 |
hogarius
Adept
Joined: 29 Jan 2003
Posts: 221
Location: islands.genesismuds.org
|
 Posted: Sun Jun 13, 2010 3:09 pm
[3.19c] Database trigger no longer works
|
I play Simutronics' Modus Operandi. The following trigger no longer creates database records for my perpetrator tracking system. I am currently running CMUD version 3.19c. I believe this worked in 3.16, but I'm honestly not sure when it stopped working.
| Code: |
<trigger name="BoardInfo" priority="940" id="3">
<pattern><![CDATA[Police are asking for your help to bring &{Perp.FirstName} &{Perp.LastName} to justice *. &{Perp.Pronoun} appears to be in %w &%w{Perp.AgeRange} &%w{Perp.AgeDecade}%p standing about &%d{Perp.HFt} feet and &%d{Perp.HIn} inch* tall with %w &{Perp.Build} build. %w has %w &{Perp.Fitness} fitness level, with a weight of about &%d{Perp.Weight} pounds. %w skin is &{Perp.SkinTone}. %w eyes are &{Perp.EyeColor}, and %w has &{Perp.Hair} hair.]]></pattern>
<value>#if (@Perp.Pronoun="He") {#var Perp.Gender "M"} {#var Perp.Gender "F"}
#switch @Perp.AgeDecade
"teens" {#var Perp.AgeNum "10's"}
"twenties" {#var Perp.AgeNum "20's"}
"thirties" {#var Perp.AgeNum "30's"}
"forties" {#var Perp.AgeNum "40's"}
"fifties" {#var Perp.AgeNum "50's"}
"sixties" {#var Perp.AgeNum "60's"}
"seventies" {#var Perp.AgeNum "70's"}
"eighties" {#var Perp.AgeNum "80's"}
"nineties" {#var Perp.AgeNum "90's"}</value>
<trigger>
<pattern>^* was recently seen &{Perp.Location}.$</pattern>
</trigger>
<trigger>
<pattern>^%w is to be considered &{Perp.ThreatLevel}%p$</pattern>
<value><![CDATA[#var Duplicate (%query("&FirstName=@Perp.FirstName && &LastName=@Perp.LastName"))
#if @Duplicate
{#dbput @Duplicate {Location=@Perp.Location}}
#if !@Duplicate
{#new Perps @Perp}
#wait 250]]></value>
</trigger>
</trigger>
|
I'd appreciate any insights anyone can give.
Also, would it be worthwhile to use the new SQL commands to recreate this database? |
|
|
|
|
hogarius
Adept
Joined: 29 Jan 2003
Posts: 221
Location: islands.genesismuds.org
|
Posted: Sun Jun 13, 2010 3:43 pm
|
Hmmm. The Perp database variable contained old data from when I ran the trigger several weeks ago. I did a #UNVAR PERP and tried to read the signpost again to populate the database. Below is a transcript from the output screen
| Code: |
[Start Of List.]
Reading a shiny notice you see:
--==**** Wanted: Markus Buchanan ****==--
Police are asking for your help to bring Markus Buchanan to justice for the crime of perjury. He appears to be in his early seventies, standing about 5 feet and 3 inches tall with a muscular build. He has an average fitness level, with a weight of about 215 pounds. His skin is bronzed. His eyes are slate, and he has dirty blonde stylish hair.
He was last seen wearing a vivid tie-dye tank top, a pair of stylish linen pants, a pair of leather walking sandals, and a floppy black beret.
Markus Buchanan was recently seen near the Foot Bridge in French Quarter.
He is to be considered not at all dangerous.
Reading a huge paper you see:
--==**** Wanted: Karina Montesa ****==--
Police are asking for your help to bring Karina Montesa to justice for the crime of carjacking. She appears to be in her mid sixties, standing about 6 feet and 4 inches tall with a trim build. She has a low fitness level, with a weight of about 180 pounds. Her skin is freckled. Her eyes are blue-gray, and she has bleach blonde braided hair.
She was last seen wearing a green and white long-sleeved baseball shirt, a pair of red jean overalls, a pair of hiking boots, and a fashionable floral-print silk scarf.
Karina Montesa was recently seen near the Avenue du Graces in French Quarter.
She is to be considered not at all dangerous.
Done!
|
After running the #UNVAR Perp command, the trigger now loads the data for the first perp (Markus Buchanan) into the database, but doesn't trigger on the following names on the signpost.
I can add a #UNVAR Perp command at the end of my script, but I'm puzzled why I would have to do that. |
|
|
|
|
Zugg
MASTER
Joined: 25 Sep 2000
Posts: 23379
Location: Colorado, USA
|
Posted: Mon Jun 14, 2010 5:04 pm
|
I was able to reproduce this and you are correct that doing #unvar perp seems to fix it. No, you shouldn't need to do that. It's probably related to the new tables/list so I'll look into this and let you know what I find.
|
|
|
|
|
hogarius
Adept
Joined: 29 Jan 2003
Posts: 221
Location: islands.genesismuds.org
|
Posted: Mon Jun 14, 2010 5:17 pm
|
Thank you, Zugg.
|
|
|
|
|
Zugg
MASTER
Joined: 25 Sep 2000
Posts: 23379
Location: Colorado, USA
|
Posted: Mon Jun 14, 2010 9:15 pm
|
This was a nasty one. Turns out the problem wasn't related to triggers at all. It was a low-level bug when replacing existing keys in existing database variables. It could cause some of the pointers in the table structure to get messed up in some cases, preventing the new value from being saved. So that's why the @Perp variable seemed "stuck" at the first set of values and it only worked again when deleting this variable.
So, this is another low-level bug that might be causing weird problems in other scripts. I'll get a quick patch out later today. |
|
|
|
|
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
