We Keep Coding

Dynamically create a TW object in IBM BPM

I am using IBM BPM 8.6
I have an input string as follows:
"\"RECORD_CONTACT\":\"Maram\" , \"DRUG\":\"Panadol\"
In a script on server side, I want to dynamically create a business object like this:
tw.local.recordContact = Maram;
tw.local.drug = Panadol;
How can I dynamically create the business object?

There are a few problems with your request. The first is that you are not creating a business object, you are creating variables. In IBM BPM the variables have to be declared at design time or you will get an error, so invoking attempting to call something like -
tw.local.myVariable = 'Bob';
Will throw an exception if tw.local.myVariable has not been declared. Base on your other question you asked here (link), I'm going to assume you actually have an ANY variable declared called "return" so that
tw.local.return.myVariable = 'Bob'
will work. Given that I based on Sven's answer I think something like the following will work (you will need to validate)
var str = "\"RECORD_CONTACT\":\"Maram\" , \"DRUG\":\"Panadol\"";
var jsonStr = "{" + str.replace(/\\\"/g,'\"') + "}";
var tempValue = JSON.parse(jsonStr);
var keyArray = Object.keys(tempValue);
var valueArray = Object.values(tempValue);
for(var keyCount=0; keyCount<keyArray.length; keyCount++{
var evalString = "tw.local.return."+keyArray[keyCount]+"="+valueArray[keyCount];
eval(evalString);
}
I'll note that doing this is a very bad idea as it would be very brittle code and that using eval() in this manner opens you up to all sorts of possible exploits. It will also fail badly if the value for one of the keys is not a simple type.
-Andrew Paier

One should know what you are going to do with dynamically created Business Objects (BO) to answer you better. Like a very generic way would be - creating JSON object instead of BO.
But if you want to stick with BO then this is only possible when you know all the BO structure (schema) beforehand during design time.
var str = "\"RECORD_CONTACT\":\"Maram\" , \"DRUG\":\"Panadol\"";
vat objArray = str.split("reg ex to split each object string")
foreach (obj in objArray ){
if(obj.indexOf( "RECORD_CONTACT")!=-1)
tw.local.recordContact = new tw.object.RECORD_CONTACT();
//below goes code get value of each attribute of BPM from string
}
else if(obj.indexOf( "DRUG")!=-1){
//similar code to create BO DRUG
}
Don't forget to create BO before using those :)

AutomationAnywhere + Javascript regular expression

I have an "Invalid character error" while using JS script to extract mails from text that I cannot handle for last 2 days.
I am getting text from web application by using Object cloning and passing it to variable which i will later pass to JS script.
And of course my JS script which I checked and it works:
var args = WScript.Arguments;
var pattern = \w+#\w+.\w;
var result = /pattern/.exec(args);
WScript.StdOut.WriteLine(result);

First and foremost lets break this down it modules and debug them.
.
First Module: Object cloning
Object Cloning is very good to build reliability and this reliability is acheived by careful selection of Properties and in your example you have selected Path,DOMXPath, HTML Tag
Its a good practice to identify the properties which are unique and therefore yield high accuracy and some of these properties depend on the context
For example in a login page some properties include:
Priority 1: Path, HTML ID, InnerText
Priority 2: DOMXPath, HTMLValue
You may chose to add properties that you think may be unique to your context
Does strResult give you the expected value ? If yes, lets proceed
Second Module: Run Script
Accepts 2 Parameters $strResult$ and $mail$
And of course my JS script which I checked and it works:
and you have confirmed the JS module also runs fine
If you have verified the results of the first 2 modules, I think there could be an Invalid character somewhere in the script, parameters, check the regular expressions used. Shouldn't the pattern be enclosed in string " " ??
=====================
EDIT:
I wanted to recreate the issue and give you the desired result but I do not know your intended input and output for Javascript. However to the best of my understanding of your javascript, I have compiled and executed this script in Automation Anywhere and works perfect.
JavaScript
var args = WScript.Arguments;
if (args.length > 0)
{
var val=0;
var str=args.item(0);
var ary = str.split(",");
//WScript.Echo(ary.length);
// for loop in case there are multiple parameters passed
for (var i=0; i < ary.length; i++)
{
//Takes the input passed as parameter
var input = (ary[i]);
// Uses the Match() Method to look for an email address in input string
var result = input.match(/\w+#\w+\.com/);
//returns the email address
}
WScript.StdOut.WriteLine(result);
}
OR
//Takes the input passed as parameter
var input = (ary[i]);
//Declares the pattern used
var pattern = /\w+#\w+\.com/
// Uses the Exec() Method to look for a match
var result = pattern.exec(input);
//returns the email address
Run Script
Input Parameter
Output Parameter

javascript how to insert new elements at different locations in page

I'm writing a chrome extension that will add helper text instructions/reminders to specific location in the "new order" form we use at work. I'm really new to coding (basically using this project as a way to learn). I've created something that works - but I'm convinced there's a more elegant solution that I just haven't been able to figure out.
var helpText = "this is the message"
var customAlert = makeAlert(helpText) //create html container for msg
function makerAlert(helpText){...} //createElem, add class/style, append children
I'm okay with that bit (above). But should i be storing information on each message in objects instead? why would/wouldn't i? what information would go in it?
function alertPlacer(customAlert){
var par = document.getElementsByClassName("class-name")[i];
var sib = par.childNodes[j];
par.insertBefore(customAlert, sib);
};
really struggling with this bit (above). I have actually made alertPlacer() functions for each message because i can't figure out how to create a function that will take different class name & index parameters. should i be breaking this up more? if i stored these bits of info in an object, would that be useful?
relevant info:
because the target locations are within a form, almost nothing has an "id" attribute. so i have to use getElementsByClassName & an index.
for each message, I know the target parent className & index and the child node index to "insert before".
i would like to stick with javascript-only solution.

