I basically need to make about 3 calls to get the data for a json object.. It basically JSON array of JSON object which have some attributes, one of which is an array of other values selected using a second query, then that one also has an array inside which is selected with another db call.
I tried using asyn.concatSeries so that I can dig down into the bottom call and put together all the information I collected for one root json object but that's creating a lot of unexpected behaviour..
Example of JSON
[
{
"item" : "firstDbCall"
"children" : [ {
"name" : "itemDiscoveredWithSecondDBCall"
"children" : [ itemsDiscoveredwith3rdDBCall]
},
]
}
]
This is really difficult using node.js. I really need to figure out how to do this properly since I have to do many of these for different purposes.
EDIT
This is the code i have. There's some strange behaviour with async.concatSeries. The results get called multiple times after each one of the functions finish for each array. So i had to put a check in place. I know it's very messy code but i've been just putting band-aids all over it for the past 2 hours to make it work..
console.log("GET USERS HAREDQARE INFO _--__--_-_-_-_-_____");
var query = "select driveGroupId from tasks, driveInformation where agentId = '"
+ req.params.agentId + "' and driveInformation.taskId = tasks.id order by driveInformation.taskId desc;";
connection.query(query, function(err, rows) {
if (rows === undefined) {
res.json([]);
return;
}
if(rows.length<1) { res.send("[]"); return;}
var driveGroupId = rows[0].driveGroupId;
var physicalQuery = "select * from drives where driveGroupId = " + driveGroupId + ";";
connection.query(physicalQuery, function(err, rows) {
console.log("ROWSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS");
console.log(rows);
async.concatSeries(rows, function(row, cb) {
console.log("-------------------------------SINGLE ROW-------------------------------------");
console.log(row);
if(row.hasLogicalDrives != 0) {
console.log("HAS LOGICAL DRIVES");
console.log(row.id);
var query = "select id, name from logicalDrives where driveId = " + row.id;
connection.query(query, function(error, drives) {
console.log("QUERY RETURNED");
console.log(drives);
parseDriveInfo(row.name, row.searchable, drives, cb);
});
}
else
var driveInfo = { "driveName" : row.name, "searchable" : row.searchable};
console.log("NO SUB ITEMS");
cb(null, driveInfo);
}, function(err, results) {
console.log("GEETTTTINGHERE");
console.log(results);
if(results.length == rows.length) {
console.log("RESULTS FOR THE DRIVE SEARCH");
console.log(results);
var response = {"id": req.params.agentId};
response.driveList = results;
console.log("RESPONSE");
console.log(response);
res.json(response);
}
});
});
});
};
parseDriveInfo = function(driveName, searchable, drives, cb) {
async.concatSeries(drives, function(drive,callback) {
console.log("SERIES 2");
console.log(drive);
console.log("END OF DRIVE INFO");
var query = "select name from supportedSearchTypes where logicalDriveId = " + drive.id;
connection.query(query, function(error, searchTypes) {
drive.searchTypes = searchTypes;
var driveInfo = { "driveName" :driveName,
"searchable" : searchable,
"logicalDrives" : drive
};
callback(null, driveInfo);
});
}, function (err, results) {
console.log("THIS IS ISISIS ISISISSISISISISISISISISISIS");
console.log(results);
if(results.length === drives.length) {
console.log("GOTHERE");
cb(null, results);
}
});
}
Getting good enough with async to use exactly the right combination of methods under the right circumstances takes a fair amount of experience. Most likely your case in particular can be handled with async.waterfall if its query1 then query2(dataFoundByQuery1) then query3(dataFoundByQuery2). But depending on the circumstances you need to mix and match async methods appropriately and sometimes have 2 levels - for example a "big picture" async.waterfall where some of the steps in the waterfall do async.parallel or async.series as needed. I've never used async.concat and given your needs I think you have chosen the wrong method. The workhorses are async.each, async.eachSeries, async.waterfall, and async.map, at least for the web app & DB query use cases I mostly encounter, so make sure you really have those understood before exploring the more specific convenience methods.
