Re: [sqlite] First beta for SQLite 3.18.0.

2017-03-20 Thread Simon Slavin 

On 20 Mar 2017, at 2:45pm, Simon Slavin  wrote:

> On 20 Mar 2017, at 2:42pm, Richard Hipp  wrote:
> 
>> More changes.  Please Reload one more time.
> 
> The changes are good but it’s still showing up as "alway" for me.  I even 
> tried a different browser.

Ignore that.  Now showing up correctly as "always".

Simon.
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] First beta for SQLite 3.18.0.

2017-03-20 Thread Richard Hipp 
On 3/20/17, Simon Slavin  wrote:
>
> On 20 Mar 2017, at 2:13pm, Richard Hipp  wrote:
>
>> All other corrections implemented.
>
> Check "alway" on the current mask sentence again.  No correction shown when
> I retrieve the page.

More changes.  Please Reload one more time.

-- 
D. Richard Hipp
d...@sqlite.org
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] First beta for SQLite 3.18.0.

2017-03-20 Thread Simon Slavin 

On 20 Mar 2017, at 2:42pm, Richard Hipp  wrote:

> More changes.  Please Reload one more time.

The changes are good but it’s still showing up as "alway" for me.  I even tried 
a different browser.

Simon.
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] First beta for SQLite 3.18.0.

2017-03-20 Thread Simon Slavin 

On 20 Mar 2017, at 2:13pm, Richard Hipp  wrote:

> All other corrections implemented.

Check "alway" on the current mask sentence again.  No correction shown when I 
retrieve the page.

Simon.
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] First beta for SQLite 3.18.0.

2017-03-20 Thread Richard Hipp 
On 3/20/17, Simon Slavin  wrote:
>
> " • (Not yet implemented) Create indexes that might have been helpful to
> recent queries."
>
> —> If I recall a previous conversation correctly, it might clarify things to
> make this sentence end "queries executed by this connection.".  On the other
> hand you may have changed your idea about how this should work and that may
> no longer be accurate.
>

The decision might also be predicated on queries from prior
connections using information saved by the 0x0004 bit.  Or this
feature may never be implemented :-)

All other corrections implemented.
-- 
D. Richard Hipp
d...@sqlite.org
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] First beta for SQLite 3.18.0.

2017-03-20 Thread Simon Slavin 

On 20 Mar 2017, at 1:06pm, Richard Hipp  wrote:

> https://www.sqlite.org/draft/index.html

In 

"   • (Not yet implemented) Create indexes that might have been helpful to 
recent queries."

—> If I recall a previous conversation correctly, it might clarify things to 
make this sentence end "queries executed by this connection.".  On the other 
hand you may have changed your idea about how this should work and that may no 
longer be accurate.

"The default MASK is and alway shall be 0xfffe. "

—> "always"

If it’s possible to specify masks using BLOB-type hex values, could you provide 
an example which shows how ?  Simply saying that "PRAGMA optimize(0x2);" works 
would be good enough.

"In the current implementation, a table is analyzed if only if"

—> "analyzed only if"

"   • Debugging mode. Do not actually perform any optimizations but instead 
return one line of text for each optimization that would have been done."

—> The section describes what is returned in debugging mode, but not what is 
returned in normal mode.  Presumably it returns that same text.

———

I see two patterns of use for this.  Some programmers will just use normal 
mode.  Others will use debugging mode first, and count the number of lines 
returned.  If the count is non-zero they will put up a dialog telling the user 
of a possible delay, allowing them to pick whether the optimizations are done 
or not.

Simon.
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users