I have a table with this structure:
CNMA_CO_PLATFORM_MESSAGE|AUDI_TI_CREATION|FIELD4|OTHER FIELDS
test-jj#2774#20210422112434957#00026129|22/04/21 11:24:34,957000000|11|..
test-jj2#2774#20210422112434957#00026129|22/04/21 11:24:34,957000000|12|..
test-jj3#2774#20210422112434957#00026129|22/04/21 11:24:34,957000000|13|..
This combination is the PRIMARY_KEY of the table:
CNMA_CO_PLATFORM_MESSAGE|AUDI_TI_CREATION
Well, I have an JdbcPagingItemReader defined like this (Pagesize is 1):
#StepScope
#Bean
public JdbcPagingItemReader<PendingNotificationDTO> pendingNotificationReader(
#Value("#{stepExecution}") StepExecution stepExecution){
final JdbcPagingItemReader<PendingNotificationDTO> reader = new JdbcPagingItemReader<>();
reader.setDataSource(daoDataSource);
reader.setName("pendingNotificationReader");
//Creamos la Query
final OraclePagingQueryProvider oraclePagingQueryProvider = new OraclePagingQueryProvider();
oraclePagingQueryProvider.setSelectClause("SELECT " +
" cegct.AUDI_TI_CREATION, "+
" CNMA_CO_PLATFORM_MESSAGE, " +
" OTHERFIELDS... ");
oraclePagingQueryProvider.setFromClause("FROM TABLE1 cegct " +
" JOIN TABLE1 notip ON cegct.field1 = notip.field1 " +
" AND notip.field2 = :frSur ");
oraclePagingQueryProvider.setWhereClause("WHERE "
+ " cegct.field3 = 0 "
+ " AND cegct.field4 in (:notifStatusList) ");
//Indicamos conjunto de campos no repetibles para poder paginar
Map<String, Order> sortKeys = new HashMap<>();
sortKeys.put("CNMA_CO_PLATFORM_MESSAGE", Order.DESCENDING);
sortKeys.put("AUDI_TI_CREATION", Order.DESCENDING);
oraclePagingQueryProvider.setSortKeys(sortKeys );
reader.setQueryProvider(oraclePagingQueryProvider);
String frSur = stepExecution.getJobExecution().getExecutionContext().getString(Constants.FM_ROLE_SUR_ZK);
String notifStatus = stepExecution.getJobExecution().getExecutionContext().getString(Constants.STATUS_REPORTS);
Map<String, Object> parameters = new HashMap<>();
parameters.put("frSur", frSur);
parameters.put("notifStatusList", Arrays.asList(StringUtils.split(notifStatus, ",")));
reader.setParameterValues(parameters );
Integer initLoaded = stepExecution.getJobExecution().getExecutionContext().getInt(Constants.RECOVER_PENDING_NOT_COMMIT);
reader.setPageSize(initLoaded);
reader.setRowMapper(new BeanPropertyRowMapper<PendingNotificationDTO>(PendingNotificationDTO.class));
return reader;
}
(I hide some irrelevant fields and table names)
Well, I run a test and my 3 records are valid to the select, these are selected one to one by the page size. Anyway, the first chunk-reader generated select my "test-jj3#..." record, my second chunk-reader select "test-jj2#.." and my third chunk-reader doesn't select doesn't recover any record (It should recover last 'test-jj#...' element.
