Pass Blue Prism AD01 Exam in First Attempt Easily

Introduction

1. Blue Prism Developer Exam AD01 Overview

All right, let's take a brief overview of the Blueprint. The voter exam is one. The exam will be made up of 60 questions. You'll have 60 minutes to complete the exam. If you're taking it in a language other than English, you may be given an extra 30 minutes to complete the exam. And finally, it has a 70% pass mark. Next up, what's covered in the exam? These topics are all covered in the exam; I'm not going to go into them. Because throughout this course we'll have lectures on each, we'll actually implement each of them in a process or object. We'll understand what the exam will test for you in these areas, and you'll also get a cheat sheet for each of these areas. And finally, after taking this course, I recommend going over these documents on Blue Prism. They're worth reading and attached to the course, and they cover exception handling, work queues, developing a best practise object layout, and finally, environment and session variables.

Hide

Exam Prep

1. Data Types

In the Blue Prism Exam, you'll be tested on seven tap functions. That is Concatenate, starts with, ends withmid left, right and finally Trend. So I'm going to run you over each of them in Blueprint, and then I'll let you know how the exam will try and catch you out when they're testing your knowledge in this area. And finally, there will be a one-pager for you to simply refer back to at the end of the course. And I'll compile all of them for each section into a simple, nice document for you. So let's get into it. So we'll add a calculation stage to our process. We're going to call this Concatenate. We're going to store the results and also create a data stage called "Concatenate Click" to automatically create a data item. And what this is going to do is thisis going to join two text items together. So remember to create a text item in an expression that is surrounded by quotation marks. So in this expression, we have hello and load, which are two separate text items. So if we want to join them, simply use anand sign which will concatenate which is joining them together. So far, okay. Drag that data item down the page to our outputs. Join the startup with concatenation. And now let's simply run that and see what happens. So we can see that Hello World has been joined together. We don't have a space between the two, so make sure you include a space. The exam might try and catch you out on that, so just be aware of that. They might ask you what the result of the concatenate function is. So make sure you check if there is aspace in there or not because that will affectwhen you join two tech times together, it wouldnot automatically put in a space for you. We've put one in here manually. And what you can also do is actually add a data item or data stage to that. So if we use this name, we're going to call the data type text. Put your name as the initial value. So mine is Andrew. Hit the reset button on your process. Then in concatenate, instead of saying, "well," we're going to drag the data item up into our expression. So introduce yourself and state your name. We hit OK. Now run that from the start. We can see the concatenated result is Hello Andrew. So the next stage starts with the Dragon calculation stage in our process. Call it "Start with." and all of these are going to be under the tax functions on the left here. And we're going to select "Starts with," and the function detail window here outlines a nice description of how it works and also gives you an example. So, for example, if we say example and do the text, the starter will hit paste. So we've got our first one. So we begin by determining whether or not example begins with the text pole, which it does not. A quick way to actually check this is to hit "evaluate expression." We can see that this result equals false. If we do exa, then we hit Evaluate expression. We can see that it is correct because the example begins with exa. But if we put Ez in front of XA, we can see that the result is false. So let's change it back to PO. We're going to store the result in an item called "Start with." To create aflag, use that little array, and it will store true or false. Connect example to PLA, and Am will drag another calculation stage called "end with like starts with." It's just checking the end of a string instead of seeing if it ends with certain characters, so you can find it under the functions. We'll type in an example. This time, we'll check against PLA and save the result in an end with. Hit that button to create it, then hit okay. Now let's connect those up to our process, and let's run both of them. So for starts with we're checking ifexample starts with PLA and ends withwe're checking example ends with PLA. So we have false force; starts with and ends with is true because the example ends with PLA. Next up are our midleft and midright calculations. So, to begin, let's check in on Midway. This is the most tricky one because you will need to actually remember the function. So if I drag that here, let's drag out a window so we can see the function details. We're going to include a text message greeting. The start point will be two. So this is the function described in the text here. Hello, we're going to start with character two, which is E and length. We're going to return the next three characters, which include character two. So this will be ell. So hit paste, store the result in mid, and hit that eraser to create that text data stage. Drag that