Hmm, even testing temporarily with 10.9.1.0 does not seem to fix this for
me, I get the same results...
On Thursday, July 12, 2012 9:51:32 AM UTC-7, Matt Hauck wrote:
>
> Hmm, it appears to have no connection to the prepared statements plugin,
> but might be related to my version of Derby (10.8.2.2), which I
> unfortunately cannot update...
>
> ruby-1.6.5.1 :016 > DB.create_table(:a) {Time :a}
>> [09:46:46] DEBUG [MDM.Dev] - (0.025000s) CREATE TABLE "A" ("A" timestamp)
>> => nil
>> jruby-1.6.5.1 :017 > t = Time.now
>> => 2012-07-12 09:46:59 -0700
>> jruby-1.6.5.1 :018 > t.to_f
>> => 1342111619.922
>> jruby-1.6.5.1 :019 > DB[:a].call(:insert, {:t => t}, :$t)
>> [09:47:39] DEBUG [MDM.Dev] - (0.005000s) PREPARE INSERT INTO "A" VALUES
>> (?)
>> [09:47:39] DEBUG [MDM.Dev] - (0.003000s) EXECUTE; [2012-07-12 09:46:59
>> -0700]
>> [09:47:39] DEBUG [MDM.Dev] - (0.001000s) SELECT IDENTITY_VAL_LOCAL() FROM
>> sysibm.sysdummy1
>> => 30
>> jruby-1.6.5.1 :020 > DB[:a].get(:a).to_f
>> [09:48:10] DEBUG [MDM.Dev] - (0.003000s) SELECT "A" FROM "A" FETCH FIRST
>> 1 ROWS ONLY
>> => 1342111619.0
>
>
>
> On Thursday, July 12, 2012 9:26:32 AM UTC-7, Jeremy Evans wrote:
>>
>> On Thursday, July 12, 2012 8:47:08 AM UTC-7, Jeremy Evans wrote:
>>>
>>> On Thursday, July 12, 2012 4:01:43 AM UTC-7, Matt Hauck wrote:
>>>>
>>>> After turning on the prepared statements plugin, I have noticed that my
>>>> timestamps are being truncated of their milliseconds when they are saved
>>>> to
>>>> the database (derby), resulting in a loss of accuracy of data that is
>>>> rather unfortunate. Is this a known issue? Is there a workaround? Perhaps
>>>> to disable prepared statements on these columns?
>>>
>>>
>>> This is going to be an adapter specific issue. The postgres adapter has
>>> specific code to keep the fractional seconds, but not all adapters do.
>>>
>>> Since you are using derby, this is going to be a jdbc adapter issue.
>>> The jdbc adapter has code that is supposed to handle the fractional parts
>>> (see set_ps_arg, java_sql_datetime, java_sql_timestamp in
>>> lib/sequel/adapters/jdbc.rb). Please post the code you are using and an
>>> SQL log.
>>>
>>> Disabling prepared statements may work around the issue.
>>>
>>> Thanks,
>>> Jeremy
>>>
>>
>> I can't replicate this issue:
>>
>> L:\sequel>jruby -rderby.jar -I lib bin\sequel -E
>> "jdbc:derby:memory:myDb;create=true"
>> Your database is stored in DB...
>> irb(main):001:0> DB.create_table(:a){Time :a}
>> I, [2012-07-12T09:16:25.661000 #3096] INFO -- : (0.188000s) CREATE TABLE
>> "A" ("A" timestamp)
>> => nil
>> irb(main):002:0> t = Time.now
>> => Thu Jul 12 09:16:25 -0700 2012
>> irb(main):003:0> DB[:a].call(:insert, {:t=>t}, :$t)
>> I, [2012-07-12T09:16:26.005000 #3096] INFO -- : (0.188000s) PREPARE
>> INSERT INTO "A" VALUES (?)
>> I, [2012-07-12T09:16:26.036000 #3096] INFO -- : (0.016000s) EXECUTE;
>> [Thu Jul 12 09:16:25 -0700 2012]
>> I, [2012-07-12T09:16:26.083000 #3096] INFO -- : (0.047000s) SELECT
>> IDENTITY_VAL_LOCAL() FROM sysibm.sysdummy1
>> => 0
>> irb(main):004:0> DB[:a].get(:a).to_f
>> I, [2012-07-12T09:16:27.895000 #3096] INFO -- : (0.015000s) SELECT "A"
>> FROM "A" FETCH FIRST 1 ROWS ONLY
>> => 1342109785.676
>> irb(main):005:0> class A < Sequel::Model(:a)
>> irb(main):006:1> plugin :prepared_statements
>> irb(main):007:1> end
>> => nil
>> irb(main):008:0> A.delete
>> I, [2012-07-12T09:19:01.103000 #3096] INFO -- : (0.015000s) DELETE FROM
>> "A"
>> => 1
>> irb(main):009:0> A.create(:a=>t).a.to_f
>> I, [2012-07-12T09:19:25.151000 #3096] INFO -- : (0.000000s)
>> Transaction.begin
>> I, [2012-07-12T09:19:25.151000 #3096] INFO -- : (0.000000s) PREPARE
>> smpsp_1: INSERT INTO "A" ("A") VALUES (?)
>> I, [2012-07-12T09:19:25.151000 #3096] INFO -- : (0.000000s) EXECUTE
>> smpsp_1 (INSERT INTO "A" ("A") VALUES (?)); [Thu Jul 12 09:16:25 -0700 2012]
>> I, [2012-07-12T09:19:25.151000 #3096] INFO -- : (0.000000s) SELECT
>> IDENTITY_VAL_LOCAL() FROM sysibm.sysdummy1
>> I, [2012-07-12T09:19:25.167000 #3096] INFO -- : (0.000000s)
>> Transaction.commit
>> => 1342109785.676
>>
>> Note in both the regular insert with bound variable and the model
>> prepared_statements plugin usage in the model, there are fractional seconds
>> on the retrieved value.
>>
>> Jeremy
>>
>
--
You received this message because you are subscribed to the Google Groups
"sequel-talk" group.
To view this discussion on the web visit
https://groups.google.com/d/msg/sequel-talk/-/GTSKhkjpMnIJ.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sequel-talk?hl=en.
