It's very important to have ability to click on cell and get array of aggregated values...
Without this it's very difficult to integrate pivottable in the any serious project...
Usualy OLAP-table is something like a filter to details table...

P.S. Good work. Best javascript pivot table! )

Sorry for my very bad english((

Currently you are caculating the pivot data table and then constructing the DOM elements.
This a violation of separation of concerns.

If you separated the functionality, it could be easily used with AngularJS applications (for example)

1. updatePivotData
2. support iterators for table header and rows (data only)
3. provide call backs for table header

Could you please confirm the sum, average and sumoversum function are working in the example? Because when i verify in examples demo with sum function there is no results.

Should add:

• array of arrays (first is header)
• function that calls back

Heatmap stops working when doctype is:

Can be reproduced by adding that doctype declaration to mps_prepop.html.

Hi Nikolas
I wonder if there is a way i can do some predefined variable combinations , i mean , i don't want user can perform centain combinations that are not logical , so i would like to create these possible combinations once i had data on client.

Thanks

The input data is a set of records with attributes (not fields or variables, because derivedAttributes is part of the public API already).

The pivot table is a function of colAttrs and rowAttrs (not colVars and rowVars) and has rows and columns whose headers are derived from rowKeys and colKeys.

Should be an internal refactoring, with some passes on the documentation to make things coherent.

Hi
i would like to know how to get a column total to use it in an agregator function, is this possible ?

Thanx

I am looking to use the built in aggregate functions via the pivot() function and am not sure how to pass them in. Is it possible to add an example of how to call them?
In particular I am looking for intSum aggregate example.

Wonderfully useful tool!

Good morning Nicolas, i'm using this tool and it is so interesting. But miss a resource of expand or collapse a cell like occurs in microsoft excel. Do you intend to develop this? Or is this already exists?
Sorry for my english.
Rafael

Hi Nicolas,

I started my project using an HTML table as input to the pivot table and decided that was somewhat unwieldy so I just converted it to a simple array.

The data seems to work, but now my row/column heatmaps as well as the table bar chart renderers don't seem to work on the rows or columns, only on the summaries.

Any ideas?

http://teamgamer.com/bf3stats_pivot.php

I'm interested in using this with a HTTP/JSON API which gives me server-side aggregation. How easy is it to swap out the aggregation logic and keep only the UI?

It is not clear how aggregate functions work. Currently, in my personal use cases, only "count" works. However, the rest of the functions (sum, avg, etc...), when selected, the cells display empty values.

It would be useful to add examples for this in the documentation.

I would like to override the col and row names for displaying.

For Example:
{color: "blue", shape: "circle"},
{color: "red", shape: "triangle"}

The above json column value is color and row value is shape. But i want to display instead of color as "Background Color" and shape as "Graph Shape".

Currently i am returning the json from java object.

Can you help me on this.

Try to set Name or Age Bin as a row and Age as a column

First, very nice work!

However I found an issue, I can't expand the <select> in Opera and Firefox, works only in Chrome:
http://nicolaskruchten.github.io/pivottable/mps_prepop.html
(using Windows 7)

BTW It would be good to write the browser compat in the README.

numbers are sorted like strings even if i declare it as numbers in my json

data to reproduce this screenshot table:
{'CUSTOMER':'A','QUANTITY':4,'ITEM':'qq','MONTH':2},
{'CUSTOMER':'B','QUANTITY':2,'ITEM':'ww','MONTH':3},
{'CUSTOMER':'C','QUANTITY':6,'ITEM':'ww','MONTH':3},
{'CUSTOMER':'A','QUANTITY':2,'ITEM':'tt','MONTH':3},
{'CUSTOMER':'A','QUANTITY':1,'ITEM':'ww','MONTH':2},
{'CUSTOMER':'C','QUANTITY':2,'ITEM':'qq','MONTH':1},
{'CUSTOMER':'B','QUANTITY':5,'ITEM':'ww','MONTH':2},
{'CUSTOMER':'C','QUANTITY':4,'ITEM':'qq','MONTH':1},
{'CUSTOMER':'A','QUANTITY':5,'ITEM':'tt','MONTH':11},
{'CUSTOMER':'B','QUANTITY':3,'ITEM':'qq','MONTH':14},
{'CUSTOMER':'B','QUANTITY':6,'ITEM':'ww','MONTH':3}

map definition:
{
rows: ['CUSTOMER'],
cols: ['MONTH'],
vals: ['QUANTITY'],
aggregatorName: 'intSum',
effectsName: 'Heatmap'
}

version: Version 28.0.1500.72 m

Console error:

