| 00:00.21 | *** join/#uphpu dataw0lf (n=dataw0lf@70.103.239.58) |
| 00:00.23 | *** join/#uphpu harm_ (n=harm@70.103.239.58) |
| 00:00.28 | *** join/#uphpu fungus (n=fungus@2001:470:1f00:645:20a:95ff:fec5:636a) |
| 00:47.35 | *** part/#uphpu fungus (n=fungus@2001:470:1f00:645:20a:95ff:fec5:636a) |
| 01:11.37 | *** join/#uphpu synic (n=synic@pdpc/supporter/student/synic) |
| 02:08.44 | *** join/#uphpu usynic (n=synic@pdpc/supporter/student/synic) |
| 04:55.12 | *** join/#uphpu bigdog_ut (n=bigdog_u@207.135.132.13) |
| 04:55.12 | *** mode/#uphpu [+o bigdog_ut] by ChanServ |
| 05:04.38 | bigdog_ut | evenin all |
| 05:15.41 | stderr | evening |
| 05:17.01 | bigdog_ut | whatz up bro? |
| 05:21.26 | RyanE | evening. |
| 05:30.50 | bigdog_ut | man it is so quiet in here now |
| 05:35.27 | *** join/#uphpu RyanE_ (n=ryan@rberick.dsl.xmission.com) |
| 06:05.26 | stderr | bigdog_ut: yea. |
| 06:05.49 | bigdog_ut | not much action..i guess everyone is a php guru now |
| 06:43.28 | diskotec|w | I isn't no guru....I'm not an english teacher either. =P |
| 06:43.46 | diskotec|w | bigdog: you still around? |
| 06:51.23 | bigdog_ut | y |
| 06:52.04 | diskotec|w | I have a question if you have time |
| 06:52.10 | bigdog_ut | shoot |
| 06:53.07 | diskotec|w | Its pretty much the same question from last night. I don't think I thoroughly understood what DISTINCT meant |
| 06:53.31 | bigdog_ut | distinct only give you one of n+1 results |
| 06:53.53 | bigdog_ut | so if you have a result that comes back that is duplicate...distinct only give you one |
| 06:54.00 | diskotec|w | I'm getting multiple listings of a med rec # when I only want records that have only one med rec number per patient. |
| 06:54.34 | bigdog_ut | then the join is messed up? |
| 06:54.43 | diskotec|w | I said that last part wrong... |
| 06:54.45 | diskotec|w | heh |
| 06:55.04 | diskotec|w | well, I'm not sure a join would fix this problem |
| 06:55.40 | diskotec|w | is there a way to compare one row to another and delete (remove) it from the results |
| 06:55.41 | bigdog_ut | explain again what you are trying to do |
| 06:55.49 | diskotec|w | ...if it's a dupe? |
| 06:58.00 | diskotec|w | ok... report that looks at all hospital visits within a given time range. I want it to return only unique patient names no matter how many visits they've had so I feel using the med rec # field is the best candidate to use |
| 07:00.37 | bigdog_ut | k |
| 07:00.56 | bigdog_ut | so then distinct on the med rec # should do it |
| 07:01.10 | bigdog_ut | but what constitues a unique patient name? |
| 07:02.22 | diskotec|w | We have so many hispanic names that are alike and if we wanted to identify a single person, we'd have to use a combination of last name, date of birth or soc sec # |
| 07:02.57 | diskotec|w | in our reporting we normally try to identify unique patient names with a medical record number |
| 07:03.48 | bigdog_ut | cool then distinct on med rec # |
| 07:06.19 | diskotec|w | doesn't the whole record have to be the same in order for it to not include another visit but with the same med rec number? |
| 07:06.37 | diskotec|w | here is the beginning of the script... |
| 07:07.41 | diskotec|w | select DISTINCT tpm300.med_rec_no, tpm300.vst_ext_id, tsm040.lst_nm, tsm040.fst_nm, tsm040.bth_ts, tpm700.plan_ds |
| 07:08.35 | diskotec|w | when I run the script with that beginning I'm getting as many dupe med rec listings as that patient had in a given date range |
| 07:09.00 | diskotec|w | as that patient had visits in a given date range |
| 07:09.44 | diskotec|w | when I use... |
| 07:09.47 | diskotec|w | <PROTECTED> |
| 07:09.49 | bigdog_ut | select DISTINCT tpm300.med_rec_no |
| 07:10.11 | diskotec|w | it works fine |
| 07:10.33 | diskotec|w | do you think it's a join issue? |
| 07:10.45 | bigdog_ut | nope |
| 07:10.58 | bigdog_ut | when you add all the other fields then it looks for the same with all of the fields |
| 07:11.22 | bigdog_ut | so those 6 fields need to be all the same before you have a dup |
| 07:11.36 | diskotec|w | ehh... when I add the rest of the fields, it's like the DISTINCT isn't even there |
| 07:11.44 | bigdog_ut | yep |
| 07:12.04 | bigdog_ut | so you want to have very limited fields...i suggest only using the primary key if you can |
| 07:12.21 | bigdog_ut | that will help eliminate dups |
| 07:13.21 | diskotec|w | hmm... I'll look into that. |
| 07:15.45 | diskotec|w | is there any quick and dirty shortcut like I mentioned earlier? comparing numbers as it goes down the list and removing either the one below or the one above if it is a dupe? |
| 07:16.50 | bigdog_ut | you could if you have a script or something |
| 07:17.35 | bigdog_ut | but if the med rec # is unique then you have your test there |
