SQLite

Timeline
Login

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

100 check-ins occurring around 955b15a020e9ea64.

2008-01-17
16:22
Registerification of the VDBE is complete. The operand stack has been removed from the code. All instructions operate out of registers only. (CVS 4718) (check-in: 706b41b7 user: drh tags: trunk)
02:36
Registerification of the WHERE clause logic. (CVS 4717) (check-in: 5581160f user: drh tags: trunk)
02:36
Registerification of the WHERE clause logic. (CVS 4716) (check-in: b3a141b2 user: drh tags: trunk)
2008-01-16
18:20
Use 1<<$x instead of pow(2,$x) in the test suite code. (CVS 4715) (check-in: 3a289b6d user: danielk1977 tags: trunk)
17:46
Get all tests running without memory leaks. (CVS 4714) (check-in: 5807921f user: drh tags: trunk)
08:24
Work around using (regexp) in the (ifcapable) function of the test suite. So that simpler versions of tcl can call it. (CVS 4713) (check-in: 3aa5606b user: danielk1977 tags: trunk)
2008-01-15
02:22
Fix for a couple of minor memory leaks. (CVS 4712) (check-in: a13d3e95 user: drh tags: trunk)
2008-01-14
15:20
Fix for ticket #2883. (CVS 4711) (check-in: 187f41f5 user: drh tags: trunk)
2008-01-13
19:02
Fix some issues with out-of-memory recovery. (CVS 4710) (check-in: 23181f86 user: drh tags: trunk)
2008-01-12
21:35
The sqlite3_trace() callback now prints a message as each trigger fires within a statement. (CVS 4709) (check-in: 110c000d user: drh tags: trunk)
19:03
Continuing work toward converting the VM to a register machine. (CVS 4708) (check-in: 426f31ec user: drh tags: trunk)
12:48
Continuing work toward converting the VM into a register machine. (CVS 4707) (check-in: a6dddebc user: drh tags: trunk)
2008-01-11
15:27
Do explicit range tests before attempting to convert a 64-bit float into a 64-bit integer. Some systems (windows) seem to throw exceptions if the conversion is out of range. Ticket #2880. (CVS 4706) (check-in: 4744257d user: drh tags: trunk)
00:06
Attempt to work around a bug in the Borland BCC 5.5.1 compiler. Ticket #2880. (CVS 4705) (check-in: 6de0ee49 user: drh tags: trunk)
2008-01-10
23:50
More work toward converting the VM into a register-based machine. (CVS 4704) (check-in: 8cbd4651 user: drh tags: trunk)
03:46
Continuing work toward registerizing the code generator. (CVS 4703) (check-in: 173698c9 user: drh tags: trunk)
00:08
Fix mixing quote excapes in an awk script. Ticket #2877. (CVS 4702) (check-in: 05fbca91 user: drh tags: trunk)
2008-01-09
23:21
Fix a test script problem associated with the new crash4.test file. Ticket #2882. (CVS 4701) (check-in: 1134c5ea user: drh tags: trunk)
23:04
All essential opcodes can now operate out of registers and completely avoid the stack. Many still optionally use the stack to support legacy code generation, but the stack is not required. The next step is to update all code generation to avoid using the stack. (CVS 4700) (check-in: e3cf1c1c user: drh tags: trunk)
18:31
Fix the build for AIX and other systems where "char" is unsigned by default. (CVS 4699) (check-in: 47672af0 user: drh tags: trunk)
02:15
Continued work toward converting to a register-based VM. (CVS 4698) (check-in: 92deff07 user: drh tags: trunk)
2008-01-08
23:54
Registerify the comparison opcodes. (CVS 4697) (check-in: 8862ce9c user: drh tags: trunk)
18:57
Finish registerizing the core logic of INSERT and UPDATE. (CVS 4696) (check-in: 5fd10367 user: drh tags: trunk)
16:03
Add crash4.test with additional crash testing. (CVS 4695) (check-in: 87b4ac4b user: drh tags: trunk)
15:18
Add additional randomness to crash tests. (CVS 4694) (check-in: 3ccce1f5 user: drh tags: trunk)
02:57
Progress toward registerification of the constraint checking logic for INSERT and UPDATE. (CVS 4693) (check-in: b9bf509e user: drh tags: trunk)
2008-01-07
19:20
Registerify the AUTOINCREMENT processing and the OP_IsNull and OP_NotNull operators. (CVS 4692) (check-in: aa48867c user: drh tags: trunk)
10:16