on to our process. Finally, let's run it. And this will receive the text message. So remember the midterm. We're grabbing the text here, which is "hello," we're starting from position two, which is the "E," and we're getting the next three characters, which is "ell." Just a heads up: in any exam, they might test you on this in a multiple-choice question with four answers. Two of them will be completely wrong; however, two might look right. And the way they'll do that is they'll give you the function, say hi, and ask you to tell you what the result will be. One of the answers might be a ll, which seems correct. Then another one might be "double L." And you could think the double L might sound correct because we take the text as "hello." We might think two is the number of characters and three is the starting position. So if we take three as a starting position, then L here and two characters would result in Ll. So you think, Hey, that does sound right. I think that is the order in which this expression is used in any exam. This useful function detail pain is not available here. So I recommend you make sure you download the cheat sheet for this course, create some flashcards, and make sure you're familiar with how these expressions actually work. Because, as you are aware, mid will return some sort of textout of text, but you may not recall the order in which these two are in, whether the start position or the length. So you will need to remember that because the exam will try to trick you on those items. It's just like that. So keep that in mind when you're going into the exam. So let's hit OK, and let's jump to the left and right. So first of all, I'm going to add to the left. So repeat the text: function, dofunctions, left, and hello. and we'll do length too. And we'll score the results in the left column. So this is going to get the leftmost two characters, which is a length. So this will be it, and paste that in for the expression. Hit okay. And we'll just copy that stage because it's exactly the same for right, except you can edit the expression directly to be right. And this will get the rightmost two characters, which is Lo. We'll save the results in the appropriate folder. So enter it correctly, and we'll call this calculation stage correctly. Hit okay. Let's connect a link to both and just keep running up until it returns what we expect. So, for hello returns, left of a length of two; right of a length of two; and rightmost to text characters, which are L and O. And the final item is a trim data stage. So we're going to call this Trim. I'm going to store the results in trim. Let's create that. Now, before I actually show you what the function does, let's say "Hello World" with a bunch of spaces in between it.Hit okay, then connect that up to Trim. Connect that up. Let's run that We can see if we store that in the current value. If we highlight it, we've got all our spaces from start to finish, and we've got a yellow world surrounded by white spaces, which we don't want. Trim works by adding trim to our expression surrounded by brackets, then hitting okay to run that trim stage again. That will get rid of all the white space around the left and right. So if we can see, we can see the current value. Use controllate to select it all. We've got no white space on the left or right. This is extremely useful in real life because you frequently encounter people entering a username or extracting an email. You might be working with an Excel sheet that has a bunch of white space in the text for some reason, and that might be messing up your formulas, or it might be messing up your concatenate tech function to actually build up some sort of tech strings, such as an email response to a customer. So that trim function is super handy for removing the white space characters from the beginning and the end of a string. It won't remove any from the middle, so it's only from the beginning and end. So hit. Okay, let's save that. And you can actually download this process from all of them for discussion in this lecture or at the end. And I'm going to jump onto a cheat sheet fortext functions and each group that I cover will haveits own cheat sheet and they'll all be put intoone PDF at the end of the course. So it'll be super handy reference material, along with some quizzes you can actually run yourself, that will prepare you for the exam. Okay, so just recapping off everything around text functions that could be in a Blue Prism exam. First of all, this is a concatenate, whereas you use the ampersand to join two things together. Then you have the starts with an end with a text on the left side and a text on the right side. You want to check if the first bit of text you pass in either starts with it or ends with it. Next up, you've got the middle. And remember the ordering for this function. You pass in the text, and you pass in the starting point. Remember that number one will be the first character, number two will be the second character, and so on. And the third parameter here is length. So, how many text bits do you want to get from the start? So, remember mid-hello from the beginning? 2 starts at the first E in terms of three characters. So ell. Then you've got the left and right; they're pretty self explanatory.You pass in the text and the length you want to get either from the left side or the right side of a piece of text. Finally, the trim option will remove any whitespace from the beginning or end of the text, rather than in the middle. So this is a cheat sheet for the text functions.

