On the web page at http://sqlite.org/csv.html the following text appears:
The CVS virtual table is not built into the SQLite amalgamation.
The acronym CVS should be CSV.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users
aller than 0x24, the initial byte of a text value"
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
d
to be the same thing.
Which of these is the correct name, or are they equivalent?
--
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
(or the ™symbol
for an unregistered trademark for that matter) on your website, so I
assumed you hadn't registered the trademark.
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
rs considerably. One of its
claimed novelties was that "Multics implemented a single level store for
data access" which could easily be confused with database functionality.
I'm not a lawyer, but I don't think you would have much ground to stand
on in trademark dispute, especially if
te.org/cvstrac/wiki?p=UndoRedo. It explains how to do
this sort of undo system using triggers.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
misnamed since it is really an interface to
whole underlying operating system, *NOT* just the filesystem."
Or something similar. Maybe the NOT doesn't need that much emphasis, but
it should be there.
HTH
Dennis Cote
___
sqlite-users maili
s a recommended optimization for
applications were it makes sense."
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
nd uncheck Open
the Last Used Database. That may clear the saved database name.
After that you can try re-enabling the option after opening the new
database file.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/
t; & _
"FROM TmpTable WHERE Day = '11'"
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
/www.sqlite.org/cintro.html for
more details).
You do not need to clear the old bindings if you are going to bind new
values for each of the variables before you execute the command again.
The clear binding call basically binds a NULL to each variable, which
you are overwriting anyway.
ate table t(a,b);
sqlite> pragma empty_result_callbacks=1;
sqlite> .header on
sqlite> select * from t;
a|b
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
natural join user
where movies.id = 'tt0426459';
select title,my_rating
from movies natural join user
where user.id = 'tt0426459';
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi
source file shell.c from
http://www.sqlite.org/sqlite-source-3_6_16.zip. Compile and link shell.c
and sqlite3.c from the amalgamation together like this (untested):
gcc sqlite3.c shell.c -o sqlite3.exe
HTH
Dennis Cote
___
sqlite-users mailing list
TE TRIGGER IF NOT EXISTS log_entry
AFTER UPDATE OF creator, editor ON "a"
BEGIN
--Insert log record for newly inserted or updated record
INSERT INTO log ...
END;
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
INTO a(code,name,identity) VALUES('V','abc',1);
why don't you do this:
INSERT INTO a(code,name,creator) VALUES('V','abc',1);
I think you may need to expand on your description of "shooting
new.identity into another field" to clarify what you are trying to
accomplish.
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
red to allow concatenation and other function calls. I assume that
the other possible values (join keyword etc) would be caught and
reported as errors.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
be used for each insert, execute
the insert, then reset the statement to be run again for the next insert.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
okups in table2 using the rowid should be very fast, and
once the page with the required record has been read into the cache the
subsequent value lookup queries
should execute very quickly as well.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-
hen only change again when switching to values that
required 4 bytes.
This may be a part of the answer even if it is not the complete answer.
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
or when it starts or
ends. The values shown for those other fields are randomly selected from
the set of rows in the matching group.
SQLite and some other database programs don't complain when you to break
this rule because it is sometimes useful to get a randomly selected
value for a column in the group. Usually it just leads to the confusion
you are seeing.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
the code as posted.
Perhaps the OP should file a bug report using this example. If there is
a bug it will probably be fixed in short order.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
tained automatically by triggers.
Having the paths available turns many common tree queries into pattern
matches against the path using like conditions on standard SQL queries.
Unless your tees are very large I find this method to be a very good
alternative to simple adjacen
me, or do others find jibberish like "wat, jus, wanna, i,
wen, etc..." to be very distracting and not the least bit "cool"?
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
vetted by users and the developers before
the changes are implemented. I think it could help to avoid future
errors like the sqlite_column_text return type issue.
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/
e implemented in the
next breaking release of SQLite?
