Re: [persistence.js] Filter QueryCollection with LIKE Operator

[Zef Hemel]

You could actually try: YourEntity.all().filter("column_name", "LIKE"
"%,number_value,%"), it may just work.

On Wed, Oct 3, 2012 at 4:05 PM, devwsdi <devw...@gmail.com> wrote:
> Hello,
>
> Is it possible to make a filter with LIKE operator on a QueryCollection?
> I try to translate this query SQL with persistence.
> "SELECT column_name FROM table_name WHERE column_name LIKE
> '%,number_value,%';
>
> Thank you for your help.



--
-- Zef

[Mark M. Young]

devwsd1,

My suggestions would be
1) Make sure you perform all your inserts in one transaction (so it's
not wasting time opening and closing/committing them).
2) If speed is really that big of an issue,
2A) have a flag that indicates whether all the JSON has been persisted
and until it's true, pull the data directly from a parsed instance of
the JSON as an object.
2B) try not putting the indexes on the table(s) until after the JSON
is persisted.

In the end, it should be more of an annoyance to testers than users
because it's only on first load/install.

Regards,
Mark M. Young

On 10/14/12, devwsdi <devw...@gmail.com> wrote:
> Hello,
> Thank you, it works well.
>
> i need your opinion : in my app, i need to insert huge rows in a different
> table at the first launch. (data are in json file) i have a loop on my json
>
> to insert my data.
>
> My problem is that the application is too slow on the first launch. (too
> many javascript insert statement )
>
> for the time I put the most important data in a Table and least important
> in json files associated with the id of the element stored in the database.
>
> when i need to access to a complete data information, i read the json file.
>
> but I still slowdowns in the first launch with instruction inserts.
>
> Is there a solution to build database and table with existing data without
> insert statement? and keep my application accessible to user?

>
> Thank you for your help.
>

> Le mercredi 3 octobre 2012 19:02:02 UTC+2, Zef Hemel a écrit :
>>
>> You could actually try: YourEntity.all().filter("column_name", "LIKE"
>> "%,number_value,%"), it may just work.
>>

>> On Wed, Oct 3, 2012 at 4:05 PM, devwsdi <devw...@gmail.com <javascript:>>

>>
>> wrote:
>> > Hello,
>> >
>> > Is it possible to make a filter with LIKE operator on a QueryCollection?
>> >
>> > I try to translate this query SQL with persistence.
>> > "SELECT column_name FROM table_name WHERE column_name LIKE
>> > '%,number_value,%';
>> >
>> > Thank you for your help.
>>
>>
>>
>> --
>> -- Zef
>>
>

--
"When pride comes, then comes disgrace, but with humility comes wisdom."
—Proverbs 11:2

[Mark M. Young]

devwsdi,

I believe in your case, it would as simple as:

// Use forEach on arrays. When dealing with asynchronous code, it can
save you many headaches.
dataObjects.forEach( function( dataObject )
{
var dataPersistence = new TableTest(
{
test_id: dataObject["test_id"],
test_name: dataObject["test_doc"],
test_desc : dataObject["test_id"]
});
persistence.add( dataPersistence );
});
// 'flush' is called just once.
persistence.flush( callback );

Let me know if that doesn't help,
Mark M. Young

On 11/23/12, devwsdi <devw...@gmail.com> wrote:
> Hello Mark,

>
> Thank you for your help.
>

> How can i make One transaction for all my insert transaction?
> because my insert code like this:
>
> for(var i=0, len= dataObject.length;i < len ; i++){
> dataPersistence = new TableTest({
> test_id: dataObject["test_id"],
> test_name: dataObject["test_doc"],
> test_desc : dataObject["test_id"]
> });
> persistence.add(dataPersistence);
> persistence.flush();
> }
>
> So i have N flush iteration.
> For the moment, my solution is to insert all my data with this method and
> when is finish, i make a dump of my dataBase to have a dump files who was
> load at the first install.
>
> Best regards

>
> Le vendredi 9 novembre 2012 17:52:09 UTC+1, Mark M. Young a écrit :
>>
>> devwsd1,
>>
>> My suggestions would be
>> 1) Make sure you perform all your inserts in one transaction (so it's
>> not wasting time opening and closing/committing them).
>> 2) If speed is really that big of an issue,
>> 2A) have a flag that indicates whether all the JSON has been persisted
>> and until it's true, pull the data directly from a parsed instance of
>> the JSON as an object.
>> 2B) try not putting the indexes on the table(s) until after the JSON
>> is persisted.
>>
>> In the end, it should be more of an annoyance to testers than users
>> because it's only on first load/install.
>>
>> Regards,
>> Mark M. Young
>>