I have a visualization table that has an event listener on select.
The need: I want the user to be able to delete documents on the google drive without having to leave the webpage
The set up: I added a button so that when clicked, I get a confirm alert box that includes the value. Once I click OK, it runs the scripts from the client-side with an event handler. This works perfectly!
The problem: I can move one document at a time but if I need to move 20+ documents it gets really tedious to click rows one after the other. Is it possible to pass multiple values to the successhandler?
google.visualization.events.addListener(archiveChart.getChart(), 'select', function () {
$("#docArchive").on("click", function() {
var selection = archiveChart.getChart().getSelection();
var dt = archiveChart.getDataTable();
if (selection.length > 0) {
var item = selection[0];
var docurl = dt.getValue(item.row, 2);
var docname = dt.getValue(item.row, 1);
var folder = dt.getValue(item.row, 4);
if(confirm("Are you sure you want to archive " + docname + "?") == true) {
archiveChart.getChart().setSelection([]);
return google.script.run.withSuccessHandler(onSuccessArchive).withFailureHandler(function(err) {
alert(err);
}).archiveDoc(docurl,folder);
} else {
archiveChart.getChart().setSelection([]);
}
}});
})
I feel like I might need to add this:
for (var i = 0; i < selection.length; i++) {
var item = selection[i];
I'm struggling a little with understanding what I might need to change (still learning). Any help or guidance is appreciated!
recommend confirming once, for all documents
then loop the selection to archive each document
google.visualization.events.addListener(archiveChart.getChart(), 'select', function () {
$("#docArchive").on("click", function() {
var selection = archiveChart.getChart().getSelection();
var dt = archiveChart.getDataTable();
var docNames = selection.map(function (item) {
return dt.getValue(item.row, 1);
}).join('\n');
if (selection.length > 0) {
if(confirm("Are you sure you want to archive the following document(s)?\n" + docNames) == true) {
for (var i = 0; i < selection.length; i++) {
var item = selection[i];
var docurl = dt.getValue(item.row, 2);
var docname = dt.getValue(item.row, 1);
var folder = dt.getValue(item.row, 4);
return google.script.run.withSuccessHandler(onSuccessArchive).withFailureHandler(function(err) {
alert(err);
}).archiveDoc(docurl, folder);
}
}
archiveChart.getChart().setSelection([]);
}
});
});
I'm developing the spark List with Large Images in flex mobile.
Each ItemRenderer has a large image.
like this.
<s:ItemRenderer>
...
<s:BitmapImage source="{data.bmpData}" />
</s:ItemRenderer>
In dataProvider, there is BitmapData as name "bmpData".
The problem is performance while scrolling.
While scrolling, it stopped for a while when new Image is rendered.
help me please.
If the problem is you render too many bitmapdata in same time, you can render them one by one in different frames.
Here is an example.
Make a custom ItemRenderer
class YourItemRenderer
{
override public function set data(value:Object):void
{
if (super.data != value)
{
super.data = value;
yourBitmapImage.source = null;
//when the data change, don't call the render function directly
EnterFrameManager.getInstance().addRenderFunction(render)
}
}
private function render():void
{
if (yourBitmapImage != null && data != null)
{
yourBitmapImage.source = data.bmpData;
}
}
}
EnterFrameManager is used to control the render functions.
class EnterFrameManager
{
import mx.core.FlexGlobals;
public function EnterFrameManager()
{
FlexGlobals.topLevelApplication.addEventListener( Event.EnterFrame, onEnterFrameHandler)
}
private var _instance:EnterFrameManager;
public static function getInstance():EnterFrameManager
{
if (_instance == null)
{
_instance = new EnterFrameManager();
}
return instance;
}
//save the render functions
private var renderQueue:Array = [];
private var nowIntervalFrame:int = 0;
//change it to small value when you don't feel lag
private const UPDATE_INTERVAL_FRAMES:int = 6;
private function onEnterFrameHandler(e:Event):void
{
nowIntervalFrame++;
if (nowIntervalFrame >= UPDATE_INTERVAL_FRAMES)
{
nowIntervalFrame = 0;
//change renderQueue by waitQueue
for each (var f:Function in waitQueue)
{
addFunctionToQueue(f, renderQueue);
}
waitQueue.length = 0;
if (renderQueue.length > 0)
{
var f:Function = renderQueue.shift();
f();
}
}
}
private var waitQueue:Array = [];
public function addRenderFunction(f:Function):void
{
addFunctionToQueue(f, waitQueue);
}
private function addFunctionToQueue(f:Function, queue:Function):void
{
var index:int = queue.indexOf(f);
if (index == -1)
{
queue.push(f);
}
else
{
var temp:Function = queue.splice(index, 1);
queue.push(temp);
}
}
}
I'm working with Chart.js and I'm wondering if there's a way when you click on part of a pie chart, it filters the bar chart.
