of the recognizer engine.
[forum:/forumpost/f50aecd5e8|Forum post f50aecd5e8].
FossilOrigin-Name: 0772ddf56713d013cd1bd44f9c75977ca14f852e3a8f038b0a6b9814f6519d79
This folder contains a collection of smaller loadable extensions.
See https://www.sqlite.org/loadext.html for instructions on how
to compile and use loadable extensions.
Each extension in this folder is implemented in a single file of C code.
Each source file contains a description in its header comment. See the
header comments for details about each extension. Additional notes are
as follows:
carray.c — This module implements the
carray table-valued function.
It is a good example of how to go about implementing a custom
table-valued function.
dbdump.c — This is not actually a loadable extension, but
rather a library that implements an approximate equivalent to the
".dump" command of the
command-line shell.
json1.c — Various SQL functions and table-valued functions
for processing JSON. This extension is already built into the
SQLite amalgamation. See
https://sqlite.org/json1.html for additional information.
memvfs.c — This file implements a custom
VFS that stores an entire database
file in a single block of RAM. It serves as a good example of how
to implement a simple custom VFS.
rot13.c — This file implements the very simple rot13()
substitution function. This file makes a good template for implementing
new custom SQL functions for SQLite.
series.c — This is an implementation of the
"generate_series" virtual table.
It can make a good template for new custom virtual table implementations.
shathree.c — An implementation of the sha3() and
sha3_query() SQL functions. The file is named "shathree.c" instead
of "sha3.c" because the default entry point names in SQLite are based
on the source filename with digits removed, so if we used the name
"sha3.c" then the entry point would conflict with the prior "sha1.c"
extension.
unionvtab.c — Implementation of the unionvtab and
swarmvtab virtual tables.
These virtual tables allow a single
large table to be spread out across multiple database files. In the
case of swarmvtab, the individual database files can be attached on
demand.