Macro Recording and Automating Windows with Macro Scheduler – Tips & News
We seem to be having some problems with our voicemail system at the moment. Messages are being truncated. If you’ve left a voicemail for us during the last week or so and haven’t heard back from us that will be the reason. If so please send us an email with your contact details and we’ll get back to you right away.
Update 20/04/06 – Problem solved!
By Bruce Davis.
My name is Bruce Davis. I am a Customer Relationship Manager at Eastman Kodak supporting our Electronics and Warehouse Club channels. My organization is responsible for managing a wide variety of styles of customer orders. Some of the orders come in through electronic methods and require little manual intervention. However, our high value equipment business orders cannot be sent electronically because they require so many manual overrides. The Kodak sales teams have much of the basic order data captured in spreadsheets. Other override information is a constant value across all orders for a sales deal. The lead time between our receiving the spreadsheets and when orders must be entered is usually short. The challenge was to quickly and accurately transfer the data from spreadsheet to our sales order system and to add the constant data to the orders.
We receive Excel spreadsheets from our sales team with hundreds of rows representing sales orders. The orders have similar patterns of the same catalog number, quantities, expected delivery dates, etc. for many different ship to addresses. We use Macro Scheduler to copy and paste the data from Excel into our order management system. There are many other overrides and text messages that have to be done on each order. We built a control macro that loops through several modular macros with a counter set to stop after the last row is entered.
Prior to using Macro Scheduler, we attempted to enter these orders manually. This was very time consuming and error prone. Since our order managers are often interrupted by other phone calls and tasks, they would sometimes forget which row of data they were working on or which overrides were done on a given order.
We have entered thousands of orders using Macro Scheduler. The tool can enter orders faster than a person, particularly when we have two computers running the macros. This frees up our order managers’ time so they can attend meetings, take calls and work on other tasks. More importantly, the orders are entered much more accurately than when done by a person. In many cases, we could not have met our shipping deadlines if we didn’t have the tool entering orders.
Over time we have learned to use the tool more wisely. Common steps done across different roll outs are built into re-usable modules. We only need to create or modify a few macros for each roll out. The rest of the steps are performed by having the control program run the common macros.
Someone sparked an interesting thread in the Macro Scheduler forums recently when they asked if it was possible to activate a macro by detecting a click on a specific area of the screen. They wanted to know if you could define a portion of the screen so that if a click is detected in that portion of screen a macro could be fired.
Macro Scheduler can do most things and I very rarely respond negatively by saying it can’t be done. Even when I can’t see a solution immediately, I usually have a good think about it, try a few ideas and come back with a solution. In this case I missed a nice solution and replied saying it isn’t yet possible. I don’t know what I was thinking! This sparked a few forum regulars to suggest some solutions ranging from maximised custom dialogs with image maps, to active desktop configurations. Then Dick suggested using the OnEvent function and VK1 – the virtual key code for a left mouse click. Of course! I kicked myself. Why didn’t I think of that?
So I set about writing an example to test the theory. The script actually turns out to be quite simple and elegant. Here it is:
// Set Bounds Here
Let>X1=440
Let>X2=630
Let>Y1=40
Let>Y2=200
// Only detect clicks on Desktop, or Anywhere?
Let>OnlyDesktop=1
OnEvent>KEY_DOWN,VK1,0,MouseClick
Label>MainLoop
Wait>0.2
Goto>MainLoop
SRT>MouseClick
GetActiveWindow>title,xact,yact
// OnlyDesktop/all window check
Let>OkToContinue=FALSE
If>OnlyDesktop=1
If>title=Program Manager
Let>OkToContinue=TRUE
Endif
Else
Let>OkToContinue=TRUE
Endif
If>OkToContinue=TRUE
// Check Cursor is Within Bounds
GetCursorPos>X,Y
If>{((%X% >= %X1%) AND (%X% <= %X2%)) AND ((%Y% >= %Y1%) AND (%Y% <= %Y2%))}
// Replace next line with call to macro using Macro> command or call to a subroutine
MessageModal>Hello
// Macro>your_macro
// Pass LClick on?
