> 1. If you define a column with NOCASE and later an index without, it
> won't be possible to insert two values differing only by case,
> because the column will reject it.
Of course it will accept the value, unless you declared the column unique so
that a unique index is created using the nocas
On Tue, 3 Sep 2013 23:50:09 +0200
Eduardo Morras wrote:
> Don't know if column collation overrides index collation or viceversa.
It's probably simpler to think of them as two things, table and index.
Neither "overrides" the other.
Ideally, they use the same collation. In the event not, I wou
On Mon, 2 Sep 2013 13:48:02 +0100
Simon Slavin wrote:
>
> On 2 Sep 2013, at 8:25am, Eduardo Morras wrote:
>
> > Or create the index with collate
> >
> > CREATE INDEX idx_collated_column ON myTable ( column COLLATE NOCASE )
>
> The problem with doing it in the index is that it's hard to predi
> Plus, of course, index will only ever be used for operations where you have
> overridden the default collating sequence for the operation, for example by
> specifying collate nocase in the join expression, or adding the collate
> nocase to
> the order by or group by.
I assume this explains why
age-
> From: sqlite-users-boun...@sqlite.org [mailto:sqlite-users-
> boun...@sqlite.org] On Behalf Of Simon Slavin
> Sent: Monday, 2 September, 2013 06:48
> To: General Discussion of SQLite Database
> Subject: Re: [sqlite] Query problems
>
>
> On 2 Sep 2013, at 8:25am, Eduar
On 2 Sep 2013, at 8:25am, Eduardo Morras wrote:
> Or create the index with collate
>
> CREATE INDEX idx_collated_column ON myTable ( column COLLATE NOCASE )
The problem with doing it in the index is that it's hard to predict when SQLite
will use a particular index. Better to think about the n
On Mon, 2 Sep 2013 02:12:05 +0100
Simon Slavin wrote:
>
> On 2 Sep 2013, at 2:03am, Joseph L. Casale wrote:
>
> > I am using LIKE as the columns are indexed NOCASE and I need the
> > comparison case insensitive.
>
> Have you tried using '=' ?
>
> Also if you declare the columns as COLLATE NO
> Have you tried using '=' ?
>
> Also if you declare the columns as COLLATE NOCASE in your table definition,
> then using '=' will definitely work the way you want it to. An example would
> be
>
> CREATE TABLE myTable (myName TEXT COLLATE NOCASE)
Simon,
That took this query from not finishing
> > 0 0 1 SCAN TABLE d_table_b AS da (~10 rows)
> >
>
> Is this the index you referenced in you reply to Simon?
> Maybe you are using wrong index/column?
I'll recheck, I am also reading up on indexes as they relate to optimizing
queries. Could be I made a mistake.
> I had t
> LIKE is used when comparing strings with wildcards. For example, val LIKE
> 'abra%' (which will match 'abraCaDAbra' and 'abrakadee'.
>
> If there are no wildcards you should be using =, not LIKE. LIKE will/should
> always indicate that a table or index scan is required, perhaps of the whole
>
Hi, Joseph,
On Sun, Sep 1, 2013 at 6:21 PM, Joseph L. Casale
wrote:
> > Hi,
> > Can you do "DESCRIBE QUERY PLAN " and post results here?
> >
> > Also, what do you mean by "unbearable at scale"? Did you measure it? What
> > is the result?
> >
> > Thank you.
>
> It doesn't finish with maybe 4 or 5
> Have you tried using '=' ?
>
> Also if you declare the columns as COLLATE NOCASE in your table definition,
> then using '=' will definitely work the way you want it to. An example would
> be
>
> CREATE TABLE myTable (myName TEXT COLLATE NOCASE)
>
> Simon.
I did and it excluded the compariso
> I am using LIKE as the columns are indexed NOCASE and I need the
> comparison case insensitive. I suspect this is where is breaks down
> but I don't know enough sql to really appreciate the ways I could
> approach this better.
LIKE is used when comparing strings with wildcards. For example, va
> Hi,
> Can you do "DESCRIBE QUERY PLAN " and post results here?
>
> Also, what do you mean by "unbearable at scale"? Did you measure it? What
> is the result?
>
> Thank you.
It doesn't finish with maybe 4 or 5 hours run time.
Sorry, do you mean "explain query plan ..."?
0 0 1
On 2 Sep 2013, at 2:03am, Joseph L. Casale wrote:
> I am using LIKE as the columns are indexed NOCASE and I need the
> comparison case insensitive.
Have you tried using '=' ?
Also if you declare the columns as COLLATE NOCASE in your table definition,
then using '=' will definitely work the wa
Hi,
Can you do "DESCRIBE QUERY PLAN " and post results here?
Also, what do you mean by "unbearable at scale"? Did you measure it? What
is the result?
Thank you.
On Sun, Sep 1, 2013 at 6:03 PM, Joseph L. Casale
wrote:
> I have a query that is unbearable at scale, for example when
> s_table_a and
On Thu, Sep 29, 2011 at 8:42 PM, David Garfield
wrote:
> If the value in key is a blob, then like matches it and = does not.
Thanks. This suggestion helped me track down the issue. I was actually
going through ActiveRecord (the ORM for Ruby on Rails). Was upgrading
my app for Ruby 1.9 and my unit
On Thu, Sep 29, 2011 at 08:32:04PM -0400, Eric Anderson scratched on the wall:
> The below statement returns records:
>
> SELECT * FROM recordings WHERE "key" LIKE
> '4df0247ce1a97685a782d2cb051b48ed952e666c';
>
> But this one does not:
>
> SELECT * FROM recordings WHERE "key" =
> '4df0247ce1a97
If the value in key is a blob, then like matches it and = does not.
Because like has to do a string conversion on key, it also doesn't use
the index.
Try:
SELECT * FROM recordings WHERE "key" =
cast('4df0247ce1a97685a782d2cb051b48ed952e666c' as blob);
Or try inserting the key as text in the fir
19 matches
Mail list logo