This is my grid of nodes:
I'm moving an object around on it using the A* pathfinding algorithm. It generally works OK, but it sometimes acts wrongly:
When moving from 3 to 1, it correctly goes via 2. When going from 1 to 3 however, it goes via 4.
When moving between 3 and 5, it goes via 4 in either direction instead of the shorter way via 6
What can be wrong? Here's my code (AS3):
public static function getPath(from:Point, to:Point, grid:NodeGrid):PointLine {
// get target node
var target:NodeGridNode = grid.getClosestNodeObj(to.x, to.y);
var backtrace:Map = new Map();
var openList:LinkedSet = new LinkedSet();
var closedList:LinkedSet = new LinkedSet();
// begin with first node
openList.add(grid.getClosestNodeObj(from.x, from.y));
// start A*
var curNode:NodeGridNode;
while (openList.size != 0) {
// pick a new current node
if (openList.size == 1) {
curNode = NodeGridNode(openList.first);
}
else {
// find cheapest node in open list
var minScore:Number = Number.MAX_VALUE;
var minNext:NodeGridNode;
openList.iterate(function(next:NodeGridNode, i:int):int {
var score:Number = curNode.distanceTo(next) + next.distanceTo(target);
if (score < minScore) {
minScore = score;
minNext = next;
return LinkedSet.BREAK;
}
return 0;
});
curNode = minNext;
}
// have not reached
if (curNode == target) break;
else {
// move to closed
openList.remove(curNode);
closedList.add(curNode);
// put connected nodes on open list
for each (var adjNode:NodeGridNode in curNode.connects) {
if (!openList.contains(adjNode) && !closedList.contains(adjNode)) {
openList.add(adjNode);
backtrace.put(adjNode, curNode);
}
}
}
}
// make path
var pathPoints:Vector.<Point> = new Vector.<Point>();
pathPoints.push(to);
while(curNode != null) {
pathPoints.unshift(curNode.location);
curNode = backtrace.read(curNode);
}
pathPoints.unshift(from);
return new PointLine(pathPoints);
}
NodeGridNode::distanceTo()
public function distanceTo(o:NodeGridNode):Number {
var dx:Number = location.x - o.location.x;
var dy:Number = location.y - o.location.y;
return Math.sqrt(dx*dx + dy*dy);
}
The problem I see here is the line
if (!openList.contains(adjNode) && !closedList.contains(adjNode))
It may be the case that an adjNode may be easier(shorter) to reach through the current node although it was reached from another node previously which means it is in the openList.
Found the bug:
openList.iterate(function(next:NodeGridNode, i:int):int {
var score:Number = curNode.distanceTo(next) + next.distanceTo(target);
if (score < minScore) {
minScore = score;
minNext = next;
return LinkedSet.BREAK;
}
return 0;
});
The return LinkedSet.BREAK (which acts like a break statement in a regular loop) should not be there. It causes the first node in the open list to be selected always, instead of the cheapest one.
Related
I hope everyone is in good health health and condition.
Recently, I have been working on Google Docs hyperlinks using app scripts and learning along the way. I was trying to get all hyperlink and edit them and for that I found an amazing code from this post. I have read the code multiple times and now I have a good understanding of how it works.
My confusion
My confusion is the recursive process happening in this code, although I am familiar with the concept of Recursive functions but when I try to modify to code to get only the first hyperlink from the document, I could not understand it how could I achieve that without breaking the recursive function.
Here is the code that I am trying ;
/**
* Get an array of all LinkUrls in the document. The function is
* recursive, and if no element is provided, it will default to
* the active document's Body element.
*
* #param {Element} element The document element to operate on.
* .
* #returns {Array} Array of objects, vis
* {element,
* startOffset,
* endOffsetInclusive,
* url}
*/
function getAllLinks(element) {
var links = [];
element = element || DocumentApp.getActiveDocument().getBody();
if (element.getType() === DocumentApp.ElementType.TEXT) {
var textObj = element.editAsText();
var text = element.getText();
var inUrl = false;
for (var ch=0; ch < text.length; ch++) {
var url = textObj.getLinkUrl(ch);
if (url != null) {
if (!inUrl) {
// We are now!
