xref: /sqlite-3.40.0/README.md (revision e89feee5)
1<h1 align="center">SQLite Source Repository</h1>
2
3This repository contains the complete source code for the
4[SQLite database engine](https://sqlite.org/).  Some test scripts
5are also included.  However, many other test scripts
6and most of the documentation are managed separately.
7
8SQLite [does not use Git](https://sqlite.org/whynotgit.html).
9If you are reading this on GitHub, then you are looking at an
10unofficial mirror. See <https://sqlite.org/src> for the official
11repository.
12
13## Obtaining The Code
14
15SQLite sources are managed using the
16[Fossil](https://www.fossil-scm.org/), a distributed version control system
17that was specifically designed to support SQLite development.
18If you do not want to use Fossil, you can download tarballs or ZIP
19archives or [SQLite archives](https://sqlite.org/cli.html#sqlar) as follows:
20
21  *  Lastest trunk check-in as
22     [Tarball](https://www.sqlite.org/src/tarball/sqlite.tar.gz),
23     [ZIP-archive](https://www.sqlite.org/src/zip/sqlite.zip), or
24     [SQLite-archive](https://www.sqlite.org/src/sqlar/sqlite.sqlar).
25
26  *  Latest release as
27     [Tarball](https://www.sqlite.org/src/tarball/sqlite.tar.gz?r=release),
28     [ZIP-archive](https://www.sqlite.org/src/zip/sqlite.zip?r=release), or
29     [SQLite-archive](https://www.sqlite.org/src/sqlar/sqlite.sqlar?r=release).
30
31  *  For other check-ins, substitute an appropriate branch name or
32     tag or hash prefix for "release" in the URLs of the previous
33     bullet.  Or browse the [timeline](https://www.sqlite.org/src/timeline)
34     to locate the check-in desired, click on its information page link,
35     then click on the "Tarball" or "ZIP Archive" links on the information
36     page.
37
38If you do want to use Fossil to check out the source tree,
39first install Fossil version 2.0 or later.
40(Source tarballs and precompiled binaries available
41[here](https://www.fossil-scm.org/fossil/uv/download.html).  Fossil is
42a stand-alone program.  To install, simply download or build the single
43executable file and put that file someplace on your $PATH.)
44Then run commands like this:
45
46        mkdir ~/sqlite
47        cd ~/sqlite
48        fossil clone https://www.sqlite.org/src sqlite.fossil
49        fossil open sqlite.fossil
50
51After setting up a repository using the steps above, you can always
52update to the lastest version using:
53
54        fossil update trunk   ;# latest trunk check-in
55        fossil update release ;# latest official release
56
57Or type "fossil ui" to get a web-based user interface.
58
59## Compiling
60
61First create a directory in which to place
62the build products.  It is recommended, but not required, that the
63build directory be separate from the source directory.  Cd into the
64build directory and then from the build directory run the configure
65script found at the root of the source tree.  Then run "make".
66
67For example:
68
69        tar xzf sqlite.tar.gz    ;#  Unpack the source tree into "sqlite"
70        mkdir bld                ;#  Build will occur in a sibling directory
71        cd bld                   ;#  Change to the build directory
72        ../sqlite/configure      ;#  Run the configure script
73        make                     ;#  Run the makefile.
74        make sqlite3.c           ;#  Build the "amalgamation" source file
75        make test                ;#  Run some tests (requires Tcl)
76
77See the makefile for additional targets.
78
79The configure script uses autoconf 2.61 and libtool.  If the configure
80script does not work out for you, there is a generic makefile named
81"Makefile.linux-gcc" in the top directory of the source tree that you
82can copy and edit to suit your needs.  Comments on the generic makefile
83show what changes are needed.
84
85## Using MSVC
86
87On Windows, all applicable build products can be compiled with MSVC.
88First open the command prompt window associated with the desired compiler
89version (e.g. "Developer Command Prompt for VS2013").  Next, use NMAKE
90with the provided "Makefile.msc" to build one of the supported targets.
91
92For example:
93
94        mkdir bld
95        cd bld
96        nmake /f Makefile.msc TOP=..\sqlite
97        nmake /f Makefile.msc sqlite3.c TOP=..\sqlite
98        nmake /f Makefile.msc sqlite3.dll TOP=..\sqlite
99        nmake /f Makefile.msc sqlite3.exe TOP=..\sqlite
100        nmake /f Makefile.msc test TOP=..\sqlite
101
102There are several build options that can be set via the NMAKE command
103line.  For example, to build for WinRT, simply add "FOR_WINRT=1" argument
104to the "sqlite3.dll" command line above.  When debugging into the SQLite
105code, adding the "DEBUG=1" argument to one of the above command lines is
106recommended.
