<4> Enter p***word:
<4> mysqladmin: connect to server at 'localhost' failed
<4> error: 'Access denied for user: 'dba_user@localhost' (Using p***word: YES)'
<4> #
<4> why is that?
<4> im following the manual of this drupal
<1> you're using the wrong p***word
<4> wrong p***word of my root p***word?
<1> wrong p***word for dba_user
<1> make sure the dba_user exists
<4> does this mean dba_user is alrady created
<1> and that you are using the proper p***word
<1> is it?
<1> you should know
<4> my first time to use mysql though
<4> ok ill create a new one
<1> then dba_user is most likely an EXAMPLE
<0> so this is the info i get from status: mysql Ver 11.18 Distrib 3.23.56, can i ***ume i'm on version 3.blah.blah?
<1> try root
<4> ok
<1> man mysql, see what -u does
<4> wow
<4> i dont have mysqld_safe
<1> wow what
<1> you do
<1> find / -name mysqld_safe
<4> # mysqld
<4> mysqld mysqld-max mysqld_multi mysqldump mysqldumpslow
<4> #
<1> find / -name mysqld_safe
<4> finding..
<1> qwertz: let me guess.... debian?
<4> nowhere to be found
<4> :((
<4> redhat 8.0
<1> ugh
<1> even worse
<1> i have no idea how to start it then
<4> heheeheh /usr/bin/safe_mysqld
<1> bleh
<1> red hat stupidity strikes again.
<4> hehehe fraggeln
<4> frank
<1> what
<5> i thought the joke normally starts with Knock Knock
<1> zircu: haha
<6> hi =)
<1> hi
<0> is there any reason that i shouldn't user auto_increment since you know that does what i need it to?
<6> I want to work with the COUNT(*) value from a query like: SELECT COUNT(*), somevalue FROM table1 GROUP BY somevalue; ... like COUNT(*)/MAX(COUNT(*))*100 ... is there a way to do it or do I have to use multiple queries ?
<1> you can do what you said
<0> me or him?
<1> however, never do count(*). instead , do something like count(PK)
<1> feindbild.
<7> Why not do COUNT(*)?
<1> possible performance penalty
<7> Why?
<5> if you specify a Primary key or single unique id, mysql knows exactly what you want, no guessing
<6> frank-: no =/ MAX(COUNT(*)) results in 'SQL State: HY000] Invalid use of group function'
<1> if your table has a lof of columns, then maybe it'll make a difference
<4> im in mysql> now
<1> but I find the count(PK) cleaner
<4> how do i change p***word of my dru_user username?
<4> im using mysql -u root -p
<5> feindbild: what is your like statment exaclyt?
<4> and now im in mysql >
<6> zircu: SELECT COUNT(*), MAX(COUNT(*)), lang FROM as_lang GROUP BY lang
<6> frank-: COUNT(*) is twice as fast as COUNT(lang) here ..
<1> I find that hard to believe.
<5> feindbild: is lang a Primary Key or Unique Key?
<1> is lang indexed?
<5> yeah, is it even indexed?
<6> frank-: lang is indexed ...
<1> is it a varchar?
<6> zircu: neither, just indexed ..
<6> frank-: yes
<8> Would you guys be able to point me to the appropriate resources that explain how to move data from one table to another?
<1> dont count on a varchar column
<5> feindbild: you are saying mysql is complaining about that statement?
<6> frank-: ... uhm ... sorry .. no ... its a char ..
<1> my point still stands
<6> zircu: i do ...
<1> count an int column, or indexed
<6> frank-: <1> however, never do count(*). instead , do something like count(PK)
<1> your PK is a char?
<1> ew
<5> jbrimble: create table dbname.tablename ( /* definition */ ) insert into dbname.tablename (dbname.tablename.fields) select tablename.fields from tablename
<6> frank-: no, my pk doesnt appear in the query ... my pk is a int ..
<1> I guess you missed my point.
<8> cheers, zircu.
<5> feindbild: use the pk field as the field you use instead of * inside count()
<9> on MyISAM count(*) is instantaneous
<1> it might do an internal optimisation, Goplat
<1> like only count the PK
<1> I'm not sure.
<9> myisam just keeps track of the current number of rows so it returns that
<1> I guess.
<1> the original point was based on the fact that select (*) is slower than select (PK)
<5> that sort of makes sense, since PK is a 1 to 1 to how many rows are involved
<4> Fatal error: Call to undefined function: mysql_real_escape_string() in /home/httpd/norminet/agencies/neda-x/includes/database.mysql.inc on line 239
<4> what does that mean?
<4> im using 4.2.2 php
<4> is that function supported?
<1> incompatible versions
<5> qwertz: you probably are using mysql client for 3.x
<5> qwertz: as i noted your upgrade to a more current according to your package system will be a little work, mysql should probably be updated as well
<5> hmm.. trying to edit a multiline string can lead to confusing statements, i hope you got what i was trying to say
<10> hi
<10> when should I use innodb over mysam ?
<11> when you need transactions, mainly. Or when you need some other feature that's not supported by myisam tables (yet): http://dev.mysql.com/doc/refman/5.1/en/roadmap.html
<10> what about speed and reliability ?
<11> innodb is significantly slower, but the only reason myisam is so fast is because it's not a real database. Only the simplest of apps don't really need transactions, which means a crashed (or killed) apache process will leave the database in an inconsistent state. But apps have to be written to use transactions with innodb, you don't get that magically.
<10> vandemar: thanks
<12> How do I restrict the number of preforked mysql processes?
<12> I looked all through the server system variables but didn't see anything.
<12> I want to prefork 2 or 3 processes, not 8! Memory is a bit scarce on this machine.
<13> mysql does not prefork processes
<13> if you're counting memory down to the KB it might be time for an upgrade
Return to
#mysql
or
Go to some related
logs:
sudo lecture_file
#oe
#perl
#gentoo
gtk change-theme ubuntu
ati x1600 second display problem
#debian
#asm
gentoo permissions portage group
#lgp
|
|