// LClick
Endif
Endif
END>MouseClick
This script sets up a continuous loop and an OnEvent function which responds to the key down event of virtual keycode 1, which happens to be a mouse left click. Yes, perhaps because this is considered a keydown event and uses a virtual key code it is not immediately obvious that you can detect a mouse click. The subroutine that is called when the event occurs checks which window is active and gets the mouse position. If the active window is one in which we should check for mouse clicks (in this case the script works either only on the desktop or on any window) and if the mouse cursor position is within the defined bounds, our routine is called. In this case the event is just a message box, but that can be replaced with a call to another macro or an external application, a subroutine call, or any code you like. If necessary the left click can then be passed on to the underlying window as it was intended – but this should only be necessary if your routine steals focus or does something to prevent the underlying window getting the click. The code could be modified to make it work only against a specific window.
This macro needs to run all the while you want the screen clicks to be detected. If that is all the time then you could set the macro to run on startup, or compile it to an executable and have that run on startup. Or maybe you only need to start the script for a short while when the screen click detection is required.
I’ve also posted the code and an explanation to the forum here. I’ve also added this to Scripts & Tips.
I’ve only recently started using iTunes. Funnily enough I had to use it to help a Macro Scheduler customer, but that’s another story. I was really impressed with how easy it is to buy tracks and download them with iTunes. So I bought a few albums and the first thing I did was burn them to CDs for use at home and in the car. But I was particularly looking forward to copying the tracks to my HP iPaq so that I could listen to them when I’m travelling. Well it was at that point when I found out that iTunes uses a format called m4p which includes some form of copy protection. m4p files won’t work with Windows Media Player on Windows Mobile and iTunes wouldn’t let me copy them to MP3. Apparently you can play them on an iPod of course! Should have figured.
Well, I had just burnt the files to CD, and I can rip CDs with iTunes can’t I? So all I did was use iTunes to import the tracks on the CD to MP3 files. Voila. Then I could copy them to the iPaq.
So, let’s recap shall we? iTunes won’t let you convert an m4p file to an mp3 directly, but it will let you burn an m4p file to CD and then rip the CD to mp3 files. So, er, what exactly is the point of the copy protection then? What exactly is the reason for the m4p files? I can only assume it’s to make using anything other than an iPod a little more complicated. Because it sure isn’t preventing anyone from making copies – the DRM (Digital Rights Management) system is ineffective.
I seem to remember Microsoft got into trouble over pre-installing Internet Explorer and making it difficult for vendors to provide alternatives. Seems Apple is doing something similar here by making you jump through extra hoops to get an iTunes file onto anything other than an iPod. Or maybe I’m just being too cynical and overreacting.
I really wish the record industry & Apple etc, would just get a grip and trust people. Those that have no intention of paying for something won’t pay for it whatever you do. And the only DRM and copy protection systems I’ve seen cause more hassle than good. Sooner or later someone will find a way round them. If all it ends up doing is annoying the genuine customer then really what is the point?
ZDNet Executive Editor David Berlind prefers to call DRM C.R.A.P (Content, Restriction, Annulment, and Protection). Watch his video to see why he thinks DRM is a load of c.r.a.p.
I stumbled across this transcript of a support issue today. It’s somewhat amusing but you can also see how frustrating it must have been for both parties. We get our fair share of awkward support issues, but so far, touch wood, nothing quite as agonising as this one!
However, I think there’s a reason for that. In situations like this when it becomes clear that we aren’t being understood, or the issue is just turning into a fruitless game of email tennis, we pick up the phone. I’m quite convinced that the issue in the transcript above could have been nipped in the bud in a matter of minutes had the support guy picked up the phone and politely explained why it wasn’t their problem and what the ‘customer’ had to do. There are some things that are much more efficiently dealt with on the phone. Someone suggests this in the comments at the end of the transcript. But because the software in question is free, the support guy claims he should not have had to provide phone support and that he shouldn’t have to pay for the phone call.