inUrl = true;
var curUrl = {};
curUrl.element = element;
curUrl.url = String( url ); // grab a copy
curUrl.startOffset = ch;
}
else {
curUrl.endOffsetInclusive = ch;
}
}
else {
if (inUrl) {
// Not any more, we're not.
inUrl = false;
links.push(curUrl); // add to links
curUrl = {};
}
}
}
if (inUrl) {
// in case the link ends on the same char that the element does
links.push(curUrl);
}
}
else {
var numChildren = element.getNumChildren();
for (var i=0; i<numChildren; i++) {
links = links.concat(getAllLinks(element.getChild(i)));
}
}
return links;
}
I tried adding
if (links.length > 0){
return links;
}
but it does not stop the function as it is recursive and it return back to its previous calls and continue running.
Here is the test document along with its script that I am working on.
https://docs.google.com/document/d/1eRvnR2NCdsO94C5nqly4nRXCttNziGhwgR99jElcJ_I/edit?usp=sharing
I hope you will understand what I am trying to convey, Thanks for giving a look at my post. Stay happy :D
I believe your goal as follows.
You want to retrieve the 1st link and the text of link from the shared Document using Google Apps Script.
You want to stop the recursive loop when the 1st element is retrieved.
Modification points:
I tried adding
if (links.length > 0){
return links;
}
but it does not stop the function as it is recursive and it return back to its previous calls and continue running.
About this, unfortunately, I couldn't understand where you put the script in your script. In this case, I think that it is required to stop the loop when links has the value. And also, it is required to also retrieve the text. So, how about modifying as follows? I modified 3 parts in your script.
Modified script:
function getAllLinks(element) {
var links = [];
element = element || DocumentApp.getActiveDocument().getBody();
if (element.getType() === DocumentApp.ElementType.TEXT) {
var textObj = element.editAsText();
var text = element.getText();
var inUrl = false;
for (var ch=0; ch < text.length; ch++) {
if (links.length > 0) break; // <--- Added
var url = textObj.getLinkUrl(ch);
if (url != null) {
if (!inUrl) {
// We are now!
inUrl = true;
var curUrl = {};
curUrl.element = element;
curUrl.url = String( url ); // grab a copy
curUrl.startOffset = ch;
}
else {
curUrl.endOffsetInclusive = ch;
}
}
else {
if (inUrl) {
// Not any more, we're not.
inUrl = false;
curUrl.text = text.slice(curUrl.startOffset, curUrl.endOffsetInclusive + 1); // <--- Added
links.push(curUrl); // add to links
curUrl = {};
}
}
}
if (inUrl) {
// in case the link ends on the same char that the element does
links.push(curUrl);
}
}
else {
var numChildren = element.getNumChildren();
for (var i=0; i<numChildren; i++) {
if (links.length > 0) { // <--- Added or if (links.length > 0) break;
return links;
}
links = links.concat(getAllLinks(element.getChild(i)));
}
}
return links;
}
In this case, I think that if (links.length > 0) {return links;} can be modified to if (links.length > 0) break;.
Note:
By the way, when Google Docs API is used, both the links and the text can be also retrieved by a simple script as follows. When you use this, please enable Google Docs API at Advanced Google services.
function myFunction() {
const doc = DocumentApp.getActiveDocument();
const res = Docs.Documents.get(doc.getId()).body.content.reduce((ar, {paragraph}) => {
if (paragraph && paragraph.elements) {
paragraph.elements.forEach(({textRun}) => {
if (textRun && textRun.textStyle && textRun.textStyle.link) {
ar.push({text: textRun.content, url: textRun.textStyle.link.url});
}
});
}
return ar;
}, []);
console.log(res) // You can retrieve 1st link and test by console.log(res[0]).
}
Update #2: Okay, I'm pretty sure my error in update #1 was because of indexing out of bounds over the array (I'm still not used to JS indexing at 0). But here is the new problem... if I write out the different combinations of the loop manually, setting the page index to 1 in moveItem() like so:
newForm.moveItem(itemsArray[0][0], 1);
newForm.moveItem(itemsArray[0][1], 1);
newForm.moveItem(itemsArray[0][2], 1);
newForm.moveItem(itemsArray[1][0], 1);
newForm.moveItem(itemsArray[1][1], 1);
newForm.moveItem(itemsArray[1][2], 1);
newForm.moveItem(itemsArray[2][0], 1);
...
