#perl
22 October 2007
Total 35 pages. You are browsing page 3/35.
First :: Prev :: [1] [2] [3] [4] [5] [...] :: Next :: Last
01:39 <****> mst: well, as of yet, the methods I've been calling haven't seemed to have worked... ideally i want to stick to rails conventions and not work with spaced column names
01:39 <****> but i might be missing the whole thing in rails anyway
01:40 <****> apeiron: But there are still gotchas lurking around--such as mysql liking to automatically update TIMESTAMP columns
01:40 <****> jagerman, I *HATEHATEHATEHATE* SQL, so I want something that'll solve both problems at once. :)
01:40 <****> its really all the fault of the access database which has all the spaced out column names
01:40 <****> neshmi: right.
01:40 <****> ... access database?
01:40 <****> neshmi: have you considered learning Catalyst and DBIx::Class instead?
01:40 <****> You're... writing a RoR application that frobs an *Access* database?
01:40 <****> neshmi: rails doens't really handle existing databases
01:40 <****> apeiron: Where I worked, it was worse, because my boss *liked* how TIMESTAMP worked, so instead of a policy of "don't use mysql-specific things like timestamp" the policy was "let's write a translation layer that makes things work like mysql"
01:40 <****> neshmi: or, indeed, databases at all.
01:41 <****> jagerman, ... .ewwww
01:41 <****> neshmi: DHH is a PHP developer who used mysql, so his database understanding is shit.
01:41 <****> well, rails seems to be doing pretty well with the databases i'm working on with other projects, but the issue i'm having here is trying to run the whole process of migration through
01:41 <****> apeiron: what is an alternative to SQL?
01:41 <****> Patterner, That's the problem. There isn't one... yet.... /me prods dduncan.
01:41 <****> neshmi: so long as your data getting corrupt doesn't matter, rails is fine.
01:42 <****> mst: "doesn't handle existing databases" seems like a really bad restriction!
01:42 <****> mst, Hurray for lack of ACID compliance!
01:42 <****> Or is it not even that?
01:42 <****> Khisanth: rails only believes in databases with single-column auto-inc primary keys
01:42 <****> Khisanth: and no foreign key constraints
01:42 <****> Khisanth: basically it's "the MyISAM framework"
01:42 <****> hm, are there any other pragmas besides use strict; use warnings; and use diagnostics; ? I get a "my variable redefined" warnings but I just don't see where it is redefined and the warning line number points at the end of the file (almost)
01:42 <****> mst: Exactly what I had to work with :(
01:42 <****> Khisanth: it doesn't even use bind parameters, activerecord support ? but it interpolates+quotes at the ruby level
01:42 <****> jagerman, At least the PB guys don't use auto-increment.
01:43 <****> PB?
01:43 <****> auto-inc has its place.
01:43 <****> Zoffix: non sequitur
01:43 <****> jagerman, Plain Black
01:43 <****> it's often very useful.
01:43 <****> Prison Break
01:43 <****> it's -requiring- it that's retarded
01:43 <****> Autoinc.
01:43 <****> so no composite primary keys?!
01:43 <****> I have nothing against auto-increment
01:43 <****> jagerman, I mean the MySQL keyword.
01:43 <****> mst: even for DBs that support placeholders?
01:43 <****> apeiron: Oh
01:44 <****> mauke, what does that mean?
01:44 <****> Khisanth: yep
01:44 <****> I think DBD::mysql just does interpolation as well ... at least by default
01:44 <****> Khisanth: no non-autoinc primary keys. at all.
01:44 <****> Khisanth: no, DBD::mysql is smarter than that
01:44 <****> jagerman, I was at the conference last week, the issue came up, they mentioned that they prefer to avoid it 'cause of some other reason but also 'cause it's MySQL-specific.
01:44 <****> Khisanth: DBD::mysqlPP does interpolation, but that's a different matter
01:44 <****> Zoffix: "does not follow". what do pragmas have to do with your error message?
