You should learn how to use "explain query plan".
Indexes are a trade-off between insert speed (slow) and select speed (fast).
If you have a static database there is no such thing as too many indexes as
they never would get changed. Most of us live in the world between.
Generally, don't
You go from 2 minutes to 30 minutes just based on clock speed.
I would imagine you don't have much disk cache on the ARM device, so every time
you read the database it has to hit the file systemouch
Did you benchmark your PC system without using a cached database? Just edit
the
You don't say how big your database is.
My guess is when you see the server using a lot of RAM (and exactly how are you
measuring this?) that it's flushing its disk cache. If you're on Unix use
vmstat to see what your OS cache is doing.
So...perhaps if you increase SQLite's internal cache it
: Sunday, March 13, 2011 8:40 AM
To: General Discussion of SQLite Database
Subject: EXT :Re: [sqlite] Optimising a query with several criteria
Hi Michael, thanks for this. My database is 40 megabytes (and growing
slowly) - is that a reasonable cachesize?
On 13/03/2011 13:07, Black, Michael
Try CodeBlocks
Cross-platform and works with gcc or MSVC or pretty much whatever.
I put in gcc 64-bit for it.
http://www.codeblocks.org/
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From:
Try this.
BEGIN TRANSACTION;
CREATE TABLE sales(SaleDate date,SaleVolume int);
INSERT INTO "sales" VALUES('2010-01-01', 10);
INSERT INTO "sales" VALUES('2010-01-02', 20);
INSERT INTO "sales" VALUES('2011-01-01', 15);
INSERT INTO "sales" VALUES('2011-01-02', 30);
INSERT INTO "sales"
But I thought he said he dropped the indexes (meaning they aren't there during
inserts).
That should make sorting irrelevant.
3 Things.
#1 Test with :memory: database and see what the speed is. That tells you if
it's SQLite or disk I/O as the bottleneck.
#2 Try WAL mode "pragma
I hope you know what you're doing with trying to preserve that much
significance. Ths first time you stick it in a double or long double variable
you'll lose it. You can use the HPAlib to get 32 digits
http://www.nongnu.org/hpalib/
// Example showing digit loss -- doesn't matter double or
Hmmm...the docs do say that...but how do you get that value back out?
Retreiving it as text doesn't work.
You still don't say what you're planning on doing with these number...just
displaying them?
I think the docs may be misleading...here is the comment in sqlite3.c
/*
** Try to convert a
Blob may be better if you need speed -- then no conversion is necessary inside
your Pascal code to/from a string.
But if you want to be able to see and understand your database text is better
(or you have to write a special Pascal program to decode your database to look
at any problems).
Unless you're running multiple SQLite apps you don't gain anything by using a
DLL. Plus, is your Pocket PC a i386 CPU?
So try downloading the amalgamation and include sqlite3.c and sqlite3.h in your
project.
http://www.sqlite.org/sqlite-amalgamation-3070500.zip
You'll also find the code will
Hmmm...according to my math...
Max 64-bit unsigned integer is
18446744073709551615
Drop the last digit as it can't hold 0-9
1844674407370955161
Make two decimal positions
18446744073709551.61
Now some commas so we can see better
18,446,744,073,709,551.61
That' $18 quadrillion dollars by my
When you say "All an index does" don't forget that an index is also usually
smaller than the data, thereby increase cache performance and reducing disk
seeks.
For a good chunk of typical uses (large tables with simple lookups) an index is
notably faster.
I'll admit my use of sqtlite3 hasn't
Base 10 multiplication is needed. Although money is the main reason for doing
BCD due to cumulative errors it's not the only reason. So I'd recommend adding
decimal*decimal -> decimal just to be complete. It's easy enough to implement
using the + function so could just be noted as "slow".
Wouldn't the addition of "Empty result sets will return SQLITE_DONE on the
first call to sqlite3_step." add some clarity?
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From: sqlite-users-boun...@sqlite.org
Just to help clarify (hopefully) the Unix/Windows "reserved filename".
