assert() that was added to sqlite3_declare_vtab() by [eb94f4a8174436b1].
This check-in fixes the problem.
FossilOrigin-Name: 857d26a68cf439e9cba4f8a3b326c69366fc486a876b76835538709ee39b8713
virtual tables) before trying to load them
(dbsqlfuzz bc02a0cde82dee801a8d6f653d2831680f87dca1). This prevents
sqlite3_declare_vtab() from running with db->init.busy turned on. Even so,
enhance sqlite3_declare_vtab() to be able to deal with db->init.busy being on,
in case there are undiscovered paths to that state.
Each of these two changes are independently sufficient to prevent the problem
fixed by the previous check-in [c7560c1329965ab5] but there
is no harm in keeping that third layer of protection in place.
FossilOrigin-Name: eb94f4a8174436b1f0deed0a43618a20018387bb815be658314ca6b454c446fb
because it now uses %lld, and some older C compilers do not understand that
construct.
FossilOrigin-Name: 53dadf80672a4fe05341fe40873b6cfa73f6bc1e07cc41fcf10fe290c9156363
SQLITE_DEBUG. Surround the implementation of all test-only SQL functions
with #ifndef SQLITE_UNTESTABLE.
FossilOrigin-Name: b7e00ef8059f6fb5658c6ad6f337cfdf065a5f1b1130452122282f3a69e98a93
calls to sqlite_rename_column(). Let the function filter them itself.
The WHERE clause can mistakenly let virtual tables through if the
sqlite_schema table is corrupt. The sqlite_rename_column() function must
deal with them. We might as well let it deal with them always.
FossilOrigin-Name: 7c82b43b7cdb825f431d3a778f8b34e32666ccb3b067f1f2533d3b3dcca7bdc9
resulting from check-in [23ca23894af352ea]. Problem reported by
[forum:/forumpost/e2ea1a3f61|forum post e2ea1a3f61]. Also change
the datatype of Pager.pageSize to i64 even though page size never
exceeds 65536, in order to help prevent future problems of this kind.
FossilOrigin-Name: f4a552ed9f4ab35520b634954c39748cc7bda535f426280b79da1b99f70599ac
associated branches with NEVER()) and add a single new test to
btreeOverwriteCell() that detects when an overflow pages is also mapped
into a b-tree page and raises and immediate SQLITE_CORRUPT error before
making any changes.
dbsqlfuzz 81791bd980fe6935ff2c7334ec8bef11c1c12b82 and others.
FossilOrigin-Name: 32210fa4ac4f06e1705ef808731c7be040a23f9a8630986440100c5d4e76dc07
of the btree rebalancer.
dbsqlfuzz f035b1b5a100a5f0cffb95c958b6c67d7c5eaf1f.
FossilOrigin-Name: 00286ca5d998d80214d5e8d501a3b305f432cda1c0ef67ca530e1219565909b7
problem was introduced by the addition of generated columns
in check-in [b855acf1831943b3] (SQLite version 3.31.0, 2020-01-22).
Reported by Wang Ke in
[forum/forumpost/699b44b3ee|forum post 699b44b3ee].
FossilOrigin-Name: b9417d400f4585004f434837022709f818044d5844fe208fe01595a6b226ef7d
its datatype. Fix for the problem reported by
[forum:/forumpost/e5c76b738e|forum post e5c76b738e]. Test cases
in TH3.
FossilOrigin-Name: c7f0813cabf9d8ab367bead5ba8cf20132b8bb9274d8e47b76ad66a10517dd2a
that capability with an ANY type for strict tables that will accept any
datatype with BLOB affinity.
FossilOrigin-Name: d8fd1a2bfd51848ea612142d23475b268b3f5269e558f2e09713d1ac18b18274
end of code generation, in order to avoid deleting expressions out from under
the aggregation function sanity checking assert()s that occur near the
end of SELECT code generation. This fixes the assertion fault described by
[forum:/forumpost/cfcb4b461d|forum post cfcb4b461d].
FossilOrigin-Name: 600f1991e5c0a5d89cd8776a157b6fd72c7489791085876925e8dd7ab146fe1f