...I don't get any errors but the items end up on different pages! What is going on?
Update #1:: Using Sandy Good's answer as well as a script I found at this WordPress blog, I have managed to get closer to what I needed. I believe Sandy Good misinterpreted what I wanted to do because I wasn't specific enough in my question.
I would like to:
Get all items from a page (section header, images, question etc)
Put them into an array
Do this for all pages, adding these arrays to an array (i.e: [[all items from page 1][all items from page 2][all items from page 3]...])
Shuffle the elements of this array
Repopulate a new form with each element of this array. In this way, page order will be randomized.
My JavaScript skills are poor (this is the first time I've used it). There is a step that produces null entries and I don't know why... I had to remove them manually. I am not able to complete step 5 as I get the following error:
Cannot convert Item,Item,Item to (class).
"Item,Item,Item" is the array element containing all the items from a particular page. So it seems that I can't add three items to a page at a time? Or is something else going on here?
Here is my code:
function shuffleForms() {
var itemsArray,shuffleQuestionsInNewForm,fncGetQuestionID,
newFormFile,newForm,newID,shuffle, sections;
// Copy template form by ID, set a new name
newFormFile = DriveApp.getFileById('1prfcl-RhaD4gn0b2oP4sbcKaRcZT5XoCAQCbLm1PR7I')
.makeCopy();
newFormFile.setName('AAAAA_Shuffled_Form');
// Get ID of new form and open it
newID = newFormFile.getId();
newForm = FormApp.openById(newID);
// Initialize array to put IDs in
itemsArray = [];
function getPageItems(thisPageNum) {
Logger.log("Getting items for page number: " + thisPageNum );
var thisPageItems = []; // Used for result
var thisPageBreakIndex = getPageItem(thisPageNum).getIndex();
Logger.log( "This is index num : " + thisPageBreakIndex );
// Get all items from page
var allItems = newForm.getItems();
thisPageItems.push(allItems[thisPageBreakIndex]);
Logger.log( "Added pagebreak item: " + allItems[thisPageBreakIndex].getIndex() );
for( var i = thisPageBreakIndex+1; ( i < allItems.length ) && ( allItems[i].getType() != FormApp.ItemType.PAGE_BREAK ); ++i ) {
thisPageItems.push(allItems[i]);
Logger.log( "Added non-pagebreak item: " + allItems[i].getIndex() );
}
return thisPageItems;
}
function shuffle(array) {
var currentIndex = array.length, temporaryValue, randomIndex;
Logger.log('shuffle ran')
// While there remain elements to shuffle...
while (0 !== currentIndex) {
// Pick a remaining element...
randomIndex = Math.floor(Math.random() * currentIndex);
currentIndex -= 1;
// And swap it with the current element.
temporaryValue = array[currentIndex];
array[currentIndex] = array[randomIndex];
array[randomIndex] = temporaryValue;
}
return array;
}
function shuffleAndMove() {
// Get page items for all pages into an array
for(i = 2; i <= 5; i++) {
itemsArray[i] = getPageItems(i);
}
// Removes null values from array
itemsArray = itemsArray.filter(function(x){return x});
// Shuffle page items
itemsArray = shuffle(itemsArray);
// Move page items to the new form
for(i = 2; i <= 5; ++i) {
newForm.moveItem(itemsArray[i], i);
}
}
shuffleAndMove();
}
Original post: I have used Google forms to create a questionnaire. For my purposes, each question needs to be on a separate page but I need the pages to be randomized. A quick Google search shows this feature has not been added yet.
I see that the Form class in the Google apps script has a number of methods that alter/give access to various properties of Google Forms. Since I do not know Javascript and am not too familiar with Google apps/API I would like to know if what I am trying to do is even possible before diving in and figuring it all out.
If it is possible, I would appreciate any insight on what methods would be relevant for this task just to give me some direction to get started.