CON: is similar to Unix's /dev/zero or /dev/null for example -- Files that
already exist and have OS meaning.
stdout is NOT a reserved filename...it's a predefined variable of FILE *. You
cannot say "cp file stdout" on
I don't understand how your column C works...so I'll assume it's pre-known for
now. But here's how to get A/B to work.
drop table t if exists;
create table t(a int,b int,c int);
create trigger trig1 after insert on t
begin
update t set a=(select count(b) from t where b=new.b) where a=0;
end;
Since it apperas you're running your commit in a separate thread and are
therefore muilti-threaded I do belive you need:
SQLITE3_THREADSAFE=2
>From http://www.sqlite.org/compile.html#threadsafe
To put it another way, SQLITE_THREADSAFE=1 sets the default threading mode to
Serialized.
You apparently don't understand "strings" in C. Or are you actually reading in
binary data?
#1 Since you said "image" I assume you're reading binaary. So get rid of
buffer[fsize]=0. You don't null terminate binary data and that statement is
1-beyond the end of the array (which is from 0 to
How are you trying to view the ouitput.result.txt (and I"ll note that it'sNOT a
text file...it's an image according to what you said.). What's the size of the
file.
And you should be able to post a COMPLETE example to show your testing. What
you say you want to do has been done by many
I ran your code with my test file and I get this...which is perfectly correct.
Do you get something different? What makes you think the stream is truncated
in the database?
Also...change SQLITE_STATIC to SQLITE_TRANSIENT...that could be your culprit if
you are still seeing truncation.
Plus
l
003
I must be doing something fundamentally wrong.;-(
Thanks for the help so far, I really appreciate it.....
Lynton
On 03/04/2011 13:44, Black, Michael (IS) wrote:
> How are you trying to view the ouitput.result.txt (and I"ll note that it'sNOT
> a text file...it's an image
Care to show us your SPL_mallocstr() function?
Sounds like you've corrupted data if that line dies. All it's doing is
checking an array value which is used all over the place in sqlite3.c
If you put a break point there and on the first time it's hit put a watch on
the address for
Seems to behave OK for me on 3.7.5 on Windows. What version are you using on
what OS with what compile flags?
You also "said" it didn't work but you didnt' actually what what you did.
Like this...
SQLite version 3.7.5
Enter ".help" for instructions
Enter SQL statements terminated with a ";"
That's the nice thing about standards...there's so many to choose from :-)
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From: sqlite-users-boun...@sqlite.org [sqlite-users-boun...@sqlite.org] on
behalf of Simon
Ummm...are we forgetting about swap space?
If you exceed RAM you hit swap. If you exceed RAM+SWAP you start failing.
Or does sqlite monitor physical memory usage?
So if you exceed RAM you just start slowing down towards a disk-based equialent
database.
Michael D. Black
Senior
Assuming your database ONLY contains the log entries this should workand be
pretty fast too since rowid is already indexed and there areYou no other
lookups.
You can add your own rowid to make this work otherwise. Just do a
max(myrowid)+1 on your insert.
PRAGMA foreign_keys=OFF;
BEGIN
create table t (d default CURRENT_DATE,i number);
insert into t (i) values(1);
select * from t;
2011-04-21|1
Use CURRENT_TIME if you want hours/minutes too.
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From:
A slight mod on my solution makes it work for DST changes too. Again...rowid
must be maintained.
PRAGMA foreign_keys=OFF;
BEGIN TRANSACTION;
CREATE TABLE log(d date);
INSERT INTO "log" VALUES('2011-03-13 01:55');
INSERT INTO "log" VALUES('2011-03-13 01:56');
INSERT INTO "log"
3.75 on Redhat 5.6 does not have a problem.
I get
8.88
16.88
32.88
64.88
Fixed in 3.7.5 maybe?
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From: sqlite-users-boun...@sqlite.org
Sounds like soundex might work for you. Not sure how it handles diacritics
though.
http://www.mail-archive.com/sqlite-users@sqlite.org/msg35316.html
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From:
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From: sqlite-users-boun...@sqlite.org [sqlite-users-boun...@sqlite.org] on
behalf of Nick [maill...@css-uk.net]
Sent: Sunday, May 08, 2011 3:52 PM
To:
You do realize the number they quote is a MAXnot necessarily what you'll
get.
With 16 transactions per second you're writing on transaction every 63ms. My
understanding is that the hard drive in the netbook is 15ms access time so
you're seeing 4 accesses per transaction with the way
#1 I don't see where you're freeing m_szErrorString (not real sure if it gets
malloc'd on success) -- but you do need to free it on errors for sure.
