[sqlite] Doc issue: No explanation of record counts for EXPLAIN QUERY PLAN

In the documentation of EXPLAIN QUERY PLAN[1], I found the following: > sqlite> EXPLAIN QUERY PLAN SELECT a, b FROM t1 WHERE a=1; > 0|0|0|SCAN TABLE t1 > The example above shows SQLite estimating that the full-table scan will visit > approximately 100,000 records. There’s no

Re: [sqlite] Doc issue: No explanation of record counts for EXPLAIN QUERY PLAN

Thanks. Those sentences are obsolete and need to be removed from the documentation. On 2/3/17, Jens Alfke wrote: > In the documentation of EXPLAIN QUERY PLAN[1], I found the following: > >> sqlite> EXPLAIN QUERY PLAN SELECT a, b FROM t1 WHERE a=1; >> 0|0|0|SCAN TABLE

Re: [sqlite] Retrieve INTEGER PRIMARY KEY

Last_insert_rowid() https://www.sqlite.org/c3ref/last_insert_rowid.html On Fri, Feb 3, 2017 at 1:51 PM, Clyde Eisenbeis wrote: > For OLE DB SQL, I have retrieved the primary key: > > - > using (System.Data.OleDb.OleDbConnection oledbConnect = new

Re: [sqlite] Simple web query tool

> > Running the sqlite3 command-line shell via cgi works way better than you > may expect. > ​Yay verily, and that is really not doing a great tool justice. I've done a lot of similar things to what Lindsay outlines above both with web and application targets, Often these procedures are set up as

Re: [sqlite] Simple web query tool

> "Sqlite just does it's thing with a minimum amount of fuss and minimal impact on system resources, can't rave enough" Yes! A while back, when I first tried what I outlined it worked so well I took it a step further. With nodejs async io, streams and a bit of javascript glue code, it is

Re: [sqlite] "DISTINCT" makes a query take 37 times as long

On Fri, Feb 3, 2017 at 12:27 AM, Kevin O'Gorman wrote: > But the big thing is that I took a look at EXPLAIN QUERY PLAN using this > ... Maybe somebody can explain them to me, but it doesn't really matter whether > I ever understand them. Perhaps Mr. Hipp can make use of

Re: [sqlite] IS NULL and IS NOT NULL constraints not passed into virtual table's xBestIndex

Thanks a ton for the replies. This pretty much confirms what I've been suspecting. On 3 February 2017 at 11:24, Clemens Ladisch wrote: > The FTS and R-tree virtual tables do not contain NULL values, so > I guess it was never seen as necessary. Does this make it an accidental

Re: [sqlite] IS NULL and IS NOT NULL constraints not passed into virtual table's xBestIndex

Daniel Kamil Kozar wrote: > I'd like to ask why using a IS NULL or IS NOT NULL constraint in a > query made to a virtual table does not result in these constraints > being present in the sqlite3_index_info structure passed to the > virtual table's xBestIndex. Because there is no

Re: [sqlite] "DISTINCT" makes a query take 37 times as long

EXPLAIN dumps SQLite "machine code". Use EXPLAIN QUERY PLAN for the human readable version. -Ursprüngliche Nachricht- Von: sqlite-users [mailto:sqlite-users-boun...@mailinglists.sqlite.org] Im Auftrag von Kevin O'Gorman Gesendet: Freitag, 03. Februar 2017 00:27 An: SQLite mailing list

Re: [sqlite] IS NULL and IS NOT NULL constraints not passed into virtual table's xBestIndex

The supported constraints are: #define SQLITE_INDEX_CONSTRAINT_EQ 2 #define SQLITE_INDEX_CONSTRAINT_GT 4 #define SQLITE_INDEX_CONSTRAINT_LE 8 #define SQLITE_INDEX_CONSTRAINT_LT 16 #define SQLITE_INDEX_CONSTRAINT_GE 32 #define SQLITE_INDEX_CONSTRAINT_MATCH 64 #define

Re: [sqlite] Does SQLite use field definitions?

It will if you add AUTOINCREMENT -Ursprüngliche Nachricht- Von: sqlite-users [mailto:sqlite-users-boun...@mailinglists.sqlite.org] Im Auftrag von Clyde Eisenbeis Gesendet: Donnerstag, 02. Februar 2017 15:12 An: SQLite mailing list Betreff: Re:

[sqlite] Retrieve INTEGER PRIMARY KEY

For OLE DB SQL, I have retrieved the primary key: - using (System.Data.OleDb.OleDbConnection oledbConnect = new System.Data.OleDb.OleDbConnection(stConnectString)) { using (System.Data.OleDb.OleDbCommand oledbCmd = oledbConnect.CreateCommand()) { ...

Re: [sqlite] BUG: Illegal initialization in icu.c : sqlite3IcuInit

On Wed, 01 Feb 2017 17:48:48 -0500 "Keith Medcalf" wrote: > Microsoft products require multiple versions of multiple Microsoft > and third-party compilers to compile any of their products. They > will simply add whatever is needed to their internal compilers suite > and use

Re: [sqlite] "DISTINCT" makes a query take 37 times as long

Un-analyzed here's what I'm getting while looking at the db: With distinct: sqlite> explain query plan select distinct ppos from move join pos on mto = pnum where pcensus = 18 and pmin < pmax; selectid|order|from|detail 0|0|1|SCAN TABLE pos USING INDEX pipos 0|1|0|SEARCH TABLE move USING

Re: [sqlite] BUG: Illegal initialization in icu.c : sqlite3IcuInit

On 3 Feb 2017, at 2:35pm, James K. Lowden wrote: > When I was > using Microsoft's C++ intensively, I would find a compiler bug about > once a month. A different set of bugs depending on the optimization level you were using. Oh man, you brought it all back. My

Re: [sqlite] BUG: Illegal initialization in icu.c : sqlite3IcuInit

On 3 Feb 2017, at 4:24pm, Warren Young wrote: > On Feb 3, 2017, at 9:15 AM, Simon Slavin wrote: > >> My solicitors will be in the post. > > How much does it cost to mail a lawyer where you are? Do they charge by > weight or unusual shape? Oh, the

Re: [sqlite] BUG: Illegal initialization in icu.c : sqlite3IcuInit

On Feb 3, 2017, at 9:15 AM, Simon Slavin wrote: > > My solicitors will be in the post. How much does it cost to mail a lawyer where you are? Do they charge by weight or unusual shape? ___ sqlite-users mailing list