Brett Harvey schrieb:
which method is better to do.
I have 5 tables. They represent sections/parts of a companies
standards. There are 13 main categories, each of those categories has
subsections (some with 3, some with 10 or more), those subsections have
subsections, etc.
Which table des
Shawn,
I agree with you that the tables can have different info with regard to
the requirements.
But for storing only addresses for specific students, this 4 table design
seems weirdish to me... I think it makes more sense to keep a
student_id in the Addresses table...
With regards,
Martijn Ton
"Martijn Tonies" <[EMAIL PROTECTED]> wrote on 05/24/2005 02:32:05 PM:
>
>
>
> > Something like this would make more sense to me and provide greater
> flexibility;
>
> It doesn't to me...
>
> > student
> >
> > student_id
> > name
> > age
> >
> > address
> > ---
> >
You probably want to add type to both the address and phone tables. Then you
can be selective in your reporting and still get 1 row per student in your
result set. Just remember if your data has the possibility of not having the
information for a student you want to use LEFT JOIN's vs INNER JOIN's
> Something like this would make more sense to me and provide greater
flexibility;
It doesn't to me...
> student
>
> student_id
> name
> age
>
> address
> ---
> address_id
> street_name
> city
> state
> zip
What addresses are these? Random addresses where a studen
From: Mike Johnson [mailto:[EMAIL PROTECTED]
> From: Koon Yue Lam [mailto:[EMAIL PROTECTED]
>
> > the problems is, when I want to query both student, address
> > and phone num, the sql will be
> >
> > select * from student s, address a, phone_num n
> > where s.student_id = a.sudent_id
> > and
From: Koon Yue Lam [mailto:[EMAIL PROTECTED]
> Hi, here is the case:
>
> one student may have more than one address, and one student
> may have more than one phone number
>
> so the db would be:
>
> student
>
> student_id
> name
> age
>
> address
> ---
> address_
Koon Yue Lam,
If you running your MySQL on Windows, you may try to use one of the
reporting tools, like Crystal Report, to create your reports.
Generally these tools allow to hide repetitive data in its reports
Mikhail Berman
-Original Message-
From: Koon Yue Lam [mailto:[EMAIL PROTECTE
Something like this would make more sense to me and provide greater flexibility;
student
student_id
name
age
address
---
address_id
street_name
city
state
zip
phone_num
--
phone_num_id
num
extension
type (cell, home, etc)
primaryNumber (yes/no)
stud
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Martijn, et al --
...and then Martijn Tonies said...
%
% > Since I want the ability to mirror, it seems that I'll probably want one
...
% > really don't want to keep the files in the DB itself). I'm open to ideas
% > of why I wouldn't, though.
%
%
> Since I want the ability to mirror, it seems that I'll probably want one
> single DB replicated across my hosts so that comments and so on stay up
> to date (I still haven't crossed the bridge of how to keep the library
> itself in sync thru something like unison or rsync, but I do know that I
>
Which is better? I guess that all depends on your entity relationships.
If I assume that each PERSON would have only 1 USER record and each USER
would match to only 1 PERSON record. That is a "1-to-1 relationship" or
(1..1) and is typically stored with all of the necessary column in just one
tabl
That worked like a charm, thanks so much! I don't know why I didn't try
that before!
Julian
At 02:46 PM 11/21/2003 -0600, Paul DuBois wrote:
At 10:56 -0500 11/21/03, Julian Zottl wrote:
Andy,
Thanks for responding. I think that I am going to go with the idea of
creating a tale for each day. M
ql List
Subject: RE: DB design question - shell scripting...
At 10:56 -0500 11/21/03, Julian Zottl wrote:
Andy,
Thanks for responding. I think that I am going to go with the idea
of creating a tale for each day. My thoughts were to write a shell
script to do this for me, but I am running into a pr
Wouldn't this also work?:
mysql -u root -p -e "CREATE TABLE t$date(...)" yourdatabase
-Original Message-
From: Paul DuBois [mailto:[EMAIL PROTECTED]
Sent: Friday, November 21, 2003 12:46 PM
To: Julian Zottl; Andy Eastham; Mysql List
Subject: RE: DB design question -
At 10:56 -0500 11/21/03, Julian Zottl wrote:
Andy,
Thanks for responding. I think that I am going to go with the idea
of creating a tale for each day. My thoughts were to write a shell
script to do this for me, but I am running into a problem: I wrote
the following:
#!/bin/sh
date=`date "+%m%
[EMAIL PROTECTED]>
To: "Andy Eastham" <[EMAIL PROTECTED]>; "Mysql List"
<[EMAIL PROTECTED]>
Sent: Friday, November 21, 2003 7:56 AM
Subject: RE: DB design question - shell scripting...
