How to search Zero Width Joiner (ZWJ in SQL Server - unicode 200d)? - sql

Below query condition doesn't work
For eg :
SELECT * FROM NAME WHERE (NAME_MA LIKE N'%‍%')

SELECT * FROM NAME WHERE HEX(NAME_MA) LIKE '%‍E2808D%')
E2808D is Hex representation for ZWJ character.
Similarly you can use E2808C for ZWNJ character if you are interested.

Related

How to use LIKE in WHERE clause to get first 5 characters of variable?

I have a variable varchar that always takes in 10 digits. How can I use the LIKE operator to find/use only the first 5 digits of the variable?
my query:
variable IN VARCHAR2
SELECT * FROM items WHERE name LIKE SUBSTRING(variable, 1, 5)
... WHERE name LIKE '12345%'
will match any string that starts 12345. the '%' is a wildcard. You can also use the wildcard to match anywhere in the string: ... WHERE name LIKE '%12345%' will match a string with 12345 anywhere within it.
Edit for completeness: WHERE name LIKE '%12345' will match any string that ends with those five characters.
Try this:
SELECT * FROM items WHERE name LIKE (SUBSTRING(variable, 1, 5) + '%')
I guess you can use LEFT() like this:
SELECT * FROM items WHERE LEFT(name,5)=LEFT(variable,5);
Or if you you want to use LIKE with a wildcard, you can do this:
SELECT * FROM items WHERE name LIKE CONCAT(LEFT(variable,5),'%')
A few more example in the Demo fiddle
Edit: The above solution is for MySQL/MariaDB because earlier the tag of this question have MySQL but it's also my fault for not recognizing OP description of the datatype VARCHAR2. I might as well just post a suggestion related to the rdbms.
So, my first suggestion there using LEFT() however Oracle don't have that function, therefore:
SELECT * FROM items WHERE SUBSTR(name,1,5)=SUBSTR(variable,1,5);
or using concatenation operator
SELECT * FROM items WHERE name LIKE SUBSTR(variable,1,5)||'%'
Demo fiddle

How to avoid selecting row in sql server which has special symbol

How to avoid selecting row which has special symbol like mentioned below .
We can use range of ASCII character as below. CHAR(n) returns character value of integer ASCII code n
SELECT *
FROM yourTable
WHERE ID NOT LIKE '%['+CHAR(32) +'-'+CHAR(126)+']%'
OR Name NOT LIKE '%['+CHAR(32) +'-'+CHAR(126)+']%';
Refer ASCII characters
You could use SQL Server's enhanced LIKE operator:
SELECT *
FROM yourTable
WHERE ID NOT LIKE '%[^A-Za-z0-9_-]%' AND Name NOT LIKE '%[^A-Za-z0-9_-]%';
This would select only rows where both ID and Name columns do not contain any special characters. Special characters here are defined as anything other alphanumeric, underscore, and hyphen.

What does 'tableName.column like '?*'' represent Microsoft Access

Something like
SELECT * FROM table WHERE table.column LIKE '?*'
What does that mean?
The question mark ? character will only search for a single character, and asterisk * character will search for all characters.
This combination will search for table.column containing atleast 1 character.

Regular Expression Pattern for Search in SQL

I want to search a table which has file name(s) with a {Numerical Pattern String}.PDF.
Example: 1.PDF, 12.PDF, 123.PDF 1234.PDF etc.....
select * from web_pub_subfile where file_name like '[0-9]%[^a-z].pdf'
But above SQL Query is resulting even these kind of files
1801350 Ortho.pdf
699413.processing2.pdf
15-NOE-301.pdf
Could any one help me what I am missing here.
One way to do it is getting the substring before the file extension and checking if it is numeric. This solution only works well if there is only one . character in the file name.
select * from web_pub_subfile
where isnumeric(left(file_name,charindex('.',file_name)-1)) = 1
Note:
ISNUMERIC returns 1 for some characters that are not numbers, such as plus (+), minus (-), and valid currency symbols such as the dollar sign ($).
To handle file names with mutliple . characters and if there is always a .filetype extension, use
select * from web_pub_subfile
where isnumeric(left(file_name,len(file_name)-charindex('.',reverse(file_name)))) = 1
and charindex('.',file_name) > 0
Sample demo
As suggested by #Blorgbeard in the comments, to avoid the use of isnumeric, use
select * from web_pub_subfile
where left(file_name,len(file_name)-charindex('.',reverse(file_name))) NOT LIKE '%[^0-9]%'
and len(left(file_name,len(file_name)-charindex('.',reverse(file_name)))) > 0
You can't really do what you are trying to do using plain out of the box sql. The reason you are seeing those results is that the % character matches any character, any number of times. It's not like * in a regex which matches the pervious character 0 or more times.
Your best option would probably be to create some CLR functions that implement regex functionality on the SQL Server side. You can take a look at this link to find a good place to start.
Depending on your version if 2012+, you could use Try_Convert()
select * from web_pub_subfile where Try_Convert(int,replace(file_name,'.pdf',''))>0
Declare #web_pub_subfile table (file_name varchar(100))
Insert Into #web_pub_subfile values
('1801350 Ortho.pdf'),
('699413.processing2.pdf'),
('15-NOE-301.pdf'),
('1.pdf'),
('1234.pdf')
select * from #web_pub_subfile where Try_Convert(int,replace(file_name,'.pdf',''))>0
Returns
file_name
1.pdf
1234.pdf

SQL statement with LIKE

I would like to select all records that have an underscore character in their 11th character,
so i try this:
SELECT * FROM "BOM_SUB_LEVEL" where TOP_CODE like '%%%%%%%%%%_%%%'
but this doesnt work as expected, can someone help?
Just use the "SUBSTRING" function :
SELECT * FROM "BOM_SUB_LEVEL" where SUBSTRING(TOP_CODE, 11, 1) = "_"
Marc
For a single character wildcard use _. For multiple characters wildcards, use %. To escape a "real" appearance of _, use \_ (thanks Bill!).
Try the following code:
SELECT * FROM "BOM_SUB_LEVEL" where TOP_CODE like '___________\_%'
To further elaborate following Dav's comment, note that '%%%' is exactly the same as '%', since by definition '%' covers multiple characters.
pervasive uses _ to match any single character and \_ to actually match an underscore.
so the select would be:
SELECT * FROM "BOM_SUB_LEVEL" where TOP_CODE like '___________\_%'
LIKE % can mean any number of characters, use LIKE _ to mean just one. Since you're looking for an underscore, you need to escape it with !.
SELECT * FROM BOM_SUB_LEVEL WHERE TOP_CODE LIKE '__________!_%'
The % is not a per character wildcard, its a beginning and end of string wild card.
i.e. if I want to find all rows that have "car" in them, I would do this:
Select * from myTable where myCol LIKE '%car%'
If I wanted just the rows that STARTED with car:
Select * from myTable where myCol LIKE 'car%'
and ended with car:
Select * from myTable where myCol LIKE '%car'
% is a wildcard and can replace an character, or combination of characters. Use ? instead which replaces a single character.
You can try something like: (play with the numbers, I don't have pervasive to test with)
SELECT *
FROM BOM_SUB_LEVEL
where SUBSTRING(TOP_CODE, 11,1) = '-'