I'm thinking of things like renaming the _v2 API functions (in
particular prepare_v2) to drop the suffix, and changing the name of the
CURRENT_TIMESTAMP default value to CURRENT_DATETIME so that
CURRENT_TIMESTAMP can be used for an julian da
>
From this it's easy (relatively) to see that inserts each row as it
find them while executing the select.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
te-users@sqlite.org/msg12121.html
for an example.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
command shell to store the result of the first query and pass it back
into a second sqlite command. You would be replacing Python programming
with bash shell scripting.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:808
e. It would eliminate the need to
combine them for this type of query. The tables already have a type
field to distinguish the email adresses from the phone numbers, so there
is no need to put them in separate tables.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
. If my memory serves me
correctly, I seem to recall it added some kind of a COMMIT opcode to the
end of a statement when it was compiled outside a transaction. This
opcode would incorrectly close the transaction when executed inside a
transaction.
HTH
Dennis Cote
___
at for functions.
You can get a list of registered collations using the "pragma
collation_list", but there is no equivalent for functions.
SQLite seems to assume that only the application that creates a function
will ever use the function, so there is no need for this information.
HTH
Denni
NSERT ...
COMMIT;
CREATE UNIQUE INDEX probe on probes (probe);
Of course this won't work if you are relying on the unique constraint to
eliminate duplicate strings in your data.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
htt
t add the sqlite3.c file to your project and include sqlite3.h in
your source files that call sqlite functions.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
ntf(stderr, "Error: %d : %s\n", rc, sqlite3_errmsg(db));
break;
}
}while( rc==SQLITE_ROW );
// finalize the statement to release resources
sqlite3_finalize(stmt);
}
// close the database file
sqlite3_close(db);
}
}
return rc!=SQLITE
nt
where starttime =
(select starttime from event
where starttime <= windowstart
order by starttime desc
limit 1);
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
/pragma.html#modify.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
, but it really is
strange to the vast majority of the worlds population. :-)
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
erential integrity triggers on the database, those
triggers will prevent inserting illegal values into your table.
You can then display the restricted data values by joining the main
table with the domain table(s), possibly using a view.
create view tv as
select id, flag, s.size as size, data
from t join sizes as s on t.size=s.id;
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
define a
collating sequence using a string, and a suitable string is defined.
He has already said that it won't be suitable for multiple byte
characters or many other languages. For those cases where it is not
suitable, a user could continue to use the ICU extension just as they
can now.
Dennis
le t (
a varchar(15) check (length(a) <= 15),
b integer check (typeof(b) = 'integer')
);
insert into t values('one', 1);
insert into t values('two', 'three');
insert into t values('one hundred twenty three million...'
expect.
select x from
(
select x from
(select x from a
union
select x from b)
as ab
intersect
select x from
(select x from c
union
select x from d)
as cd
Tom,
You should add a column for the Spatialite GUI which can be found at
http://www.gaia-gis.it/spatialite/
It is a free open source Mac OS X native GUI DB management tool.
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http:/
ave to upgrade to a newer version to get access to that function.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
Jibin Scaria wrote:
> I am facing problem with spaces in the database path, able to open database
> but queries are returning "no such table: table name".
>
>
>
If your table names contain embedded spaces you must quote the name in
your SQL queries
select * from
in
floating point division.
sqlite> select TOTAL(1);
1.0
sqlite> select TOTAL(1)/3;
0.333
Alternatively, if you want to use SQL that is more portable, you could
also cast the result of the SUM function to a floating point value to
ensure that a floating point
es('one');
sqlite> select * from test_table;
test_field two
-- --
-1 one
You can report the bug at
http://www.sqlite.org/cvstrac/captcha?nxp=/cvstrac/tktnew
HTH
Dennis Cote
___
sqlite-users mailing
On Sun, Oct 5, 2008 at 7:59 AM, Gilles Ganault <[EMAIL PROTECTED]>wrote:
>
> This is for an accounting program: As an exemple, all accounts
> starting with 6 must be listed together, which means that eg. 6001
> must come after 601, and not after, say, 801.
It sounds like you want the account
t would be best if you
could post the code you are using to prepare and execute the query.
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
it is now doing case sensitive
comparisons. This may lead to a change in behavior if the string
zConflict is not constrained to be lower case elsewhere in the code.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.o
P Kishor wrote:
>
> Still, you have a point, and maybe DRH will expound and enlighten us
> on his reticence to enable locking style equal to one.