And where's your Callback function? Why are you calling SaveResultSet (which
you also don't show)? That should probably be done inside the
Do I understand you're still seeing a segfault? I assume you're not seeing
your "Year retrieved..." statement?
You haven't showed us your table definition.
Change your strcmp to strcasecmp and see if that fixes it for you.
Michael D. Black
Senior Scientist
NG Information Systems
Welcome to DLL hell...Microsoft keeps changing it...
http://msdn.microsoft.com/en-us/library/ms682586(v=vs.85).aspx
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From: sqlite-users-boun...@sqlite.org
as the EXE file.
I do not know if there is a link but the user who experiences this
problem is running my application under WIndows 7 32bit. Is there any
possible link?
Thanks again
Cyrille
Le 29/05/2011 20:33, Black, Michael (IS) a écrit :
> Welcome to DLL hell...Microsoft keeps changing it...
>
My radar says having a TEXT field as a primary key is bad (your userTable).
String compares are horrendously slow.
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From: sqlite-users-boun...@sqlite.org
And do you wrap all your updates inside a transaction?
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From: sqlite-users-boun...@sqlite.org [sqlite-users-boun...@sqlite.org] on
behalf of Ian Hardingham
return 0;
}
return 0;
}
On 03/06/2011 15:28, Black, Michael (IS) wrote:
>
> And do you wrap all your updates inside a transaction?
>
> Michael D. Black
>
> Sen
}
else
{
// error occured
Con::executef(this, 2, "onQueryFailed", m_szErrorString);
delete pResultSet;
return 0;
}
return 0;
}
On 03/06/2011 15:28, Black, Michael (IS) wrote:
>
> And do you wrap all your updates inside a transactio
That's the nice thing about standards...there's so many to choose from... :-)
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From: sqlite-users-boun...@sqlite.org [sqlite-users-boun...@sqlite.org] on
behalf of
Ummm...wouldn't it be a whole lot simpler and faster to have an "Online" table?
One id is all that's needed unless you want other info.
Then you're just inserting and deleting records and the whole table is "online"
at any moment.
I don't see any reason why this online status needs to be
You do this:
INVOKE sqlite3_prepare_v2,__hSql,__lpszQuery,-1,eax,edx
Then this:
INVOKE sqlite3_step,_lpSQLStatment
You probably want
INVOKE sqlite3_prepare_v2,__hSql,__lpSQLStatement,-1,eax,edx
At least from what I tell of your logic.
You're not using the same variable fo the statement
Oops...sorry...you're correct...been far too long since I've done assembler.
Are we to understand that if you simply give this table a different name it
works?
That seems to be what you said in your initial email.
Are you inserting records in the table in your assembler too?
Are you
...@phrio.biz]
Sent: Monday, June 13, 2011 12:08 PM
To: General Discussion of SQLite Database
Subject: EXT :Re: [sqlite] A simple SELECT
If I change the table name it works.
I tryed the same query using SqLite3Explorer and it does not give any result
!
-Message d'origine-
From: Black, Michael
) [m...@phrio.biz]
Sent: Monday, June 13, 2011 12:45 PM
To: General Discussion of SQLite Database
Subject: Re: [sqlite] EXT :Re: A simple SELECT
For example "Test"
-Message d'origine-----
From: Black, Michael (IS)
Sent: Monday, June 13, 2011 7:30 PM
To: General Discussion of SQLit
sqlite> create table t(d date);
sqlite> insert into t values('2011-12-31 09:00');
sqlite> insert into t values('2011-12-31 12:15');
sqlite> select d,substr(datetime(d,'-12 hours'),1,16) from t;
2011-12-31 09:00|2011-12-30 21:00
2011-12-31 12:15|2011-12-31 00:15
Michael D. Black
Senior
See if this thread helps you -- a couple of different solutions. Both of which
you should be able to solve for your users.
http://forums.asp.net/t/939729.aspx/1
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
Also...try dependency walker to ensure you have all the DLLs you need deployed
with your package that aren't "standard".
http://www.dependencywalker.com/
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From:
questions are stupid but I cannot figure it out.
