Re: [GENERAL] histogram

2011-05-01 Thread Thomas Markus
hi, group by 1 means group by first output column order by 2 means order by second output column ascending Am 30.04.2011 19:00, schrieb Joel Reymont: What is the meaning of group by 1 order by 2 e.g. what to the numbers 1 and 2 stand for? What would change if I do the follo

Re: [GENERAL] histogram

2011-04-30 Thread Rob Sargent
David Johnston wrote: Given that you are actively implementing the code that uses the 1 and 2 I don't see how it is that egregious. When generating calculated fields it is cleaner than the alternative: Select trunc(distance * 10.)/10., count(*) From doc_ads Group by (trunc(distance * 10.)) Or

Re: [GENERAL] histogram

2011-04-30 Thread David Johnston
Given that you are actively implementing the code that uses the 1 and 2 I don't see how it is that egregious. When generating calculated fields it is cleaner than the alternative: Select trunc(distance * 10.)/10., count(*) >From doc_ads Group by (trunc(distance * 10.)) Order by (trunc(distance *

Re: [GENERAL] histogram

2011-04-30 Thread Rob Sargent
re: 1 and 2. They're horrible (imho) reference to the attributes of the returned tuple. Or at best an exposure of the implementation. :) Order by "2" if you want the most frequent (highest counts) of your distances at the bottom of the output (or ordery by 2 desc) if you want them at the top o

Re: [GENERAL] histogram

2011-04-30 Thread Joel Reymont
I think this should do what I want select trunc(distance * 10.)/10., count(*) from doc_ads group by 1 order by 1 Thanks, Joel -- - for hire: mac osx device driver ninja, kernel extensions and usb d

Re: [GENERAL] histogram

2011-04-30 Thread Joel Reymont
What is the meaning of group by 1 order by 2 e.g. what to the numbers 1 and 2 stand for? What would change if I do the following? group by 1 order by 1 On Apr 30, 2011, at 5:48 PM, Thomas Markus wrote: > Hi, > > try something like this: > > select >trunc(random(

Re: [GENERAL] histogram

2011-04-30 Thread Joel Reymont
Thank you Thomas! Is there a way for the code below to determine the number of rows in the table and use it? Thanks, Joel On Apr 30, 2011, at 5:48 PM, Thomas Markus wrote: > Hi, > > try something like this: > > select >trunc(random() * 10.)/10. >, count(*) > from >generat

Re: [GENERAL] histogram

2011-04-30 Thread Thomas Markus
Hi, try something like this: select trunc(random() * 10.)/10. , count(*) from generate_series(1,200) group by 1 order by 2 regards Thomas Am 30.04.2011 18:37, schrieb Joel Reymont: I have a column of 2 million float values from 0 to 1. I would like to figure out how many valu

[GENERAL] histogram

2011-04-30 Thread Joel Reymont
I have a column of 2 million float values from 0 to 1. I would like to figure out how many values fit into buckets spaced by 0.10, e.g. from 0 to 0.10, from 0.10 to 0.20, etc. What is the best way to do this? Thanks, Joel

Re: [GENERAL] Histogram generator

2010-07-28 Thread Sam Mason
On Tue, Jul 27, 2010 at 09:25:05PM -0400, Patrick May wrote: > On Jul 27, 2010, at 9:21 PM, Steve Atkins wrote: > > select date_trunc('hour', foo) + interval '30 minutes' * > > floor(extract(minute from foo) / 30) as start, event, count(*) from bar > > group by 1, 2 order by 1 asc; > > Tha

Re: [GENERAL] Histogram generator

2010-07-27 Thread Patrick May
On Jul 27, 2010, at 9:21 PM, Steve Atkins wrote: > On Jul 27, 2010, at 6:07 PM, Patrick May wrote: >> On Jul 27, 2010, at 7:12 PM, Dann Corbit wrote: I have a table containing events with timestamps. I would like to generate a histogram of the number of each type of event for each >>>

Re: [GENERAL] Histogram generator

2010-07-27 Thread Steve Atkins
On Jul 27, 2010, at 6:07 PM, Patrick May wrote: > On Jul 27, 2010, at 7:12 PM, Dann Corbit wrote: >>> I have a table containing events with timestamps. I would like >>> to generate a histogram of the number of each type of event for each >>> half-hour period from 8:00 am to 6:00 pm. Are the

Re: [GENERAL] Histogram generator

2010-07-27 Thread Patrick May
On Jul 27, 2010, at 7:12 PM, Dann Corbit wrote: >> I have a table containing events with timestamps. I would like >> to generate a histogram of the number of each type of event for each >> half-hour period from 8:00 am to 6:00 pm. Are there any tools that >> will help me do this? > > GROUP

Re: [GENERAL] Histogram generator

2010-07-27 Thread Dann Corbit
> -Original Message- > From: pgsql-general-ow...@postgresql.org [mailto:pgsql-general- > ow...@postgresql.org] On Behalf Of Patrick May > Sent: Tuesday, July 27, 2010 3:59 PM > To: pgsql-general@postgresql.org > Subject: [GENERAL] Histogram generator > > Hi, &

[GENERAL] Histogram generator

2010-07-27 Thread Patrick May
Hi, I have a table containing events with timestamps. I would like to generate a histogram of the number of each type of event for each half-hour period from 8:00 am to 6:00 pm. Are there any tools that will help me do this? Thanks, Patrick smime.p7s Description: S/MIME cryptograp