Chromium Code Reviews| Index: third_party/sqlite/README.chromium |
| diff --git a/third_party/sqlite/README.chromium b/third_party/sqlite/README.chromium |
| index aeacf146741a8e1aa644f1d5d6f0f10ab41c87f6..412649f11cb691de34e660e220d983bcff166a2e 100644 |
| --- a/third_party/sqlite/README.chromium |
| +++ b/third_party/sqlite/README.chromium |
| @@ -5,194 +5,244 @@ Included In Release: Yes |
| Security Critical: Yes |
| License: Public domain |
| -Instructions for importing a new release of SQLite from sqlite.org. |
| - |
| -Note: our current base version is 3.7.6.3. |
| - |
| -First, you need to be on Linux. |
| - |
| -# Determine the versions of the release you want and the release we currently |
| -# have. (See the VERSION file to determine which release we currently have.) |
| -# You may wish to consult http://www.sqlite.org/changes.html to find out what |
| -# changes have been made in each release. |
| -# Note - this is just an example. Always refer to the version above for our |
| -# real current version. |
| -# Set some variables to remember the versions, e.g.: |
| -BASE=3.7.6.3 |
| -LATEST=3.7.6.4 |
| - |
| -# Get to the src/third_party directory in your Chromium client: |
| -cd src/third_party |
| - |
| -# Download the .tar.gz files for the releases: |
| -# (If the URL changes you might need to find the new one.) |
| -# TODO(shess): Rewrite this to track the new naming format. Meanwhile, |
| -# manually navigate to www.sqlite.org and find downloads, use "legacy" version. |
| -wget http://www.sqlite.org/sqlite-$BASE.tar.gz |
| -wget http://www.sqlite.org/sqlite-$LATEST.tar.gz |
| - |
| -# Extract the vanilla current and desired versions: |
| -tar xzf sqlite-$BASE.tar.gz |
| -tar xzf sqlite-$LATEST.tar.gz |
| - |
| -# Use kdiff3 to merge the changes: |
| -kdiff3 -m sqlite-$BASE sqlite-$LATEST sqlite |
| - |
| -# Resolve any conflicts. Figure out if we've got everything we should |
| -# have (see below), or if we can omit any changes we no longer need. |
| - |
| -# Change to the sqlite directory: |
| -cd sqlite |
| - |
| -# Run the google_generate_amalgamation.sh script: |
| +1) Managing differences between SQLite core and Chromium's version. |
| +2) Making changes to Chromium SQLite. |
| +3) Import new release of SQLite. |
| +4) Running SQLite's test suite within the Chromium checkout. |
| + |
| +--- |
| + |
| +1) Managing differences between SQLite core and Chromium's version. |
| + |
| +Chromium maintains some differences WRT SQLite, for reasons beyond this |
| +document's remit. Some differences are bugs we have found and fixed (and |
| +hopefully upstreamed), some are fixed we've backported from a later version of |
|
michaeln
2015/01/29 02:40:09
fixed s/b fixes
Scott Hess - ex-Googler
2015/01/29 22:08:33
Done.