2. Date Functions

The next group of functions that you'll be tested on in the exam are data types. How do you convert something from a number to text? How do you convert a data text? How do you convert a time span to days, hours, and minutes? So let's cover this off. Now, it's super easy. There are a number of built-in functions in Blue Prism. So first of all, we're going to do a data item. We'll call this number a data type number. We'll do the initial value of ten. Then we're going to do another data item called text. Data type as text, do the initial value as hello. Then we're going to check on another data item, and we're going to make this a time span. You might not have used this before, but it stores lengths of time in days, hours, minutes, and seconds. So we'll do one day, and this time span will be called hit. Now, once you've got this set up, let's look at how we convert these two different data types. So first of all, let's look at how we can convert some text to a number. So our text up here is low. So what we'll actually do is put the text inside the calculation stage in the process. So we're going to do the conversion. You'll say a two-digit number, and you'll simply put in a text you want to convert to a number. Let's hit paste. If you go evaluate expression, it saysresult equals ten, which is a number. However, if we try to put "hello" in there, it will actually return an error because the value we supplytext is too low to be converted to a number. So remember that for dysfunction. If you don't provide it with a proper number, it will throw an error. We're going to call this two numbers, and we'll also store the results in an array. Hit the eraser stage to create it. Hit okay. And of course, when we run that stage from the start of the process, we can see this is actually stored as a number data type now with a value of tan instead of a text value. So how would we convert this input number up here to an actual string or text? Let's drudge through another calculation stage. This will be referred to as two texts. And what you would expect is that there would be a two-text conversion function, but there isn't any. There are ten. How do we convert this to text? Well, it's quite simple. If you want to convert anything to text, simply use an ampersand and an empty text item here at the very end of the expression, and then it will convert the whole thing to text. So, if we evaluate and test, we get a text data item type. So we go ahead and save the results as text. Hit that eraser. Hit okay. Connect that up. Just make sure that runs through. All right, we can see here. When we open up to text, we have "ten" as a text data type. Okay, we're just going to drag one more data item. We're going to call it a state text. We're going to create a data type text and enter, say, the 26th of the 62000 as the initial value. Okay, hit reset. And we are now going to look at converting this date text to an actual date item. We'll do this to date if we drag another calculation stage. We do date. Let's go to conversion and say we've got the date. You inserted some text there. So we're dragging a date text, hitting paste, copy, saving the result in two days, and then hitting the eraser again to create it automatically. And then let's see what happens. As the process runs through, you can see that it successfully converted our text for the 27th of the 60,000 to an actual date data value. Let's copy and paste the calculation stage. There's also a similar function called date time. What the date time does is that we drag the date text in, hit paste, and store the results in two day time.Don't forget to update that. Hit the eraser to automatically create it. Let's link that up. Now you'll notice it will actually store the time as well. So since we didn't provide any in our actual text, it will default to 12:00 A.M. But if we had written this text today at 07:00 p.m., it would actually be stored in the date/time data type as 07:00 p.m. Okay. Finally, there is the time span function. We're going to convert this one day to an hour's worth of minutes. You can also convert it to the days that are already in the calendar. So we don't need to do that, but I'll show you how to use it. So first of all, we'll simply do 2 hours. We drag out the time span here. Hit paste. We'll call this calculation stage 2 hours.Copy, paste that result, and hit that eraser to automatically create it. It's going to convert to hours, whatever the time span is here. So let's press okay and connect that up. We do that. You can clearly see that one day equals 24 hours. Similarly, if we copy this twice, you can also do two days, which will return just whatever days are in this time span. And you can just type in the expression or drop down that conversion to see all the options. We're going to store this result for two days. Instead, hit those arrays, create a name calculation stage, wait two days, and we're going to do the same thing in two minutes. So if we do two minutes, call this calculation stage two minutes. And finally, in two minutes, you can already guess what they're going to do. It does the same thing. I forgot to hit that arrays to create the data item. Make sure you hit that. So yeah, as I mentioned, it will do the same thing as the 2 hours does, except convert it to days and minutes. So we can see two days and two minutes. So we can see obviously there's one day in one day inour time span, there's 14 to 40 minutes in one day. We could also say one day, let's say 15 hours, and reset our process so we can see the timespan has updated to one day and 15 hours hit run.You can see we're running through the conversions—there are 39 hours in one day, 15 hours. It looks like it will round down. And we've also got the total minutes in one day, 15 hours. So you won't need to worry about the routing; you just need to worry about the functions in the Blueprint Developer One exam. So let's go on to the cheat sheet. All right, this is a cheatsheet for data type conversion functions. We've already done them. So I'm just going to run over three key factors to look out for in the Blue Prison Exam. The first one is converting a value to text. There is no function; you simply add an ampersand and an empty quotation at the end of an expression to convert something to text. So, for example, 200 and an empty string will output 200 as text. A second key point to remember is that when you're using two-day time, if no time is provided in the text, it will default to 12:00 a.m. So in this example, today, the 26th of January, it will default to 12:00 a.m. Finally, the two days, hours andminutes will always round down. So say you give the two-day function 1.5 days, and it's going to round down to one day. And remember, it takes in the time span, whichis in the format days colon, hours colon min. So, three key factors to remember, and this will be available for download along with the rest of the cheap seats and actual source code for the Blueprint processes used in this example.

