sqldiff.exe: Database Difference Utility

1. Usage

The sqldiff.exe binary is a command-line utility program that displays the differences between SQLite databases. Example usage:

sqldiff [options] database1.sqlite database2.sqlite

The usual output is an SQL script that will transform database1.sqlite (the "source" database) into database2.sqlite (the "destination" database). This behavior can be altered using command-line switches:

--changeset FILE

Do not write changes to standard output. Instead, write a (binary) changeset file into FILE. The changeset can be interpreted using the session extension to SQLite.

--lib LIBRARY

-L LIBRARY

Load the shared library or DLL file LIBRARY into SQLite prior to computing the differences. This can be used to add application-defined collating sequences that are required by the schema.

--primarykey

Use the schema-defined PRIMARY KEY instead of the rowid to pair rows in the source and destination database. (See additional explanation below.)

--schema

Show only differences in the schema not the table content

--summary

Show how many rows have changed on each table, but do not show the actual changes

--table TABLE

Show only the differences in content for TABLE, not for the entire database

--transaction

Wrap SQL output in a single large transaction

--vtab

Add support for handling FTS3, FTS5 and rtree virtual tables. See below for details.

2. How It Works

The sqldiff.exe utility works by finding rows in the source and destination that are logical "pairs". The default behavior is to treat two rows as pairs if they are in tables with the same name and they have the same rowid, or in the case of a WITHOUT ROWID table if they have the same PRIMARY KEY. Any differences in the content of paired rows are output as UPDATEs. Rows in the source database that could not be paired are output as DELETEs. Rows in the destination database that could not be paired are output as INSERTs.

The --primarykey flag changes the pairing algorithm slightly so that the schema-declared PRIMARY KEY is always used for pairing, even on tables that have a rowid. This is often a better choice for finding differences, however it can lead to missed differences in the case of rows that have one or more PRIMARY KEY columns set to NULL.

3. Limitations

1. The sqldiff.exe utility does not compute changesets for either: rowid tables for which the rowid is inaccessible; or tables which have no explicit primary key. Given the --changeset option, sqldiff omits them from the comparison. Examples of such tables are:

   CREATE TABLE NilChangeset (
      -- inaccessible rowid due to hiding its aliases
      "rowid" TEXT,
      "oid" TEXT,
      "_rowid_" TEXT
   );
   

   and

   CREATE TABLE NilChangeset (
      -- no explicit primary key
      "authorId" TEXT,
      "bookId" TEXT
   );
   

   When sqldiff is made to compare only such tables, no error occurs. However, the result may be unexpected. For example, the effect of this invocation:

   sqldiff --changeset CHANGESET_OUT --table NilChangeset db1.sdb db2.sdb
   

   will be to produce an empty file named "CHANGESET_OUT". See session limitations for details.

2. The sqldiff.exe utility does not (currently) display differences in TRIGGERs or VIEWs.

3. By default, differences in the schema or content of virtual tables are not reported on.

   However, if a virtual table implementation creates real tables (sometimes referred to as "shadow" tables) within the database to store its data in, then sqldiff.exe does calculate the difference between these. This can have surprising effects if the resulting SQL script is then run on a database that is not exactly the same as the source database. For several of SQLite's bundled virtual tables (FTS3, FTS5, rtree and others), the surprising effects may include corruption of the virtual table content.

   If the --vtab option is passed to sqldiff.exe, then it ignores all underlying shadow tables belonging to an FTS3, FTS5 or rtree virtual table and instead includes the virtual table differences directly.