These are the generated sqls (I hide some sensible no relevant fields)
First chunk, Select 1 register
SELECT * FROM (
SELECT
cegct.AUDI_TI_CREATION
CNMA_CO_PLATFORM_MESSAGE, [otherfields]
FROM [FROM]
WHERE [where]
ORDER BY CNMA_CO_PLATFORM_MESSAGE DESC, AUDI_TI_CREATION DESC
) WHERE ROWNUM <= 1;
Second chunk, Select 1 register (Here, the rownum filter by the sortkeys)
SELECT * FROM (
SELECT
cegct.AUDI_TI_CREATION
CNMA_CO_PLATFORM_MESSAGE, [otherfields]
FROM [FROM]
WHERE [where]
ORDER BY CNMA_CO_PLATFORM_MESSAGE DESC, AUDI_TI_CREATION DESC
) WHERE
ROWNUM <= 1 AND (
(CNMA_CO_PLATFORM_MESSAGE < 'test-jj3#2774#20210422112434957#00026129')
OR
(CNMA_CO_PLATFORM_MESSAGE = 'test-jj3#2774#20210422112434957#00026129' AND AUDI_TI_CREATION < TO_DATE('2021-04-22 11:24:34', 'YYYY-MM-DD HH24:MI:SS'))
);
Third chunk, select 0 registers
SELECT * FROM (
SELECT
cegct.AUDI_TI_CREATION
CNMA_CO_PLATFORM_MESSAGE, [otherfields]
FROM [FROM]
WHERE [where]
ORDER BY CNMA_CO_PLATFORM_MESSAGE DESC, AUDI_TI_CREATION DESC
) WHERE
ROWNUM <= 1 AND (
(CNMA_CO_PLATFORM_MESSAGE < 'test-jj2#2774#20210422112434957#00026129')
OR
(CNMA_CO_PLATFORM_MESSAGE = 'test-jj2#2774#20210422112434957#00026129' AND AUDI_TI_CREATION < TO_DATE('2021-04-22 11:24:34', 'YYYY-MM-DD HH24:MI:SS'))
);
Sorry for my english, I hope you can understand my problem.
Logs for the Prepared SQL Statement
Executing prepared SQL statement [SELECT * FROM (
SELECT
cegct.AUDI_TI_CREATION,
CNMA_CO_PLATFORM_MESSAGE,
OTHERFIELDS...
FROM TABLE1 cegct
JOIN TABLE2 notip ON cegct.field1 = notip.field1
AND notip.field2 = ?
WHERE cegct.field3 = 0
AND cegct.field4 in (?, ?, ?)
ORDER BY CNMA_CO_PLATFORM_MESSAGE DESC, AUDI_TI_CREATION DESC) WHERE ROWNUM <= 1]
20221116 12:52:43.560 TRACE org.springframework.jdbc.core.StatementCreatorUtils [[ # ]] - Setting SQL statement parameter value: column index 1, parameter value [1], value class [java.lang.String], SQL type unknown
20221116 12:52:43.560 TRACE org.springframework.jdbc.core.StatementCreatorUtils [[ # ]] - Setting SQL statement parameter value: column index 2, parameter value [11], value class [java.lang.String], SQL type unknown
20221116 12:52:43.560 TRACE org.springframework.jdbc.core.StatementCreatorUtils [[ # ]] - Setting SQL statement parameter value: column index 3, parameter value [12], value class [java.lang.String], SQL type unknown
20221116 12:52:43.560 TRACE org.springframework.jdbc.core.StatementCreatorUtils [[ # ]] - Setting SQL statement parameter value: column index 4, parameter value [13], value class [java.lang.String], SQL type unknown
A bind variable is a single value; therefore when you use:
AND cegct.field4 in (:notifStatusList)
Then :notifStatusList is a single string and is NOT a list of values and you effectively doing the same as:
AND cegct.field4 = :notifStatusList
If the bind variable :notifStatusList is a single value then it will work; however, when you try to pass in multiple values then it will not match those multiple values but will try to match field4 to the entire delimited list (which fails and will filter out all the rows).
If you want to pass a delimited string then use:
AND ',' || :notifStatusList || ',' LIKE '%,' || cegct.field4 || ',%'
Alternatively, pass the values as an array (rather than a delimited string) into an Oracle collection and then test to see if it is in that collection.
Related
I have this query, and one of its column is a subquery that should be bringing a list of values using a listagg function. This list has its starting point as the S.ID_ORGAO_INTELIGENCIA value. The list is a should be, it always has values.