3. Debugging

Alright, now it's time to take a look at the date functions that you'll see in the exam. That will be date add, date diff, format, and date make for today and now. So first of all, you'll notice I've already gotten two inputs here. I've got date one of the 27th of 2019, and day two is a month later, which is the 27th of the 11th month of 2019. So the first two functions we are going to look at are date add, whereby we add some time to the state, and date diff, which will give us a difference between two days in a certain interval. So first of all, let's take a look at the date added. So let's do a calculation. Let us refer to it as date + add. You'll notice the date we have listed, and let's just run over how this function works because it's quite important for this one. Date add and date div define the order and how the interval works. So first of all, we'll start at the bottom. So give the date to which we want to add some sort of time to.So we'll give it date one, and let's just say we want to add one week to it. How do we add one week? We have this number, an interval. It's a bit confusing, but the number—let's just go with one because we want to add another week—and the intervaluses numbers or codes of what sort of time to add are clear. So number one corresponds to a week, zero corresponds to a year, four corresponds to a quarter, five corresponds to a month, and so on. So I've got a handy dandy table up on the screen right now, which actually tells us what the interval means for both date add and date div. It's very important to memorise this at the end when we go to the cheat sheet. I'll share with you a few techniques that I use that really help solidify this, because they will test you on this. Sadly, it's just one of those things you have to remember, unfortunately.So the interval can be found in documentation; otherwise, just use a table on screen. So we're going to paste that in. We're going to store the results in days plus one week plus one week. Hit the arrays to create that date item. Let's turn that on and see what happens. adding a week to our date. So we can see here that we've got the third of the 11th, 2019, which is a week after the 27th of the 10th, 2019. All right, let's move on to the next one, which is quite similar, which is called date diff. And this is going to give us the difference between two dates. So at the calculation stage, we'll link it up. We're going to call this date diff. In the functions under date, we have this date if, where you'll notice the structure is almost the same as date add, except we have the interval we want to get. So remember, one is for a week. We'll use that for now. Once again, the tables are up on screen for reference. And this will be at the end of this lecture. In the cheat sheet, we're going to calculate the difference between day one and date two. We're going to paste that inand we'll call this date difference. Hit that eraser icon to create it. Now let's run it. Yes, we know there is four weeks difference between the27th of the 10th and the 27th 11th 2019. Let's just say we want to find the difference in days. If we return to our date calculation, if the interval is nine to get days to evaluate, we can quickly test that we get a 31-day difference. If we want to get the difference in months, thatinterval code is number five, go test it's one. So let's change it back to week. So once a week we evaluate tests; that's four hits, okay? And we're done with that. So out of everything, those are some of the most tricky ones to get tested on in the exam out of everything.Just because you have to remember the interval and the order in which the actual function is set up, because I'll shade for the ad. The first part is the interval; the second part is the number you want to get, which is, say, one week. You want to add one week to the state. At the end, they'll tell you that the same function will say the week and the amount you want to add have been switched around to fool you into thinking the result is something else. So let's close that Next up, we'll show you format Date. Format Date. This one. If we go by dates, go by format date. This will actually format a date in a given format. so that it can be formatted. So for the date to be formatted, we'll take in day one. And let's just say we only want to get a few days out of it. You can use DD to do things like getpaste go evaluate expression go test. It just returns the date, which is 2000. So you might have used similar functions in Excel before the format date. It's the same with Blueprint, and you just use the same keywords to actually extract the exact format type you want. So if we do, for example, we'll add that to ourexpression If you hit the evaluate test, you can see we've got the number, but if we do two more Ms and hit the evaluate expression test, we actually get the month name. So that's how you can take a date in Blueprint and format it in whatever format you want. There'll be a cheat sheet on the screen now on what the actual formats you can use are. and also at the end of the lecture in that slide, which you will get as a download. So we'll save this result in format state. Hit that eraser icon to create it. Okay, we're just going to get the month over with. Let's do that later. Once you say the next two functions, the next stop is "make date." And I called it the "make dates" stage of a calculation stage.Going to go to date. We're having a make-day here. And what this does is actually make an accurate data item for a given day, month, and year. So day of the month, let's say the 15th. It's the 1st of January. Number one is the 1 January, no worry. Two is February, and so on. Keep this warning in mind: do not use a two-digit year; instead, use a full year. So let's just say in the year 2000, hitting post makes a day out of the day, which is the 15th month, which is the first in a year, which is 2000. We're