While working with one of the oracle query, when i am firing a select command with where condition something like:
select * from table_name where contains(address,'"IN%" AND "BÖDEN%"') >0
No results are shown as part of this query though results are there in database.
Again if i introduce one space between "IN" and "%" i am getting results in proper:
select * from table_name where contains(address,'"IN %" AND "BÖDEN%"') >0
There is one more scenario :
Same type of query but if "IN" is replaced with "IM" the corresponding results are displayed.
select * from table_name where contains(address,'"IM%" AND "BÖDEN%"') >0
or
select * from table_name where contains(address,'"IM %" AND "BÖDEN%"') >0
Both above queries are giving same results. which is quite obvious too.
Hence i suspect there is something speacial with "IN" in oracle.
Can anybody suggest me proper reason for this problem and any resolution for the problem ?
The IN keyword is a stop word in the default Stoplist which is used to specify the words which are not be indexed in Oracle Text.
You need to remove it from the stop list in order to search such stop words.
Check the link for Stoplist
Also please find the other default Stop words for English in oracle for your reference Default Stop words
Related
We have front-end app where user enters customer and/or PO to retrieve data.
For instance, if user want to retrieve all POs for a customer, he will enter '%' in PO field.
If user wants to retrieve all data, he will enter '%' in every field.
I am trying this, but it does not work
SELECT *
FROM PO_INFO
WHERE customer_id = case when '%' then customer_id else 'Macys' end
AND purchase_order = case when '%' then purchase_order else '79124' end
What am I missing?
You should not (some would say must not) just plug the user-entered search values into your SQL query as literals. That is,
AND purchase_order = case when '%' then purchase_order else '79124' end
... is not going to perform or scale well because every single search looks to Oracle like a brand new SQL query that has to get parsed and optimized (i.e., "hard parsed). This is an expensive process that also requires a lot of latches, meaning multiple users trying to run searches at the same time will have to wait for each other.
Instead, you should construct your SQL using bind variables. So,
AND purchase_order = :1 -- or something. We'll discuss the exact logic later...
That :1 is a bind variable, a placeholder for a value your code will supply when it submits the query. Now, everyone doing a search is using the same query (just each supplying different values for the bind variables). No more excessive hard parsing and performance disaster averted.
But, here is the next problem. One query for all means it only gets hard parse one time (good) but it also means everyone runs using the same execution plan. But in a search like this, one size does not fit all. Suppose the execution plan Oracle comes up with uses an index on column 'ABC'. If a user does not supply a bind variable value for column 'ABC', that execution plan will still be followed, but with terrible results.
So, what we want really is one SQL for each set of bind variables that have values or don't, but not one SQL for each distinct set of literal search values.
Build your SQL in code by starting with this string:
SELECT * FROM PO_INFO WHERE 1=1
Then, for each search condition add this (if the value is %)
AND (:1 IS NULL) -- and pass `NULL`, not "%" as the value for :1
(Aside: the reason for this condition, which is essentially NULL IS NULL is to make to so the number and order of the bind variables that have to be passed in is always the same, regardless of what the end user does or does not give you a value for. This makes it much easier to submit the SQL in some languages, like PL/SQL).
If the search condition is not %, add this:
AND (customer_id /* or whatever column */ = :1) -- and pass the user-entered value
So, for example, if the user specified values for customer_id and po_date but not purchase_order, your final SQL might look like:
SELECT *
FROM PO_INFO
WHERE 1=1
AND customer_id = :1
AND po_date := 2
AND :3 IS NULL -- this is purchase order and :3 you will pass as null
If you do all this, you'll get the least amount of hard-parsing and the best execution plan for each search.
I have two broad questions:
Q1. Are executing the lines of code in Oracle SQL Developer and executing the same code in sqlplus command prompt same things?
(reason I ask this that I heard that not all the sqlplus commands are executable in SQL Developer. If answer to above question is yes then please then few useful links will help).
Q2. I am spooling the results of a sql query to a .csv file, but the thing is that columns names are truncated according the maximum length of the values in that column.
My code is:
set colsep ","
spool C:\Oracle\sample.csv
select * from acct_clas_rule;
spool off;
Output of above code is (middle column is having null values)
ABCD_,ABCD_,ABC
-----,-----,---
AB , ,WSD
ABCD , ,WSD
ABCD , ,WSD
SG , ,WSD
KD , ,WSD
WD , ,LKJ
KLHGF, ,LKO
WSDFG, ,LOK
WSDF , ,LKO
WS , ,GH
In above output, columns names have been truncated. I want full names of the columns to be displayed. Can anyone help?
I have seen the question in this link, but I didn't understand how to apply the answers provided there as there was no particular example cited. I am new to these things so I couldn't understand.
Original names of my columns (from left to right in above table) are :
ABCD_DFGT_SDF, ABCD_EDF_GH, ABCD_DFRE
PS -
1. I am using Oracle SQL developer to run sqlplus commands. I think because of which few of my commands are not working (like set underline, set linesize etc.).Please let me know if this is the case. I actually want remove those underlines beneath the columns names.