Based on comments from Sandy Good and two SE questions found here and here, this is the code I have so far:
// Script to shuffle question in a Google Form when the questions are in separate sections
function shuffleFormSections() {
getQuestionID();
createNewShuffledForm();
}
// Get question IDs
function getQuestionID() {
var form = FormApp.getActiveForm();
var items = form.getItems();
arrayID = [];
for (var i in items) {
arrayID[i] = items[i].getId();
}
// Logger.log(arrayID);
return(arrayID);
}
// Shuffle function
function shuffle(a) {
var j, x, i;
for (i = a.length; i; i--) {
j = Math.floor(Math.random() * i);
x = a[i - 1];
a[i - 1] = a[j];
a[j] = x;
}
}
// Shuffle IDs and create new form with new question order
function createNewShuffledForm() {
shuffle(arrayID);
// Logger.log(arrayID);
var newForm = FormApp.create('Shuffled Form');
for (var i in arrayID) {
arrayID[i].getItemsbyId();
}
}
Try this. There's a few "constants" to be set at the top of the function, check the comments. Form file copying and opening borrowed from Sandy Good's answer, thanks!
// This is the function to run, all the others here are helper functions
// You'll need to set your source file id and your destination file name in the
// constants at the top of this function here.
// It appears that the "Title" page does not count as a page, so you don't need
// to include it in the PAGES_AT_BEGINNING_TO_NOT_SHUFFLE count.
function shuffleFormPages() {
// UPDATE THESE CONSTANTS AS NEEDED
var PAGES_AT_BEGINNING_TO_NOT_SHUFFLE = 2; // preserve X intro pages; shuffle everything after page X
var SOURCE_FILE_ID = 'YOUR_SOURCE_FILE_ID_HERE';
var DESTINATION_FILE_NAME = 'YOUR_DESTINATION_FILE_NAME_HERE';
// Copy template form by ID, set a new name
var newFormFile = DriveApp.getFileById(SOURCE_FILE_ID).makeCopy();
newFormFile.setName(DESTINATION_FILE_NAME);
// Open the duplicated form file as a form
var newForm = FormApp.openById(newFormFile.getId());
var pages = extractPages(newForm);
shuffleEndOfPages(pages, PAGES_AT_BEGINNING_TO_NOT_SHUFFLE);
var shuffledFormItems = flatten(pages);
setFormItems(newForm, shuffledFormItems);
}
// Builds an array of "page" arrays. Each page array starts with a page break
// and continues until the next page break.
function extractPages(form) {
var formItems = form.getItems();
var currentPage = [];
var allPages = [];
formItems.forEach(function(item) {
if (item.getType() == FormApp.ItemType.PAGE_BREAK && currentPage.length > 0) {
// found a page break (and it isn't the first one)
allPages.push(currentPage); // push what we've built for this page onto the output array
currentPage = [item]; // reset the current page to just this most recent item
} else {
currentPage.push(item);
}
});
// We've got the last page dangling, so add it
allPages.push(currentPage);
return allPages;
};
// startIndex is the array index to start shuffling from. E.g. to start
// shuffling on page 5, startIndex should be 4. startIndex could also be thought
// of as the number of pages to keep unshuffled.
// This function has no return value, it just mutates pages
function shuffleEndOfPages(pages, startIndex) {
var currentIndex = pages.length;
// While there remain elements to shuffle...
while (currentIndex > startIndex) {
// Pick an element between startIndex and currentIndex (inclusive)
var randomIndex = Math.floor(Math.random() * (currentIndex - startIndex)) + startIndex;
currentIndex -= 1;
// And swap it with the current element.
var temporaryValue = pages[currentIndex];
pages[currentIndex] = pages[randomIndex];
pages[randomIndex] = temporaryValue;
}
};
// Sourced from elsewhere on SO:
// https://stackoverflow.com/a/15030117/4280232
function flatten(array) {
return array.reduce(
function (flattenedArray, toFlatten) {
return flattenedArray.concat(Array.isArray(toFlatten) ? flatten(toFlatten) : toFlatten);
},
[]
);
};
// No safety checks around items being the same as the form length or whatever.