107
108SQLite does not require [Tcl](http://www.tcl.tk/) to run, but a Tcl installation
109is required by the makefiles (including those for MSVC).  SQLite contains
110a lot of generated code and Tcl is used to do much of that code generation.
111
112## Source Code Tour
113
114Most of the core source files are in the **src/** subdirectory.  The
115**src/** folder also contains files used to build the "testfixture" test
116harness. The names of the source files used by "testfixture" all begin
117with "test".
118The **src/** also contains the "shell.c" file
119which is the main program for the "sqlite3.exe"
120[command-line shell](https://sqlite.org/cli.html) and
121the "tclsqlite.c" file which implements the
122[Tcl bindings](https://sqlite.org/tclsqlite.html) for SQLite.
123(Historical note:  SQLite began as a Tcl
124extension and only later escaped to the wild as an independent library.)
125
126Test scripts and programs are found in the **test/** subdirectory.
127Addtional test code is found in other source repositories.
128See [How SQLite Is Tested](http://www.sqlite.org/testing.html) for
129additional information.
130
131The **ext/** subdirectory contains code for extensions.  The
132Full-text search engine is in **ext/fts3**.  The R-Tree engine is in
133**ext/rtree**.  The **ext/misc** subdirectory contains a number of
134smaller, single-file extensions, such as a REGEXP operator.
135
136The **tool/** subdirectory contains various scripts and programs used
137for building generated source code files or for testing or for generating
138accessory programs such as "sqlite3_analyzer(.exe)".
139
140### Generated Source Code Files
141
142Several of the C-language source files used by SQLite are generated from
143other sources rather than being typed in manually by a programmer.  This
144section will summarize those automatically-generated files.  To create all
145of the automatically-generated files, simply run "make target&#95;source".
146The "target&#95;source" make target will create a subdirectory "tsrc/" and
147fill it with all the source files needed to build SQLite, both
148manually-edited files and automatically-generated files.
149
150The SQLite interface is defined by the **sqlite3.h** header file, which is
151generated from src/sqlite.h.in, ./manifest.uuid, and ./VERSION.  The
152[Tcl script](http://www.tcl.tk) at tool/mksqlite3h.tcl does the conversion.
153The manifest.uuid file contains the SHA3 hash of the particular check-in
154and is used to generate the SQLITE\_SOURCE\_ID macro.  The VERSION file
155contains the current SQLite version number.  The sqlite3.h header is really
156just a copy of src/sqlite.h.in with the source-id and version number inserted
157at just the right spots. Note that comment text in the sqlite3.h file is
158used to generate much of the SQLite API documentation.  The Tcl scripts
159used to generate that documentation are in a separate source repository.
160
161The SQL language parser is **parse.c** which is generate from a grammar in
162the src/parse.y file.  The conversion of "parse.y" into "parse.c" is done
163by the [lemon](./doc/lemon.html) LALR(1) parser generator.  The source code
164for lemon is at tool/lemon.c.  Lemon uses the tool/lempar.c file as a
165template for generating its parser.
166
167Lemon also generates the **parse.h** header file, at the same time it
168generates parse.c. But the parse.h header file is
169modified further (to add additional symbols) using the ./addopcodes.tcl
170Tcl script.
171
172The **opcodes.h** header file contains macros that define the numbers
173corresponding to opcodes in the "VDBE" virtual machine.  The opcodes.h
174file is generated by the scanning the src/vdbe.c source file.  The
175Tcl script at ./mkopcodeh.tcl does this scan and generates opcodes.h.
176A second Tcl script, ./mkopcodec.tcl, then scans opcodes.h to generate
177the **opcodes.c** source file, which contains a reverse mapping from
178opcode-number to opcode-name that is used for EXPLAIN output.
179
180The **keywordhash.h** header file contains the definition of a hash table
181that maps SQL language keywords (ex: "CREATE", "SELECT", "INDEX", etc.) into
182the numeric codes used by the parse.c parser.  The keywordhash.h file is
183generated by a C-language program at tool mkkeywordhash.c.
184
185The **pragma.h** header file contains various definitions used to parse
186and implement the PRAGMA statements.  The header is generated by a
187script **tool/mkpragmatab.tcl**. If you want to add a new PRAGMA, edit
188the **tool/mkpragmatab.tcl** file to insert the information needed by the
189parser for your new PRAGMA, then run the script to regenerate the
190**pragma.h** header file.
191
192### The Amalgamation
193
194All of the individual C source code and header files (both manually-edited
195and automatically-generated) can be combined into a single big source file
196**sqlite3.c** called "the amalgamation".  The amalgamation is the recommended
197way of using SQLite in a larger application.  Combining all individual
198source code files into a single big source code file allows the C compiler
199to perform more cross-procedure analysis and generate better code.  SQLite
200runs about 5% faster when compiled from the amalgamation versus when compiled
201from individual source files.