The listagg function is consuming an inline view that uses a window function to create the list.
select *
from (
SELECT DISTINCT S.ID_SOLICITACAO,
S.NR_PROTOCOLO_SOLICITACAO,
S.DH_INCLUSAO,
S.ID_USUARIO,
U.NR_CPF,
OI.ID_MODULO,
OI.ID_ORGAO_INTELIGENCIA,
OI.NO_ORGAO_INTELIGENCIA,
R.ID_ATRIBUICAO,
P.ID_PERMISSAO,
1 AS TIPO_NOTIFICACAO,
(
select LISTAGG(oc6.ID_ORGAO_INTELIGENCIA || '-' || oc6.ord || '-', '; ') WITHIN GROUP (ORDER BY oc6.ord) eai
from (
SELECT oc1.ID_ORGAO_INTELIGENCIA,
oc1.ID_ORGAO_INTELIGENCIA_PAI,
oc1.SG_ORGAO_INTELIGENCIA,
rownum as ord
FROM TB_ORGAO_INTERNO oc1
WHERE oc1.DH_EXCLUSAO is null
-- THE VALUE FROM S.ID_ORGAO_INTELIGENCIA IS NOT AVAILBLE HERE
START WITH oc1.ID_ORGAO_INTELIGENCIA = S.ID_ORGAO_INTELIGENCIA
CONNECT BY prior oc1.ID_ORGAO_INTELIGENCIA_PAI = oc1.ID_ORGAO_INTELIGENCIA
) oc6) aproPrec
FROM TB_SOLICITACAO S
INNER JOIN TB_ORGAO_INTERNO OI ON S.ID_ORGAO_INTELIGENCIA = OI.ID_ORGAO_INTELIGENCIA
INNER JOIN TB_RELACIONAMENTO_ATRIBUICAO R
ON (R.ID_MODULO = OI.ID_MODULO AND R.ID_ORGAO_INTELIGENCIA IS NULL AND
R.ID_SOLICITACAO IS NULL)
INNER JOIN TB_PERMISSAO P
ON (P.ID_USUARIO = :usuario AND P.ID_ORGAO_INTELIGENCIA = :orgao AND
P.ID_ATRIBUICAO = R.ID_ATRIBUICAO)
INNER JOIN TB_USUARIO U ON (U.ID_USUARIO = S.ID_USUARIO)
WHERE 1 = 1
AND U.DH_EXCLUSAO IS NULL
AND P.DH_EXCLUSAO IS NULL
AND S.DH_EXCLUSAO IS NULL
AND OI.DH_EXCLUSAO IS NULL
AND R.ID_ATRIBUICAO IN :atribuicoes
AND P.ID_STATUS_PERMISSAO = 7
AND OI.ID_MODULO = 1
AND S.ID_STATUS_SOLICITACAO IN (1, 2, 5, 6)
and s.ID_ORGAO_INTELIGENCIA in (SELECT DISTINCT o.ID_ORGAO_INTELIGENCIA
FROM TB_ORGAO_INTERNO o
WHERE o.DH_EXCLUSAO IS NULL
START WITH o.ID_ORGAO_INTELIGENCIA = 3
CONNECT BY PRIOR o.ID_ORGAO_INTELIGENCIA = o.ID_ORGAO_INTELIGENCIA_PAI)
);
The problem is that the aproPrec column is always returning null as its result.
If I force the criteria to have the S.ID_ORGAO_INTELIGENCIA hardcoded, the list returns its true value.
If I chance this:
START WITH oc1.ID_ORGAO_INTELIGENCIA = S.ID_ORGAO_INTELIGENCIA
To this:
START WITH oc1.ID_ORGAO_INTELIGENCIA = 311
where 311 is the value that the S.ID_ORGAO_INTELIGENCIA column really has.
Is there a way to make this query works as 'I think' it should work?