Best regards,
Cyrille
Le 18/06/2011 15:50, Black, Michael (IS) a écrit :
> Also...try dependency walker to ensure you have all the DLLs you need
deployed with your package that aren't "standard".
>
> http://www.dependencywalk
As I said before...welcome to DLL hell...
You guessed itthe original DLL will give an error if any of the DLLs it
depends on are not loadable.
Since sqlite.interop.dll is non-standard you should include ALL DLLs that it
depends on in your installation. Subject to Microsoft's
I believe this will work if you put the SQL-required semi-colons at the end of
your statements.
sql = "BEGIN;"; //you need to add newline here
sql += "create table episodes (id integer primary key, season int, name
text);";
sql += "insert into episodes(id, season, name) Values(1,2,'bill');";
Any reason you can't add another field to your database?
0=equal
1=n1n2
The create an index on that field.
Then your query plan would look like this:
sqlite> explain query plan select n1,n2 from table2 where flag = 1 intersect
select n2,n1 from table2 where flag = 2;
sele order
[sqlite-users-boun...@sqlite.org] on
behalf of Black, Michael (IS) [michael.bla...@ngc.com]
Sent: Thursday, June 23, 2011 2:03 PM
To: General Discussion of SQLite Database
Subject: EXT :Re: [sqlite] Query with UNION on large table
Any reason you can't add another field to your database?
0=
Ucaching?
Try doing your first select last and see what happens. I'm betting your first
"select min(*) from test" will be a lot slower.
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From:
NG Information Systems
Advanced Analytics Directorate
From: sqlite-users-boun...@sqlite.org [sqlite-users-boun...@sqlite.org] on
behalf of Luuk [luu...@gmail.com]
Sent: Saturday, June 25, 2011 7:37 AM
To: sqlite-users@sqlite.org
Subject: EXT :Re: [sqlite] ques
s-boun...@sqlite.org] on
behalf of Black, Michael (IS) [michael.bla...@ngc.com]
Sent: Saturday, June 25, 2011 8:15 AM
To: General Discussion of SQLite Database
Subject: EXT :Re: [sqlite] question about sqlite aggregate funcitons
OK...I generated 10M ints and imported them. I get times that I would
Can the optimizer recognize these "instant" functions and perhaps lean towards
using them?
Are there cases where this would NOT be good?
I did show you can construct a query which runs at the "correct" speed.
sqlite> select * from (select min(i) from test) as a,(select max(i) from test)
Slavin [slav...@bigfraud.org]
Sent: Saturday, June 25, 2011 10:38 AM
To: General Discussion of SQLite Database
Subject: EXT :Re: [sqlite] question about sqlite aggregate funcitons
On 25 Jun 2011, at 4:12pm, Black, Michael (IS) wrote:
> Can the optimizer recognize these "instant&qu
It's not obvious but this works
CREATE TABLE x (s string);
INSERT INTO "x" VALUES('ab:cdef');
INSERT INTO "x" VALUES('ghij:klmn');
sqlite> select
ltrim(ltrim(s,'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ'),':') from
x;
ltri
cdef
klmn
Just make sure your char set contains all
I'd recommend NOT relying on the system sqlite3. That way you can control your
changes.
Get the amalgamation and put sqlite3.c and sqlite3.h in your project.
And, you forgot to put in the name for "-o" -- so you would get a file named
"-lsqlite3" in your directory.
And you'll probably
Use single quotes instead of double
sqlite> select (select v from t1 where n='a') wrong,* from a1;
wrong|a|b
2000|123|456
2000|999|999
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From:
You're getting closeif you don't use a field in a table you don't HAVE to
create it.
Also...if you want to make your database a bit more bullet proof you want
foreign keys to help ensure you maintain the relationship between users and
tournaments (otherwise you can accidentally delete
e: [sqlite] Defining a relationship as unique
On 06/30/11 02:31 PM, Black, Michael (IS) wrote:
> sqlite> create table user(userid integer primary key autoincrement,name
> varchar, login varchar);
> sqlite> create unique index index1 on user(userid);
Isn't userid already unique by &q
If you can build your SQL string with column_name exactly what is the problem
with adding it a 2nd time?
How do you build your SQL statement?
And I get the same answer with either of your queries in 3.7.4 -- what version
are you running and what answer are you getting?