Failed to load resource: Origin null is not allowed by Access-Control-Allow-Origin. file:///C:/Users/Impel123/Desktop/issues%20fixes/pivot/pivottable-master/pivottable-master/examples/mps.json

XMLHttpRequest cannot load file:///C:/Users/Impel123/Desktop/issues%20fixes/pivot/pivottable-master/pivottable-master/examples/mps.json. Origin null is not allowed by Access-Control-Allow-Origin.

Hi, I'm having trouble implementing a rather large data source through the pivot table. The source is currently about 300,000 rows with 12 columns.

As I begin to increase the number of rows I'm selecting, it gets slower and slower, eventually not loading at all. I mean, the data is returned from the server, but the page is just blank. Here are some numbers:
1,000 1.12MB 581ms
10,000 2.4MB 939ms
15,000 3.6MB 1.46s
50,000 12.0MB 4.76s

This is # of records, size of response and response time in Chrome x64.

Anything after 15,000 does not render a chart or table at all.

I see someone posted in #29 a similar issue and he said that he resolved it by breaking his ajax requests apart. But I'm not really sure how to do this. I can't find anything in the documentation that lets data be loaded in piecemeal, after the initial instantiation of the table. Is this possible?

it would be useful to be able to add the operator '% of column' or '% of row' to the results table.

can i post general comments and requests here? where?

I've noticed that the cols option parameter does not seem to work for pivotUI (rows is fine), whereas it works fine for pivot.

Here is a snippet of code which illustrates what I'm trying to do:

rows = ['Amount Bin100'];
cols = ["Scheme", "Code"];

var derivers = $.pivotUtilities.derivers;
$("#pivottable").pivotUI(filteredLocations, {
    derivedAttributes: {
        "Amount Bin100": derivers.bin("Amount", 100),
        "Amount Bin1000": derivers.bin("Amount", 1000)
    },
    hiddenAttributes: hiddenAttributes,
    cols:cols, rows: rows
});

});

Any idea why cols is not working for pivotUI, but works if I replace pivotUI with pivot ?

Some sort of function that holds a mapping of disaggregated to aggregated?

Is there a way to create subtotals on rows and columns when exists more than one field in each axis?

Rafael

Is there a simple way of sorting by totals? I know right now it sorts by the attribute name... but is there a way to sort by the attribute value? Thanks!

This happens when handling a 90mb data object on the pivot table. For smalled datasets, it works fine.

Unfortunately, the application that I use this for return a high number of json objects. Is there a workaround/fix for this behaviour?

Draggable elements should have probably cursor:move in CSS instead of the default, text selection cursor (and perhaps some color/bgcolor change when hovered/active, or sth like that), to make it more discoverable that you can move it around.

I think move is the most appropriate, I initially wrote here pointer but this is rather good for links and buttons.
https://developer.mozilla.org/en-US/docs/Web/CSS/cursor

A nice enhancement would be make the table html markup compatible with http://www.fixedheadertable.com/
i will try to and possibly make a fork

Hi, I was wondering if it is possible to write an aggregator that shows sums as percentages of the row or column total?

Hi
Is there a way to calculate percetages over rows colums or grand total ??
Can I use the symbol % as format ?

Thanx

--- initially ---
$("#cube").pivotUI(mps, {
rows: ["Year", "Name"],
cols: ["Category"],
vals: ["Number"],
effectsName: "Heatmap",
aggregatorName: "intSum"
});

--- since i need to reload the data due to some reasons ---
$("#cube").pivotUI(mps);

The cube forgets all the current options (user might have changed the rows, columns and vals).

