Pattern matching and regular expressions in PostgreSQL - Das

In this paper , I will discuss PostgreSQL Pattern matching and regular expressions in .

An important aspect of writing queries is that users should be able to filter and select the data that the system needs to access . Additional or unnecessary data will increase bandwidth and reduce query performance . therefore ,SQL Provides a filter predicate “WHERE”, Users can use it to filter queries and select results that match only filter predicates .

 

PostgreSQL Regular expressions in

In the normal SQL In case of operation ,LIKE Operators are good , But when filtering large databases ,LIKE Operator seems to have some performance problems . Besides , Use LIKE The filter condition of the operator is limited to including wildcards (%) To find patterns .

To overcome this problem ,PostgreSQL Provides a high-level method of pattern matching using regular expressions . For a long time , Regular expressions are widely used in programming languages , However , stay SQL Use these regular expressions in the statement , Make the query highly dynamic , And perform better in large databases .

PostgreSQL The regular expression in is using TILDE ( ~ ) Operator , And use '.* ” As a wildcard operator .

SELECT * from GreaterManchesterCrime WHERE CrimeID = '^.*$';

As you can see , We are PostgreSQL Regular expressions are used in , Used TILDE（~） Operators and wildcards '.*'. This query will start from GreaterManchesterCrime Select all in the table that have valid CrimeID The record of . Because the mode condition is only a wildcard , It will get all records from the table .

in addition , stay PostgreSQL Another thing to note when writing regular expressions in is , Pattern matching statements always begin with "^" Operator starts , With "$" End of symbol . These two operators mark the beginning and end of a regular expression statement . Summarize the following figure , When using wildcards to filter all records , The expression can be implemented as "^.*$".

 

Regular expressions - A string beginning with a number or character

In the last section , We have learned how to implement regular expressions by using wildcards . Now? , We will take a step forward , Try for column CrimeID Get records that begin with characters or numbers .

stay PostgreSQL in , have access to "[a-z]" or "[A-Z]" Pattern to match alphanumeric characters , It depends on what we are trying to match . One thing to note here is that , because PostgreSQL It's case sensitive , therefore , The exact case we want to match must be specified in the pattern .

similarly , For matching numbers , We can use "[0-9]" or "/d". Now we use Regex Filter characters and numbers and have some understanding , Let's continue to implement it in the database .

For matching characters -

SELECT * FROM GreaterManchesterCrime WHERE CrimeID ~ '^[AZ].*$'

CrimeID The field has been filtered , Contains only ID Records beginning with capital letters . Please note that wildcards are used again “ .* ” To express SQL Anything after the first character in the statement .

To filter numbers in statements , We can simply write the following query ：

SELECT * FROM GreaterManchesterCrime WHERE CrimeID ~ '^[0–9].*$'

Only those beginning with numbers CrimeID Records of are filtered to match regular expression conditions .

 

Regular expressions —— String starting with repeated numbers or characters

In the previous section of this article , We have learned how to be in PostgreSQL Using regular expression syntax SQL Inquire about , And how to filter characters and numbers in the first position of the filtered string .

Now? , Let's add some more complex content to the requirements , And filter records that start with multiple characters or numbers .PostgreSQL Just provide the count in curly brackets , You can easily specify the number of repetitions . perhaps , You can also repeat the same character pattern matching multiple times to match your criteria . for example , We want to filter all records that start with two characters . This query can be written as follows .

SELECT * FROM GreaterManchesterCrime WHERE CrimeID ~ '^[AZ] [AZ].*$'

perhaps

SELECT * FROM GreaterManchesterCrime WHERE CrimeID ~ '^[AZ] {2}.*$'

We use numbers in curly braces to indicate the number of times to find patterns in the column . This makes the query very dynamic , Because you can specify any number of characters to search in the query , You will get results .

Again , You can also implement the same logic for repeated values . under these circumstances , Queries can be written as follows .

SELECT * FROM GreaterManchesterCrime WHERE CrimeID ~ '^[0–9]{2}.*$'

This shows the implementation of pattern matching using regular expressions while filtering duplicate values .

 

Regular expressions —— String starting with characters and numbers

In the last part of this article , We will combine what we have seen so far . We will build queries and match patterns that will filter records that begin with characters and are followed by numbers . To form a standard query , You can combine the conditions of alphabetic and numeric characters in regular expressions .SQL The statement can be written as follows .

SELECT * FROM GreaterManchesterCrime WHERE CrimeID ~ '^[AZ][0–9].*$'