Since this is a Chart.js question :-), this is how you do it Chart.js (and it's not too complex either)
Setting up the Pie Chart
// pie
var data = [
{
value: 300,
color: "#F7464A",
highlight: "#FF5A5E",
label: "Red",
subData: [28, 48, 40, 19, 86, 27, 190]
}, {
value: 50,
color: "#46BFBD",
highlight: "#5AD3D1",
label: "Green",
subData: [90, 28, 48, 40, 19, 86, 127]
}, {
value: 100,
color: "#FDB45C",
highlight: "#FFC870",
label: "Yellow",
subData: [28, 48, 40, 19, 86, 27, 190]
}
]
var canvas = document.getElementById("chart");
var ctx = canvas.getContext("2d");
var myPieChart = new Chart(ctx).Pie(data);
Setting up the Bar Chart using Pie Data
// bar using pie's sub data
var bardata = {
labels: ["Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun"],
datasets: [
{
label: "My Second dataset",
fillColor: "rgba(151,187,205,0.5)",
strokeColor: "rgba(151,187,205,0.8)",
highlightFill: "rgba(151,187,205,0.75)",
highlightStroke: "rgba(151,187,205,1)",
data: data[0].subData.map(function (point, i) {
var pointTotal = 0;
data.forEach(function (point) {
pointTotal += point.subData[i]
})
return pointTotal;
})
}
]
};
var subcanvas = document.getElementById("subchart")
var subctx = subcanvas.getContext("2d");
var myBarChart = new Chart(subctx).Bar(bardata);
Updating Bar data when Clicking Pie
// connect them both
canvas.onclick = function (evt) {
var activeSector = myPieChart.getSegmentsAtEvent(evt);
myBarChart.datasets[0].bars.forEach(function (bar, i) {
var pointTotal = 0;
data.forEach(function (point, j) {
if (activeSector.length === 0 || point.label === activeSector[0].label)
pointTotal += data[j].subData[i]
})
bar.value = pointTotal;
});
myBarChart.update();
};
Clicking outside the pie (but in the pie chart's canvas) resets the bar chart.
Fiddle - http://jsfiddle.net/0zwkjv8a/
Other answers posted already cover what I would generally advise here which is to use dc-js if you want crossfilter enabled charts out of the gate. I would have commented on this answer, but I don't have enough reputation so I'm posting this as option 'c.)' where 'a.)' is using dc-js and 'b.)' is making some modifications to an existing Chart.js chart.
Option 'c.)' is to extend the Chart.js chart type and make the child chart work like a dc-js chart. Chart.js chart types follow an inheritance hierarchy, so if you like a chart that already exists you can wrap its prototype methods with some of your own. Additionally important to this option, in the selected answer to the stack overflow question with heading 'dc.js - Listening for chart group render', it is described how the current implementation of dc-js's chartRegistry object is fairly decoupled from d3 or dc internals, so any chart implementing chartRegistry's interface can be part of a chartGroup.
I was in the position of wanting very much to use Polar Area Charts in a dataset where I was already using a chart group full of dc-js charts to crossfilter the data. I wrote an extension for Polar Area charts that could serve as an example of one way (I'm going to go ahead and say probably not the best way) to extend a chart type with dc-js like behaviors. The repo for this is at https://github.com/nsubordin81/Chart.dc.js, Licensed under an MIT License, and in case that ever goes anywhere, all of the code is copied into the example fiddle: http://jsfiddle.net/nsubordin81/3w725o3c/1/
Chart.dc.js v. 0.1.0
MIT Licensed: opensource.org/licenses/MIT
Copyright (c) 2015 Taylor Bird
(function () {
"use strict";
var root = this,
Chart = root.Chart,
dc = root.dc,
helpers = Chart.helpers,
//class for data structure that manages filters as they relate to chart segments. This should probably be generalized to chart elements of all kinds.