To facilitate some sort of 'save state' functionality.

When creating a pivot table on a page that already contains one, the new one inherits some of the old one's parameters. I've confirmed this with cols and rows. For example, if I add the attribute "Sex" as a column, then the new pivot table will be created with Sex as a column (see attached image). Even worse, the user can't drag the column to remove it.

It looks like the issue is with the use of #id selectors like $("#rows li nobr"). If there's more than one pivot table on the page, then data from an existing one will be selected.

Hi,
actually when i drop the two fields in aggregate area,then it performs the aggregate function only on one field.but i need it to perform that functionality on all the fields that is present in aggregation area.
is there any way to do that.
thanks in advance.

Is there an option to "hide" the aggregator/renderer drop-down through the init code?

Something like:

pivotOpts : {
rows : [ 'Name' ],
vals : [ 'Age' ],
rendererName : 'Heatmap',
aggregatorName : 'intSum',
hideAggregator : true
}

Currently after it gets rendered I'm doing something like:

  if(hideAggregator){
        $("#aggregator").hide();
    }

Pretty standard requirement

I have a project where we would like to calculate % of row and column, sorting, and other features not currently in your great pivottable tool. Can we work on this and submit code to you for possible inclusion? if so, i guess i can put you in touch with the programmer to discuss standards and such?

Once the pivot() function has been split into render(model()), create a PivotChart renderer

On failed render, show error in UI?

Hello,

First of all, thanks for such a wonderful work!. I was playing around with the derived attributes and I found the following error when you are trying to use it.

Uncaught TypeError: Object [object global] has no method 'splice'

My solution was to omit the removeAt key in the deriveAttributes function and it worked.

deriveAttributes = function(record, derivedAttributes, f) {
var k, v, _ref, _ref2;
console.log(derivedAttributes);
for (k in derivedAttributes) {
if(k!="removeAt"){ // I Included this
v = derivedAttributes[k];
record[k] = (_ref = v(record)) != null ? _ref : record[k];
}
}
for (k in record) {
if (!__hasProp.call(record, k)) continue;
if ((_ref2 = record[k]) == null) {
record[k] = "null";
}
}
return f(record);
};

Hope its the right approach,

Thanks,

Andrés.

Using this: https://gist.github.com/dergachev/4627207

• time&space data (space includes stuff like campaigns, budget categories)
• binary trial data (click through rate, dropped call rate)
• ratio data (average order value, normalized data)

it would be useful to be able to sort the output table columns (A-Z, Z-A) such that the end result is something like 'top 10 provinces who voted NDP' or similar result.

Hi, I am struggling to get a pivot table working using an HTML table as input.

I have it laid out properly, as far as I can tell with thead->th elements as my attributes and tbody->tr->td elements as my data.

My table id is "players" and I instantiate the pivottable as such:

<script>
    var input = $("#players");
     $(function(){
                    $("#output").pivotUI(input,
                                                { 
        rows: ["Weapon"], 
        cols: ["PlayerName"] 
    });
     });
</script>

I am not a very experience jQuery guy, so I might be doing something very silly.

Any chance I could get a more complete example than is currently in the input formats docs?

cheers

$("#output").pivot(
    [
        {a: "1-0", b: "0"},
        {a: "1", b: "0-0"}
    ],
    {
        rows: ["a", "b"],
        cols: []
    }
);

Gives the following table

which is incorrect, expected two separate rows.

I propose you to use String.fromCharCode(0) or String.fromCharCode(0) + "-"

I'm wondering if it's possible to output the records that have been binned into a given pivot table entry.

The use case is the following:

• I have set up a pivot table, for instance with heatmap colouring.
• I now want to click on a given pivot table entry of interest in order to filter on/ inspect / output those records, creating a filtered view of the initial input records.

i.e. I want to answer the question "Which records contribute to this pivot table entry?", and make use of them elsewhere in coordinated views.

Another approach would be to output the necessary row and column data that describes the pivot table entry in order to perform filtering externally (perhaps with crossfilter.js).

I have not looked into the source code to see how to approach this. It would be great to hear from you to know the difficultly level of this enhancement.