| 07:17.38 | diskotec|w | ok...that sounds like more work. thought there might have been some sort of function that took care of something like that. |
| 07:17.50 | bigdog_ut | distinct |
| 07:23.09 | diskotec|w | well...what I meant was asking about a function that after sorting by med_rec_no started at the med_rec_no field in record 1 and compared itself to the med_rec_no field in record 2 and removed record 2 IF those two numbers matched. |
| 07:23.42 | bigdog_ut | distinct would do that for you |
| 07:24.01 | bigdog_ut | it would use a cursor to move through and create the result set for you doing that |
| 07:24.22 | bigdog_ut | you could do it too with stored procedures and cursors |
| 07:24.46 | diskotec|w | not there yet on knowing how to work with stored procs or cursors |
| 07:25.49 | diskotec|w | I'll see what I can do about making use of a primary key |
| 07:26.32 | bigdog_ut | what is your primary key on the table(s)? |
| 07:33.39 | diskotec|w | It looks like this table has multiple keys. med_rec_no has multiple indexes. I can't quite tell which is the primary from the documentation I have. |
| 07:33.54 | bigdog_ut | k |
| 07:34.14 | bigdog_ut | i would just suggest using the primary(s) as part of the distinct fields |
| 07:34.19 | bigdog_ut | that will help out good |
| 07:35.38 | diskotec|w | thanks...I think I have an idea on how to tackle this now...as soon as I can get some clarification on the primary key(s). |
| 07:35.53 | bigdog_ut | good luck on that |
| 07:36.01 | bigdog_ut | which db are you using? |
| 07:36.01 | diskotec|w | Thanks for the pointer. |
| 07:36.09 | diskotec|w | sql server 2k |
| 07:36.14 | bigdog_ut | k |
| 07:36.18 | diskotec|w | made by www.mckesson.com |
| 07:36.36 | diskotec|w | or at least the apps are |
| 07:36.57 | bigdog_ut | y |
| 07:39.04 | diskotec|w | y? |
| 07:39.13 | bigdog_ut | yep |
| 07:39.23 | diskotec|w | not sure what you mean? |
| 07:39.36 | bigdog_ut | just saying yep i have heard of them |
| 07:39.43 | diskotec|w | oh ok |
| 07:41.36 | diskotec|w | sorry....I just hadn't seen that before (y=yep) and it's probably widely used. heh. |
| 07:41.39 | diskotec|w | well.... goodnight |
| 07:41.49 | bigdog_ut | night |
| 15:23.51 | *** join/#uphpu wps (n=wps@209.63.57.116) |
| 15:23.51 | *** mode/#uphpu [+o wps] by ChanServ |
| 15:25.03 | *** join/#uphpu synic (n=synic@pdpc/supporter/student/synic) |
| 16:17.17 | *** join/#uphpu _psychic_ (n=john@71.32.228.156) |
| 17:00.12 | *** join/#uphpu Alphahawk (n=Alphahaw@70.56.105.158) |
| 17:16.42 | *** join/#uphpu _psychic_ (n=john@71.32.228.156) |
| 17:19.01 | *** join/#uphpu Utah_Dave (n=boucha@0-1pool139-198.nas28.salt-lake-city1.ut.us.da.qwest.net) |
| 17:20.18 | *** join/#uphpu fungus (n=fungus@2001:470:1f00:645:20a:95ff:fec5:636a) |
| 17:41.55 | *** join/#uphpu xilch|w (n=duh@66.239.17.226.ptr.us.xo.net) |
| 19:52.06 | *** part/#uphpu _psychic_ (n=john@71.32.228.156) |
| 20:22.47 | *** join/#uphpu _psychic_ (n=john@c-24-2-78-210.hsd1.ut.comcast.net) |
| 20:39.25 | *** join/#uphpu _psychic_ (n=john@c-24-2-78-210.hsd1.ut.comcast.net) |
| 23:27.18 | *** join/#uphpu xilch|w (n=duh@66.239.17.226.ptr.us.xo.net) |
| 23:32.53 | diskotec|w | anyone around and have time to answer a sql question? I've asked the question in #sql but nobody is responding. |
| 23:41.54 | macnewbold | diskotec|w: shoot |
| 23:43.54 | diskotec|w | I'm trying to convert datetime to smalldatetime with the style of 1 (http://msdn.microsoft.com/library/default.asp?url=/library/en-us/tsqlref/ts_ca-co_2f3o.asp). According to the website it should be mm/dd/yy but the results that keep coming back are like "2004-01-05 13:40:00" Do I need to do anything else to convert or cast? I just want the date, not the time. |
| 23:48.50 | diskotec|w | this is the part I have that isn't working... days = convert(smalldatetime, tpm300.dschrg_ts, 1) |
| 23:52.23 | macnewbold | smalldatetime? never heard of it. |
| 23:52.35 | macnewbold | are you trying to change the field type in the table, or just query the date only? |
| 23:52.36 | diskotec|w | mssql thing I guess |
| 23:52.44 | macnewbold | oh, MSSQL. duh. okay. |
| 23:52.56 | macnewbold | in mysql, there's a FORMAT() function that lets you format it how you want |
| 23:53.02 | macnewbold | so you could pull results back with just the date |
| 23:53.07 | diskotec|w | yeah...just trying to change it for the query....not trying to change the data type in the table. |
| 23:53.20 | macnewbold | look for something like FORMAT() then |
| 23:53.32 | macnewbold | give it a format like "%m/%d/%Y" or whatever, |
| 23:53.38 | macnewbold | and you won't get the time in your result |
| 23:54.07 | diskotec|w | Ok...I'll check that out thanks |