FilterManager = function (segmentList) {
//private member variable
var filterMap = [];
//constructor
//accepts a list of SegmentArcs that have had the extra properties added to them
for (var i = 0; i < segmentList.length; i++) {
add(segmentList[i].segmentID);
}
//private methods
function testOnAll(test) {
var testResult = true;
for (var i = 0; i < filterMap.length; i++) {
//one failure of test means testOnAll fails
if (!test(filterMap[i])) {
testResult = false;
}
}
return testResult;
}
//add a filter, pretty much just a wrapper for push
function add(segmentID) {
filterMap.push({
"segmentID": segmentID,
"active": false
});
}
//remove a filter by id, returns removed filter
function remove(segmentID) {
var removed = filterMap.find(segmentID);
filterMap = filterMap.filter(function (elem) {
return elem.segmentID !== segmentID;
});
return removed;
}
//return this segment if it is filtered
function find(segmentID) {
for (var i = 0; i < filterMap.length; i++) {
if (filterMap[i].segmentID === segmentID) {
return filterMap[i];
}
}
return -1;
}
//public methods
return {
//tell me if the filter for this segment is active
isActive: function (segmentID) {
var filter = find(segmentID);
if (filter === -1) {
console.error("something went wrong, the filter for this segment does not exist");
}
return filter.active;
},
//for the given segment, activate or deactivate its filter. return whether the filter is on or off.
flip: function (segmentID) {
var filter = find(segmentID);
if (filter === -1) {
console.error("something went wrong, the filter for this segment does not exist");
}
filter.active ? filter.active = false : filter.active = true;
return filter.active;
},
//if all filters are on, we want to be able to quickly deactivate them all
turnAllOff: function () {
for (var i = 0; i < filterMap.length; i++) {
filterMap[i].active = false;
}
},
//tell me if all of the filters are off
allOff: function () {
return testOnAll(function (elem) {
return !elem.active;
});
},
//tell me if all the filters are on
allOn: function () {
return testOnAll(function (elem) {
return elem.active;
});
}
}
};
//utility function, Takes an array that has some property as its key
//and forms a javascript object with the keys as properties so we can get O(1) access
function createKeyMap(arr, propName) {
var keyMap = {}
for (var i = 0; i < arr.length; i++) {
keyMap[arr[i][propName]] = arr[i];
}
return keyMap;
}
Chart.types.PolarArea.extend({
name: "PolarAreaXF",
//this will have to be a member
dimension: undefined,
colorTypes: {
"NORMAL": 0,
"HIGHLIGHT": 1,
"FILTER": 2,
"FILTER_HIGHLIGHT": 3
},
chartGroup: undefined,
filters: undefined,
originalDataKeys: undefined,
initialize: function (data) {
//--PRE--
var that = this;
//Polar Area initialize method is expecting (data, options) in arguments,
//but we pass in an array of components to merge. Let's clean this up.
var argsArray = Array.prototype.slice.call(arguments);
//remove the first element of arguments which is our array, then we do a bunch of Chartjs converison on it . . .
argsArray.splice(0, 1);
//TODO - check if data is an array, if not, put a message in a console explaining how you are supposed to send data in an array
this.dimension = data.dimension;
data.chartGroup ? this.chartGroup = data.chartGroup : this.chartGroup = 0;
//short but magical line. Now we are linked with all dc charts in this group!
dc.registerChart(this, this.chartGroup);
var data = this.setupChartData(data.colors, data.highlights, data.labels);
//... and push the result in its place.
argsArray.unshift(data);
//originalDataArray -- this is used as a reference to the original state of the chart, since segments can come and go,
//we use this to track what a segment's original colors were when adding it back in. This would mess up adding a truly new segment, but who
//is gonna do that? Assumption here is dimensions start with so many groups and that is it.
this.originalDataKeys = createKeyMap(data, "key");
//parent's initialize
Chart.types.PolarArea.prototype.initialize.apply(this, argsArray);
//--modify SegmentArcs--
//assign colors and ids to all existing segment arcs
var mySegments = this.segments;
for (var i = 0; i < mySegments.length; i++) {
mySegments[i].colorList = [undefined, undefined, "#777", "#aaa"];
mySegments[i].colorList[this.colorTypes.NORMAL] = mySegments[i].fillColor;
mySegments[i].colorList[this.colorTypes.HIGHLIGHT] = mySegments[i].highlight;
mySegments[i].segmentID = i;
mySegments[i].key = data[i].key;
}
//add methods to SegmentArc objects that will color them one way or the other depending on their filter
this.SegmentArc.prototype.setIncluded = function (include) {
if (include) {
this.fillColor = this.colorList[that.colorTypes.NORMAL];
this.highlight = this.colorList[that.colorTypes.HIGHLIGHT];
} else {
this.fillColor = this.colorList[that.colorTypes.FILTER];
this.highlight = this.colorList[that.colorTypes.FILTER_HIGHLIGHT];
}
}
//--initialize filters--
this.filters = new FilterManager(this.segments);
//handle clicks on segments as filter events, do the styling and crossfilter changes at the Chart level in the filter method.