functions can take multiple arguments:
function alertPlacer(customAlert,className,parIndex,childIndex){
var par = document.getElementsByClassName(className)[parIndex]; var sib = par.childNodes[childIndex];
par.insertBefore(customAlert, sib);
};
And you call your function like
alertPlacer(yourAlert,"class-name",6,9);

Wrap around javascript object to .Net usercontrol

I have a question i don't know how to start with...
We all know that DOM elements have their own events such as 'onChange'. Well a TextBox can fire onChange events and anyone can register to them.
I have a .NET usercontrol with 2 textboxes (country code + phone). The textboxes are validated using a .NET CustomValidator. I have server-side methods to handle the control such as 'Value' witch gives-me a unique string with cc+phone.
What about client side? ...
I would like to wrap around the 2 textboxes and the customvalidator in a "javascript control". I would love to create functions to work with my control as an unique entity.
What i would like to do was things like:
var x = document.getElementById('myControlId'); // where my control id was the wrap around
x.value("+351875647356); // witch will fill the two text boxes respectively.
Is this possible? What the way to follow?
Thank U All.

It's possible to do wnat you want by creating a javascript object ala a c#/vb.net 1class to represent your control. Your javascript instance would keep references to the individual elements and you would use custom methods to do the heavy lifting.
<script>
function MyUserControl (ccID, phoneID) {
// set references to elements
var _ccInput = document.getElementById(ccID);
var _phoneInput = document.getElementById(phoneID);
this.setValues = function (value) {
// TODO process your input here
// _ccInput.value = ... ;
// _phoneInput .value = ... ;
}
}
</script>
You could create a new instance of the object by generating the javascript from the server side.
string script = string.format("var myControl = new MyUserControl('{0}','{1}');",
ccTextbox.ClientID, phoneTextbox.ClientID);
ClientScript.RegisterStartupScript(this.GetType(), "uniqueKeyName", script, true);
after it's created on the client side, you can reference it from some javascript.
<script>
myControl.setValues("+351875647356");
</script>

Locating text and performing operation based on its existence

I'm trying to learn jQuery, but it's coming slowly as I really don't know any JavaScript.
My site is in VB.NET and I'm putting jQuery code on both my actual .ascx UserControl and in a separate file (something like myscripts.js). This is because I'm using webforms as I still don't know MVC well enough to implement it, so I have to get the clientID's on the page.
What I would like to do is the following:
Grab text from a textbox and make it all lowercase
Get the username from the login info. I've done this like so on my actual page:
var userName = "<%=Split(System.Web.HttpContext.Current.User.Identity.Name.ToLowerInvariant, '|')%>";
Check to see if the username is in the text. If it IS in the text, I want to set a variable to "false", othewise to true.
How do I do this?

I am completely ignorant of the ASP.NET side of it, but as far as jQuery and Javascript....
To get the value of a text field, you use the jQuery function val():
var value = $('#mytextbox').val();
To turn a string to lower case, you use the string method toLowerCase():
var value = $('#mytextbox').val().toLowerCase();
Since val() returns a string we can throw that at the end.
To check if a string is within another string, you use the string method indexOf():
var needle = 'Hello';
var haystack = 'Hello World';
var match = haystack.indexOf(needle); // -1 if no matches, 0 in this case

Another thing to remember is that ASP.NET renames all your control ID's. To access your controls in JavaScript, you should use the following in place of the Control ID <%= txtUserName.ClientID %>.
In jQuery, here is what my selector would look like for a textbox with the ID "txtUserName".
$('#<%= txtUserName.ClientID %>')
Enjoy,
Zach

var userName = "username as it comes out of your web app";
// stuff happens
var $myTextbox = $('#ID_of_textbox');
var userNameIsContained = $myTextbox.val().toLowerCase().indexOf(userName) >= 0;
Short explanation:
$('#ID_of_textbox') // fetches the jQuery object corresponding to your textbox
.val() // the jQuery function that gets the textbox value
.toLowerCase() // self explanatory
.indexOf() // returns the position of a string in a string (or -1)
See the JavaScript String object reference at w3schools.
Alternative (to check if the textbox value equals the username):
var userNameIsEqual = $myTextbox.val().toLowerCase() == userName;

The basics of JQuery are like so: Find a list of dom elements, and perform actions on them.
In your case, you should start off by finding the dom element that is your testbox. For example's sake, we'll choose $('#userName'). The selector # means "id" and together with the name "userName" it finds all elements with the id of "userName". (Ids on a page should be unique if you're following best practices.)
Once you have that list (in this case, a list of one element), you can ask it what the value is.
$('#userName').val()
This gets you the value of the value="" attribute of the input tag.
You can then assign it to a variable and use standard javascript String functions to do the rest!

function checkLogin(userName){
return $('#mytextbox').val().toLowerCase() == userName
}

if ($("#textBoxID").val()) != "") { /*Do stuff*/ }