Firstly I think this highlights one of the big problems facing open source or free software compared to commercial software. Had the software been commercial the support guy would have thought nothing of picking up the phone and sorting the issue out straight away. Companies that use open source software need to think about how it is going to be supported and how to get support quickly should a major problem arise. Ok, in this case the issue should really have been with the web hosting company, and the customer was emailing the wrong people. But the problem was further exacerbated by support refusing to just pick up the phone and deal with it.
Secondly, with cheap calling plans and VoIP, how much is a phone call really? In the case above both parties were in the USA. I find it hard to believe that the cost of a phone call to sort this issue out would be worth more than the time all those emails took to write! I can call a number in the USA from the UK for less than 2 cents a minute.
The version of AutoLogon that ships with Macro Scheduler does not support the optional message box that can be set up in Windows Security Policies and appears just after issuing CTRL+ALT+DEL before log on.
To address this we have released an update which will be provided in the next maintenance release of Macro Scheduler. In the mean time you can download just the updated AutoLogon service here: http://www.mjtnet.com/usergroup/viewtopic.php?t=2735
By Ron Lee.
Pacejet makes a Transportation Management (TMS) software suite that helps manufactures/distributors ship product via LTL, Truckload, Parcel, etc, and executes their warehouse pick-pack-ship as part of the process. Our software integrates with ERP systems so the customer has one end-to-end process flow in their business from customer to invoice.
One of the challenges we had with one customer was to benchmark our shipment planning performance for large blanket orders with their specific ERP. They would enter orders with 600-800 lines into the ERP system all at one particular time during the day – where most of the lines went to different ship-tos on different dates, etc. Our software would look at the orders and schedule all the shipments and then handle all the execution when the orders were ready to ship (wireless/RF, etc).
The technical issue was how to be able to “pound” multiple 800 line orders into the ERP system repeatedly to push a continuous wave of transactions at our software. That’s where Macro Scheduler came into play. I put together a relatively small set of macros to simulate the user’s data entry into the ERP’s order entry screen so we could validate our planning performance.
The input parameters on the orders could be varied enough to stress test different planning scenarios. With Macro Scheduler we also were able to drive the ERP application screens directly to make this as real-world a test as possible. And lastly, we didn’t have to spend a lot of time coding this as a more complex application.
The result was great for our planning performance, since we can plan orders almost faster than they can be entered into the ERP. But the result was also great for Macro Scheduler since we were able to use it to test out the scenario(s), make our customer comfortable with the result, and ensure we’re in a good spot with our product for the future.
Macro Scheduler is a great product. I appreciate all the flexibility in it. And the option to compile macros is awesome!
We’ve updated IEAuto.DLL with the ability to identify elements by their TITLE, ID and numeric INDEX in the ClickTag function. Some people have found websites where a number of ‘A’ elements (links) have identical text and no unique identifiers. WebRecorder will record these but the script may not play back – or only the first link will get clicked. The code can now be modified to identify such elements with either their title or id. Or if they have no title or id you can now resort to using the numeric index. WebRecorder Info.
Software bugs sometimes have drastic consequences, but check this bug report out:
http://bugzilla.mozilla.org/show_bug.cgi?id=330884
Microsoft have just announced that Vista will be delayed and won’t be generally available until the beginning of 2007. Read the official announcement.
This doesn’t really surprise me. Since November 2005 we have been working with the Vista beta releases to ensure Macro Scheduler will be Vista ready. Finding the information we needed to achieve this has been difficult and it is clear that much of the SDK and documentation is still not complete. In the end I had to get in touch directly with the Vista UAC team at Microsoft in order to make any real progress. It is clear that there is still a lot of work to be done and I just couldn’t see Vista being ready by the end of the year. Vista makes a complete paradigm shift in security by restricting default user accounts to standard level privileges. The discussion in the developer newsgroups makes it painfully clear that this will have huge implications on many existing applications and the way Windows software has been written in the past.
Delaying the general availability of Vista until 2007 has got to be a good thing. It is better that Microsoft concentrate on quality and get things working properly than rush the release out. Developers also need time to assess the implications and work on compatibility changes and for many this delayed release will cause a sigh of relief.