going to store this results in May date. Hit that little eraser icon to create it. Let's look it up and check it out. Working now. So, if we open up the data item, we can see that we have a data type date, which is the 15th of the first month of the year, $2,000 So just keep a note of this function: it takes in numbers in the actual expression. So if you have, say, text from the year 2000, you'll need to convert that text to a number and then stick it in the make date" function to actually make a date out of a given text item. So it only takes numbers here. Keep that in mind. Number two, the ordering is important. It's a day, month, and year. They'll try and switch around in orderingany exam and catch you out. So that's another key thing to remember. So let's hit, okay? And the final two items are today and now for dates. So let's just do them together. We go under date; let's say we get today. We're going to cause calculations today. We'll store the results today. Hit that eraser icon to create it. And then let's also make the present one. And then I'll cover the difference between them in a moment. So let's call this stage of the calculation over now. Under function, we'll take in the now function, store the results in now, and hit the array icon to create it. Connect those to our process. Let's run it so you can clearly see the difference here. Today just simply gets the day, month, and year of today's date. Now, in addition to the day, month, and year, we'll get you the time, which currently is quoted as twelve on my actual system. Okay, we just got a few cheat sheets to go over, which are the date functions, what the intervals actually mean, and how to format a date. So first of all, the first two functions are recall with date add and date difference. They both start with a specified interval, which is a code word for week, month, year, and so on. For date add, you pass in the number and the date you want to add. So, for example, you might say the interval is in a week. You add four weeks to a specified date for date.If you pass in the interval, then the start date and end date will return the difference between two dates for you. The next function we've got is "format date," where you pass in a date and the format string you want to format it in. So in this example, we've got a format date of January 27 of the 10th, 2019, and we're forming it in the Mmstring, which will output the value ten. Then we have to make a date whereby you pass in a day, month, and year, and it will make an actual date data item type. And finally, you have today. Simply today will return the current date as a datevalue, and now we will return the current date as a datetime value, which will include the current time of day. Next up, we've got the intervals and what they mean. So remember, you need this for date addition and date deletion. This is quite a lot to remember, so I've highlighted all the ones in bold and with a blue background colour that will likely be in the exam. The rest, I didn't see in the exam today, and I have a list to remember on the right side. So this is my simple memorization technique. It is entirely up to you how you remember it. So zero is for a year; one is for a full week. Four means four quarters in a year. Number five is for a month, which is after a quarter. And number nine means the ninth day of Christmas. So that's how I remembered them. I simply had flash cards and I just record thesememorization techniques to actually get them in my head becausethe task might say, hey, we're adding to a certaindate with say interval of one of number two, whichmeans we're adding two weeks to the state. And you'll notice that, for date addition, some of these intervals don't actually do anything. We've got an NA, whereas date divyou actually have some more functions. So just remember those for the test, okay? And now finally, we have our date function formats. Remember that these are case-sensitive, and I've got the most common examples here, so you can use these to remember what they'll actually do. The first is formatted as 2 November 2016 with a capital D. This will output the 2 November 2016. If we pass in a lower case "D" for the actual formatting, it will be just a second "11" of 2016. If we pass in an uppercase M, it will be due in November. And the next few are going tobe a few key patterns to remember. First of all, is the format date just DDD? If you just see DDD anywhere, this will put the actual day in the string or text. So this is Wednesday. Next up, if we pass in Y Y in lowercase D), it will be 2016 of the actual month, which is the 11th of two. So, actually, put a zero before the actual day. And you see these dashes here? We would see aesthetics if we included a forward slash in our output. So you can put whatever you want in the middle. If it doesn't match a code, it will actually put that in the output. So you can replace those dashes with a four-slash comma, and that would actually reflect in the output. In the final example we've got, and you can see here, we've got Nov. So the short month description, two comma, 2016. So there's a link here that is super-technically detailed on how this actually works. I recommend just playing around with it in Blueprints and getting comfortable with the concept. They likely won't test you in detail at all, theymight just test you on, say, if we pass inYY and then DD, what would the output be? And as long as you remember, Y is generally for year, M is for month, and D is for Dave, but the keynote for the exam is that anything but Y means a year, M means a month, and D means a day. And if they give you an example question on this, it should be pretty obvious as to what the actual output will look like. But once again, go to Blue Prism, test around with this format date, function with some of these examples, and change it up and see what happens with the output, and you'll get comfortable.