01:44 <****> I consider auto-inc equiv. to SERIAL
01:44 <****> if it doesn't matter which, it doesn't matter which :)
01:44 <****> apeiron: Well, it sort of is, but there are equivelants in just about everything else.
01:44 <****> It is, in a few ways.
01:45 <****> jagerman, True.
01:45 <****> mauke, to perhaps at least give me the sub name of where the redefinition occurs, or some more info. Because the line number is obviously wrong.
01:45 <****> apeiron: In postgres, for instance, using a serial/bigserial instead of int/bigint, or creating the sequence yourself.
01:45 <****> er.
01:45 <****> DBIC handles 'is_auto_increment' via SQLT for pretty much every db
01:45 <****> jagerman: right. DBIC Just Handles that sort of thing.
01:46 <****> mst, Indeed. Am very pleased with that part of DBIC. :)
01:46 <****> Unfortunately, though, you sometimes get stuck with the lowest-common-denominator approach.
01:46 <****> Zoffix: bytes, bigint, integer, charnames, constant, lib, overload, re, utf8 ... are all pragmas
01:46 <****> Zoffix: I don't see how that helps you
01:46 <****> ooh
01:46 <****> For instance, we had to support mysql 3.23, so subselects were out of the question.
01:47 <****> ... wows
01:47 <****> That's just not cool.
01:47 <****> Does anyone know why the if block isn't entered when $text is "enable" but the other 2 if statements aren't met. It works for each of the other 2 if statements, just not for the last one .. And I'm sure $text is 'enable'
01:47 <****> if (($text =~ /^$info{'me'},? (.*)$/) || ($enabled{$target} == 1 && $text =~ /^(?:$info{'me'},? )?(.*)$/) || ($text eq 'enable')) {
01:47 <****> jagerman: they STILL are!
01:47 * mst wanders off to watch TV
01:48 <****> Timmy: $text isn't 'enable'
01:48 <****> Khisanth: What still are what?
01:48 <****> I have this text file i want to parse http://pastebin.com/m629081b8 , I want the lines from FCST ASH CLOUD +6H till till just before FCST ASH CLOUD +12H, how would i extract these lines?
01:48 <****> Foxhacker: The paste m629081b8 has been copied to http://erxz.com/pb/4887
01:48 <****> It is, I checked it just before the if block
01:48 <****> Khisanth: Doesn't everything since mysql 4.1 support subselects?
01:48 <****> jagerman: subselects
01:48 <****> Timmy: the if block also checks it
01:49 <****> jagerman: sure but the performance sucks
01:49 <****> Yes, but if "if statement 1 and 2" are false, it won't even enter the if block, even when $text equals enable
01:49 <****> Khisanth: Ah, but see, at least in that case you can say "Not our fault, switch to a better database"
01:49 <****> Well, here is my 263 lines of code: ( http://sial.org/pbot/28160 ) and here is the warnings message ( http://sial.org/pbot/28161 ) the code doesn't seem to load any factoids and I just can't find where $factoid and $value are redefined :|
01:49 <****> Timmy: no, wrong
01:49 <****> ok guys
01:49 <****> Timmy: that's not what "||" does
01:50 <****> Well, I know it should, but for some odd reason, it isn't
01:50 <****> || = OR, right ?
01:50 <****> what does ref return if argument is not reference ?
01:50 <****> Khisanth: Also, isn't mysql's subselect performance improving?
01:50 <****> empty list or undef ?
01:50 <****> Timmy: yes, and I'm pretty sure it works right, you your assumptions are wrong
01:50 <****> s/you/so/
01:50 <****> in prof perl programming it's stated it return undef in perldoc it says empty string
01:50 <****> who's lying?
01:50 <****> jagerman: no idea
01:51 <****> eval: ref "doom"
01:51 <****> SubStack: ''
01:51 <****> TrueFX: perldoc is always right except when it's wrong. in which case you've found a bug.
01:51 <****> Khisanth: I turned my back on commercial software development when I left, though, so I never have to worry about mysql sucking again.
Total 35 pages. You are browsing page 3/35.
First :: Prev :: [1] [2] [3] [4] [5] [...] :: Next :: Last