Comment changes in select.c. (CVS 4691) (check-in: 38020592 user: danielk1977 tags: trunk)
2008-01-06
00:25
Registerify the SRT_Subroutine destination for SELECT results. (CVS 4690) (check-in: 8201f717 user: drh tags: trunk)
2008-01-05
18:48
Modify OP_RegMakeRec to take a base register and count and optionally store results in the register specified by P3. (CVS 4689) (check-in: 6bb1b1bc user: drh tags: trunk)
18:44
Fix a memory leak introduced with #4687. (CVS 4688) (check-in: 2b98b0fc user: danielk1977 tags: trunk)
17:39
First pass at optimizing max()/min() as described in #2853. Some refinements to come. (CVS 4687) (check-in: c449e04f user: danielk1977 tags: trunk)
16:29
Registerify binary operators. Add register tracing to debugging output. (CVS 4686) (check-in: 66396d2f user: drh tags: trunk)
06:51
Expression code generator takes advantage of recent opcode changes. (CVS 4685) (check-in: 6c78d2a4 user: drh tags: trunk)
05:38
Register-ify the OP_ForceInt opcode. (CVS 4684) (check-in: 32380dca user: drh tags: trunk)
05:20
Register-ify the OP_AddImm and all casting opcodes. Omit the OP_MemIncr opcode. (CVS 4683) (check-in: 3e8a07dd user: drh tags: trunk)
04:06
Get rid of OP_Dup, OP_MemStore, OP_MemLoad, and OP_MemMove. Replace with OP_Copy, OP_SCopy, and OP_Move. Add the infrastructure for operation properties in1, in2, in3, out2, and out3 but do not yet use any of these. (CVS 4682) (check-in: cc149eb9 user: drh tags: trunk)
2008-01-04
22:01
Implement the out2-prerelease opcode design pattern. (CVS 4681) (check-in: fe057a88 user: drh tags: trunk)
19:33
Allow the P2 operand to be negative on opcodes that are not jumps. (CVS 4680) (check-in: 717bcd11 user: drh tags: trunk)
19:12
Fix mkopcodeh.awk so that it works on a mac. (CVS 4679) (check-in: 59d3dfa4 user: drh tags: trunk)
19:10
Some modifications to insert.c to work without using the stack. (CVS 4678) (check-in: d9ac6bee user: danielk1977 tags: trunk)
16:50
Replace the NOPUSH_MASKs with a bit-vector mechanism that can contain several different properties about each opcode. (CVS 4677) (check-in: 042dcb96 user: drh tags: trunk)
13:57
Modify FifoRead and FifoWrite to work exclusively with memory cells. (CVS 4676) (check-in: 2c913908 user: danielk1977 tags: trunk)
13:24
Modify the code generated for a DELETE to use registers instead of the vdbe stack. (CVS 4675) (check-in: 173f2813 user: danielk1977 tags: trunk)
11:01
Add OP_MemSet, for setting a memory cell to a string value. (CVS 4674) (check-in: 8bb9f970 user: danielk1977 tags: trunk)
2008-01-03
23:44
Change the sqlite3ExprCode() function so that callers can request that the result of the expression be left on the stack or in a register. (CVS 4673) (check-in: 61bfb77c user: drh tags: trunk)
18:56
Fix typo in comment. Ticket #2870. (CVS 4672) (check-in: e97e4578 user: danielk1977 tags: trunk)
18:44
Update OP_Rowid, OP_Column and related opcodes to use registers. (CVS 4671) (check-in: 4f396707 user: drh tags: trunk)
18:39
Modify VFilter and VRename to use registers instead of the vdbe stack for inputs. (CVS 4670) (check-in: 253ed40a user: danielk1977 tags: trunk)
18:03
Registers (aka memory cells) in the VM are now numbered starting with 1 instead of 0. A register number of 0 means "no such register". (CVS 4669) (check-in: 0b849805 user: drh tags: trunk)
17:31
Modify OP_VUpdate to read arguments from a range of memory cells instead of from the stack. (CVS 4668) (check-in: 955b15a0 user: danielk1977 tags: trunk)
11:50
Change the VdbeOp.p4 union to include specific pointer types for the various values of VdbeOp.p4type. (CVS 4667) (check-in: 7e8330c8 user: danielk1977 tags: trunk)
09:51
Change the OP_Insert opcode to read the key and data to insert from memory cells, not the stack. (CVS 4666) (check-in: 46501f49 user: danielk1977 tags: trunk)
08:18
Change the output of vdbe_trace etc. to include operands p3 and p5. (CVS 4665) (check-in: 76f2040b user: danielk1977 tags: trunk)
08:08
