SQL queries

SQL (Structured Query Language) manages data stored in relational databases through simple, declarative statements. Use the ANSI SQL-92 standard for querying ODBC databases and embedded databases.

SQL grammar

The top-down SQL parser is based on a simplified version of standard SQL grammar because:
  • FactoryTalk Optix supports only a subset of standard SQL.
  • The simplified standard SQL grammar best fits to the top-down parser implementation.
  • The simplified standard SQL grammar facilitates the readability of the resulting AST.

Statements

TIP: Escape column names with special characters in this way:
SELECT MAX(
"/Variable1"
) FROM mystore.recipeschema1;
FactoryTalk Optix Studio
 supports these statements in SQL queries:
  • Use the
    SELECT
     statement to select data from a database. This statement supports both column names and value expressions, as well as the
    ALL
     and
    DISTINCT
     set quantifiers.
    Examples: 
    SELECT *, FROM Table1
SELECT *, Timestamp AS T FROM Table1
SELECT Column1 FROM Table1
SELECT *, 10 FROM Table1
SELECT 10, * FROM Table1
SELECT *, 'text value' FROM Table1
SELECT COUNT(*) FROM Table1
SELECT DINSTINCT Column1 FROM Table1
    NOTE: The
    CASE
     and
    WHEN
     expressions are not supported.
    IMPORTANT:
    The SQL standard does not allow combining
    *
     with derived columns.
    SELECT *, Timestamp FROM Table1
    SELECT *, ROW_NUMBER() OVER (ORDER BY Timestamp) FROM DataLogger
    FactoryTalk Optix
     supports this standard to allow any query manipulation for Sparkline charts.
  • Use the
    DELETE
     statement to delete any existing records in a table.
    Examples: 
    SELECT * FROM Table1 WHERE ...
DELETE FROM Table1 WHERE ...
UPDATE Table1 SET column1 = 10 WHERE column2 > 20 ...
  • Use the
    UPDATE
     statement to modify any existing records in a table.
    Examples: 
    SELECT * FROM Table1 WHERE ...
DELETE FROM Table1 WHERE ...
UPDATE Table1 SET column1 = 10 WHERE column2 > 20 ...
FactoryTalk Optix Studio
 partially supports these statements in SQL queries:
  • Use the
    CREATE
     statement to create tables or tables in databases. This statement partially supports
    TABLE
     and
    INDEX
    .
    Examples: 
    CREATE UNIQUE INDEX Table_Index ON Table1 (Column1)
CREATE INDEX IF NOT EXISTS Table_Index ON Table1 (Column1)
    CREATE TABLE TempTable AS SELECT Column1 FROM MyTable WHERE Column1 > 5
CREATE TEMPORARY TABLE TempTable AS SELECT Column1 FROM MyTable WHERE Column1 > 5
    CREATE TEMPORARY TABLE Table2 AS SELECT Column1 FROM Table1 WHERE Column1 > 5
CREATE UNIQUE INDEX Column1_Index ON Table1 (Column1)
DROP TABLE Table1
  • The
    DROP
     statement drops any existing databases or tables in databases.
    Examples: 
    DROP TABLE TemporaryTable
DROP TABLE Table1

Literal values

Supported literal values
Literal value type
Example
Notes
Integral
1520
Do not use thousands separators.
Use the Int32 type.
Real
15.025
Use
.
 as the decimal separator.
Use the Double type.
Real scientific
2.7000001072883606E-1
Boolean
  • true
  • false
Use lowercase for Boolean values.
String
'string value'
Use single quotes to delimit text.
Insert a single quote (
'
) in a text string by escaping the single quote with another single quote. For example, to write
'string'
, use two single quotes
''string''
 in the strings.
Date and time
"2004-05-23T14:25:10.487"
Use the ISO8601 format delimited by double quotes.
The date and time notation is non-standard ANSI SQL used to avoid the
DATE
,
TIME
, and
TIMESTAMP
 keywords without ambiguity.
Regular identifier
  • Column1
  • Grüße
Do not use spaces.
Start with an uppercase or a lowercase letter.
You can use numeric digits, letters, and the underscore character
_
.
Delimited identifier
  • "Water level"
  • "SELECT"
Use double quotes (
"
) as the delimiter.
In the identifier, use any character except for the
"
 character.
Any spaces at the end of the identifier are ignored.