|
| +SQLite, and some our local changes unlikely to ever be upstreamed. New versions |
| +of SQLite are imported every year or two, at which point the changes need to be |
| +reviewed for continued applicability, and sometimes adjusted to reflect upstream |
| +code changes. |
| + |
| +To this end, the repository contains a reference copy of the SQLite source code |
| +as of the last import, plus a series of patches which can be applied to |
| +re-create the current trunk code. These patches are generated and processed by |
| +git, with the intention of re-creating a changelist series so that importers can |
| +use their regular revision-control knowledge to manage import merges. |
| + |
| +--- |
| + |
| +2) Making changes to Chromium SQLite. |
| + |
| +third_party/sqlite/src is the patched source from SQLite. This is used to |
| +generate the amalgamation, a concatenation of all of the files into a giant |
| +sqlite3.c. To prototype, edit in src/, then call |
| + ./google_generate_amalgamation.sh |
| +to regenerate sqlite3.c. The code in src/ is much easier to edit, and the |
| +SQLite test framework can easily be run. During development it may be |
| +convenient to modify sqlite.gyp (or BUILD.gn) based on src/main.mk to just pull |
| +in the src/ files rather than sqlite3.c. |
| + |
| +Once your patch is complete, squash it down into a reasonable CL, then |
| +re-generate the patches. This is a truncated version of the import flow. The |
| +following is written like a shell script to allow copy/paste to a shell, ignore |
| +comments and change the obvious lines. These instructions should work on Linux |
| +or OSX. They may assume a modern version of git (I'm using 2.2.1). |
| + |
| +# Everything based in sqlite subdir. |
| +cd third_party/sqlite |
| + |
| +BASE=3070603 |
| + |
| +#### Create a reference branch. |
| +git checkout -b sqlite_${BASE} master |
| +rm -rf src |
| +cp -a sqlite-src-${BASE} src |
| +# -f includes ignored files, of which there are a couple. |
| +git add -f src/ |
| +git commit -m "Reset to sqlite-src-${BASE}" |
| +# This branch is unlikely to build. |
| + |
| +#### Create a reference branch with patches applied. |
| +git checkout -b sqlite_${BASE}_patched master |
| +git rebase sqlite_${BASE} |
| +git am --keep-non-patch patches/*.patch |
| +git diff master |
| +# This branch should be identical to master, unless someone forgot to export |
| +# their changes into a patch. If so, do that as a separate CL and start over. |
| + |
| +#### Cherry-pick your change. |
| +git cherry-pick <your change> |
| +# This branch should be identical to your development branch, except |
| +# amalgamation. |
| + |
| +# Rebuild the patch set. |
| +rm patches/* |
| +git format-patch --output-directory=patches sqlite_${BASE}..HEAD |
| +git add patches/*.patch |
| +git commit -m "Rebuild patches for sqlite_${VERSION}" patches/*.patch |
| + |
| +# Re-generate the amalgamation. |
| ./google_generate_amalgamation.sh |
| +git commit -m 'google_generate_amalgamation.sh' amalgamation/ |
| +# At this point everything should build and work. |
| -# Find a sucker. Send review. |
| -# TODO(shess) Describe an appropriate comment style. Seems like it |
| -# should at least include the SQLite version number. |
| +# Do a squash upload. This should add your single patch to patches/, and apply |
| +# the changes your patch represents to src/ and amalgamation/. Other patches |
| +# will have hash changes. A sensible check-in comment would be something like |
| +# the patch's checkin comment, plus "regenerate amalgamation and generate patch |
| +# file." |
| +# TODO(shess): Should hash changes be checked in, or backed out? |
| --------------------------------------------- |
| +# Find a sucker. Send review. |
| -For reference, all of our local patches are also kept as .patch files in the |
| -sqlite directory. Here is a list of the patches, in the order they should be |
| -applied to a vanilla SQLite (of the version we currently have) to get, in |
| -principle, exactly what is checked in: |
| - |
| -misc.patch |
| -safe-tolower.patch |
| -fts2.patch |
| -fts3.patch |
| -fts3_85522.patch |
| -icu-regexp.patch |
| -icu-shell.patch |
| -attach-integer.patch |
| -webdb.patch |
| -test.patch |
| -mac_time_machine.patch |
| -system-sqlite.patch |
| -sqlite-3.7.6.3-fix-out-of-scope-memory-reference.patch |
