As Willy-Peter pointed out, a lot of my evenings have been filled with a type of visual cloud computing, that being PowerShell’s white text on a blue (not quiet azure) background that make me think of clouds for the purpose of automating the virtual machines that the VSTS Rangers are building.
So how are we doing this? Well there is a great team involved, it’s not just me and there are a few key scripts we are trying build which should come as no surprise it’s configuration before and after you install VSTS/VS/Required software, tailoring of the environment and installing software.
The Structure
So how have structured this? Well since each script is developed by a team of people I didn’t want to have one super script that everyone is working on and fighting conflicts and merges all day and yet at the same time I don’t want to ship 100 scripts that do small functions and call each other - I want to ship one script but have development done on many. So how are we doing that? Step one was to break down the tasks of a script into functions, assign them to people and assign a number to them - like this:
*Note this is close to reality, but the names have been changed to protect the innocent team members and tasks at this point.
| Task | Assigned To | Reference Number |
| Install SQL | Team Member A | 2015 |
| Install WSS | Team Member B | 2020 |
| Install VSTS | Team Member C | 2025 |
And out of that people produce the scripts in the smallest way and they name them beginning with the reference number - so I get a bunch of files like:
- 2015-InstallSQL.ps1
- 2020-InstallWSS.ps1
- 2025-InstallVSTS.ps1
The Build Script
These all go into a folder and then I wrote another PowerShell script which combines them into a super script which is what is used. The build script is very simply made up of a number PowerShell commands that get the content and outputs it to a file, which looks like:
Get-Content .\SoftwareInstall\*.ps1 | Out-File softwareInstall.ps1
That handles the combining of the scripts and the reference number keeps the order of the scripts correct.
As an aside I didn’t use PowerShell for the build script originally, I used the old school style DOS copy command - however it had a few bugs.
The Numbering
What’s up with that numbering you may ask? Well for those younger generation who never coded with line numbers and GOTO statements it may seem weird to leave gaps in the numbering and should rather have sequential numbering - but what happens when someone realises we have missed something? Can you image going to each file after that and having to change numbers -EEK! So leaving gaps is leaving the ability to deal with mistakes in a non-costly way.
Next why I am starting with 1015? Well each script is given a 1000 numbers (so pre install would be 1000, software install 2000 etc…) so that I can look at script and know what it’s for and if it’s in the wrong place. I start at 15 as 00, 05 and 10 are already taken:
- 00 - Header. The header for the file explaining what it is.
- 05 - Functions. Common functions for all scripts.
- 10 - Introduction. This is a bit of text that will be written to the screen explaining the purpose of the script and ending with a pause. The pause is important because if you ran the wrong script you can hit Ctrl+C at that point and nothing will have been run.
So that is part I. In future parts I will be looking at some of the scripts and learning's I have been getting.