EDIT: This is a more in depth example based on use of the connection library you seem to be using. Please note, some of this is javascript psuedo code. Things like adding objects to the resultsArray are clearly not complete, the only thing I took time to make sure was correct is the "flow of logic" as it pertains to callbacks. Everything else is for you to implement. In order to support multiple calls to the same callback function and maintain state from call to call, the best way is to wrap the set of callbacks in a closure. This allows the callbacks to share some state with the main event loop. This allows you to pass arguments to the callbacks, without actually having to pass them as arguments, much like class variables in c++, or even globals in javascript, but we haven't poluted the global scope :)
function queryDataBase(query) {
//wrap the whole query in a function so the callbacks can share some
//variables with similar scope. This is called a closure
int rowCounter = 0;
var dataRowsFromStep2;
var resultsArray = {};
connection.query(query, dataBaseQueryStep2);
function dataBaseQueryStep2(err, rows) {
//do something with err and rows
dataRowsFromStep2 = rows;
var query = getQueryFromRow(dataRowsFromStep2[rowCounter++]);//Always zero the first time. Might need to double check rows isn't empty!
connection.query(query, dataBaseQueryStep3);
}
function dataBaseQueryStep3(err, rows) {
//do something with err and rows
if(rowCounter < dataRowsFromStep2.size) {
resultsArray.add(rows);//Probably needs to be more interesting, but you get the idea
//since this is within the same closure, rowCounter maintains it's state
var query = getQueryFromRow(dataRowsFromStep2[rowCounter++]);
//recursive call query using dataBaseQueryStep3 as it's callback repeatedly until
//we run out of rows to call it on.
connection.query(query, dataBaseQueryStep3)
} else {
//when the if statement fails we have no more rows to run queries on so return to main program flow
returnToMainProgramLogic(resultsArray);
}
}
}
function returnToMainProgramLogic(results) {
//continue running your program here
}
I personally like the above logic better than the syntax async produces... I believe the heart of your problem rests in your nested calls to async, and the fact that ASYN itself, runs the series of functions asynchronously, but in order(confusing I know). If you write your program like this, you won't have to worry about it!
I would strongly suggest using sequelize.js It provides a really powerful orm that allows you to chain queries together. It also allows you to directly load your data into js objects, write dynamic sql, and connect to many different databases. Picture ActiveRecord from the Ruby world for Node.
Related
I was recently building a scraper module to get some information with nodejs until I encountered this "little" problem. The modules that I'm using are cheeriojs and request.
Actually the module works like a charm if I call only one method at a time. It contains three function and only two of them are exported, this is the code:
'use strict';
var request = require('request'),
cheerio = require('cheerio'),
counter = 0;
function find(term, cat, callback) {
// All the check for the parameters
scrape("http://.../search.php?search=" + encodeURIComponent(term), cat, callback);
}
function last(cat, callback) {
// All the check for the parameters
scrape("http://google.com/", cat, callback);
}
function scrape(url, cat, callback) {
request(url, function (error, response, body) {
if (!error && response.statusCode == 200) {
var $ = cheerio.load(body);
var result = [];
var items = $('.foo, .foo2').filter(function() {
// Condition to filter the resulted items
});
items.each(function(i, row) {
// Had to do another request inside here to scrape other information
request( $(".newpagelink").attr("href"), function(error, response, body) {
var name = $(".selector").text(),
surname = $(".selector2").text(),
link = cheerio.load(body)('.magnet').attr('href'); // This is the only thing that I'm scraping from the new page, the rest comes from the other "cheerio.load"
// Push an object in the array
result.push( { "name": name, "surname": surname, "link": link } );
// To check when the async requests are ended
counter++;
if(counter == items.length-1) {
callback(null, result);
}
});
});
}
});
}
exports.find = find;
exports.last = last;
The problem now, as I was saying, is that if I create a new node script "test.js" and I call only last OR find, it works perfectly! But if I call both the methods consecutively like this:
var mod = require("../index-tmp.js");
mod.find("bla", "blabla", function(err, data) {
if (err) throw err;
console.log(data.length + " find");
});
mod.last(function(err, data) {
console.log(data.length + " last");
});
The results are completely messed up, sometimes the script doesn't even print something, other times print the result of only "find" or "last", and other times returns a cheeriojs error (I won't add here to not mess you up, because probably it's my script's fault). I thought also to repeat the same function two times for both the methods but nothing, the same problems occur... I don't know what else to try, I hope you'll tell me the cause of this behavior!