4. Debugging Cheatsheet

Alright, what do you need to cover in debugging for the exam? Let's run over it now. First of all, I've got a basic process set up here. I've got pages one and two connected up to the end stage.and I've got a simple exception recovery where it will recover and throw an exception. For now, let's just focus on page one and page two. Let's go into them. You can see they're empty. Page one has nothing in it; so does page two. And you can download this starter kit for this lecture or for the course. And you'll also be able to download the completed one at the end. The first part of the debugging is the step, step over, and step out debugging options. So let's have a look at what each does. First of all, step what this will do. This will step through the process one stage at a time. And if it's set on a page, you can see that this yellow indicates that the next stage is about to begin. If we click on step, it's actually going togo into that page and we can step throughit likewise with page two until the end. And we don't want to do anything if we just say we'll add a calculation stage but let's connect it up. Let's reset the process, hit step one at a time, and see how it goes for each stage one at a time. That stage of the calculation had nothing in it. But if we do say hello, store the results. in "hello," click here to create that data item. Let's go back to the start. Let's step over each item and see we go calculation. It goes to an end. So that's what Step does. It will go into every page or stage of the process. Now let's look at the next one. Step over. What this does, if you hit step over and it's ona page, it will simply execute everything in that page one. Move onto page two. So if we had 50 stages inside this page here, it would automatically go through all of them for you and then pause at the next stage on the current page you're debugging. So if we go over again, you can see it gets to page two and so on. So the main thing to remember about step-over is that if it's on a page, it will completely bypass that page. It will execute it, but it will not do so one by one as Step does. If we go back to the beginning, this is the final one. So go one step, then step into page one, then step out. What this does is, if you're debugging and you're inside a page and you step out, it's going to automatically jump back up to the parent page. So it's going to jump out into the parent page, executing everything you have left on that page. So remember these three key debugging items. It might ask you: "You've got a process, you hit step. Where does it go next?" Alternatively, you must step over or out. Next up is a breakpoint. If you right click on any stage and select breakpoint, it's going to create a red box around the stage. Then if we go to the start of the process and hit run, it's going to run the hot process until that breakpoint. It will tell you that a breakpoint has been reached, and it will stop running that process completely until you hit the run button again. And then the process will keep on running until it hits another break point or finishes the process. Finally, the recover and resume. We got this bad process page here, so let's jump into it now, and I'll run over it. We've got the start stage, we've got the number calculation stage, and you might notice there's actually something wrong with it. We're trying to convert the text ABC to a number and store the results in this number output here. Of course, that's not going to work at all, so that's going to throw an error. So how do we recover from that error? This is where the recovery stage comes in handy. What happens is if you have this box here at any stage where it has a recovery stage in that box, it will automatically jump to this recovery stage here and then try and actually recover the process. Then we've got this retry loop here. We have this data item called Retry account, which is set to one by default. If that's less than three, we're goingto increment that Retry account by one. So add one to it, then resume and try to convert to a number again. And that's going to repeat until this retrycount is called three times, in which case it's going to throw an exception. So keep in mind that this "retry account," in which you retry apart ten times, is a standard recommended blueprint process. So they will likely ask you about this. So remember, when you throw an exception and recover, go into the retry count. If it's three times, throw an exception. Otherwise, keep on retrying to do yoursteps in your process where you can. And finally, you need to make sure you connect up to a resume stage when you go to continue running your