202
203The amalgamation is generated from the tool/mksqlite3c.tcl Tcl script.
204First, all of the individual source files must be gathered into the tsrc/
205subdirectory (using the equivalent of "make target_source") then the
206tool/mksqlite3c.tcl script is run to copy them all together in just the
207right order while resolving internal "#include" references.
208
209The amalgamation source file is more than 200K lines long.  Some symbolic
210debuggers (most notably MSVC) are unable to deal with files longer than 64K
211lines.  To work around this, a separate Tcl script, tool/split-sqlite3c.tcl,
212can be run on the amalgamation to break it up into a single small C file
213called **sqlite3-all.c** that does #include on about seven other files
214named **sqlite3-1.c**, **sqlite3-2.c**, ..., **sqlite3-7.c**.  In this way,
215all of the source code is contained within a single translation unit so
216that the compiler can do extra cross-procedure optimization, but no
217individual source file exceeds 32K lines in length.
218
219## How It All Fits Together
220
221SQLite is modular in design.
222See the [architectural description](http://www.sqlite.org/arch.html)
223for details. Other documents that are useful in
224(helping to understand how SQLite works include the
225[file format](http://www.sqlite.org/fileformat2.html) description,
226the [virtual machine](http://www.sqlite.org/opcode.html) that runs
227prepared statements, the description of
228[how transactions work](http://www.sqlite.org/atomiccommit.html), and
229the [overview of the query planner](http://www.sqlite.org/optoverview.html).
230
231Years of effort have gone into optimizating SQLite, both
232for small size and high performance.  And optimizations tend to result in
233complex code.  So there is a lot of complexity in the current SQLite
234implementation.  It will not be the easiest library in the world to hack.
235
236Key files:
237
238  *  **sqlite.h.in** - This file defines the public interface to the SQLite
239     library.  Readers will need to be familiar with this interface before
240     trying to understand how the library works internally.
241
242  *  **sqliteInt.h** - this header file defines many of the data objects
243     used internally by SQLite.  In addition to "sqliteInt.h", some
244     subsystems have their own header files.
245
246  *  **parse.y** - This file describes the LALR(1) grammar that SQLite uses
247     to parse SQL statements, and the actions that are taken at each step
248     in the parsing process.
249
250  *  **vdbe.c** - This file implements the virtual machine that runs
251     prepared statements.  There are various helper files whose names
252     begin with "vdbe".  The VDBE has access to the vdbeInt.h header file
253     which defines internal data objects.  The rest of SQLite interacts
254     with the VDBE through an interface defined by vdbe.h.
255
256  *  **where.c** - This file (together with its helper files named
257     by "where*.c") analyzes the WHERE clause and generates
258     virtual machine code to run queries efficiently.  This file is
259     sometimes called the "query optimizer".  It has its own private
260     header file, whereInt.h, that defines data objects used internally.
261
262  *  **btree.c** - This file contains the implementation of the B-Tree
263     storage engine used by SQLite.  The interface to the rest of the system
264     is defined by "btree.h".  The "btreeInt.h" header defines objects
265     used internally by btree.c and not published to the rest of the system.
266
267  *  **pager.c** - This file contains the "pager" implementation, the
268     module that implements transactions.  The "pager.h" header file
269     defines the interface between pager.c and the rest of the system.
270
271  *  **os_unix.c** and **os_win.c** - These two files implement the interface
272     between SQLite and the underlying operating system using the run-time
273     pluggable VFS interface.
274
275  *  **shell.c.in** - This file is not part of the core SQLite library.  This
276     is the file that, when linked against sqlite3.a, generates the
277     "sqlite3.exe" command-line shell.  The "shell.c.in" file is transformed
278     into "shell.c" as part of the build process.
279
280  *  **tclsqlite.c** - This file implements the Tcl bindings for SQLite.  It
281     is not part of the core SQLite library.  But as most of the tests in this
282     repository are written in Tcl, the Tcl language bindings are important.
283
284  *  **test*.c** - Files in the src/ folder that begin with "test" go into
285     building the "testfixture.exe" program.  The testfixture.exe program is
286     an enhanced Tcl shell.  The testfixture.exe program runs scripts in the
287     test/ folder to validate the core SQLite code.  The testfixture program
288     (and some other test programs too) is build and run when you type
289     "make test".
290
291  *  **ext/misc/json1.c** - This file implements the various JSON functions
292     that are build into SQLite.
293
294There are many other source files.  Each has a succinct header comment that
295describes its purpose and role within the larger system.
296
297
298## Contacts
299
300The main SQLite webpage is [http://www.sqlite.org/](http://www.sqlite.org/)
301with geographically distributed backups at
302[http://www2.sqlite.org/](http://www2.sqlite.org) and
303[http://www3.sqlite.org/](http://www3.sqlite.org).
304