Your counter variable is global, not specific to each scrape call. It wouldn't work if you called find twice at the same time either, or last.
Move the declaration and initialisation of var counter = 0; into the scrape function, or even better right next to the result and items declarations.
From scanning your code quickly, this is probably due to the variable counter being global. These are asynchronous functions, so they will both act on counter at the same thing. Move the declaration inside of the scrape function.
If you need more information about asynchronous programming, refer to Felix's great answer in this question.
I have only recently started developing for node.js, so forgive me if this is a stupid question - I come from Javaland, where objects still live happily sequentially and synchronous. ;)
I have a key generator object that issues keys for database inserts using a variant of the high-low algorithm. Here's my code:
function KeyGenerator() {
var nextKey;
var upperBound;
this.generateKey = function(table, done) {
if (nextKey > upperBound) {
require("../sync/key-series-request").requestKeys(function(err,nextKey,upperBound) {
if (err) { return done(err); }
this.nextKey = nextKey;
this.upperBound = upperBound;
done(nextKey++);
});
} else {
done(nextKey++);
}
}
}
Obviously, when I ask it for a key, I must ensure that it never, ever issues the same key twice. In Java, if I wanted to enable concurrent access, I would make make this synchronized.
In node.js, is there any similar concept, or is it unnecessary? I intend to ask the generator for a bunch of keys for a bulk insert using async.parallel. My expectation is that since node is single-threaded, I need not worry about the same key ever being issued more than once, can someone please confirm this is correct?
Obtaining a new series involves an asynchronous database operation, so if I do 20 simultaneous key requests, but the series has only two keys left, won't I end up with 18 requests for a new series? What can I do to avoid that?
UPDATE
This is the code for requestKeys:
exports.requestKeys = function (done) {
var db = require("../storage/db");
db.query("select next_key, upper_bound from key_generation where type='issue'", function(err,results) {
if (err) { done(err); } else {
if (results.length === 0) {
// Somehow we lost the "issue" row - this should never have happened
done (new Error("Could not find 'issue' row in key generation table"));
} else {
var nextKey = results[0].next_key;
var upperBound = results[0].upper_bound;
db.query("update key_generation set next_key=?, upper_bound=? where type='issue'",
[ nextKey + KEY_SERIES_WIDTH, upperBound + KEY_SERIES_WIDTH],
function (err,results) {
if (err) { done(err); } else {
done(null, nextKey, upperBound);
}
});
}
}
});
}
UPDATE 2
I should probably mention that consuming a key requires db access even if a new series doesn't have to be requested, because the consumed key will have to be marked as used in the database. The code doesn't reflect this because I ran into trouble before I got around to implementing that part.
UPDATE 3
I think I got it using event emitting:
function KeyGenerator() {
var nextKey;
var upperBound;
var emitter = new events.EventEmitter();
var requesting = true;
// Initialize the generator with the stored values
db.query("select * from key_generation where type='use'", function(err, results)
if (err) { throw err; }
if (results.length === 0) {
throw new Error("Could not get key generation parameters: Row is missing");
}
nextKey = results[0].next_key;
upperBound = results[0].upper_bound;
console.log("Setting requesting = false, emitting event");
requesting = false;
emitter.emit("KeysAvailable");
});
this.generateKey = function(table, done) {
console.log("generateKey, state is:\n nextKey: " + nextKey + "\n upperBound:" + upperBound + "\n requesting:" + requesting + " ");
if (nextKey > upperBound) {
if (!requesting) {
requesting = true;
console.log("Requesting new series");
require("../sync/key-series-request").requestSeries(function(err,newNextKey,newUpperBound) {
if (err) { return done(err); }
console.log("New series available:\n nextKey: " + newNextKey + "\n upperBound: " + newUpperBound);
nextKey = newNextKey;
upperBound = newUpperBound;
requesting = false;
emitter.emit("KeysAvailable");
done(null,nextKey++);
});
} else {
console.log("Key request is already underway, deferring");
var that = this;
emitter.once("KeysAvailable", function() { console.log("Executing deferred call"); that.generateKey(table,done); });
}
} else {
done(null,nextKey++);
}
}
}
I've peppered it with logging outputs, and it does do what I want it to.
