I took me a while to find out that it’s not very easy to find out how big a sqlite table is on disk. But luckily the sqlite creators have a simple tool for that. It’s called sqlite3_analyzer and is available in precompiled format for Linux, Windows and OSX on SQLite’s Download section

Confusing: There also seems to be a commercial graphical sqlite tool in a windows only variant that has the same name, but it doesn’t have functionality to see how big tables are. http://www.kraslabs.com/sqlite_analyzer.php

When you run sqlite3_analyzer it will create a .sql file with info about the database. You can insert the .sql file in a new database to get more detailed info about the tables sizes and some other metrics.

./sqlite3_analyze path_to_db/database.sqlite > dbinfo.sql

It will produce a sql file with a big sql comment on top. After the comment some SQL statements are printed that will create a table “space_used” with more detailed info.

Example:
the example DB is 381Mb on disk. The Binary table is used to store some binary data.

Page size in bytes................................ 1024      
Pages in the whole file (measured)................ 390022    
Pages in the whole file (calculated).............. 390022    
Pages that store data............................. 390022     100.0% 
Pages on the freelist (per header)................ 0            0.0% 
Pages on the freelist (calculated)................ 0            0.0% 
Pages of auto-vacuum overhead..................... 0            0.0% 
Number of tables in the database.................. 51        
Number of indices................................. 2         
Number of defined indices......................... 0         
Number of implied indices......................... 2         
Size of the file in bytes......................... 399382528 
Bytes of user payload stored...................... 390583107   97.8% 

*** Page counts for all tables with their indices *****************************

BINARY............................................ 301833      77.4% 
INSTALL........................................... 70156       18.0% 
MAIL_QUEUE........................................ 15847        4.1% 
ACCOUNT_INFO...................................... 602          0.15% 
NODE_info......................................... 591          0.15% 
etc.

I used it to clean up some DB’s with a lot of tables. After deleting a lot of records I noticed that the file size didn’t shrink, so as a reminder for myself; you have to execute one SQLite SQL statement to free up disk space after a lot of delete statements: VACUUM;