| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Comments:
<0> anyone into the job/hiring business? how do i make a good coversheet for my resume with salary req's and whatever else i need on it
<1> by storing it in a mysql database in a text field
<0> ofcourse
<0> but what is it i need to store exact;y
<0> thats the problem
<2> To use mysqldump to produce a CVS text file, you use --tab and also specify a --fields... and --lines... option, right?
<3> when someone says CVS instead of CSV, I always wonder if it's muscle memory (which it is when I do it) or if they just get their acronyms confused often
<4> hehe
<2> I meant SVN ;-)
<3> and yeah, that should work
<2> Thanks!
<5> Hi guys. I need to form a query that selects 5 posts from todays date, using a date field. How can that be done ?
<6> LIMIT 5
<6> select * from tableName where date='dateFormat' LIMIT 5
<6> you can even order by if you wish
<5> date='dateFormat' , what does that mean ?
<5> dateFormat, is that the name of the field ?
<7> he means, a date in MySQL format
<5> hmm
<8> hello, I'm trying to run a hulltext with the keyword 'ed' but apparently that's not allowed. how can I search on these small keywords?
<8> where hulltext = fulltext
<6> hulltext=`fulltext`
<6> fulltext is a mysql reserved word
<9> fulld: you need to lower the minimum fulltext size search in my.cnf, restart mysql and rebuild your fulltext indexes
<9> fulld: ft_min_word_len=
<9> default is 4 if I remember well.
<8> oh wow
<8> i'm writing a PHP/MYSQL program for the m***es, and many people wont be able to do this, is there an alternative solution?
<9> fulld: not that I know off.
<10> fulld: you can specify minimum length of lexeme when you create fulltext index...
<10> ah, zeitoun is here
<10> :)
<9> hello domas :)
<8> domas: much better :-)
<10> Hi!
<11> I don't know if "SET GLOBAL ft_min_word_len=2" would work, but it's worth trying
<11> er, or that
<5> Can't get this to work....
<5> It won't select anything
<12> mendel: you would need to rebuild the indexes after.
<12> mendel: you may also have stopwordlist issues with 2-3 character matches.
<8> i'll looking specifically for names
<8> like ed
<8> tom
<8> would be popular against my database
<8> I want to avoid seting a global since the people running this script wont have permissions on their shared hosts
<8> I'm trying to find how to set minium length when i create the index
<8> that would be optimal
<13> hello! I am trying to write a file that is about 1Mb large to a Blob with jdbc. But when I am looking in the database all I see is a 65535Bytes large Blob in there. what can be the mistake?
<12> fulld: set ft_min_word_len = 2 before you create the index and set it back to 4 when done.
<14> replication sometimes does it and sometimes dont't what could be the problem?
<8> do you need any specific privs to do that?
<12> fulld: if you have an existing index set ft_min_word_len then "REPAIR TABLE <t1> quick"
<12> fulld: yes, super.
<8> is that something people are goin to get from shared hosts
<12> fulld: no not likely.
<8> oh, so it's not like a database specific priv its server-wide?
<12> yea
<12> it's "SUPER" user privilege.
<9> bright: you should use mediumblob or largeblob. blob is limited to 2^16
<9> bright: http://dev.mysql.com/doc/refman/5.0/en/storage-requirements.html
<12> fulld: actually... I just tried setting that value on my test server and it gave me Unknown system variable 'ft_min_word_len'
<12> even as 'root'
<8> ulth
<12> it's probably one of those variables you need to set at server startup.
<8> i'm looking in the documentation for alter
<12> which makes it a whole lot harder to get your host to do. ;)
<8> you can specify a parser_name when defning the index
<8> but I can't find anything about the parser_name
<9> fulld: http://dev.mysql.com/doc/refman/5.0/en/fulltext-fine-tuning.html
<9> fulld: I tried hard for a site of mine and setting the var at server startup was really the only solution I could use.
<9> domas: by the way... where did you see you could set the min word length on fulltext index creation ? If it is really possible, I'm much interested.
<7> zeitoun - its in the doc you referenced
<15> I've forgotten my mysql root p***word, how do i get it bacK?
<7> The minimum and maximum lengths of words to be indexed are defined by the ft_min_word_len and ft_max_word_len system variables. (See Section 5.2.2, Server System Variables.) T
<15> It's on my local windows machine
<7> !m nash reset root p***word
<16> nash: (How to Reset the Root P***word) : http://dev.mysql.com/doc/mysql/en/Resetting_permissions.html
<8> it appears that are alternative parsers for fulltext searches
<12> nash: http://dev.mysql.com/doc/refman/5.0/en/resetting-permissions.html
<7> fulld - but it's probably not a good idea to set it to '2'
<10> zeitoun: same ft_ values.