>
It seems he isn't so reticent after all. See checkin [5737] from this
morning at http://www.sqlite.org/cvstrac/chngview?cn=5737.
itted, or on every update? I am
> assuming the former, but, in that case, how will know how many rows
> were inserted?
>
Same for the insert trigger. It executes after each row is inserted.
Always add 1.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
ere number = '12345678')
order by a.address_id;
I am still working on a complete query, but I have managed to trigger a
crash in SQLite in the process.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
or the select * query.
>> Have you filed a bug report to see in any of these can be resolved?
>
I will file a bug report for this one.
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
>>
I get column names of
select|select
Yes, that would be another bug. The second columns name should include
the back ticks that were escaped by doubling them inside the outer back
tick quotes. SQLite seems to get confused because t
ever you need it.
You could also do a join to the view using the id column whenever you
want do get the mac address string in a query that still needs to use
the original mac address as an integer.
select data, mac_addr
from t
join tv on tv.id = t.id
where t.mac in (select ...)
6789ABCDEF', ((mac >> 32) & 15) + 1, 1) ||
':' ||
substr('0123456789ABCDEF', ((mac >> 44) & 15) + 1, 1) ||
substr('0123456789ABCDEF', ((mac >> 40) & 15) + 1, 1)
as 'MAC Address'
from t;
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
15) + 1, 1) ||
':' ||
substr('0123456789ABCDEF', ((mac >> 4) & 15) + 1, 1) ||
substr('0123456789ABCDEF', ((mac >> 0) & 15) + 1, 1)
as 'MAC Address'
from t;
This assumes that the table t has an integer column mac that hods the
mac address to b
ve, and what format do you really want to
use to display it?
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
t column names (at least
that I am aware of), so this sort of code is inherently database engine
specific.
> I also hope, however, that you may see the current behavior as
> undesirably inconsistent, even if it is justifiable by your design
> philosophy.
>
You should perhaps file a bug report to see if this can be corrected.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
ment, and it returns SQLITE_DONE, then it
completed and hence would have closed any transaction it opened in auto
commit mode. If you still have a journal file then you must have a
manual transaction open.
You will probably get better, more detailed, help if you post the code
you are using
Clark Christensen wrote:
>
> Long setup for a simple question: Is null the expected result when
> one column of a concatenation operation is null?
>
Yes, that is the result required by the SQL standard. The result of a
concatenation operator is NULL if either argument is NULL.
create view foo_B as
select B.* from A join B on B._ID = A.b where A.name = 'foo%';
With this view defined you can now get the same results using a simpler
query.
select * from foo_B;
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
he id_list data.
For fastest operation the mt.table.id should also be an "integer primary
key" column as this will eliminate a rowid lookup operation if it is an
indexed column.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
Lothar Behrens wrote:
> Am 10.09.2008 um 17:37 schrieb Dennis Cote:
>
>> Lothar Behrens wrote:
>>> What is the function to rollback a transaction or commit ?
>>> I want also to break into these functions. If there is no way I try
>>> to implement the
ce.customer_id with a compound index on invoice.customer_id and
invoice.status.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
rent version.
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
pdate mytable
set myfield = trim(myfield)
where rowid = new.rowid;
end;
Now your application can insert untrimmed data, but the database will
only store trimmed data, and therefore you will only ever retrieve
trimmed data.
HTH
Dennis Cote
___
ing after a busy return. This will probably
cause problems if you use this to execute a query.
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
ion operator. This coercion works in the same way as cast
operator, ie it ignores any non numeric suffix.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
30868 in the amalgamation source for version 3.6.2.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
If you close the database without committing
this transaction, the changes that you can see in your application will
be rolled back and lost (see H12019 at
http://www.sqlite.org/c3ref/close.html).
Can you add a function to check the auto commit status in your main l
commit mode).
See http://www.sqlite.org/c3ref/get_autocommit.html for details.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
in database and ensure that is the
same file you are looking at with your database browser.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
e function to reverse the string you are trying
to match and use a like comparison to locate the strings quickly using
the index (since the search string is now the prefix of string).
select field from sometable
where reversed like reverse(:somestring) || '%';
HTH
Dennis Cote
_
uld not be resetting the prepared statement on a busy return. You
should simply sleep and then continue to retry.
If you want to ensure you don't loop forever, you could add a retry
counter and do a reset and return if the retry limit is exceeded.