// This mutates form.
function setFormItems(form, items) {
items.forEach(function(item, index) {
form.moveItem(item, index);
});
};
I tested this code. It created a new Form, and then shuffled the questions in the new Form. It excludes page breaks, images and section headers. You need to provide a source file ID for the original template Form. This function has 3 inner sub-functions. The inner functions are at the top, and they are called at the bottom of the outer function. The arrayOfIDs variable does not need to be returned or passed to another function because it is available in the outer scope.
function shuffleFormSections() {
var arrayOfIDs,shuffleQuestionsInNewForm,fncGetQuestionID,
newFormFile,newForm,newID,items,shuffle;
newFormFile = DriveApp.getFileById('Put the source file ID here')
.makeCopy();
newFormFile.setName('AAAAA_Shuffled_Form');
newID = newFormFile.getId();
newForm = FormApp.openById(newID);
arrayOfIDs = [];
fncGetQuestionID = function() {
var i,L,thisID,thisItem,thisType;
items = newForm.getItems();
L = items.length;
for (i=0;i<L;i++) {
thisItem = items[i];
thisType = thisItem.getType();
if (thisType === FormApp.ItemType.PAGE_BREAK ||
thisType === FormApp.ItemType.SECTION_HEADER ||
thisType === FormApp.ItemType.IMAGE) {
continue;
}
thisID = thisItem.getId();
arrayOfIDs.push(thisID);
}
Logger.log('arrayOfIDs: ' + arrayOfIDs);
//the array arrayOfIDs does not need to be returned since it is available
//in the outermost scope
}// End of fncGetQuestionID function
shuffle = function() {// Shuffle function
var j, x, i;
Logger.log('shuffle ran')
for (i = arrayOfIDs.length; i; i--) {
j = Math.floor(Math.random() * i);
Logger.log('j: ' + j)
x = arrayOfIDs[i - 1];
Logger.log('x: ' + x)
arrayOfIDs[i - 1] = arrayOfIDs[j];
arrayOfIDs[j] = x;
}
Logger.log('arrayOfIDs: ' + arrayOfIDs)
}
shuffleQuestionsInNewForm = function() {
var i,L,thisID,thisItem,thisQuestion,questionType;
L = arrayOfIDs.length;
for (i=0;i<L;i++) {
thisID = arrayOfIDs[i];
Logger.log('thisID: ' + thisID)
thisItem = newForm.getItemById(thisID);
newForm.moveItem(thisItem, i)
}
}
fncGetQuestionID();//Get all the question ID's and put them into an array
shuffle();
shuffleQuestionsInNewForm();
}
From the https://gist.github.com/oshliaer/d468759b3587cfb424348fa722765187 , It is possible to select a particular word from the findText, I want to implement the same for bold words only
I have a function to find bold. How do I modify the above gist?
var startFlag = x;
var flag = false;
for (var i = x; i < y; i++) {
if (text.isBold(i) && !flag) {
startFlag = i;
flag = true;
} else if (!text.isBold(i) && flag) {
flag = false;
rangeBuilder.addElement(text, startFlag, i - 1);
doc.setSelection(rangeBuilder.build());
return;
}
}
if (flag) {
rangeBuilder.addElement(text, startFlag, i - 1);
doc.setSelection(rangeBuilder.build());
return;
}
Let's assume another algorithm
/*
* #param {(DocumentApp.ElementType.LIST_ITEM | DocumentApp.ElementType.PARAGRAPH)} element
*/
function hasBold(element, start) {
var text = element.editAsText();
var length = element.asText().getText().length;
var first = -1;
var end = -1;
while (start < length) {
if (first < 0 && text.isBold(start)) {
first = start;
}
if (first > -1 && !text.isBold(start)) {
end = start - 1;
return {
s: first,
e: end
}
}
start++;
}
if (first > -1) {
return {
s: first,
e: length - 1
}
}
return false;
}
It's not clean but I've tested it and it works fine.
hasBold lets us finding bolds in the current element.
Finally, we have to loop this feature within document.getBody().
You could to get the full code here find next bold text in google document.
