This may be slightly confusing so I'll try to be detailed. I need to query several months of daily records based on a certain variable. In this case a file name. There were several failed files that were rerun, and the rerun will include the original job run number in the new file name. For instance:
run 4145 failed. File name of "ORIGINALFILENAME"
run 4150 file name will look like "YYYYMMDDHHMMSS_4145_ORIGINALFILENAME"
There are 50+ failed files over the course of several months. Is there a way to build my where statement such that it will return both the failed runs (easily done via status_code = 'FAILURE') as well as the reruns? I imagine something like where file_name like '%_(select run_number where status_code = 'FAILURE')_ORIGINALFILENAME
The % character will work for any or no characters. So
where file_name like '%ORIGINALFILENAME'
will work for the failures and the reruns
This script return all the tuples which contain status_code = 'FAILURE' and where the file_name contains the sub-string "_ORIGINALFILENAME".
SELECT file_name FROM tblYourTable
WHERE status_code = 'FAILURE'
AND file_name LIKE '%[_ORIGINALFILENAME]%'
but... on a second look the requirement seems to be: "...statement such that it will return both the failed runs (easily done via status_code = 'FAILURE') as well as the reruns..." which if I understand correctly implies to adjust the predicate to return either scenario i.e.:
SELECT file_name FROM tblYourTable
WHERE status_code = 'FAILURE'
OR file_name LIKE '%[_ORIGINALFILENAME]%'
Related
I am trying to find a way to capture relevant errors from oracle alertlog. I have one table (ORA_BLACKLIST) with column values as below (these are the values which I want to ignore from
V$DIAG_ALERT_EXT)
Below are sample data in ORA_BLACKLIST table. This table can grow based on additional error to ignore from alertlog.
ORA-07445%[kkqctdrvJPPD
ORA-07445%[kxsPurgeCursor
ORA-01013%
ORA-27037%
ORA-01110
ORA-2154
V$DIAG_ALERT_EXT contains a MESSAGE_TEXT column which contains sample text like below.
ORA-01013: user requested cancel of current operation
ORA-07445: exception encountered: core dump [kxtogboh()+22] [SIGSEGV] [ADDR:0x87] [PC:0x12292A56]
ORA-07445: exception encountered: core dump [java_util_HashMap__get()] [SIGSEGV]
ORA-00600: internal error code arguments: [qercoRopRowsets:anumrows]
I want to write a query something like below to ignore the black listed errors and only capture relevant info like below.
select
dae.instance_id,
dae.container_name,
err_count,
dae.message_level
from
ORA_BLACKLIST ob,
V$DIAG_ALERT_EXT dae
where
group by .....;
Can someone suggest a way or sample code to achieve it?
I should have provided the exact contents of blacklist table. It currently contains some regex (perl) and I want to convert it to oracle like regex and compare with v$diag_alert_ext message_text column. Below are sample perl regex in my blacklist table.
ORA-0(,|$| )
ORA-48913
ORA-00060
ORA-609(,|$| )
ORA-65011
ORA-65020 ORA-31(,|$| )
ORA-7452 ORA-959(,|$| )
ORA-3136(,|)|$| )
ORA-07445.[kkqctdrvJPPD
ORA-07445.[kxsPurgeCursor –
Your blacklist table looks like like patterns, not regular expressions.
You can write a query like this:
select dae.* -- or whatever columns you want
from V$DIAG_ALERT_EXT dae
where not exists (select 1
from ORA_BLACKLIST ob
where dae.message_text like ob.<column name>
);
This will not have particularly good performance if the tables are large.
i have this query
insert into changes (id_registro)
select d2.id_registro
from daily2 d2
where exists (
select 1
from daily d1
where
d1.id_registro = d2.id_registro
and (d2.origen, d2.sector, d2.entidad_um, d2.sexo, d2.entidad_nac, d2.entidad_res,
d2.municipio_res, d2.tipo_paciente,d2.fecha_ingreso, d2.fecha_sintomas,
d2.fecha_def, d2.intubado, d2.neumonia, d2.edad, d2.nacionalidad, d2.embarazo,
d2.habla_lengua_indig, d2.diabetes, d2.epoc, d2.asma, d2.inmusupr, d2.hipertension,
d2.otra_com, d2.cardiovascular, d2.obesidad,
d2.renal_cronica, d2.tabaquismo, d2.otro_caso, d2.resultado, d2.migrante,
d2.pais_nacionalidad, d2.pais_origen, d2.uci )
<>
(d1.origen, d1.sector, d1.entidad_um, d1.sexo, d1.entidad_nac, d1.entidad_res,
d1.municipio_res, d1.tipo_paciente, d1.fecha_ingreso, d1.fecha_sintomas,
d1.fecha_def, d1.intubado, d1.neumonia, d1.edad, d1.nacionalidad, d1.embarazo,
d1.habla_lengua_indig, d1.diabetes, d1.epoc, d1.asma, d1.inmusupr, d1.hipertension,
d1.otra_com, d1.cardiovascular, d1.obesidad,
d1.renal_cronica, d1.tabaquismo, d1.otro_caso, d1.resultado, d1.migrante,
d1.pais_nacionalidad, d1.pais_origen, d1.uci ))
it results in an insersion data that doesn't exist in another table, that's fine. but i want know exactly which field has changed to store it in a log table
You don't mention precisely what you expect to see in your output but basically to accomplish what you're after you'll need a long sequence of CASE clauses, one for each column
e.g. one approach might be to create a comma-separated list of the column names that have changed:
INSERT INTO changes (id_registro, column_diffs)
SELECT d2.id_registro,
CONCAT(
CASE WHEN d1.origen <> d2.origen THEN 'Origen,' ELSE '' END,
CASE WHEN d1.sector <> d2.sector THEN 'Sector,' ELSE '' END,
etc.