As another answer mentions, you will potentially end up with results different from what you want. Taking things in order:
function KeyGenerator() {
// at first I was thinking you wanted these as 'class' properties
// and thus would want to proceed them with this. rather than as vars
// but I think you want them as 'private' members variables of the
// class instance. That's dandy, you'll just want to do things differently
// down below
var nextKey;
var upperBound;
this.generateKey = function (table, done) {
if (nextKey > upperBound) {
// truncated the require path below for readability.
// more importantly, renamed parameters to function
require("key-series-request").requestKeys(function(err,nKey,uBound) {
if (err) { return done(err); }
// note that thanks to the miracle of closures, you have access to
// the nextKey and upperBound variables from the enclosing scope
// but I needed to rename the parameters or else they would shadow/
// obscure the variables with the same name.
nextKey = nKey;
upperBound = uBound;
done(nextKey++);
});
} else {
done(nextKey++);
}
}
}
Regarding the .requestKeys function, you will need to somehow introduce some kind of synchronization. This isn't actually terrible in one way because with only one thread of execution, you don't need to sweat the challenge of setting your semaphore in a single operation, but it is challenging to deal with the multiple callers because you will want other callers to effectively (but not really) block waiting for the first call to requestKeys() which is going to the DB to return.
I need to think about this part a bit more. I had a basic solution in mind which involved setting a simple semaphore and queuing the callbacks, but when I was typing it up I realized I was actually introducing a more subtle potential synchronization bug when processing the queued callbacks.
UPDATE:
I was just finishing up one approach as you were writing about your EventEmitter approach, which seems reasonable. See this gist which illustrates the approach. I took. Just run it and you'll see the behavior. It has some console logging to see which calls are getting deferred for a new key block or which can be handled immediately. The primary moving part of the solution is (note that the keyManager provides the stubbed out implementation of your require('key-series-request'):
function KeyGenerator(km) {
this.nextKey = undefined;
this.upperBound = undefined;
this.imWorkingOnIt = false;
this.queuedCallbacks = [];
this.keyManager = km;
this.generateKey = function(table, done) {
if (this.imWorkingOnIt){
this.queuedCallbacks.push(done);
console.log('KG deferred call. Pending CBs: '+this.queuedCallbacks.length);
return;
};
var self=this;
if ((typeof(this.nextKey) ==='undefined') || (this.nextKey > this.upperBound) ){
// set a semaphore & add the callback to the queued callback list
this.imWorkingOnIt = true;
this.queuedCallbacks.push(done);
this.keyManager.requestKeys(function(err,nKey,uBound) {
if (err) { return done(err); }
self.nextKey = nKey;
self.upperBound = uBound;
var theCallbackList = self.queuedCallbacks;
self.queuedCallbacks = [];
self.imWorkingOnIt = false;
theCallbackList.forEach(function(f){
// rather than making the final callback directly,
// call KeyGenerator.generateKey() with the original
// callback
setImmediate(function(){self.generateKey(table,f);});
});
});
} else {
console.log('KG immediate call',self.nextKey);
var z= self.nextKey++;
setImmediate(function(){done(z);});
}
}
};
If your Node.js code to calculate the next key didn't need to execute an async operation then you wouldn't run into synchronization issues because there is only one JavaScript thread executing code. Access to the nextKey/upperBound variables will be done in sequence by only one thread (i.e. request 1 will access first, then request 2, then request 3 et cetera.) In the Java-world you will always need synchronization because multiple threads will be executing even if you didn't make a DB call.
However, in your Node.js code since you are making an async call to get the nextKey you could get strange results. There is still only one JavaScript thread executing your code, but it would be possible for request 1 to make the call to the DB, then Node.js might accept request 2 (while request 1 is getting data from the DB) and this second request will also make a request to the DB to get keys. Let's say that request 2 gets data from the DB quicker than request 1 and update nextKey/upperBound variables with values 100/150. Once request 1 gets its data (say values 50/100) then it will update nextKey/upperBound. This scenario wouldn't result in duplicate keys, but you might see gaps in your keys (for example, not all keys 100 to 150 will be used because request 1 eventually reset the values to 50/100)
This makes me think that you will need a way to sync access, but I am not exactly sure what will be the best way to achieve this.
