PostgreSQL Query Catalog

Find row level update triggers that incorrectly implement update of password hash. It should not be that the new password hash is calculated based on the existing hash.

If the name of an object has the prefix that refers to the type of the object (for instance, primary key constraint or foreign key constraint), then you should use references to the correct object type. Find prefixes of constraint names and index names that incorrectly refer to the type of the object. For instance, incorrect would be to use chk_ as the prefix of an index name or pk_ as the prefix of a check constraint name.

This query identifies ineffectual CHECK constraints on base and foreign table columns that incorrectly attempt to prohibit empty or whitespace-only strings using the predicate trim(column_name) IS NOT NULL. Due to PostgreSQL's strict distinction between an empty string ('') and NULL, this check is a tautology; trim('') evaluates to '', and the condition '' IS NOT NULL is always true. The query finds these logically flawed constraints, which fail to provide any data validation and permit the insertion of the exact values they were intended to prevent.

Find user-defined routines that possibly use incorrect name of a system-defined function (currenttimestamp (correct is current_timestamp), currentdate (correct is current_date), currenttime (correct is current_time), local_time (correct is localtime), local_timestamp (correct is localtimestamp),localdate (there is no such function),local_date (there is no such function), sessionuser (correct is session_user), ucase (correct is upper), lcase (correct is lower)). The problem can arise only if the routine uses dynamic SQL. In case of static SQL the DBMS checks the SQL statemen at the creation time and finds out that for instance, SELECT Count(*) AS cnt FROM Emp WHERE hiredate<=currentdate; is incorrect statement because currentdate is not a function name and there is no column currentdate in the table Emp.

Find the use of regular expressions where logical or is incorrectly specified, i.e., (| or |).

If the name of an object has the suffix that refers to the type of the object (for instance, primary key constraint or foreign key constraint), then you should use references to the correct object type. Find suffixes of constraint names and index names that incorrectly refer to the type of the object. For instance, incorrect would be to use _chk as the suffix of an index name or _pk as the suffix of a check constraint name.

This query identifies user-defined routines and derived tables (views or materialized views) that use the COUNT(*) aggregate function (without a FILTER clause) alongside an OUTER JOIN and a GROUP BY clause. This combination is a common SQL anti-pattern. When grouping with an outer join, COUNT(*) counts the rows themselves, including the NULL-filled rows generated for non-matching records. Consequently, it inaccurately reports a count of 1 for completely empty groups instead of the correct count of 0. To get accurate results, developers should count a specific non-null column from the joined table.

This query identifies CHECK constraints that use non-deterministic time functions (such as now or current_date) in a way that causes initially valid data to become invalid over time. For example, a constraint like localtimestamp(0) > end_date is an anti-pattern because advancing time will eventually violate it. However, safe implementations—such as birth_date < CURRENT_DATE—are permitted, because once the condition is met, it remains true indefinitely as time moves forward.

Make sure that you do not drop INFORMATION_SCHEMA schema. In this case most of the design checking queries will not work. This schema automatically exists in all databases.

For the sake of making code better understandable follow naming conventions.

This query analyzes the semantic consistency of routine signatures by identifying named input parameters that share an identical identifier but possess divergent data types across different routines. While this pattern is a prerequisite for valid routine overloading (polymorphism), it can also indicate a lack of standardization in the data dictionary. For example, using the parameter name status to denote an INTEGER in one context and TEXT in another creates ambiguity regarding the parameter's expected domain. The query results should be audited to distinguish intentional overloading from naming inconsistencies.

This query identifies varchar columns in base tables whose names suggest they store national identification numbers (personal codes), specifically in tables that also include a country code column. This combination implies that the table stores personal codes from various countries. To safely accommodate different international formats, the field size of the personal code column should be at least 20 characters. The query flags any such columns that are too short.

This query performs a basic structural assessment of the database schema by counting the number of user-defined base tables. It verifies whether the data model meets a minimum complexity threshold, requiring the existence of at least 7 distinct base tables. This metric serves as a proxy for the scope and depth of the implemented domain model.

This query assesses the utilization of user-defined domains within the database schema. It verifies a specific structural requirement: the database must contain at least one user-defined domain that is referenced by at least two distinct non-foreign key columns in base tables. This metric serves as an indicator of proper domain reuse and data type standardization. The query validates whether the schema design effectively leverages domains to enforce consistent data definitions across multiple attributes.

This query assesses the utilization of PostgreSQL's Foreign Data Wrapper (FDW) capabilities. It verifies the existence of at least 2 user-defined foreign tables within the database schema. This requirement ensures that the implemented solution demonstrates the ability to integrate and query data from external sources, extending the data model beyond local storage.

This query performs a quantity-based assessment of the database's procedural logic. It verifies whether the schema meets a minimum complexity requirement by counting the number of user-defined, non-trigger routines (functions and procedures). The check passes only if the count is equal to or greater than the threshold of 4. This metric is used to ensure a baseline level of backend logic implementation.

This query assesses the database's adherence to modern SQL standards regarding procedural code. It verifies whether the schema contains a minimum of 3 user-defined, non-trigger routines that are specifically written in the SQL language and utilize a SQL-standard body (defined using BEGIN ATOMIC ... END). This excludes routines written in PL/pgSQL or those using the legacy string-literal body definition. The metric ensures a baseline adoption of the standardized, portable syntax for SQL routines introduced in newer PostgreSQL versions.

This query assesses the extent of active logic implementation within the database schema. It verifies that there are at least 3 user-defined triggers or rewrite rules present in the database. These objects represent automated tasks that for example enforce complex integrity constraints. Meeting this threshold indicates a baseline competency in utilizing PostgreSQL's event-driven capabilities.

This query assesses the extent of active logic implementation within the database schema. It verifies that there are user-defined triggers or rewrite rules present for at least 3 tasks. These objects represent automated tasks that for example enforce complex integrity constraints. Meeting this threshold indicates a baseline competency in utilizing PostgreSQL's event-driven capabilities.

This query assesses the use of abstraction layers within the database schema. It verifies that there are at least 4 user-defined views present. Views are essential for simplifying complex queries, restricting data access, and presenting specific data perspectives to applications. Meeting this threshold indicates a sufficient implementation of data abstraction and query encapsulation.