To make it work, I changed the subquery by this another one:
(
select qt_.*
from (
SELECT QRY_NAME.*,
rownum as ord
FROM (
SELECT oc1.ID_ORGAO_INTELIGENCIA,
oc1.ID_ORGAO_INTELIGENCIA_PAI,
connect_by_root (oc1.ID_ORGAO_INTELIGENCIA) as root
FROM TB_ORGAO_INTERNO oc1
CONNECT BY NOCYCLE PRIOR oc1.ID_ORGAO_INTELIGENCIA_PAI = oc1.ID_ORGAO_INTELIGENCIA
) QRY_NAME
WHERE root = s.ID_ORGAO_INTELIGENCIA
) qt_
)
I want to run a multi-column correlated update of this kind:
UPDATE t1 t1_alias
SET (table_name, tablespace_name) = (
SELECT table_name, tablespace_name
FROM t2 t2_alias
WHERE t1_alias.table_name = t2_alias.table_name
);
But my attempt:
update customer up
set (customer_name, account, active) = (
select tmp.name, tmp.account, case when tmp.active = 'Yes' then 1 else 0 end
from customer_temp tmp
where up.agent = substr(tmp.agent, -4) and up.customer_code = tmp.code
);
... throws:
ORA-01407: cannot update ("FOO"."CUSTOMER"."CUSTOMER_NAME") to NULL
The source table customer_temp has no null values so I must be getting matches wrong. What is my error or misconception?
Presumably, there are some rows in the target table that have no match in the subquery.
You can avoid this with by adding an exists condition that filters out "unmatched" rows:
update customer up
set (customer_name, account, active) = (
select tmp.name, tmp.account, case when tmp.active = 'Yes' then 1 else 0 end
from customer_temp tmp
where up.agent = substr(tmp.agent, -4) and up.customer_code = tmp.code
)
where exists (
select 1
from customer_temp tmp
where up.agent = substr(tmp.agent, -4) and up.customer_code = tmp.code
);
Overall Task :- I need to retrieve data from 45 fields in system A and dump that data into a temp table which is then picked up by a unix process which produces an xml data file to be imported into system B.
Specific Question : What would be the best way of retrieving the data to be written into the 45 fields. Majority of the data is independent and can't be retrieved using a single statement. The way i currently retrieve this data is as follows (example below)
My temp tables hold the affected properties ID that i need to extract data for. i.e PROP_LIST_TEMP and ASSOC_PROP_TEMP.
SELECT SUBSTR (pro.pro_propref, 1, 25) UPRN,
(SELECT SUBSTR (adr_building, 1, 100)
FROM addresses, address_usages
WHERE aus_adr_refno = adr_refno
AND aus_aut_far_code = 'PHYSICAL'
AND aus_aut_fao_code = 'PRO'
AND (aus_end_date IS NULL OR aus_end_date > SYSDATE)
AND aus_pro_refno = pro.pro_refno)
BUILDING_NAME,
(SELECT CASE
WHEN (adr_street_number like 'BLOC%'
OR adr_street_number like '%-%'
OR adr_street_number like '%/%')
THEN NULL
ELSE regexp_replace (adr_street_number, '[^[:digit:]]+')
END
FROM addresses, address_usages
WHERE aus_adr_refno = adr_refno
AND aus_aut_far_code = 'PHYSICAL'
AND aus_aut_fao_code = 'PRO'
AND (aus_end_date IS NULL OR aus_end_date > SYSDATE)
AND aus_pro_refno = pro.pro_refno)
