Related
I am trying to parse the yaml string to a column in SQL.
I have a yaml string in one of the columns like
country_cd:GB
postal_town_city:London
zip:CT11 A1Cr
I have to write a query to pass the key and get a value. When I pass 'country_cd', I want to receive 'GB'.
I have tried the below script but it is returning 'country_cd:G'. What am I doing incorrectly ?
select trim(regexp_substr('---
country_cd:GB
postal_town_city:London
zip:CT11 A1Cr', '\n?'||'postal_town_city'||'*[:=] *(.+?) *\n?', 1, 1, 'i')) from dual ;
You can use REGEXP_SUBSTR with the m match parameter to parse the string as separate lines:
SELECT REGEXP_SUBSTR(yaml, '^country_cd:(.*)$', 1, 1, 'm', 1) AS value
FROM table_name;
Which, for the sample data:
CREATE TABLE table_name (yaml) AS
SELECT 'country_cd:GB
postal_town_city:London
zip:CT11 A1Cr' FROM DUAL;
Outputs:
VALUE
GB
or, you can split the string using simple string functions (which are often quicker than using regular expressions) and then filter for the correct key:
WITH lines (yaml, line_start, separator_pos, line_end) AS (
SELECT yaml,
1,
INSTR(yaml, ':', 1),
INSTR(yaml, CHR(10), 1)
FROM table_name
UNION ALL
SELECT yaml,
line_end + 1,
INSTR(yaml, ':', line_end + 1),
INSTR(yaml, CHR(10), line_end + 1)
FROM lines
WHERE line_end > 0
)
SELECT SUBSTR(yaml, line_start, separator_pos - line_start) AS key,
CASE line_end
WHEN 0
THEN SUBSTR(yaml, separator_pos + 1)
ELSE SUBSTR(yaml, separator_pos + 1, line_end - separator_pos)
END AS value
FROM lines
WHERE SUBSTR(yaml, line_start, separator_pos - line_start) = 'country_cd'
Which outputs:
KEY
VALUE
country_cd
GB
db<>fiddle here
Using REGEXP_SUBSTR with a capture group, we can try:
SELECT yaml,
REGEXP_SUBSTR(yaml_col, '(^|\s)country_cd:(\S+)', 1, 1, NULL, 2) AS country_cd
FROM yourTable;
Here is regex demo.
I need to convert a string to a date field. The field stores 30 characters. Dates, when present, are formatted as 'yyyymmdd' (20170202). In all cases, dates have 22 spaces after. I need to format this field as a date field like this: dd-mm-yyyy.
I've tried several formulas:
TO_CHAR(PERSACTION.NEW_VALUE_02, 'dd-mm-yyyy') ,TO_CHAR(PERSACTION.NEW_VALUE_02, 'yyyymmdd'), trim(TO_CHAR(PERSACTION.NEW_VALUE_02, 'yyyymmdd')) with error message: invalid number format model. Your expertise is welcome and appreciated.
to_char(to_date( rtrim(new_value_02), 'yyyymmdd'), 'dd-mm-yyyy')
Should do the trick. rtrim removes spaces on right side of string. Then I convert it to date using the date format specified, and then convert it to a string again in the desired format.
Did tried to convert to date format and then to char again?
TO_CHAR(TO_DATE(PERSACTION.NEW_VALUE_02,'yyyymmdd'),'dd-mm-yyyy')
Please, please, please do not store DATEs and CHARACTER datatypes. This will only lead to issues that can be avoided when using the DATE datatype.
If you want to change the string 20170202 to another string and not actually a date (which would have no intrinsic formatted text representation), you could optionally use a regular expression to transform it, instead of converting to a date and back:
select regexp_replace('20170202 ', '^(\d{4})(\d{2})(\d{2}) +$', '\3-\2-\1')
from dual;
REGEXP_REPLACE(
---------------
02-02-2017
Or you could use substr instead of regexp_substr, which may perform better even if you have to call it three times; using a CTE just to avoid repeating the value:
with t(str) as (
select '20170202 ' from dual
)
select substr(str, 7, 2) ||'-'|| substr(str, 5, 2) ||'-'|| substr(str, 1, 4)
from t;
SUBSTR(STR
----------
02-02-2017
If you do convert to a date and back you would uncover any values which cannot be converted, as they will cause an exception to be thrown. That would imply you have bad data; which would have been avoided by using the right data type in the first place, of course. These will convert any old rubbish, with varying results depending on how far the strings stray from the pattern you expect - but including strings like '20170231' which represent an invalid date. And null value or strings of just spaces will be converted to odd things with the substr version, but you could filter those out.
You can see the kind of variation you would get with some sample data that doesn't match your expectations:
with t(str) as (
select '20170202 ' from dual
union all select '20170231 ' from dual
union all select '2017020c ' from dual
union all select '2017020 ' from dual
union all select '201702021 ' from dual
union all select ' ' from dual
union all select null from dual
)
select str,
regexp_replace(str, '^(\d{4})(\d{2})(\d{2}) +$', '\3-\2-\1') as reg,
substr(str, 7, 2) ||'-'|| substr(str, 5, 2) ||'-'|| substr(str, 1, 4) as sub
from t;
STR REG SUB
------------- ------------- -------------
20170202 02-02-2017 02-02-2017
20170231 31-02-2017 31-02-2017
2017020c 2017020c 0c-02-2017
2017020 2017020 0 -02-2017
201702021 201702021 02-02-2017
- -
--
With the anchors and whitespace expectation, the regular expression doesn't modify anything that doesn't consist entirely of 8 numeric characters. But it can still form invalid 'dates'.