Are you saying that
That would 'splain it...I should have noticed...
So we're left with this question...
What's the problem with building the query that works? I don't see where you
can't make it "column agnostic"...though perhaps there's another definition of
that
sprintf(sql,"select * from t where %s
You're going to have to create a complete stand-alone example if you want us to
debug your code for you.
Make a simple program that only does that one insert and share that with us.
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
If you hash the username too you don't have that problem.
Passwords should always be a 1-way hash.
It's actually best to collapse username/password into a single one-way hash.
That way it's very difficult to crack it.
You can use a user "account#" value that users would have to track for
I downloaded 5.3.6 from here -- I used the Thread Safe version Installer.
http://windows.php.net/download/
And you're script ran just fine and created this:
sqlite> .dump
PRAGMA foreign_keys=OFF;
BEGIN TRANSACTION;
CREATE TABLE tbl2 (one varchar(10),two varchar(10));
COMMIT;
I the modified
Any particular reason you can't build your own string and just pass one
parameter?
select * from words where "word" like ?;
And any reason why you don't create a 2nd field holds the 1st and last char?
and index that? Then your query should be blazingly fast as it will actually
use the index
3.7.4 doesn't indicate it will use an index in either case with like...but glob
seems to call the index...why does glob use the index but like does not?
sqlite> create table words(word text);
sqlite> create index idx on words(word);
sqlite> explain query plan select * from words where "word"
te> select count(*) from words where word like 'L%P';
136
CPU Time: user 0.281250 sys 0.00
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From: sqlite-users-boun...@sqlite.org [sqlite-users-boun...@s
Somebody smarter than I may be able to figure out how to use views to do the
upper levels.
But if you can afford your database to be a bit less then twice as big just use
tables.
create table level1(id int,l int,r int);
insert into level1 values(1,251,18);
insert into level1 values(2,5,91);
If CREATE has no return why do I get a return?
To be clear I'm using PHP 5.3.6 and it most definitely gets a return
But I re-wrote this according to some PDO examples I found. This is a more
complete example.
You should be able to run this with a PHP command line -- forget the web
I can see adding a forward/reverse link to the tables making it a linked-list
type structure much like your btree.
By default each node is linked to the one in front and back. Then you adjust
those pointers for cut/paste operations.
You could also do the cut/paste just by copying to a new
I thought of another way to do your copy/cut/paste...
Assuming you keep the original audio around and use the levels I showed before.
create table sequence(level int,parent int,start int,end end);
insert into seqeunce values(1,0,0,-1); // note that -1 means "until end of
data".
See
http://www.sqlite.org/compile.html
Read section 1.4 -- tells you all you need to know.
First thing you should do is see if there's any significant performance
difference with -Os instead of -O3.
You may be surprised.
Michael D. Black
Senior Scientist
NG Information Systems
Advanced
Would glob be faster than like? Since I assume like has to case-convert?
And you don't have to set any pragmas or NOCASE for it to use the index.
select * from tsamov where tsamov_code glob 'AFG*';
sqlite> explain query plan select * from tsamov where tsamov_code glob 'AFG*';
sele order
If each pid sees a completely separate set of rows then add the pid to each row
and select by pid.
If 2 pid's can see any of the same records this won't work (i.e. all pid's must
be mutually exclusive)
No views necessary but you could create a view for each pid if you wanted to
for some
Column numbers are zero-based, not one-based.
>From http://www.sqlite.org/c3ref/column_blob.html
"The leftmost column of the result set has the index 0. "
That's at least part of your problem.
So should be:
k.AAArticoli=sqlite3_column_int(statmentS, 0);
If you don't know how to get your rules yet then you don't know how to design a
solution.
Based on what you've said I see 2 more tables.
create table (pid int, rule int)
create table (rule int, record int)
You can have multiple rules per pid, reuse rules across pids, and records can
I decided to test this so here's an almost-complete example minus any
cross-checks on the tables which you should ultimately do.
This implements exclusion rules which is what you've been describing.
You can make them inclusion rules if you just remove the "not" in the select
statement below, but
If you showed a small sample code that caused your problem it would sure help a
lot.
I peeked at the library code. That message is generated when _cnn is null (the
DB connection).
private void InitializeForReader()
{
if (_activeReader != null && _activeReader.IsAlive)
Unless your caching at byte-level you don't need all the addresses.