STREET_NUMBER,
(SELECT CASE
WHEN (adr_street_number like 'BLOC%'
OR adr_street_number like '%-%'
OR adr_street_number like '%/%')
THEN SUBSTR (adr_street_number, 1, 20)
ELSE REGEXP_REPLACE (adr_street_number, '[^[:alpha:]]+', '')
END
FROM addresses, address_usages
WHERE aus_adr_refno = adr_refno
AND aus_aut_far_code = 'PHYSICAL'
AND aus_aut_fao_code = 'PRO'
AND (aus_end_date IS NULL OR aus_end_date > SYSDATE)
AND aus_pro_refno = pro.pro_refno)
STREET_NUMBER_SUFFIX,
(SELECT SUBSTR (ptv_pty_code, 1, 3)
FROM prop_type_values
WHERE ptv_refno = pro.pro_hou_ptv_refno)
HOUSE_TYPE
FROM properties pro
WHERE pro_refno IN (select * from PIMSS_PROP_LIST_TEMP
UNION
select * from PIMSS_ASSOC_PROP_TEMP)
AND pro.pro_hou_hrv_hot_code IN
(SELECT frv_code
FROM first_ref_values
WHERE frv_frd_domain IN ('ASS_OWN_REF')
AND frv_current_ind = 'Y');
Since the where clauses of the subqueries in the select statement are identical, you could simply pull that out into the where clause, like so:
SELECT SUBSTR (pro.pro_propref, 1, 25) UPRN,
SUBSTR (addr.adr_building, 1, 100) BUILDING_NAME,
CASE WHEN (addr.adr_street_number like 'BLOC%'
OR addr.adr_street_number like '%-%'
OR addr.adr_street_number like '%/%')
THEN NULL
ELSE regexp_replace (addr.adr_street_number, '[^[:digit:]]+')
END STREET_NUMBER,
CASE WHEN (addr.adr_street_number like 'BLOC%'
OR addr.adr_street_number like '%-%'
OR addr.adr_street_number like '%/%')
THEN SUBSTR (addr.adr_street_number, 1, 20)
ELSE REGEXP_REPLACE (addr.adr_street_number, '[^[:alpha:]]+', '')
END STREET_NUMBER_SUFFIX,
(SELECT SUBSTR (ptv_pty_code, 1, 3)
FROM prop_type_values
WHERE ptv_refno = pro.pro_hou_ptv_refno) HOUSE_TYPE
FROM properties pro,
(select adr_building,
adr_street_number
FROM addresses, address_usages
WHERE aus_adr_refno = adr_refno
AND aus_aut_far_code = 'PHYSICAL'
AND aus_aut_fao_code = 'PRO'
AND (aus_end_date IS NULL OR aus_end_date > SYSDATE)) addr
WHERE pro.pro_refno = aus_pro_refno
and pro_refno IN (select * from PIMSS_PROP_LIST_TEMP
UNION
select * from PIMSS_ASSOC_PROP_TEMP)
AND pro.pro_hou_hrv_hot_code IN (SELECT frv_code
FROM first_ref_values
WHERE frv_frd_domain IN ('ASS_OWN_REF')
AND frv_current_ind = 'Y');
You might possibly need an outer join if there's a chance that no rows could be returned from the addr subquery.
Following is sample query.
CREATE PROCEDURE GetModel
(
#brandids varchar(100), -- brandid="1,2,3"
#bodystyleid varchar(100) -- bodystyleid="1,2,3"
)
AS
select * from model
where brandid in (#brandids) -- use a UDF to return table for comma delimited string
and bodystyleid in (#bodystyleid)
My requirement is that if #brandids or #bodystyleid is blank, query should return all rows for that condition.
Please guide me how to do this? Also suggest how to write this query to optimize performance.
You'll need dynamic SQL or a split function for this anyway, since IN ('1,2,3') is not the same as IN (1,2,3).