<10> ;-)
<9> sjrussel domas: ho ok..... I thought there was some create index option I has missed.....
<7> a better answer is probably to do the search, then add a HAVING to limit the results by the short word
<9> had...
<7> I don't think so, but you have to rebuild the index after to get the different words
<8> sjrussel: could you make a quick example of this having syntax?
<7> !m fulld SELECT
<16> fulld: (SELECT Syntax) : http://dev.mysql.com/doc/mysql/en/SELECT.html
<7> HAVING is like WHERE
<7> but acts on the result set.. generally, it's better to use WHERE, but this might be a counterexample
<7> or just put it in where
<8> so like SELECT * FROM photos WHERE MATCH(description) AGAINST('mary') HAVING 'ed'
<7> no, it's like WHERE
<7> HAVING field like '% ed %'
<8> so like SELECT * FROM photos WHERE MATCH(description) AGAINST('mary') HAVING description LIKE 'ed'
<8> cool
<8> ed is the specific cause of this whole mess
<7> it might be better to just put in the where, actually
<8> since %ed% matches "bothered"
<7> let the optimizer figure out how to run the query
<17> wait, what would `description` contain in that particular case?
<17> for matching rows
<7> a regex might be better
<17> *mary*ed* ?
<8> i have photos descriptions like 'ed drinking beer' 'christopher and mary' 'bob at home'
<7> 'ed and mary josephson'
<8> and users will search for stuff like 'ed' 'chris' 'chris and mary'
<17> oh, ok, so it would include "ed" and "mary" in any order
<8> currently i use %term% like desc
<7> a regex is probably better than like
<17> as opposed to %ed%mary% which would only match in that order.
<17> thanks. sorry to intrude, that was confusing.
<7> /\bed\b/
<8> i could use PHP to turn 'ed and mary' to '(\W|^)ed' AND '(\W|^)mary'
<18> i have a query that does: COUNT(*) where col=x, how do it take that and divide it by the total COUNT(*) (with no where clause)?
<19> Iontas: Use IF() instead
<12> fulld: it appears that the alternate parsers are a going to be in 5.1
<18> salle: ?
<8> TodoInTX: hmm it appears that fulltext makes too many restrictions and I'll have to do a custom solution
<15> Thanks.
<18> if COUNT(*) where col=x then / COUNT(*) ??
<19> Iontas: SELECT SUM(IF(col=x, 1, 0)) AS c1, COUNT(*) AS c2,
<8> will that regex above work in my case with the (\W|^), specifically?
<12> http://dev.mysql.com/doc/refman/5.1/en/plugin-full-text-plugins.html
<19> Iontas: ... (SUM(IF(col=x, 1, 0))/COUNT(*)) * 100 AS percent ...
<19> Iontas: And so on
<8> salle: that's wild
<19> fulld: Why?
<18> kewl...lemme try
<8> just an awesome query
<19> Iontas: If you prefer you can do it with subqueries indeed
<19> fulld: Well ... that's what RDBMS are for :)
<19> fulld: To generate new data using the existing
<18> boo...where's the table go in that line?
<18> salle: i don't care how it's done...whatever is easiest. once i get one, i can do the rest
<19> Iontas: In FROM part :) I only told you about SELECT part
<18> i tried...lemme redo it...i hosed something up
<19> Iontas: I recommend you to read carefully this fantastic article. http://dev.mysql.com/tech-resources/articles/wizard/index.html
<19> fulld: Same for you
<18> w00t....thx
<18> i'm rtfm'ing the wrong page :-P
<20> any apache 1.3 and mysql_auth folks around ?
<17> don't ask to ask
<20> ok, i'll just shoot...
<18> salle: many thx...i think this will do what i need it (did i mention i hate pulling stats)
<20> when using mysql_auth with apache 1.3 it works if the database is local to the apache server (localhost..etc..) But when i change the database host to be a remote host it fails, because it's looking for the field name "mysql_auth" in my database table. I have specifically told it too look for "users" table, but it ignores that. It does this only when connecting to a remote databse host.
Return to
#mysql
or
Go to some related
logs:
running ppt viewer on ubuntu
projectm sources.list
#perl
endrazine
#mysql
innodb without mysql
#kde
symbol errno, version GLIBC_2.0 not defined in file libc.so.6 with link time ref
static.zoomr
kerb5 deb
|
|