2. Also let me know that whether you answer is applicable to Oracle SQL Developer or sqlplus.
Thank You
There are a couple of things you can do, in addition to #JChomel's approach - that will work in either SQL Develoepr or SQL*Plus, while these suggestions are specific to SQL Developer.
Let's start with a dummy query based on a CTE to get something like your situation:
set colsep ","
with acct_clas_rule (abdc_1, abcd_2, abcd_3) as (
select cast('AB' as varchar2(5)), cast(null as varchar2(5)), cast('WSD' as varchar2(4)) from dual
union all select 'ABCD', null, 'WSD' from dual
-- ...
union all select 'WS', null, 'GH' from dual
)
select * from acct_clas_rule;
When run as a script in SQL Developer (from the document+arrow icon, or F5) the output is:
ABDC_,ABCD_,ABCD
-----,-----,----
AB , ,WSD
ABCD , ,WSD
WS , ,GH
If you change the query to include the SQL Developer-specific formatting hint /*csv*/ then you get the output you want:
select /*csv*/ * from acct_clas_rule;
"ABDC_1","ABCD_2","ABCD_3"
"AB","","WSD"
"ABCD","","WSD"
"WS","","GH"
except that the strings are all enclosed in double-quotes, which might not be what you really want. (It depends what you're doing with the spooled file and whether any of your string values contain commas, which would confuse Excel for instance).
With more recent versions of SQL Developer you can get exactly the same result without modifying the query using the sqlformat option:
set sqlformat csv
select * from acct_clas_rule;
but again you get the double-quotes. You could change the double-quotes to different enclosure characters, but that probably doesn't help.
A different approach is to use the built-in export tools instead of spool. If you run the query as a statement (green 'play button' icon, or control-enter) then rather than appearing in the Script Output panel, a new Query Result panel will open next to that, showing the results as a grid.
If you right-click on the results you'll get a contextual menu, and choosing Export... from that will give you a wizard to export the results in a format of your choice, including CSV:
You can leave the left and right enclosures as double-quotes to get the same results as the options above, except that null values use the word 'null' instead of an empty string:
"ABDC_1","ABCD_2","ABCD_3"
"AB",null,"WSD"
"ABCD",null,"WSD"
"WS",null,"GH
or you can change them, or remove them by choosing 'none', which gives you:
ABDC_1,ABCD_2,ABCD_3
AB,null,WSD
ABCD,null,WSD
WS,null,G
#thatjeffsmith has commented on how to change the null-replacement text for the /*csv*/ approach. For export it looks like having the word 'null' might have been a bug in 17.2.0; in 17.3.1 that does not appear, and instead you see:
ABDC_1,ABCD_2,ABCD_3
AB,,WSD
ABCD,,WSD
WS,,GH
or enclosed empty strings ("") if you leave the enclosures set.
Q1: TI'm not an expert of SQL Developer. There might be a mode like "command line" or something where you could get similar result, but unsure.
Q2: You have to set the right option to sqlplus: here is a trick I know of (it will also remove the --- --- --- that will cause other issue):
SET HEADING OFF`, to avoid column name to be printed
Union all the column names at the beginning of your script:
set colsep ","
spool C:\Oracle\sample.csv
select 'ABCD_DFGT_SDF', 'ABCD_EDF_GH', 'ABCD_DFRE' from dual
UNION ALL
select * from acct_clas_rule;
spool off;
use the sql plus set command -----> set underline off
I need to write a query on an Oracle database where one of the fields is called ACCOUNT. ACCOUNT is on the reserved list http://docs.oracle.com/cd/B19306_01/em.102/b40103/app_oracle_reserved_words.htm and my query is failing.
In this DB, ACCOUNT is a VARCHAR2 and I cannot change it's name nor the structure of anything, I can only run SELECT queries.
Although ACCOUNT is a VARCHAR2, it always contains an integer and I want to get at a range of values.
I thought that this would do the trick:
SELECT *
FROM TABLE
WHERE TO_NUMBER(ACCOUNT) > 1000
AND TO_NUMBER(ACCOUNT) < 2000
but I just get an ORA-01722 invalid number error.
I have checked that ACCOUNT only contains integers and running this query with a non-reserved keyword works fine...
You can escape the reserve word using " double quote like
SELECT *
FROM TABLE
WHERE TO_NUMBER("ACCOUNT") > 1000
AND TO_NUMBER("ACCOUNT") < 2000
(OR) Better use BETWEEN construct like
SELECT *
FROM TABLE
WHERE TO_NUMBER("ACCOUNT") BETWEEN 1001 AND 1999
In case your table name really is TABLE; you need to escape that too cause that as well a reserve word.