Within the THEN part of the CASE you can build whatever detail you want to show
e.g. a string showing before and after values of the columns CONCAT('Origen: Was==> ', d1.origen, ' Now==>', d2.origen). Presumably though you'll also need to record the times of these changes if there can be multiple updates to the same record throughout the day.
Essentially you'll need to decide what information you want to show in your logfile, but based on your example query you should have all the information you need.
Not sure the best way to word this but I'm looking for a way to specify a condition when a value in a column has at least one value in a given list AND avalue not in the same list, then that column's value should show up. An example table:
email program
john#john.com program1
john#john.com program2
john#john.com program3
jeff#jeff.com program3
jeff#jeff.com program4
steve#steve.com program1
steve#steve.com program2
If I have this table and a list of (program1, program2), I would like the corresponding email to show up if the programs associated with a given email match at least one in the given list AND if the given email has a program NOT in the given list
So for the table above and the given list above all we would have show up with the correct query would be:
email
john#john.com
Any help on this would be greatly appreciated. Note: this would be in Redshift/PostgreSQL
I like doing this with group by and having. Here is a pretty general approach:
select email
from t
group by email
having sum( (program = 'program1')::int ) > 0 and
sum( (program = 'program2')::int ) = 0;
In this case, "program1" is required and "program2" is not. And, you can keep adding conditions -- as many as you like.
I forget if Redshift supports the :: syntax. You can always express this using standard SQL:
having sum( case when program = 'program1' then 1 else 0 end ) > 0 and
sum( case when program = 'program2' then 1 else 0 end ) = 0;
EDIT:
I think #dnswit is right on the parsing of the OP's question. The logic would be:
having sum( (program in ('program1', 'program2'))::int ) > 0 and
sum( (program not in ('program1', 'program2'))::int ) > 0;
if you just want a single list of emails no matter how many times they are on the list by having multiple programs
it is just select distinct email from tablename
First your Data Table is constructed wrong, you should use an unique Identifier so you can retrieve the program version you are specifying.
so your database should look like this:
> email program1 program2 program3
john#john.com ProgVersion1 ProgVersion2 ProgVersion3
steve#steve.com ProgVersion1 ProgVersion2 ProgVersion3
If you notice of the table above you can now query to get the program value you need for the specified Email. Use SQL Query, your Data Fields for your table are email, Program 1 Program 2 Program 3, when retrieving the value of the fields to be displayed, you are using redundancy you do not need to repeat the email address multiple times for each version of the program. This would not be expectable methodology.
SQL Query you can use:
instructions: you will create a parameter to use as a variable to query the data table from the list.
> CREATE PROCEDURE spLoadMyProgramVersion
>
> #email nvarchar(50),
>
> AS
>
>BEGIN
>SELECT program1,program2,program3
>FROM MyTableName
>WHERE (email LIKE #email) RETURN
This will allow you to load all your program version in a list by just specify the email address you want to load, this is a loading stored procedure just use it when you make a SQLCommand Object you can call your stored procedure.
Some background...
I have 20 + Files.
I read these file names from a prebuilt table building a subfile screen.
I select 1 file then build another screen with the contents of file selected.
I then select the record I want to delete, so far so good...
eval MySQL = stat3 + %trimr(scrwcrd) + STAT3B
my SQL Statement which reads in debug
MySQL = DELETE FROM FILESEL WHERE K00001 = ? with NC
PREPARE STAT3 from :MYSQL
EXECUTE STAT3 using :PROD
where :prod is the variable supplied from Screen selection
My sqlcod ends up at 100 with sqlstt = 2000 after the EXECUTE indicating ROW not found for Delete.
Now for a fact this is not the case. I see the record on the file selected and I see the value of PROD using debug any ideas...
What datatypes and length are the K00001 field and :PROD host variable?
Equality could be an issue. If they are character fields you may need to TRIM/%TRIM the values in order to match.
We would like to store both rails query string and table name in db and retrieve them for execution at run time. Here is the scenario:
Retrieve active customer records from customers table. Let's say we have 2 variable defined as:
table_name = 'Customer'
query_string = ':active => true'
In rails, the query could be:
records = Customer.where(:active => true)
Now with table name and query string stored in variables table_name and query_string, is it possible to assemble a query string with 2 variables like:
records = table_name.where(query_string) ?
Thanks for the help.
You could do this, but it's not generally recommended to evaluate a string as a hash. Also, table_name is an unfortunate name for the variable, because you actually are storing the class name (table would be 'customers'). In any event, what you are missing is the eval of these strings:
records = class_name.constantize.where(instance_eval(query_string))
Note that running instance_eval on a user-inputted string can be disasterous for security and the well-being of your application. Use with care, and stick to building an actual hash.
The definition of instance_eval is: Evaluates a string containing Ruby source code, or the given block, within the context of the receiver (obj).
Another way to eval a query string is to include where in the string, like:
table_name = 'Customer'
query_string = 'where(:active => true)'
Then the record could be retrieved by:
records = table_name.constantize.instance_eval(query_string)
By putting where into the string, we can use the full power of instance_eval instead of just returning the source code to where as in the question above.