Combine the OP_Column and OP_DfltValue opcodes. (CVS 4664) (check-in: c30f8206 user: danielk1977 tags: trunk)
07:54
Change OP_OpenRead and OP_OpenWrite so that the database number is read from the P3 operand, not the stack. (CVS 4663) (check-in: 35da261d user: danielk1977 tags: trunk)
07:09
Fix EXPLAIN and EXPLAIN query plan to work with new opcode format. (CVS 4662) (check-in: b166c33a user: danielk1977 tags: trunk)
01:28
Revamp sqlite3VdbeAddOpList() so that the structure holds integer values for P1, P2, and P3. (CVS 4661) (check-in: 4dc507d8 user: drh tags: trunk)
00:01
Refactoring the VM. The P3 parameter is changed to P4. A P5 parameter is added (though not currently used.) Add routines sqlite3VdbeAddOpX() where X is one of 0, 1, 2, 3, or 4. (CVS 4660) (check-in: 027875e4 user: drh tags: trunk)
2008-01-02
17:25
Add operands p4 and p5 to the VdbeOp structure. (CVS 4659) (check-in: ad528735 user: drh tags: trunk)
17:11
Minor optimizations for sub-selects in EXISTS(...) expressions: Discard any DISTINCT or ORDER BY clauses and do not calculate any column values. (CVS 4658) (check-in: fbd17a89 user: danielk1977 tags: trunk)
16:27
Combine the eDest and iParm arguments to sqlite3Select() into a single type - "SelectDest". (CVS 4657) (check-in: 86dcdfe6 user: danielk1977 tags: trunk)
14:28
Add opcodes OP_ResultRow and OP_RegMakeRec which are register-based equivalents to OP_Callback and OP_MakeRecord. Use the new opcodes. (CVS 4656) (check-in: 4c7f35da user: drh tags: trunk)
13:05
Remove a surplus OP_Close from delete.c. Fixes a problem with (4654). (CVS 4655) (check-in: 03cc91b3 user: drh tags: trunk)
11:50
Take advantage of the fact that b-tree cursors need not be closed while other cursors modify their tables to simplify trigger program generation code. (CVS 4654) (check-in: f1966a8a user: danielk1977 tags: trunk)
04:41
Add some extra tests for the recent trigger compilation optimization. (CVS 4653) (check-in: de54dad8 user: danielk1977 tags: trunk)
00:34
Begin setting a foundation on which to convert the VM from a stack-based to a register-based machine. Everything is still mostly stack based with this check-in. This change merely begins adding infrastructure to support a register-based architecture. (CVS 4652) (check-in: 051ec01f user: drh tags: trunk)
2008-01-01
19:02
Optimize trigger compilation to avoid populating the OLD.* and NEW.* pseudo-tables with data that will never be used. Some testing to come. (CVS 4651) (check-in: e08a33ed user: danielk1977 tags: trunk)
06:19
Silence a harmless unitialised variable warning in btree.c. Ticket #2862. (CVS 4650) (check-in: b51782cc user: danielk1977 tags: trunk)
05:49
Define SQLITE_CORE=1 at the start of the amalgamation file for the benefit of any extension source code (i.e. fts3) appended to it. Ticket #2858. (CVS 4649) (check-in: 6dea8c16 user: danielk1977 tags: trunk)
2007-12-30
23:38
fix case in os2GetTempname() where none of the usual environment variables are set to not overwrite two unrelated bytes (CVS 4648) (check-in: 9719a063 user: pweilbacher tags: trunk)
23:35
simplify os2FullPathname() and at the same time make the allocations more robust (CVS 4647) (check-in: 6f8952a8 user: pweilbacher tags: trunk)
23:29
move variable declaration to be able to compile sqlite3_mutex_alloc() with older compilers (like EMX/gcc 2.8.1) (CVS 4646) (check-in: a568a9cf user: pweilbacher tags: trunk)
2007-12-29
13:39
Add experimental pragma "quick_check", a reduced version of integrity_check that runs without most of the overhead of the full integrity_check. (CVS 4645) (check-in: 2ddc8d72 user: danielk1977 tags: trunk)
13:18
Mem3.c enhanced so that an allocation of N bytes only requires (N+11)&~7 bytes instead of (N+15)&~7 bytes of heap storage. Minimum heap usage per allocation is still 16 bytes. 8-byte alignment is preserved. (CVS 4644) (check-in: d027f91c user: drh tags: trunk)
2007-12-27
15:12
Fix a race condition that can occur when reloading the database schema in shared-cache mode. (CVS 4643) (check-in: b37babef user: danielk1977 tags: trunk)
