I am not sure what you mean by "the data remains the same".
If you filter only the rows with salary < 1000, then the data is not the
same.... You remain with only part of the rows of the original table.
This is just what would have happened if you used a query on top of that
table and used "select * from employees where salary < 1000".

The only difference I can see is the syntax of the request, that is not
langauce based, but more of a function calls, which I find more simple and
easy to use (and surely less error prone).


VizGuy



On Mon, Aug 3, 2009 at 7:32 PM, toti <[email protected]> wrote:

>
> Hi,
>
> Thanks for your answer.
>
> I understand what you mean.
>
> Unluckily, I don't have the resources (mainly, technical skills) to
> build an js API on top of the DataTable/DataView object that
> aggregates the data in a structured way (aka through an sql-like
> language).
>
> At first, I thought the "DataView" object could do this, as it seems
> natural that a View represents the same data but shown in a different
> way. "natural" as in SQL :)
>
> Well, it does allow to sort and filter rows and columns, but the data
> remains always the same.
>
> Cheers
>
> On Aug 3, 2:51 am, VizGuy <[email protected]> wrote:
> > Hi,
> >
> > While the query language cannot be used to query a data table that
> already
> > resides in teh browser, and is simply a javascript object, there are
> other
> > things you can do in this case.
> > Since you are already in js environment, you can use the various js
> methods
> > to create a new data table, or a data view, to sort and filter rows, and
> > soon to group and join data tables.
> > This way, from your js code, instead of querying a data table, you can
> > manipulate it, or create another table or view based on this one, that
> will
> > have the data that you want.
> > It will take less time indeed than to send a remote query, and since no
> > query language parsing and execution is involved, it will (probably) take
> > less time.
> >
> > VizGuy
> >
> > On Mon, Aug 3, 2009 at 12:35 AM, toti <[email protected]> wrote:
> >
> > > Hi,
> >
> > > I joined this group just to ask if it is posible to query a datatable.
> > > It seems that the answer is no.
> >
> > > I'd like to know if, in future releases, this will be posible or maybe
> > > the whole idea of querying directly a datatable is wrong from an
> > > architectural point of view.
> >
> > > I don't like having to requery a servlet (which works as datasource)
> > > an transfer +50kb each time. Sometimes is +300kb.
> >
> > > In my case, I need to aggregate the same information I'm using in a
> > > PieChart to build another Visualization.
> >
> > > Cheers,
> > > Alejandro
> >
> > > On Jul 17, 5:55 pm, chrissky <[email protected]> wrote:
> > > > Hi Scarlett,
> >
> > > > I can give you a bit of guidance on your idea - since I just did
> > > > something very similar for a blog article.
> > > > "Feeding the Google Annotated Timeline gadget from a GAQL query"
> > > > You can read it herehttp://
> > > blog.shufflepoint.com/2009/07/feeding-the-google-annotated-tim...
> >
> > > > Basically, I requery the datasource when the user changes the date
> > > > range across some thresholds.
> >
> > > > If you upload your CSV to a google spreadsheet, then you could use
> the
> > > > approach I presented in my article.
> >
> > > > I haven't seen info on the RC 1.1  - do you have a link?
> >
> > > > Chris Harrington,
> > > > ShufflePoint, Inc.
> >
> > > > On Jul 16, 5:28 pm, Scarlett <[email protected]> wrote:
> >
> > > > > I'm still trying to find a solution for handling large sets of data
> > > > > with the annotated timeline.
> >
> > > > > Issue:
> > > > > Plotting a large number of data point takes a long time ... and it
> > > > > doesn't look good if there is variability.
> >
> > > > > Idea:
> > > > > Load data at the appropriate resolution (~between 50 and 200 data
> > > > > points?) based on getVisibleChartRange(). Something similar to how
> the
> > > > > time axis labels are adjusted at different resolutions. When chart
> > > > > range is <30 days plot in hours, 30-365 days plot in days, 1-5
> years
> > > > > plot in months ...
> >
> > > > > Wrinkle:
> > > > > Even with this method the loading of the non-visible data slows the
> > > > > whole chart down. My idea is to use the Query language (~SQL) to
> > > > > "filter" the non visible data to a lower resolution and keep the
> > > > > visible range at the higher resolution. I can do this in Oracle SQL
> by
> > > > > SELECT column_name GROUP BY to_char(date_column_name, 'YYYY-MM')
> ...
> > > > > where this creates monthly data set. I hope this can be done in the
> > > > > Query language so there are three DataTables that are joined back
> > > > > together: (1) one for the low resolution prior to the
> > > > > VisibleChartRange, (2) one higher resolution during the
> > > > > VisibleChartRange, and (3) one lower resolution after the
> > > > > VisibleChartRange.
> >
> > > > > Problem:
> > > > > 1. I'm a hack when it comes to programming.
> > > > > 2. I can't see any examples of how to use the Query language on a
> > > > > DataTable (which I have loaded from a csv file.) not a Google
> > > > > spreadsheet. Anyone have an example? Or can I use the Query
> language
> > > > > against the csv file right away?
> > > > > 3. I think with the release candidate 1.1 there are more methods
> > > > > available for the Annotated timeline, but I can find any
> documentation
> > > > > on what they might be. Where might I get more info on the
> development
> > > > > version?
> >
> > > > > Cheers,
> > > > > Scarlett
> >
>

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"Google Visualization API" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/google-visualization-api?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to