I'm struggling with getting a query to work, and I could really use some help. We have an in house app that we use for building small web apps where I work. It's basically a drag and drop GUI. There's functionality built in to access query string values using the key.
I'm passing a comma separated list of values into a page through the query string. I'm then trying to use the list of values as part of an in clause in a query.
I can see that the value is correct in the query string.
orders=1,2,3
Here's the specific part of the query
"AND OrderNumber IN (this is where it maps from the query string)
I've tried running similar queries in Toad, and I think I've found the issue. It's giving an invalid number error, and I think it's wrapping the query string value in single quotes. I can replicate the error when I do "AND OrderNumber IN ('1,2,3')" in Toad.
Here's where I get really confused. The following works in Toad.
"AND OrderNumber IN ('1','2','3')"
So I tried recreating that by doing
select replace('1,2,3', ',', chr(39)||','||chr(39)) from dual;
I have confirmed that returns '1','2','3' in Toad.
However, I still get an Invalid Number error when I run the following in Toad.
AND OrderNumber IN (replace('1,2,3', ',', chr(39)||','||chr(39))
I've been racking my brain over this, and I can't figure it out. It seems to me that if "AND OrderNumber IN ('1','2','3')" works, and replace('1,2,3', ',', chr(39)||','||chr(39)) returns '1','2','3', that "AND OrderNumber IN (replace('1,2,3', ',', chr(39)||','||chr(39))" should work.
Any help you might be able to offer on this would be greatly appreciated. I know the rest of the query works. That's why I didn't post it. I'm just stuck on trying to get this IN clause working.
A change to phonetic_man's answer that will allow for NULL elements in the list. The regex format of '[^,]+' for parsing delimited lists does not handle NULL list elements and will return an incorrect value if one exists and thus its use should be avoided. Change the original by deleting the number 2 for instance and see the results. You will get a '3' in the 2nd element's position! Here's a way that handles the NULL and returns the correct value for the element:
SELECT TRIM(REGEXP_SUBSTR(str, '(.*?)(,|$)', 1, LEVEL, NULL, 1)) str
FROM ( SELECT '1,,3,4' str FROM dual )
connect by level <= regexp_count(str, ',') + 1;
See here for more info and proof: https://stackoverflow.com/a/31464699/2543416
Can you try the following query.
SELECT * FROM orders
WHERE orderno IN
(
SELECT TRIM(REGEXP_SUBSTR(str, '[^,]+', 1, LEVEL)) str
FROM ( SELECT '1,2,3,4' str FROM dual )
CONNECT BY INSTR(str, ',', 1, LEVEL - 1) > 0
)
The inline query splitting the string in different rows. So, on executing it you will get the following result.
SELECT trim(regexp_substr(str, '[^,]+', 1, LEVEL)) str
FROM ( SELECT '1,2,3,4' str FROM dual )
CONNECT BY instr(str, ',', 1, LEVEL - 1) > 0
1
2
3
4
Now, passing this result to the main query IN clause should work.
I think the desired clause to be built is:
AND OrderNumber IN (1,2,3)
A numeric list. The example you tested:
AND OrderNumber IN ('1','2','3')
works because an implicit conversion from a VARCHAR2 to a NUMBER is occurring for each element in the list.
The following clause will not work because no implicit conversion of the string '1,2,3' can be made (note the clause has a single string element):
AND OrderNumber IN ('1,2,3')
When doing a replace, you are converting the single string: 1,2,3 with the single string: 1','2','3 and this single string cannot be implicitly converted to a number.
I have a query that uses a connect by statement to order the recursive data. The problem I have is that there is occasionally a one to many or a many to one relationship and I dont know how to deal with it.
SELECT *
FROM (SELECT * FROM bdTable WHERE parentek = t_parKey)
START WITH source is null
CONNECT BY PRIOR target = source
So to explain. I have a source and target columns. About 99% of the time these are a single unique ID. Unfortunately the other 1% of the time there are a grouping of IDs in one of the columns. This table is a flat representation of a flowchart type tool, so there are splits and decisions which can have many outputs and merges which can have many inputs.
To deal with this in loading the data for the table, the unique IDs are concatenated together using the listagg function. So I end up with a Target value of something like '1254143,2356334,6346436,3454363,3462354,442356'.
So when my connect by statement is executed, it works perfectly until it comes to one of these scenarios, at which point it just stops (which is expected of course).
I thought I might be able to use IN or INSTR in some way to get it working, but haven't had any luck yet and I can't find anything on it online.
Any help would be appreciated.....
If you want to join target and source with use of some logic, based on intersection of set of values, listed in each column, then most reliable way to do so is to split strings to collections and operate on collection from prior row and collection from current row to build a tree.