2007-12-21
04:47
Fix for #2854. "BEGIN EXCLUSIVE" excludes other shared cache users from using the database. (CVS 4642) (check-in: 2e59b1d0 user: danielk1977 tags: trunk)
00:02
Fix some issues with lemon. Tickets #2852 and #2835. (CVS 4641) (check-in: 5283e0d1 user: drh tags: trunk)
2007-12-19
09:20
Add the ioerr3.test file, that found an error in br3317. The error is not present in this branch. (CVS 4640) (check-in: d3efec14 user: danielk1977 tags: trunk)
2007-12-18
17:50
Fix a problem with SQLITE_MAX_SQL_LENGTH introduced by check-in (4636). Ticket #2851 (CVS 4639) (check-in: cdd866f5 user: drh tags: trunk)
15:41
In the CLI, quote strings that contain the separator character. Ticket #2850. (CVS 4638) (check-in: 493a17c4 user: drh tags: trunk)
11:19
Fix for typo in main.mk: the output of target sqlite3 was testcli. (CVS 4637) (check-in: 15675dc5 user: danielk1977 tags: trunk)
2007-12-17
16:20
When SQLITE_MAX_SQL_LENGTH is 0, disable the limit. Make the default 0. Once again build the sqlite3 CLI from individual source files so that it can be built on systems that lack tclsh. Tickets #2845 and #2846. (CVS 4636) (check-in: 07aeca3b user: drh tags: trunk)
2007-12-14
17:39
Version 3.5.4 (CVS 4635) (check-in: cf4a11b2 user: drh tags: trunk)
17:24
Update library version number to 3.5.4. (CVS 4634) (check-in: 5c653bfc user: drh tags: trunk)
17:24
Fix a C++-ism in recent changes to select.c. (CVS 4633) (check-in: 4a008050 user: drh tags: trunk)
17:22
Change to the crash simulator to give better diagnostics in case it runs out of memory while using the mem3.c allocator. (CVS 4632) (check-in: 0d5747db user: drh tags: trunk)
16:11
Fix a typo in the previous check-in. Also simplify a line in select.c. (CVS 4631) (check-in: 86f45d7b user: drh tags: trunk)
15:12
Fix an out-of-memory NULL pointer defer in the code generator. Ticket #2843. (CVS 4630) (check-in: b821b6ed user: drh tags: trunk)
2007-12-13
21:54
Change all instances of "it's" in comments to either "its" or "it is", as appropriate, in case the comments are ever again read by a pedantic grammarian. Ticket #2840. (CVS 4629) (check-in: 4e91a267 user: drh tags: trunk)
19:15
Fix a memory leak that can occur following a malloc() failure. (CVS 4628) (check-in: 993a2130 user: danielk1977 tags: trunk)
18:29
Fix the location of a #include in test_thread.c. ticket #2826. (CVS 4627) (check-in: 6129fce8 user: drh tags: trunk)
18:24
Add test cases for errors in "IN(SELECT ...)" expressions where the SELECT statement is a compound SELECT. No faults found. (CVS 4626) (check-in: 49b67adf user: danielk1977 tags: trunk)
18:20
Add new makefile target for "testcli" that builds a CLI using individual source files instead of the amalgamation. Ticket #2838. (CVS 4625) (check-in: b6f04e5b user: drh tags: trunk)
17:50
Additional test cases for tkt2822. Fix a related bug in printf(). (CVS 4624) (check-in: 8f184e40 user: drh tags: trunk)
08:15
Return an error if the user attempts to rename a view. Related to (but not a fix for) #2831. (CVS 4623) (check-in: 19d56d99 user: danielk1977 tags: trunk)
07:58
Adjust the test suite to account for recent changes related to #2822. Most changes are related to English language error messages only. (CVS 4622) (check-in: 2f88b9b3 user: danielk1977 tags: trunk)
03:45
ORDER BY in a compound SELECT will first match against the left-most SELECT. If there is no match there, it begins working its way to the right. (CVS 4621) (check-in: 56063ec8 user: drh tags: trunk)
02:45
Towards getting ORDER BY to match against the correctin columns. This version only looks at the left-most column in a compound SELECT. That is the correct thing to do, but not what SQLite has historically done. (CVS 4620) (check-in: bbddf16a user: drh tags: trunk)
2007-12-12
22:24
Minor cleanup changes on the OP_StackDepth opcode. Added the sidedelete test for additional testing of ticket #2832. (CVS 4619) (check-in: c0689409 user: drh tags: trunk)