I am making a new webservice where i send a curl command with JSON and the JSON contains a array as
[{tempid:1,email:abc#123,address:asd},{tempid:2,email:abc#12345,address:asd45},{tempid:3,email:abc#1234,address:asd4}]
Now when i pass and insert the array in a mysql table tempid is just to show a mapping to the user to the contact id generated in the database as tempid:1 is now inserted and in database it has cid 120 , like this for tempid2 and 3 ,
But when i am trying to show the client the updated values it shows only one value , last last change not the whole updated Array. Its becuase of the async nature of the connection.querry function , so i need help in this , here is my webservice
contactadd webservice -->
for(var i=0;i<=request.body.contact.length-1;i++)
{
if(request.body.contact[i].tempid)
{ var ardata=new Array();
var o=request.body.contact[i];
pair=Object.keys(o).map(function(a){ return [a, o[a]] });
AM.addcontact(pair,request.session.user,request.body.contact.length,function(e,o){
if(!o)
{
response.send('something went wrong'+e);
}
else
{
//response.send(o);
}
});
}
}
Here is the update function in the database.js script -->
//ContactSync-addcontact module for database
exports.addcontact=function (arr,email,addnum,callback)
{
var counter=0;
var uid;
var data=new Array();
var showinsert=new Array();
var values=new Array();
var datatable=new Array();
var inserting=new Array();
var tempid=0;
connection.query('SELECT UID FROM user where email1="'+email.email+'"',function(err,rows,fields){
if(err)
{
throw err;
}
else
{
if(rows[0]!=undefined)
{
uid=rows[0]['UID'];
}
else
{
uid="no id in database";
}
}
});// get the UID of the inserting user
// make array of user provided data
for(var j=0;j<=arr.length-1;j++)
{
if(arr[j][0]!='tempid')
{
data.push(arr[j][0]);
}
else
{
tempid=arr[j][1];
}
}
connection.query('SELECT column_name FROM information_schema.columns where table_schema="webservice" AND table_name="usercontacts"',function(err,rows,fields){
if(err)
{
throw err;
}
else
{
for(var i=0;i<=rows.length-1;i++)
{
datatable.push(rows[i]['column_name']);
}
}
for(var k=0;k<=datatable.length-1;k++)
{
if(inArray(data[k],datatable))
{
inserting.push(data[k]);
}
}
if(inserting.length>0)
{
for(var z=0;z<=arr.length-1;z++)
{
if(inArray(arr[z][0],inserting))
{
values.push('"'+arr[z][1]+'"');
}
}
// Insert tempid values and data in the usercontacts table with inserting and values
connection.query('INSERT INTO usercontacts (cid,uid,'+inserting+') VALUES("","'+uid+'",'+values+')',function(err,rows,fields){
if(err)
{
throw err;
}
else
{
connection.query('SELECT * FROM usercontacts WHERE uid="'+uid+'" ORDER BY cid DESC LIMIT 0,'+addnum+'',function(err,rows,fields){
if(err)
{
throw err;
}
else
{ showinsert.push('temp-id: '+tempid+',cid:'+rows[0].cid+',uid:'+uid);
//for(var i=0;i<=inserting.length-1;i++)
forEach(inserting,function(row,index)
{
showinsert.push(inserting[index]+":"+values[index]);
counter+=1;
});
callback(null,showinsert);
}
});
}
});
//insertion finished
}
else
{
callback("Please Provide atleast one field to enter with tempid");
}
});
}
I just need to insert all the callback in a array which has been inserted and show user that array ,please help , completely stuck and then only i am trying StackOverflow.
Thank you for reading till the end BTW
I'm not sure what the specific problem is, but there are some problems with the code you've shared that will bite you sooner or later. One of these may be causing your problem.