Split function:
CREATE FUNCTION dbo.SplitInts
(
#List VARCHAR(MAX),
#Delimiter CHAR(1)
)
RETURNS TABLE
AS
RETURN ( SELECT Item = CONVERT(INT, Item) FROM (
SELECT Item = x.i.value('(./text())[1]', 'int') FROM (
SELECT [XML] = CONVERT(XML, '<i>' + REPLACE(#List, #Delimiter, '</i><i>')
+ '</i>').query('.') ) AS a CROSS APPLY [XML].nodes('i') AS x(i)) AS y
WHERE Item IS NOT NULL
);
Code becomes something like:
SELECT m.col1, m.col2 FROM dbo.model AS m
LEFT OUTER JOIN dbo.SplitInts(NULLIF(#brandids, ''), ',') AS br
ON m.brandid = COALESCE(br.Item, m.brandid)
LEFT OUTER JOIN dbo.SplitInts(NULLIF(#bodystyleid, ''), ',') AS bs
ON m.bodystyleid = COALESCE(bs.Item, m.bodystyleid)
WHERE (NULLIF(#brandids, '') IS NULL OR br.Item IS NOT NULL)
AND (NULLIF(#bodystyleid, '') IS NULL OR bs.Item IS NOT NULL);
(Note that I added a lot of NULLIF handling here... if these parameters don't have a value, you should be passing NULL, not "blank".)
Dynamic SQL, which will have much less chance of leading to bad plans due to parameter sniffing, would be:
DECLARE #sql NVARCHAR(MAX);
SET #sql = N'SELECT columns FROM dbo.model
WHERE 1 = 1 '
+ COALESCE(' AND brandid IN (' + #brandids + ')', '')
+ COALESCE(' AND bodystyleid IN (' + #bodystyleid + ')', '');
EXEC sp_executesql #sql;
Of course as #JamieCee points out, dynamic SQL could be vulnerable to injection, as you'll discover if you search for dynamic SQL anywhere. So if you don't trust your input, you'll want to guard against potential injection attacks. Just like you would if you were assembling ad hoc SQL inside your application code.
When you move to SQL Server 2008 or better, you should look at table-valued parameters (example here).
if(#brandids = '' or #brandids is null)
Begin
Set #brandids = 'brandid'
End
if(#bodystyleid = '' or #bodystyleid is null)
Begin
Set #bodystyleid = 'bodystyleid'
End
Exec('select * from model where brandid in (' + #brandids + ')
and bodystyleid in (' + #bodystyleid + ')')
I have the following query against an Oracle database:
select * from (
select Person.pId, Person.lastName as patLast, Person.firstName as patFirst
, Person.dateOfBirth
, Medicate.mId, Medicate.startDate as medStart, Medicate.description
, cast(substr(Medicate.instructions, 1, 50) as char(50)) as instruct
, ml.convert_id_to_date(Prescrib.pubTime) as scripSigned
, max(ml.convert_id_to_date(Prescrib.pubTime)) over (partition by Prescrib.pId, Prescrib.mId) as LastScrip
, UsrInfo.pvId, UsrInfo.lastName as provLast, UsrInfo.firstName as provFirst
from ml.Medicate
join ml.Prescrib on Medicate.mId = Prescrib.mId
join ml.UsrInfo on Prescrib.pvId = UsrInfo.pvId
join ml.Person on Medicate.pId = Person.pId
where Person.isPatient = 'Y'
and Person.pStatus = 'A'
and Medicate.xId = 1.e+035
and Medicate.change = 2
and Medicate.stopDate > sysdate
and REGEXP_LIKE(Medicate.instructions
, ' [qtb]\.?[oi]?\.?[dw][^ayieo]'
|| '|[^a-z]mg?s'
|| '|ij'
|| '|[^a-z]iu[^a-z]'
|| '|[0-9 ]u '
|| '|[^tu]hs'
, 'i')
order by ScripSigned desc
) where scripSigned = lastScrip and scripSigned > date '2011-01-01'
I have a Report Parameter defined, DateBegin, defined as a DateTime, and I've associated it with a Query Parameter also called DateBegin. I just can't figure out how to replace "date '2011-01-01'" with "DateBegin" so that the blinkin' thing actually works. Thanks!
Use the Oracle format for parameters - so use the following:
) where scripSigned = lastScrip and scripSigned > :DateBegin
(The # sign is used in SQLServer to identify SQLServer variables.)