HTH
Dennis Cote
ngID = :theBuilding) + 1,
:theBuilding);
Now you only need to specify the building and sqlite will calculate the
lowest unused room number in that building.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin
e exists a room 101 in building A, and a room 101 in
building B). In this case, you do *not* want to auto increment the
roomID values. Now you require the combination of the roomID and the
buildingID to identify a particular room.
HTH
Dennis Cote
_
n could check for the existence of the journal file before
opening the database, and report that the incomplete transaction will be
rolled back as the database is opened. I doubt if many applications do
this however.
HTH
Dennis Cote
___
sql
database was opened. Did you perhaps delete the journal file before
starting sqlite or your application after the power failure?
I'm sorry I can't help you with recovering your database, except to
suggest going back to your last backup, and redoing the changes since then.
HTH
Dennis Cote
reason to evaluate the length function twice.
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
key, RNDValue real);
insert into rnd
select rowid, (random() / 9223372036854775807.0 + 1.0) / 2.0 as
RNDValue from names;
-- select name based on associated random numbers
select name, RNDValue
from names
join rnd on names.rowid = rnd.id
where rnd.RNDValue < 0.99;
-- delete temp ta
ion time is only a small percentage of the statement's
execution time (i.e complex long running queries on large tables).
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
Hardy, Andrew wrote:
> Is there any way to log conversions to highlite any issues that would
> have been hilighted by failure with strict typing?
>
Not that I am aware of.
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite
toring,
loading, or comparing the values.
See http://www.sqlite.org/datatype3.html for the column type affinity
deduction rules.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
s because the difficulty in
writing higher level, more complex, algorithms in assembler often leads
to the use of simpler slower algorithms.
In short, writing in assembly language does not guarantee that the
resulting program will be fast. Assembler can be fast, but it is by no
means certain that
Alexey Pechnikov wrote:
>
> Is it included to 3.6.1 or 3.6.2 version?
>
No, it is not included in either version. The patch was submitted by the
mozilla group, but it has not been checked in to SQLite.
You can of course apply the patch to your own customized version of SQLite.
H
"create
virtual table" statement is shown here
http://www.sqlite.org/lang_createvtab.html. The virtual table statement
does not allow the optional "if not exists" clause.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.
se premises are incorrect, and/or the logical
> deduction from the premises to the conclusion is flawed.
>
Classic. :-)
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
n the pager and the OS interface layers.
Nothing else changes, and all the data is stored securely encrypted in
the pages of the file.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
Brown, Daniel wrote:
> Interesting, I just tried that in my test application and Dennis's and I
> get access violations during the vacuum command execution when trying to
> resize the pages from 1k to 4k with my database or Dennis's test
> database.
>
Daniel,
I have found that sqlite works
ll insert a row with a rowid somewhere between 0 and 999.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
something similar that is
causing the unexpectedly large expansion.
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
ant and valuable.
Furthermore, there are other, non-rule based issues, such as "out of
memory" or "I/O error" that may cause an API function to fail. The mere
existence of such possibilities should not be grounds to avoid
documenting the errors that are returned when any of the many
org/c3ref/next_stmt.html for details.
It was introduced in 3.6.0 (see http://www.sqlite.org/changes.html), so
it was not present in 3.5.9.
HTH
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
f your code copied
all the tables in the test database out to the new database file with
the new page size. This database doesn't use any named (i.e,
non-automatic) indexes or triggers, so there was nothing else to be copied.
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
esting, both
the version with the 1K page size (17.4 MB) and the one with the 4K page
size (12.2 MB).
Where would you like me to send them? The zipped versions are each about
1.3 MB in size.
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sql
r
memory requirement, but didn't change the memory required to hold the
database after the copy was completed.
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
t 2000 to a much smaller 100
pages reduced the highwater mark to 18544 KB, which is only slightly
higher than the 18102 KB of memory used after the table is built. The
actual memory used is exactly the same (as expected since it is storing
the same tables).
Dennis Cote
_
ase file
was set when it was created.
Daniel, can you run a "pragam page_size;" query on youyr database and
let us know the results?
Dennis Cote
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
1 - 100 of 1177 matches
Mail list logo