Race conditions
If the query SELECT UID FROM user where email1= for any reason takes longer than the SELECT column_name FROM information_schema.columns just below it then you won't have a value for the variable uuid and your logic will fail. Remember that these calls are non-blocking, so you can't rely on one finishing before the other one unless they're nested or use another flow-control mechanism (As #Tracker points out, async is popular).
Catching edge cases
In the line below you're assigning a string value to the uid variable and then continuing to use that variable even though it now contains an error message.
uid="no id in database";
Doing that means that your code later on will have trouble reacting. Instead use a different variable, leave the uid = undefined or immediately return the callback with an error, e.g.
return callback(new Error("user not found"));
Reporting errors
Don't throw errors in Node unless you want to kill the process, e.g. dependency problems during server startup. It doesn't work like Java, async errors are not caught by try/catch and will kill your process or leave you in a state that's hard to reason about. Instead make the error object your first parameter to the callback and return it immediately, like this:
if ( err ) return callback(err);
Then in your client code you can always check the first parameter to see if there was a problem.
Security problem
As #Tracker mentioned, don't ever do the this:
connection.query('SELECT UID FROM user where email1="'+email.email+'"', ...
If the value of the variable is passed through as "; drop table user; or similar then you're in trouble. Instead you can use node-mysql's build in escaping like this:
connection.query('SELECT UID FROM user where email1=?', [email.email], ...
Whitelist
You're querying information_schema.columns in order to detect which fields are valid then inserting them into usercontacts. This is a clever trick, but increases a 3 query process to 4 queries, and raises questions if there are any fields that a user shouldn't be inserting data into. Using a column whitelist may seem like more code to maintain, but would actually be simpler than all the code required to match columns dynamically.
Arrays
I don't see the source for the function inArray() but it looks like it does the same as Array.prototype.indexOf() so it may be better to use that. e.g.
if ( datatable.indexOf(data[k]) > -1 ) inserting.push(data[k]);
Every line of custom code you can delete is a line of code you don't have to maintain.
Asynchronous callbacks are great but when one callback depends on the result of another I have callbacks with api calls that have callbacks, and so on.
apiCall(function () { apiCall(function () { apiCall(function () ...
I can name the callback functions instead of including them inline. That looks prettier and has less nesting but I do not find it any easier to read.
Here is an example. I need to query the local sqlite database, use the result to query a server, then use the response to update the local database.
function sync() {
db.transaction(
function (transaction) {
execute(transaction, 'SELECT max(server_time) AS server_time FROM syncs;', [],
function (transaction, results) { // Query results callback
var t = results.rows.item(0).server_time;
$.post('sync.json', { last_sync_time: (t || '1980-01-01') },
function (data) { // Ajax callback
db.transaction(
function(transaction) {
$(data.thing).each(function () {
var thing = new Thing(this.thing);
thing.insert(transaction);
});
});
});
});
});
}
Is there a way to untagle this (other than naming the callbacks)?
I think you're too quick to discard un-nesting things by naming your functions rather than writing them inline. This is pretty much the only way to clean up that mess.
Instead of:
do_a(
function () {
// more nesting...
}
);
Use names to provide a bit of clarity and purpose to each function:
function on_a_complete() {
}
do_a(on_a_complete);
I think you answered your own question. Naming your callbacks is really the only way to clean this up anymore. Something like:
execute(transaction, 'SELECT max(server_time) AS server_time FROM syncs;', [],handleLocalResults, errorHandler);
handleLocalResults = function (transaction, results)...
handleServerResults = func...
Naming the callbacks is one thing you can do, but the other thing that you need to do is to have non-overlapping SQL transactions. (1)
Your first transaction should be like this:
// The whole thing starts here
db.transaction(selectTimeCB, null, ajaxPost);
The transaction starts with the callback to select the time, and when the transaction is complete, make a call to the ajaxPost operation.
// Initial transaction to get server_time
var selectTimeCB = function(t) {
var query = 'SELECT max(server_time) AS server_time FROM syncs';
t.executeSql(query, [], postLastSyncCB);
};
// This saves the results from the above select, and nothing else.
var server_time;
var postLastSyncCB = function(t, results) {
server_time = results.rows.item(0).server_time;
};
var ajaxPost = function() {
$.post('sync.json', { last_sync_time: (server_time || '1980-01-01') }, nextDbTransaction);
};
If you have overlapped SQL transactions, that can really kill the performance of the database. I recently did a test of 200 mixed transactions on a database with a bit over 500 rows, and found that keeping the transactions separate reduced a run time of over 90 seconds to 3-5 seconds.