I looked into your issue and I was able to DUPLICATE
this error ORA-01722 invalid number error. occurs because your sql is trying to convert something like
To_NUMBER('SOMETEXT') > SOME NUMBER
so when your sql is converting the to_number('somenumber as a varchar2') it comes across ('sometext varchar2)
an example
SELECT *
FROM TABLE as tbl
WHERE TO_NUMBER('helloworld') > 1000
this will throw that error. check your column's data, somewhere in that data, there is some text in one or more row.
have you tried to add an alias to the table
SELECT *
FROM TABLE as tbl
WHERE TO_NUMBER(tbl.ACCOUNT) > 1000
AND TO_NUMBER(tbl.ACCOUNT) < 2000
So I am having a big table of AbstractID, SentenceID, WordID, and Word. Some words show in multiple files. I would like to count how many files every word shows, and then create a new table of with Word and Count. Below is a piece of my data.
My sql query is below:
CREATE TABLE WORDLIST(WORD, TOTALCOUNT) AS
(
SELECT WORD FROM ASSIGNMENT2 WHERE
(
SELECT WORD, COUNT(*) FROM (select DISTINCT FILEID, WORD FROM ASSIGNMENT2)
GROUP BY WORD ORDER BY COUNT(*) DESC));
It is giving me an error:
"Error report:
SQL Error: ORA-00936: missing expression
00936. 00000 - "missing expression"
*Cause:
*Action:"
How do I correct this? Thanks.
You're rather overcomplicating the query. The error is because your outer where clause doesn't have a condition; you're selecting multiple things, which isn't right anyway, but then not comparing those to anything. And you don't need to enclose the query in a set of parentheses.
You can just do:
CREATE TABLE WORDLIST (WORD, TOTALCOUNT) AS
SELECT WORD, COUNT(DISTINCT FILEID)
FROM ASSIGNMENT2
GROUP BY WORD;
There's no point ordering the query as it won't affect future queries against the new table - you still have to order those, as there is no inherent order of rows within the table. (And I think an order by will cause an error in a create-as statement, but can't immediately check that).
Creating a table to hold this information seems odd though as it will go out of date quickly. Perhaps you wanted a view really?
I am using oracleclient provider. I was wondering how do I use a parameter in the query.
select * from table A where A.a in ( parameter).
The parameter should be a multivalue parameter.
how do I create a data set?
Simple. Add the parameter to the report and make sure to check it off as multi-valued. Then in the data tab and go in and edit the query click the "..." button to edit the dataset. Under the parameters tab create a mapping parameter so it looks something like this (obviously you will have different names for your parameters):
#ids | =Parameters!ContractorIDS.Value
Then in the query tab use the coorelated sub-query like your example above. I have done this many times with SQL server and there is no reason it should not work with Oracle since SSRS is going to build an ANSI compliant SQL statement which it will pass to Oracle.
where A.myfield in (#ids)
You can't have a variable in list in oracle directly. You can however, break apart a comma seperated list into rows that can be used in your subquery. The string txt can be replaced by any number of values seperated by a comma.
select * from a where a.a in (
SELECT regexp_substr(txt,'[^,]+',1,level)
FROM (SELECT 'hello,world,hi,there' txt -- replace with parameter
FROM DUAL)
CONNECT BY LEVEL <= LENGTH (REGEXP_REPLACE (txt, '[^,]'))+1
)
The query works by first counting the number of commas that are in the text string. It does this by using a reqular expression to remove all non commas and then counts the length of the remainder.
It then uses an Oracle "trick" to return that number + 1 number of rows from the dual table. It then uses the regexp_substr function to pull out each occurence.
Firstly in SSRS with an Oracle OLEDB connection you need to use the colon, not the # symbol e.g. :parameter not #parameter but then you aren't able to do this as a multi-valued parameter, it only accepts single values. Worse, if you are using an ODBC connection you have to use the question mark by itself e.g. ? not #parameter and then the ordering of parameters becomes important, and they also cannot be multi-valued. The only ways you are left with is using an expression to construct a query (join() function for the param) or calling a stored proc.
The stored proc option is best because the SSRS can handle the parameters for stored procs to both SQL Server and Oracle very cleanly, but if that is not an option you can use this expression:
="select column1, column2, a from table A where A.a in (" + Join(Parameters!parameter.Value,", ") + ")"
Or if the parameter values are strings which need apostrophes around them:
="select column1, column2, a from table A where A.a in ('" + Join(Parameters!parameter.Value,"', '") + "')"
When you right-click on the dataset, you can select "dataset properties" and then use the fx button to edit the query as an expression, rather than using the query designer which won't let you edit it as an expression.
This expression method is limited to a maximum limit of about 1000 values but if you have that many this is the wrong way to do it anyway, you'd rather join to a table.
I don't think you can use a parameter in such a situation.
(Unless oracle and the language you're using supports array-type parameters ? )
The parameters in oracle are defined as ":parametername", so in your query you should use something like:
select * from table A where value in (:parametername)
Add the parameter to the paramaters folders in the report and mark the checkbox "Allow multiple values".
As Victor Grimaldo mentioned… below worked for me very fine. As soon as I use the :parameter in my SQL query in SSRS dataset1.. it asked me to enter the values for these parameters, for which I choose already created SSRS parameters.
SELECT * FROM table a WHERE VALUE IN (**:parametername**)
Thanks Victor.