Also you could try it on a copy
A new idea
The Direct searcing
The best way is to use a callback while it is checked
var assay = function (re) {
var text = re.getElement()
.asText();
for (var offset = re.getStartOffset(); offset <= re.getEndOffsetInclusive(); offset++) {
if (!text.isBold(offset)) return false;
}
return true;
}
function findNextBold() {
var sp = 'E.';
Docer.setDocument(DocumentApp.getActiveDocument());
var rangeElement = Docer.findText(sp, Docer.getActiveRangeElement(), assay);
rangeElement ? Docer.selectRangeElement(rangeElement) : Docer.setCursorBegin();
}
The Approx searching
var assay = function(re) {
var text = re.getElement().asText();
var startOffset = re.getStartOffset();
var endOffset = re.getEndOffsetInclusive() + 1;
for (var offset = startOffset; offset < endOffset; offset++) {
if (!text.isBold(offset)) return false;
}
return this.test(text.getText().slice(startOffset, endOffset));
}
function findNextBold() {
var searchPattern = '[^ ]+#[^ ]+';
var testPattern = new RegExp('^[A-Z0-9._%+-]+#[A-Z0-9.-]+\.[A-Z]{2,4}$');
Docer.setDocument(DocumentApp.getActiveDocument());
var rangeElement = Docer.findText(searchPattern, Docer.getActiveRangeElement(), assay.bind(testPattern));
rangeElement ? Docer.selectRangeElement(rangeElement) : Docer.setCursorBegin();
}
Docer
Yes. it is possible to find bold text. You need to use findText(searchPattern) to search the contents of the element for the specific text pattern using regular expressions. The provided regular expression pattern is independently matched against each text block contained in the current element. Then, use isBold() to retrieve the bold setting. It is a Boolean which returns whether the text is bold or null.
I have twenty eight instances of a two-frame MovieClip (frame1 = off - frame 2 = on) to select PDFs to send. The following code works fine, but I am looking to tighten it up and make it less verbose and easier to read. I include only one reference to an instance for space and sanity sake.
function PDFClick(e:MouseEvent):void {
targetPDF = e.target.ID;
trace("targetPDF " +targetPDF);
if (targetPDF == "PDF1")
if (pdf.pcconnectionPDF1.currentFrame == 1)
{
pdf.pcconnectionPDF1.gotoAndPlay(2);
PDF1 = 1;
trace("PDF1 is "+PDF1);
}else{
pdf.pcconnectionPDF1.gotoAndPlay(1);
PDF1 = 0;
trace("PDF1 is "+PDF1);
}
Thanks! trying to learn
You'll want to generalize your calls to your ID, that way you don't need special code for each condition.
function PDFClick(e:MouseEvent):void {
var ID:String = e.target.ID;
var mc = pdf["pcconnection" + ID];
if (mc.currentframe == 1) {
mc.gotoAndPlay(2);
this[ID] = 1;
} else {
mc.gotoAndPlay(1);
this[ID] = 0;
}
}
How about this:
function PDFClick(e:MouseEvent):void {
targetPDF = e.target.ID;
trace("targetPDF " +targetPDF);
if (targetPDF == "PDF1") {
var frame:int = pdf.pconnectionPDF1.currentFrame;
pdf.pconnectionPDF1.gotoAndPlay( frame == 1 ? (PDF1 = 1)+1 : (PDF1 = 0)+1 );
}
}
I think that's about what you are looking for.
I have a button on page - when clicked, it passes all the data to the servlet that could update each row data. My question is how to pass the whole store to the servlet as json data? Is there any easy way? Thanks
Here is some code I wrote to get the store to an object. Then it can be converted to JSON using dojo.toJson(obj);. I learned about this from the dojotoolkit website originally. (Give credit where credit is due). I realize this code is huge and nasty. When I looked for a better way about a year back I could not find one.
JsonHelper.storeToObject = function(store) {
var object = [];
var index = -1;
store.fetch({
onItem : function(item, request) {
object[++index] = JsonHelper.itemToObject(store, item);
}
});
return object;
};
JsonHelper.itemToObject = function(store, item) {
// store:
// The datastore the item came from.
// item:
// The item in question.
var obj = {};
if (item && store) {
// Determine the attributes we need to process.
var attributes = store.getAttributes(item);
if (attributes && attributes.length > 0) {
var i;
for (i = 0; i < attributes.length; i++) {
var values = store.getValues(item, attributes[i]);
if (values) {
// Handle multivalued and single-valued attributes.
if (values.length > 1) {
var j;
obj[attributes[i]] = [];
for (j = 0; j < values.length; j++) {
var value = values[j];
// Check that the value isn't another item. If
// it is, process it as an item.
if (store.isItem(value)) {
obj[attributes[i]].push(itemToObject(store,
value));
} else {
obj[attributes[i]].push(value);
}
}
} else {
if (store.isItem(values[0])) {
obj[attributes[i]] = itemToObject(store,
values[0]);
} else {
obj[attributes[i]] = values[0];
}
}
}
}
}
}
return obj;
};