Join

Any kind of joins such as
INNER
,
UNION
,
LEFT
,
RIGHT
,
FULL
,
NATURAL
 and
CROSS
 is supported.
NOTE: Some joins are not fully supported by every DBMS.

Aliases

You can use aliases in the identifiers, derived column subqueries, and tables.
The following are examples of queries with aliases: 
SELECT Timestamp AS T FROM Table1 WHERE ...
SELECT * FROM Table1 AS T WHERE ...
SELECT * FROM (SELECT * FROM Table1) AS SubQuery WHERE ...
CREATE TABLE Table1 AS SELECT Timestamp, Column1 FROM Table2

Table references

The
FROM
 clause supports the following table references.
Supported table references
Reference
Example
Fully qualified reference
... FROM table_name
... FROM schema_name.table_name
... FROM database_name.schema.name.table_name
Sub query
... FROM (SELECT * FROM Table2) WHERE ...
NOTE: Fully qualified names are also supported as column references.
Examples: 
... WHERE Table1.Column1 > 10... 
WHERE Table1."order id" > 10
SELECT * FROM Table1 AS T WHERE T."order id" > 10
... WHERE "my table"."order id" > 10

Operators

Supported operators
Operator
Example
IN
SELECT * FROM Table1 WHERE Column1 IN (10, 20, 30)
BETWEEN
SELECT * FROM Table1 WHERE Column1 BETWEEN 100 AND 200
LIKE
... WHERE column1 LIKE '%a'
... WHERE column1 LIKE '%a%'
... WHERE column1 LIKE '%bbpi!%ppo%' ESCAPE '!'
EXISTS
... WHERE EXISTS (SELECT Table1.Column1 FROM Table2)
IS
... WHERE column1 IS NULL
NOT
... WHERE column1 IS NOT NULL
... WHERE column1 NOT IS (10, 20)
... WHERE column1 NOT BETWEEN 100 AND 200
The
NOT
 operator applies to all other operators.
NOTE:
ANY
 and
ALL
 operators are not supported.

Window functions

Supported window functions
Function
Example
ROW_NUMBER
ROW_NUMBER() OVER (ORDER BY Column2)
ROW_NUMBER() OVER (PARTITION ON Column1 ORDER BY Column2)
ROW_NUMBER() OVER (window_name PARTITION ON Column1 ORDER BY Column2)
RANK
RANK() OVER (PARTITION ON Column1 ORDER BY Column2)
DENSE_RANK
DENSE_RANK() OVER (PARTITION ON Column1 ORDER BY Column2)
AVG
AVG(Column1) OVER (PARTITION BY Column2)
SUM
,
MIN
,
MAX
SUM(Column1) OVER (PARTITION BY Column2)
COUNT
COUNT(Column1) OVER (PARTITION BY Column2)
COUNT(*) OVER (PARTITION BY Column2)
NOTE: The
PARTITION BY
 clause supports only column identifiers.

Tokens

Supported tokens
Token
Regular expression
regular_identifier
[a-zA-Z][a-zA-Z0-9_]*
delimited_identifier
".*?"
real
boolean
Other trivial tokens
Regex for single char or single-word tokens
NOTE: The lexer does not recognize date and time values as tokens: they are parsed as
delimited_identifier
. The semantic analyzer extracts any date and time values from the
delimited_identifier
 according to the
r
 regular expression.
Provide Feedback
Have questions or feedback about this documentation? Please submit your feedback here.