Category: Windows

I’ve been spending some time at work this week rearranging some stuff between my two development VMs, and I hit on a few items that I thought might be worth mentioning on this blog. I have two development VMs, one with a full install of Dynamics AX 2012 R2 on it, and another with a full install of SharePoint 2013 on it. Both are running Windows Server 2012 R2. And both have Visual Studio 2013 and 2017 installed. My AX work needs to get done on the AX VM, and any old-style SharePoint development needs to get done on the SharePoint VM.

General .NET development can be done on either VM. For reasons that made sense at the time, and aren’t worth getting into, my general .NET work has all ended up on the SharePoint VM. This is fine, but not optimal really, since the SP VM has only 8 GB of RAM, and 6 GB of that is in constant use by the SP 2013 install. That’s leaves enough for VS 2017, but just barely. The AX VM has a whopping 32 GB of RAM, and the AX install generally uses less than 10 GB. And my company is gradually moving from SP 2013 to SharePoint Online, so my need for a dedicated SharePoint VM will be going away within the next year or so (hopefully).

So it makes sense to me to move my general .NET projects from the SP VM to the AX VM. That’s mostly just a case of copying the solution folder from one VM to the other. Back when we were using TFS (with TFVC) for .NET projects, it would have been more of a pain, but with git, you can just move things around with abandon and git is fine.

All of this got me looking at my tool setups on both VMs, and trying to get some stuff that worked on the SP VM to also work on the AX VM, which led me down a number of rabbit holes. One of those rabbit holes had me looking at my PowerShell profiles, which led me to refresh my memory about how those worked and how to customize the PowerShell prompt.

The official documentation on PowerShell profiles is here, and the official doc on PowerShell prompts is here. User profile scripts are generally found in %userprofile%\Documents\WindowsPowerShell. Your main profile script would be “Microsoft.PowerShell_profile.ps1”. And you might have one for the PS prompt in VS Code as “Microsoft.VSCode_profile.ps1”. (Note that I haven’t tried using PowerShell Core yet. That’s another rabbit hole, and I’m not ready to go down that one yet…)

Anyway, on to prompts: I’ve always kind of disliked the built-in PowerShell prompt, because I’m often working in a folder that’s several levels deep, so my prompt takes up most of the width of the window. The about_prompts page linked above includes the source for the default PowerShell prompt, which is:

function prompt {
    $(if (Test-Path variable:/PSDebugContext) { '[DBG]: ' }
      else { '' }) + 'PS ' + $(Get-Location) +
        $(if ($NestedPromptLevel -ge 1) { '>>' }) + '> '
}

In the past, I’ve replaced that with a really simple prompt that just shows the current folder, with a newline after it:

function prompt {"PS $pwd `n> "}

Yesterday, I decided to write a new prompt script that kept the extra stuff from the default one, but added a couple of twists:

function prompt {
    $loc = $(Get-Location).Path.Replace($HOME,"~")
    $(if (Test-Path variable:/PSDebugContext) { '[DBG]: ' } else { '' }) + 
    $loc + 
    $(if ($NestedPromptLevel -ge 1) { '>>' }) +
    $(if ($loc.Length -gt 25) { "`nPS> " } else { " PS> " })
}

The first twist is replacing the home folder with a tilde, which is common on Linux shells. The second twist is adding a newline at the end of the prompt, but only if the length of the prompt is greater than 25 characters. So, nothing earth-shattering or amazing. Just a couple of things that make the PowerShell prompt a little more usable. (I’m pretty sure that I picked up both of these tricks from other people’s blog posts, but I can’t remember exactly where.)

Anyway, this is all stuff that I’m doing in the “normal” PowerShell prompt. I also have cmder set up, which applies a bunch of customization to both the cmd.exe and PowerShell environments. Honestly, the default prompt in cmder is fine, so none of the above would be necessary if I was only using cmder. But I’ve found that certain things were only working for me in the “normal” PowerShell prompt, so I’ve been moving away from cmder a bit. Now that I’m digging in some more, though, I think some of my issues might have just been because I had certain things set up in my normal PowerShell profile that weren’t in my cmder PowerShell profile.

Cmder is basically just a repackaging of ConEmu with some extra stuff. I don’t think I’ve ever tried ConEmu on its own, but I’m starting to think about giving that a try. That’s another rabbit hole I probably shouldn’t go down right now though.

I’d love to be able to run Windows Terminal on my dev VMs, but that’s only for Windows 10. (It might be possible to get it running on Windows Server 2012 R2, but I haven’t come across an easy way to do that.) Scott Hanselman has blogged about how to get a really fancy prompt set up in Windows Terminal.

And at this point, I’ve probably spent more time messing with my PowerShell environment than I should have and I should just settle in and do some work.