| -misalignment.patch |
| -memcmp.patch |
| -separate_cache_pool.patch |
| -recover.patch |
| - |
| -So, e.g. you could do this to apply all our patches to vanilla SQLite: |
| - |
| -cd sqlite-$LATEST |
| -patch -p0 < ../sqlite/misc.patch |
| -patch -p0 < ../sqlite/fts2.patch |
| -patch -p0 < ../sqlite/fts3.patch |
| -patch -p0 < ../sqlite/fts3_85522.patch |
| -patch -p0 < ../sqlite/icu-shell.patch |
| -patch -p0 < ../sqlite/webdb.patch |
| -patch -p0 < ../sqlite/test.patch |
| -patch -p0 < ../sqlite/mac_time_machine.patch |
| -patch -p0 < ../sqlite/system-sqlite.patch |
| -patch -p0 < ../sqlite/sqlite-3.7.6.3-fix-out-of-scope-memory-reference.patch |
| -patch -p0 < ../sqlite/misalignment.patch |
| -patch -p0 < ../sqlite/memcmp.patch |
| -patch -p0 < ../sqlite/separate_cache_pool.patch |
| -patch -p0 < ../sqlite/recover.patch |
| - |
| -This will only be the case if all changes we make also update the corresponding |
| -patch files. Therefore please remember to do that whenever you make a change! |
| - |
| -Descriptions of the changes we've made can be found at the bottom of this file. |
| +--- |
| + |
| +3) Import new release of SQLite. |
| + |
| +Importing a new SQLite involves merging our local changes with SQLite's changes. |
| +Like any other merge, this may involve dropping some CLs while modifying others. |
| +The basic idea below is to generate git branches to work with: |
| + sqlite_${BASE} - current version without patches |
| + sqlite_${BASE}_patched - current version with patches applied via git CLs |
| + sqlite_${VERSION} - new version without patches |
| + sqlite_${VERSION}_patched - new version with patches applied via git CLs |
| +At this point, squashing sqlite_${VERSION}_patched to master gives exactly a CL |
| +suitable for committing. |
| + |
| +# Everything based in sqlite subdir. |
| +cd third_party/sqlite |
| + |
| +BASE=3070603 |
| +VERSION=3080704 |
| + |
| +#### Create current-SQLite reference branch. |
| +git checkout -b sqlite_${BASE} master |
| +rm -rf src |
| +cp -a sqlite-src-${BASE} src |
| +# -f includes ignored files, of which there are a couple. |
| +git add -f src/ |
| +git commit -m "Reset to sqlite-src-${BASE}" |
| +# This branch is unlikely to build. |
| + |
| +#### Convert patches into CLs. |
| +git checkout -b sqlite_${BASE}_patched master |
| +git rebase sqlite_${BASE} |
| +git am patches/*.patch |
| +git diff master |
| +# This branch should be identical to master. |
| + |
| +#### Create new-SQLite reference branch. |
| +git checkout -b sqlite_${VERSION} master |
| +git rebase sqlite_${BASE} |
| +# SQLite's download page is at <http://www.sqlite.org/download.html>. Scroll to |
| +# "Legacy Source Code Distribution Formats", and grab sqlite-src-<VERSION>.zip. |
| +# Unzip it and pull it into the repo. |
| +wget http://www.sqlite.org/2014/sqlite-src-${VERSION}.zip |
| +unzip sqlite-src-${VERSION}.zip |
| +rm sqlite-src-${VERSION}.zip |
| +# -f includes ignored files, of which there are a couple. |
| +git add -f sqlite-src-${VERSION}/ |
| +git commit -m "Begin import of sqlite-src-${VERSION}" sqlite-src-${VERSION} |
| +rm -rf src |
| +cp -a sqlite-src-${VERSION} src |
| +# -f includes ignored files, of which there are a couple. |
| +git add -f src/ |
| +git commit -m "Update src to sqlite-src-${VERSION}" src/ |
| +# This branch is unlikely to build. |
| + |
| +#### Create a branch for merging the CLs to the new SQLite. |
| +git checkout -b sqlite_${VERSION}_patched master |
| +git rebase sqlite_${VERSION} |
| + |
| +# Replay the patches onto this branch. There will be merge conflicts to fix. |
| +# My approach is generally to just accept them prefering the patch's change in |
| +# case of conflicts, and then resolve the conflicts as a second pass. |
| +git rebase --onto zsqlite_${VERSION}_patched zsqlite_${BASE} zsqlite_${BASE}_patched |
| +# Once everything is resolved, re-generate the amalgamation. |
| +./google_generate_amalgamation.sh |
| +git commit -a -m "google_generate_amalgamation.sh" |
| + |
| +# The goal is to have a set of reasonably-independent CLs which can be |
| +# understood separately, so that future importers can sensibly determine how to |
| +# handle conflicts. So use git-rebase and slipstream fixups back into their |
| +# original CL until things are relatively clean. |