For 4K page size mask off the lower 11 bits.
You should actually be able to reduce your memory usage by a LOT if you track
by pages and not bytes.
That will also speed you up along with it.
Michael D. Black
Senior
Could you define "awfully slow"? That's pretty hard to tell if your speed is
what one should expect.
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From: sqlite-users-boun...@sqlite.org [sqlite-users-boun...@sqlite.org] on
behalf of Paul Linehan
select id,count(distinct track) from tracks group by id having count(distinct
track)>=10;
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From: sqlite-users-boun...@sqlite.org [sqlite-users-boun...@sqlite.org] on
You need to normalize your database.
Though it's easy to put everything in one table it's horrendously inefficient
for indexing.
Your indexes are huge as you're putting strings in them.
Ideally you should never have a string in an index if you can avoid it. Too
much space, too long a
Try "order by code.rowid"
I think that will do what you want and fits in with what sqlite3's behavior
should be.
As long as you don't delete rows you'll be OK.
http://www.sqlite.org/autoinc.html
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From:
And don't you wish Microsoft and other's could spit out intelligent error
messages like:
"64-bit application trying to load 32-bit DLL".
I ding my students whey do stupid stuff like:
FILE *fp=fopen(filename,"r");
if (fp == NULL) {
fprintf(stderr,"Cannot open file\n");
}
Instead
An 8X speedup is a pretty good achievement...congrats...
#1 I take it your query is CPU bound the first time?
#2 Can you show us the query planner please?
#3 Can you show us the query planner minus the "INDEXED BY"?
#4 Can you show us sqlite_stat1?
#5 Can you show us your tables now?
#6
You need to normalize your "kind" value.
.pragma cache_size=15000;
drop index idxlog_kind_computer;
create table kind(id integer,kind text);
insert into kind values(1,'debug');
insert into kind values(2,'error');
insert into kind values(3,'info');
insert into kind values(4,'timing');
insert
That's because they have 2 completely different query plans.
I created the table so that id,a,b,c all had the same values so the indexing
would be indentical.
#include
main()
{
int i;
for(i=1;i<=10;++i) {
char sql[4096];
sprintf(sql,"insert into abctable(a,b,c)
If I run your sql.txt script with the normalization of kind my first time query
shows 0 seconds.
D:\x>sqlite3 sq1
> select count(*) as cnt,kind,computer
> from log
> group by kind,computer
> order by cnt desc
>
> what happens here?
>
> SELECT *
> FROM log INDEXED BY idxlog_kind_computer
>
Part of the problem is it seems you can't create an index with rowid:
3.7.5
sqlite> create table t(i int);
sqlite> create index idx1 on t(i);
sqlite> create index idx2 on t(i,rowid);
Error: table t has no column named rowid
Any particular reason it can't be included in an index?
What they don't say explicitly is that if all you need is key/value capability
then an SQL database is overkill and only slows you down (bit of a duh factor
there though not obvious to neophytes). Generally speaking that's one thing
they don't teach in college is optimization. I can't count
bject: EXT :Re: [sqlite] LevelDB benchmark
2011/7/29 Black, Michael (IS) <michael.bla...@ngc.com>:
> What they don't say explicitly is that if all you need is key/value
> capability then an SQL database is overkill and only slows you down (bit of a
> duh factor there though
I ran a few tests. First the 2 tests from Rev 45. Then NORMAL sync, and
finally bump the cache up to what sqlite3 needs vs an arbitrary 100M. I think
the question should be how well does it perform in cache -- not necessarily how
much cache it needs. How much cache is just a comparision of
This is a side-question to this thread...but has anybody every done row-level
locking for edit?
I can see it:
create table t(id int primary key,stuff text, lock l);
insert into t values(1,'stuff1',0);
select * from t where id=1 and lock=0; // or drop lock to get all and check
lock!=0 to
If it's meaningless then shouldn't it be a syntax error?
Michael D. Black
Senior Scientist
NG Information Systems
Advanced Analytics Directorate
From: sqlite-users-boun...@sqlite.org [sqlite-users-boun...@sqlite.org] on
behalf of Simon Slavin
1 - 100 of 844 matches
Mail list logo