I'm working on a bit of JavaScript that interacts with a client-side SQLite database, via the newish window.openDatabase(...), database.transaction(...) and related APIs. As most of you know when you execute a query in this way it is an asynchronous call, which is typically good. You can make the call and handle the results as appropriate with callbacks.
In my current situation I'm working on an algo for a client that does some hierarchy walking in the locally stored database. The part of the algo I'm having trouble with requires starting at some row, which has a reference to a "parent" (by id) that is another row further up in the table. I have to keep walking up this tree until I reach the root.
The problem is that I'm at a point where I'm not sure how to use an asynchronous style query with a callback to keep feeding the loop parent ids. Ideally I could get the query to block so that I can do it all in the loop. Here's the key parts of my current setup:
for (i in search.searchResults.resultsArray)
{
hierarchyArr = new Array();
pageHierarchyArr = new Array();
id = search.searchResults.resultsArray[i].ID;
while (id != null && id != "")
{
var hierarchySql = "SELECT ID, parentID, type, content FROM content WHERE ID = " + id;
// This is a prettied up call to database.transaction(...)
var rs = db.getRS(hierarchySql);
// Ideally the code below doesn't execute until rs is populated
hierarchyArr.push(rs[0]);
if (rs[0].type == "page")
{
pageHierarchyArr.push(rs[0]);
// Do some additional work
}
id = rs[0].parentID;
}
}
As you might imagine, it doesn't work well. hierarchyArr gets an "undefined" pushed into it, and then the script crashes when it tries to check the type of rs[0].
When I try to set it up with a callback (db.getRSAndCallback(sql, callbackFunc), which I used for the earlier, non-interdependent queries just fine) it's worse: the inner loop takes off like crazy because id isn't getting updated; presumably because the loop is keeping the JavaScript interpreter so busy that it never actually fills rs. In some artificial testing where I forced the inner loop to break after a few iterations all the callbacks started coming through all at the end, after the loop finished.
The "standard" (such as it is right now) at http://dev.w3.org/html5/webdatabase/#synchronous-database-api seems to indicate that there is a synchronous API, but I haven't seen any sign of it on any WebKit based browsers.
Can anyone offer suggestions on how I might either, a. properly formulate these iterative, interdependent queries using callbacks or, b. somehow get the call to actually happen in a synchronous or apparently synchronous manner.
Many thanks in advance for anyone who takes a crack at this seemingly tricky little problem.
Naim
P.S. Here's the client's implementation of db.getRS for reference:
.
.
.
getRS: function(sql)
{
var output = [];
db.database.transaction(function(tx)
{
tx.executeSql(sql, [], function(tx,rs)
{
for(i = 0; i < rs.rows.length; i++)
{
output.push(rs.rows.item(i));
}
},
function(tx, error) { ... }
)});
return output;
},
.
.
.
I used callbacks and a closure to solve a similar problem, consider:
function getFolder(id, callback) {
var data = [];
ldb.transaction(function (tx) {
tx.executeSql('SELECT * FROM folders where id=?',
[id],
function (tx, results) {
if (results.rows && results.rows.length) {
for (i = 0; i < results.rows.length; i++) {
data.push(results.rows.item(i));
}
}
if (typeof(callback) == 'function')
callback(data);
},
function (tx, error) {
console.log(error);
});
});
}
In the continuation of this example, folder has a property parent to define it's relation to other folders. As does a document. The following will get you the path of a document using a closure (success):
function getDocPath(doc, callback) {
var path = [];
var parent = doc.parent;
var success = function(folder) {
var folder = folder[0];
parent = folder.parent;
path.push({'id':folder.id,'name':folder.name});
if (parent != "undefined")
getFolder(parent, success);
else
if ( typeof(callback) == 'function' ) callback(path.reverse());
}
getFolder(parent, success);
}
You could use callbacks with a closure to your stack of remaining queries. Or you could use recursion, passing the stack as parameters.