| + |
| +# Rebuild the patch set. |
| +rm patches/* |
| +# This assumes that HEAD is still google_generate_amalgamation.sh. |
| +git format-patch --output-directory=patches sqlite_${VERSION}..HEAD^ |
| +git add patches/*.patch |
| +git commit -m "Rebuild patches for sqlite_${VERSION}" patches/*.patch |
| + |
| +# Drop the old version of SQLite. |
| +git rm -r sqlite_${BASE} |
| +git commit -m 'Remove sqlite_${BASE}' -- sqlite_${BASE} |
| + |
| +# Do a squash upload. Edit the commit message appropriately to reflect anything |
| +# from <http://www.sqlite.org/changes.html> which might be deemed important. |
| +# Don't enumerate all of the patch messages, those are assumed, but do reference |
| +# any material changes made. |
| +# TODO(shess) Describe an appropriate comment style. Seems like it should at |
| +# least include the SQLite version number. Meanwhile, look in the logs for past |
| +# commits to model things on. |
| + |
| +Find a sucker. Send review. |
| + |
| +TODO(shess): It is basically impossible to trybot the entire change, it takes |
| +forever to upload and sqlite3.c breaks things because it's too large. I have a |
| +nasty Perl script to break up sqlite3.c into pieces which are then included by a |
| +single .c file, but there must be a better way. Perhaps just have sqlite.gyp |
| +include all the .c files directly? |
| + |
| +Note that things can be broken down differently, if you prefer. For instance, |
| +adding the new version of the SQLite distro and removing the old one can be |
| +distinct CLs. |
| -------------------------------------------- |
| -How to run the SQLite tests for the Chromium version of SQLite on Linux. |
| +4) Running SQLite's test suite within the Chromium checkout. |
| -Prerequisties: On my corp Ubuntu 8.04 workstation, I needed to install the |
| -following packages: |
| -sudo apt-get install tcl8.4-dev libicu-dev |
| +Prerequisites: The test suite requires tcl-dev and libicu-dev. Install those on |
| +Ubuntu like: |
| + sudo apt-get install tcl8.5-dev libicu-dev |
| +On OSX, I use Homebrew: |
| + sudo port install tcl |
| +TODO(shess): OSX works fine with either tcl8.5 or tcl8.6, but on Ubuntu 14.04.1 |
| +with tcl8.6, I see crashes in some of the WAL tests. Revisit using tcl8.6 on |
| +next import of SQLite. |
| +TODO(shess): That Homebrew command has installed tcl8.6 for a few years, so the |
| +above may require some adaptation of the build files. |
| -cd src/third_party/sqlite/src |
| +cd third_party/sqlite/src |
| mkdir build |
| cd build |
| -make -f ../Makefile.linux-gcc testfixture |
| +make -f ../Makefile.linux-gcc testfixture sqlite3 |
| +# NOTE(shess): 64-bit builds under 3.7.6.3 show various warnings. AFAICT they |
| +# don't impact test operation. Later versions are better so I'm not fixing it. |
| make -f ../Makefile.linux-gcc test > /tmp/test.log |
| -egrep -v 'Ok$' /tmp/test.log |
| -# For an ideal test run, you would see: |
| -# 0 errors out of 57887 tests |
| -# However, the current situation on my corp Linux Ubuntu 8.04 machine, with |
| -# test run on a locally mounted directory, is the failure of: |
| -# "rollback-2.3", "tkt3457-1.4" |
| -# I do not know why, but it is not related to our fts2.c changes -- I backed |
| -# them out to check. |
| +egrep 'errors out of' /tmp/test.log |
| +# Show broken tests: |
| +egrep 'Failures on these tests:' /tmp/test.log |
| +# Broken tests will also show lines ending in "..." instead of "... Ok". |
| -Chris Evans <cevans@google.com>, Oct 1, 2009 |
| +In version Right now on OSX, I see no failures: |
| + 0 errors out of 77849 tests |
| --------------------------------------------- |
| +In version 3.7.6.3 on Ubuntu 14.04 I see some failures: |
| + 83 errors out of 78604 tests |
| + - precision in formatting numbers in enc4.test. |
| + - case sensitivity in nan.test. |
| + - oserror-1.1.[1-3] fail because there are too many fds available. Fixed with |
| + "ulimit -n 1024". |
| + |
| +-- |
| + |
| +NOTE(shess): On Ubuntu it is possible to run the tests in a tmpfs something |
| +like: |
| + |
| +TMPFS=/dev/shm/sqlite_build |
| +BUILD=$PWD |
| +mkdir $TMPFS |
| +(cd $TMPFS ; $BUILD/testfixture $BUILD/../test/veryquick.test >/tmp/test.log) |
| -As of May 07, 2010, these are our changes from sqlite_vendor: |
| - |
| - - A fix for a crash passing an integer expression to ATTACH / DETACH. See |