helpers.bindEvents(this, ["mousedown"], function (evt) {
var activeSegment = Chart.types.PolarArea.prototype.getSegmentsAtEvent.apply(this, [evt])[0];
this.handleFilter(activeSegment);
});
},
//convert crossfilter dimension into chart.js Polar Area data object array
setupChartData: function (colors, highlights, labels) {
var chartJSible = [];
//probably need checks here to make sure client actually passed in a crossfilter dimension
var grouped = this.dimension.group().reduceCount().top(Infinity);
//probably need checks here to either fail if the arrays aren't all long enough or have some way to add random colors/highlights if they are shorter.
for (var i = 0; i < grouped.length; i++) {
var dataObject = {
value: grouped[i].value,
key: grouped[i].key,
color: colors[i],
highlight: highlights[i],
label: labels ? (labels[i] ? labels[i] : grouped[i].key) : grouped[i].key
};
chartJSible.push(dataObject);
}
return chartJSible;
},
//figure out what changed between Chart.js' internally maintained data object array and crossfilter's dimension data. use the saved information
//about what colors and highlight a key has to rebuild the segmentArc list 'segments'. can't trash the old, it might mess up the animations.
redraw: function () {
var grouped = this.dimension.group().reduceCount().top(Infinity);
var currentSegmentKeys = createKeyMap(this.segments, "key");
var crossfilterGroupKeys = createKeyMap(grouped, "key");
//loop through the segment list, if the segment for a group is already there, update the value, if it is not there, add it back using the
//original data as a guide for what it's color and highlight color should be. if there are segments in the existing list
var length = Math.max(this.segments.length, grouped.length);
//going through both lists, whichever is longer
for (var i = 0; i < length; i++) {
var sList = this.segments;
var gList = grouped;
//only do this part if we still have items in the new filtered list
if (gList[i]) {
//we already have a segment for this crossfilter group, just get that segment and update its value
if (currentSegmentKeys[gList[i].key]) {
currentSegmentKeys[gList[i].key].value = gList[i].value;
} else {
//the chart doesn't have the crossfilter group item, add a new segment with the right colors and values from original data
var theSegment = this.originalDataKeys[gList[i].key];
this.addData(theSegment, 0, true);
}
}
//only do this part if we still have items in the current chart segment list
if (sList[i]) {
//we don't have this segment in the new crossfilter group, remove it from the chart
if (!crossfilterGroupKeys[sList[i].key]) {
this.removeData(i);
}
}
}
this.update();
},
filterAll: function () {
this.dimension.filterAll();
this.filters.turnAllOff();
this.colorMeIn();
this.redraw();
},
handleFilter: function (clicked) {
//after we have all of the filters figured out, change the colors to reflect what they should be and update the chart
this.filters.flip(clicked.segmentID);
this.colorMeIn();
if (this.filters.allOn()) {
this.dimension = this.dimension.filterAll();
dc.redrawAll(this.chartGroup);
this.filters.turnAllOff();
}
dc.redrawAll(this.chartGroup);
},
colorMeIn() {
var activeFilters = [];
var segments = this.segments;
for (var i = 0; i < segments.length; i++) {
var segment = segments[i];
if (this.filters.isActive(segment.segmentID) || this.filters.allOff()) {
segment.setIncluded(true);
activeFilters.push(segment.key);
} else {
segment.setIncluded(false);
}
}
this.dimension = this.dimension.filterFunction(function (d) {
for (var i = 0; i < activeFilters.length; i++) {
if (d === activeFilters[i]) {
return true;
}
}
return false;
});
}
})
}).call(this);
Use dc.js: https://dc-js.github.io/dc.js/
It has exactly the functionality asked for.
Within famo.us, I want to place a variable-length wrapping title near the top then have images etc follow. All items must be their own surface as some have click events and animations but all the positioning must be super fast to calculate and place based on the text height and of course must avoid DOM access for the 60fps. This must happen for a series of mini-posts, streaming real time and for infinite scroll.