There are a number of techniques to build collection from separated string in Oracle, one of them illustrated in this answer to another question.
Create required collection type:
create or replace type TIdList as table of varchar2(100);
Inner select in your case would look like this:
SELECT
t.*,
(
cast(multiset( -- Convert set of values into collection
select -- Build list of values from separated string
substr(
source,
decode( level, 1, 1, instr(source,',',1,level-1)+1 ),
decode( instr(source,',',1,level), 0, length(source)+1, instr(source,',',1,level) )
-
decode( level, 1, 1, instr(source,',',1,level-1)+1 )
) code
from dual
start with source is not null
connect by instr(source,',',1,level-1) > 0
) as TIdList )
) source_id_list,
(
cast(multiset( -- Convert set of values into collection
select -- Build list of values from separated string
substr(
target,
decode( level, 1, 1, instr(target,',',1,level-1)+1 ),
decode( instr(target,',',1,level), 0, length(target)+1, instr(target,',',1,level) )
-
decode( level, 1, 1, instr(target,',',1,level-1)+1 )
) code
from dual
start with target is not null
connect by instr(target,',',1,level-1) > 0
) as TIdList )
) target_id_list
FROM bdTable t
WHERE t.parentek = t_parKey
Because I don't know which column (source or target) contains separated list, I include column for each.
After building collection(s) it's possible to use multiset operators and available test functions to match target with source. E.g.
with inner_query as (
SELECT
t.*,
(
cast(multiset( -- Convert set of values into collection
select -- Build list of values from separated string
substr(
source,
decode( level, 1, 1, instr(source,',',1,level-1)+1 ),
decode( instr(source,',',1,level), 0, length(source)+1, instr(source,',',1,level) )
-
decode( level, 1, 1, instr(source,',',1,level-1)+1 )
) code
from dual
start with source is not null
connect by instr(source,',',1,level-1) > 0
) as TIdList )
) source_id_list,
(
cast(multiset( -- Convert set of values into collection
select -- Build list of values from separated string
substr(
target,
decode( level, 1, 1, instr(target,',',1,level-1)+1 ),
decode( instr(target,',',1,level), 0, length(target)+1, instr(target,',',1,level) )
-
decode( level, 1, 1, instr(target,',',1,level-1)+1 )
) code
from dual
start with target is not null
connect by instr(target,',',1,level-1) > 0
) as TIdList )
) target_id_list
FROM bdTable t
WHERE t.parentek = t_parKey
)
select
level lvl,
tree_list.*
from
inner_query tree_list
start with
source is null
connect by
nvl(cardinality(prior target_id_list MULTISET INTERSECT source_id_list),0) > 0
If only one column can contain list of values, then MEMBER OF construct are useful.
I would like to know that if I can compare a value with a list of items in an decode function. Basically I want to know that if is it possible to make a decode statement's 'search' value a list. For example,
decode(task_id, (1,2,3), 3 * task_time)
This piece of code won't compile though. Is this the only option for this case then (without using case-when) or are there alternative ways of doing this?
decode(task_id, 1, 3 * task_time,
2, 3 * task_time,
3, 3 * task_time)
I am using Oracle 10gR2. Any help is much appreciated.
If a single list of values is sufficient, you can turn it into a CASE and IN clause:
case when task_id in (1, 2, 3) then 3 * task_time else null end
I don't think its possible to use a list with decode in this way. Per the docs:
DECODE compares expr to each search value one by one. If expr is equal
to a search, then Oracle Database returns the corresponding result. If
no match is found, then Oracle returns default
So task_id is compared with a search value one by one. If search value was a list, you couldn't compare with a single value.
I found a solution :)
select
decode(
task_id,
(select task_id from dual where task_id in (1,2,3)),
3*task_time)
decode ( (taskid-1)*(taskid-2)*(taskid-3), 0, 3 * tasktime ) could do what you want
Here's a working example:
with a as (
select 1 taskid, 11 tasktime from dual union all
select 2 taskid, 11 tasktime from dual union all
select 3 taskid, 11 tasktime from dual union all
select 4 taskid, 11 tasktime from dual
)
select
taskid,
decode (
(taskid-1) *
(taskid-2) *
(taskid-3) ,
0, 3 * tasktime
) decoded
from a;
you can use union all:
select 3 * task_time from your_table where task_id in (1,2,3)
union all
select task_time from your_table where task_id not in (1,2,3)
but why ?
In if condition :
IF vVal in (3,1,2) THEN
dbms_output.put_line('FOUND');
ELSE
dbms_output.put_line('NOT FOUND');
END IF;
I've seen instr(...) and static strings used as a way to quickly determine whether a value is one of multiple you're looking for as a condition for what value you return. You may need to choose a delimiter, but with limited datasets you can even omit it. It avoids using case-when, subqueries, and PL/SQL. As far as I know, there is no shorter way to do this:
decode(instr('123', taskid), 0, null, taskid * 3)
It's also very convenient when you want to set exceptions (for instance returning taskid without multiplication if it equals 1):
decode(instr('12345', taskid), 0, null, 1, taskid, taskid * 3)