process. Because what happens is that between this recovery here and this resume, it's in a special debug mode where Blue Prism holds the exception details. So if you have another exception during this recover and resume area, it doesn't know what to do. It's tried to hold the details of the last exception, and it only has room in the box to hold the details of that one exception. So it can't hold two, three, four, and so on. It can only hold one. And a way to empty that box is by using a resume stage. It will empty it and go that way. If there's another exception later on, blueprintscan store it in that box. It has room once again. So let's take a look at this process here. So I'm going to right click and go start, hit Run. You can see it throws an exception and increments the retry count by one. So Retry account is at two. It's going to try it again, and it throws an exception. But you'll notice the exception is empty here. So why is that? Well, in exceptions, you can actually customise the message, so we could say "System Exception" and we could name it "Bad Calculation Stage." Stage four is hit. Okay, I'm just going to move that Retry account aside for now, just while I'm showing you the exception testing. We'll connect that back up at the end, so we've reset the process. So go back to the start stage and hit Run. We can see that it attempts to recover, and we have the system exception we set here as well as bar calculations, a stateful number that allows you to set customised accepts or messages. But what if you wanted to actually get a bit more detail on the error Blue Prism ran into? Well, you can use this. Preserve the type and detail of the current exception here. So tick that box, okay? Then hit reset for our process. Let's run it again. And now we can see here, this is theinternal error that Blue Prism got so it givesus more details value to evaluate expression two numberABC unable to convert to a number. So it's a lot more detailed, but it gives you better insight as to what actually went wrong. So I recommend using it unless you have areason to provide a customised exception error message. Now, finally, we're going to connect the RetryStage back up, and we're going to run over the concept of bubbling up. So let's connect that back up, hit reset for our process, go back to the main page, and what we're going to do, we're going to add a new page here, go create a reference to an existing page, select Bad Process, and hit Finish. We're just going to connect StartUp to BadProcess From start to finish, let's see what happens when that bad process freezes. So we're going to go start, we're going to speed up thedebug speed a bit so it runs a bit faster for us. Hit Run. Let's let it run, and I'm going to speed it up, and we'll see what happens with the exception happens.Okay, what you'll have noticed is the exceptionactually broke out to the main page fromthe sub page called Bad Process. This concept is called "bubbling up." That's whereby in a subpage, such as Bad Process, we find an exception, and in the main page we have this block here, and in the block we have a recovery stage. So any exceptions that are caught within any of these stages or the subpages will actually get caught up and come back to the main page. Blue Prism recommends bubbling up any exceptions that are all to the main page, so you can gracefully handle them whether you reset your entire application. So you might try and submit an order again or you might send an email to the business owner of a given process saying, "Hey, your process didn't work." Here are the exception details. Can you please manually investigate that? It's a recommended pattern by Blue Prism, and they will ask you about it. They'll have some visual examples with subpages such as Bad Process, and I'll ask, if an exception happens here, where will it bubble up to? And the answer is "main page." So keep that in mind. The final tip I want you to remember is to set the process in debug mode. There's a break on exceptions option if we use that and run our process. What this means is that as soon as we hit an exception, it will automatically cause a break point in error. We can close this, and then we can step through here one at a time to actually debug and figure out what's happening. The exception, and you'll notice the exception details are at the bottom of the logs, so you know exactly what happened. That way, if you're testing a process end-to-end and something wrong happens during development, it will automatically throw a breakpoint on an exception, and you can look at the error details straight away. So let's jump into a cheat sheet of what you need to remember for.

Hide