So far, I came up with an approach that works using an ascii character map to pixel width load off screen on init if isn't already in localStorage. It uses jquery to do the sizing for each character, then determine line height by checking a breaking word and the height of that box. After that I use this map for calculations on the fly so I don't touch DOM again. It seems to work fine but is completely dependent on knowing the style of font going to be used in a very specific way. For each new style, I would have to have a different mapping which sucks.
Is there another approach that is more built in to famo.us?
I'm answering my own question because its frankly been awhile, the new framework is going to come out June 22nd of this month (at the time of the this writing). Just going to put how I'm doing this currently which is the way I described above.
Here's the 0.3.5 way of Famou.us to do this.
This is in a file "utils/lines.js" I whipped up. Hack, yes. Works, yes.
define(function(require) {
'use strict';
var $ = require('jquery');
var cache = require('utils/cache');
function getLineInfo(id) {
var cacheKey = 'lineInfo_' + id;
var savedLineInfo = cache.get(cacheKey);
if (savedLineInfo)
return savedLineInfo;
var charStart = 32; // space
var charEnd = 126; // ~
var $charDump = $('<div>', { id: id });
$('body').append($charDump);
for (var i = charStart; i <= charEnd; i++)
$charDump.append($('<div>', { class: 'char', css: { top: (i-32) * 20 } }).html('&#' + i));
var charMap = {};
$charDump.find('.char').each(function() {
var $this = $(this);
var char = $this.text().charCodeAt(0);
var w = $this.width();
charMap[char] = w;
});
$charDump.html('<div class="line">abc<br>123<br>456</div>');
var lineHeight = $charDump.height() / 3;
$charDump.remove();
savedLineInfo = {
lineHeight: lineHeight,
charMap: charMap
};
cache.set(cacheKey, savedLineInfo);
return savedLineInfo;
}
function getLines(text, width, id, maxLines) {
var lineInfo = getLineInfo(id);
var cleaned = $.trim(text.replace(/\s+/gm, ' '));
var words = cleaned.split(' ');
var lines = 1;
var lineWidth = 0;
var spaceLength = lineInfo.charMap[32];
var allLines = '';
words.forEach(function(word) {
var wordLength = 0;
word.split('').forEach(function(char) {
var charLength = lineInfo.charMap[char.charCodeAt(0)];
if (!charLength)
charLength = spaceLength;
wordLength += charLength;
});
if (lineWidth + wordLength + spaceLength <= width) {
lineWidth += wordLength + spaceLength;
if (maxLines)
allLines += word + ' ';
}
else {
lineWidth = wordLength;
if (maxLines && lines <= maxLines)
allLines += word + ' ';
else if (maxLines)
return {text: allLines.substr(0, allLines.length - 3) + '...', lines: lines};
lines++;
}
});
if (maxLines)
return {text: allLines, lines: lines};
return lines;
}
function getTextHeight(text, width, id, maxLines) {
var lineInfo = getLineInfo(id);
var info;
if (maxLines) {
info = getLines(text, width, id, maxLines);
info.height = lineInfo.lineHeight * info.lines;
return info;
}
return lineInfo.lineHeight * getLines(text, width, id);
}
return {
getLines: getLines,
getLineInfo: getLineInfo,
getTextHeight: getTextHeight
};
});
I've made 4 rectangles in raphael.js using the for loop. When I apply events such as onmouseover or onmouseout it applies only to the last rectangle created. I know something is wrong in my code. Please provide a solution and is there a way to simplify the code?
JS Fiddle Link
window.onload = function(){
var paper = Raphael(0,0,640,540);
for (i=0;i<2;i++){
for (j=0;j<2;j++){
var boxes = paper.rect(0+(j*320),0+(i*270),320,270).attr({fill:'#303030',stroke:'white'});
boxes.node.onmouseover = function () {
boxes.attr("fill", "blue");
};
boxes.node.onmouseout = function () {
boxes.attr("fill", "#303030");
};
}
}
}
This is an extremely common mistake in javascript. You reuse the boxes variable, so when any of the handlers are executed, it points to the last value it had.
The common way of overcoming this is to wrap the code inside the loop in a function call:
window.onload = function() {
var paper = Raphael(0, 0, 640, 540);
for (i = 0; i < 2; i++) {
for (j = 0; j < 2; j++) {
(function(i, j) {
var boxes = paper.rect(0 + (j * 320), 0 + (i * 270), 320, 270).attr({
fill: '#303030',
stroke: 'white'
});
boxes.node.onmouseover = function() {
boxes.attr("fill", "blue");
};
boxes.node.onmouseout = function() {
boxes.attr("fill", "#303030");
};
})(i, j);
}
}
}