| - attach-integer.patch |
| - - A fix for a crash mis-calling the REGEXP() function of the ICU extension. |
| - See icu-regexp.patch |
| - - A large number of fts2 robustness fixes against corrupt data in its metadata |
| - tables. |
| - - fts2.c disables fts2_tokenizer(). |
| - - fts3.c disables fts3_tokenizer(). |
| - - Tweak to SQLITE_EXTENSION_INIT* in sqlite3ext.h. |
| - - That implied a change in src/test_autoext.c for testing. |
| - - Added fts.test in tests, modified quick.test. |
| - - Modifications to Makefile.linux-gcc and main.mk for compiling |
| - SQLite tests. |
| - - Compile warning (cast to void* for sqlite3_free) fixed in func.c. |
| - - Avoid using tolower() in fts code which causes problem in some locales, see: |
| - safe-tolower.patch |
| - http://crbug.com/15261 |
| - http://www.sqlite.org/src/tktview/991789d9f3136a0460dc83a33e815c1aa9757c26 |
| - - Check that the third argument to memset() is nonzero in expr.c to avoid |
| - a linker warning when the compiler can optimize it to a constant zero |
| - (e.g. see http://www.sqlite.org/cvstrac/tktview?tn=3765,39) |
| - |
| -Changes from Chrome: |
| - - I marked all changes I made with "evanm", so you can find them with |
| - "grep evanm *". |
| - - Most files include sqlite3ext.h with SQLITE_CORE #defined, but two don't: |
| - fts2_tokenizer.c and icu.c. Without this #define, the calls in |
| - fts2_tokenizer.c try to go through some pointer to the sqlite API instead |
| - of calling the functions directly (to work as a loadable module), but then |
| - crash (because the other files never initialize that loadable module |
| - support). As a hack I #defined it in these files, but it'd be nice to |
| - figure out what really ought to happen here (perhaps this file is new and |
| - hasn't been tested to verify it works right). Update: Seems this is an |
| - issue we get because we're using fts2 instead of fts3. |
| - - shell_icu_win.c and shell_icu_linux.c are Chrome-specific files used to load |
| - our ICU data. shell.c has been modifed to call into these files. |
| - - fts2_icu.c and fts3_icu.c have a critical bug. U8_NEXT is used over |
| - a UTF-16 string. It's rep$ by U16_NEXT (jungshik) |
| - - Added a new function chromium_sqlite3_initialize_win_sqlite3_file() |
| - at the end of os_win.c. It allows the Windows-specific Chromium VFS |
| - to reuse most of the win32 SQLite VFS. |
| - - Added a new function |
| - chromium_sqlite3_initialize_unix_sqlite3_file() and made |
| - fillInUnixFile() non-static in os_unix.c. It allows the |
| - Linux-specific Chromium VFS to reuse most of the unix SQLite VFS. |
| - - Exposed three functions that deal with unused file descriptors in |
| - os_unix.c, to allow Chromium's Posix VFS implementation in |
| - WebKit/WebCore/platform/sql/chromium/SQLiteFileSystemChromiumPosix.cpp |
| - to correctly implement the "unused file descriptors" logic in the |
| - xDlOpen() method. The new functions are |
| - chromium_sqlite3_get_reusable_file_handle(), |
| - chromium_sqlite3_update_reusable_file_handle() and |
| - chromium_sqlite3_destroy_reusable_file_handle(). Also, added the |
| - chromium_sqlite3_fill_in_unix_sqlite3_file() function that calls |
| - fillInUnixFile(), which will be made static again as soon as a |
| - WebKit patch using the new function lands. |
| - - From mac_time_machine.patch: |
| - When __APPLE__ and when creating a -journal file with any unix-type vfs, |
| - determine if the database for which the journal is being created has been |
| - excluded from being backed up using Apple's Time Machine and if so then also |
| - exclude the journal. These changes were made in pager.c with includes of |
| - Apple interfaces being made in sqliteInt.h. In order to eliminate a symbol |
| - conflict with an Apple library after amalgamation it was also necessary to |
| - rename fts3_porter.c's 'cType' to 'vOrCType'. |
| - - fts3_85522.patch allows fts3 to work if PRAGMA is not authorized. |
| - - src/recover.c file implements a virtual table which can read |
| - through corruption. |
| - - Enable the macro 'SQLITE_TEMP_STORE=3' for Android. |
| - - memcmp.patch backports ASAN-related fixes from SQLite trunk. |
| +This is faster, but it is plausible that different things are being tested than |
| +real-world use. |