> Andy,
> Thanks for responding. I think that I am going to go with the ide
Andy,
Thanks for responding. I think that I am going to go with the idea of
creating a tale for each day. My thoughts were to write a shell script to
do this for me, but I am running into a problem: I wrote the following:
#!/bin/sh
date=`date "+%m%d%Y"`
export date
mysql -u root -p < createdb.
Julian,
Your design is sound in my opinion. An area you probably need to consider
is when you need to search across a day boundary.
You will need to make the application aware that it needs to search across a
day boundary, so that it searches two tables with a union where necessary.
It will also
Give this more thought. I think you have more options that the two you
proposed.
With really large tables, you can collect data in them for a fixed time
period (monthly) then run a batch that removes the data for the time period
after moving it to an archive table.
Try making a staging table that
Whats the goal with the Data? If it is graphing it then go with MRTG
with RRDtool, which will keep about 550 days of SNMP data and produce
graphs
displaying a daily, weekly, monthly and yearly timeframe... Could do
1500 devices with probably less than 10 Gigs of space...
You could then load the
Mahesh,
The best advice, from what I've heard around the list is to base your decision on this
based on your filesystem. Some filesystems handle large # of files well (1 db, many
tables) some don't. Some handle many directories well (many db's 1 tbl each), some
don't (ala your example).
P
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On Sunday 20 July 2003 09:03, Andreas wrote:
> Hi folks,
>
> how would you design a 1:1 relation ?
>
>
> I'd like to split an entities's attributes because they won't get
> equally frequently requested. So I can save memory and disk access time.
> The
r Brawley
> [mailto:[EMAIL PROTECTED]
> >Sent: 19 July 2003 17:50
> >To: MySQL-Lista
> >Subject: Re: DB Design
> >
> >
> >You probably don't mean the $Nk tools
> (PowerDesigner, ERWin, Rational Rose
> >&c). MS Visio does it, but Dezign
>
this is exactly what I am looking for :) Is there a free one of these guys
hanging around somewhere?
Andrew
>-Original Message-
>From: Peter Brawley [mailto:[EMAIL PROTECTED]
>Sent: 19 July 2003 17:50
>To: MySQL-Lista
>Subject: Re: DB Design
>
>
>You probably
You probably don't mean the $Nk tools (PowerDesigner, ERWin, Rational Rose
&c). MS Visio does it, but Dezign (http://www.datanamic.com/dezign/) does it
better & cheaper IMO (no I'm not a principal). PB
-
- Original Message -
From: Andrew
To: MySQL-Lista
Sent: Saturday, July 19,
inetly agree specially if you consider that the number of
interests in table may increase at anytime.
Siomara
>
>Hope that wasn't too long winded. :)
>
>Roger
>
>- Original Message -
>From: "Siomara Pantarotto" <[EMAIL PROTECTED]>
>To: <[EMAI
asn't too long winded. :)
Roger
- Original Message -
From: "Siomara Pantarotto" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>
Sent: Friday, July 06, 2001 5:07 PM
Subject: Re: db design questions
> Hi,
>
> Try to keep the simple
sorry I typed my website wrong. The right url is:
www.geocities.com/hisiomara
>From: "Siomara Pantarotto" <[EMAIL PROTECTED]>
>To: [EMAIL PROTECTED], [EMAIL PROTECTED]
>Subject: Re: db design questions
>Date: Fri, 06 Jul 2001 18:07:36 -0300
>
>Hi,
>
>
Hi,
Try to keep the simple attributes in one table and the repetitions in
separated tables>
For example:
A person has only one gender, one name, one marital stauts , etc
however a person can have none,one or more than one kid. Create another
table and create a relationship between both.
30 matches
Mail list logo
