I came across a Sitemap generation code), the idea interested me enough to do a bit more research into the topic, and try to work out the details.
So here's some notes on the hows and whys of using code to change layout details. (I'm working on Sitecore 6.6 here, but this information is relevant to many versions of Sitecore)
Another helpful addition to the "scripted installs" functions I've been looking at for the last few weeks is the ability to trigger a full rebuild of a search index. Last week we looked at deferring the indexing of items installed by a package to try and help speed up the scripted install of packages. So it makes sense to be able to trigger a build as well...
Last week I looked at some simple PowerShell script to automate installing packages into Sitecore instances. Having spent a bit of time testing this process, I found a couple of useful tips for speeding it up:
After you've finished installing an instance of Sitecore for development or test (as I've been discussing in my previous few posts – Installing, Adding DMS, Hosts & SSL), chances are you'll need to deploy some modules or packages onto it. Things like Web Forms for Marketers are regularly needed as a basis for development work. So what can you do if you'd like to automate this with PowerShell as well?
Continuing my current theme of installation automation via PowerShell, the next task I found myself looking at was how to enable Sitecore's analytics for a newly installed Sitecore instance. This is a task that the standard .exe installer doesn't do for the v6.6 sites I'm working with, but is relatively simple to automate.
Ages back I wrote up some
work on automating the Sitecore installer. I noted back then that I wasn't sure what the purpose of the
SC_IISSITE_ID
parameter to the MSI was. Having done some more work on the topic of install automation recently, I've got some more detail on what it's for now, as well as another parameter that's more important than I had realised...
When you install an instance of Sitecore via the .exe installer you get a single host name added to your machine, which matches the instance name you chose. But for real world projects this is rarely enough. Most websites need to respond to other host names, and they often also need SSL certificates installed so that these hosts can be accessed via HTTPS. Developers need these features set up so that they can code and test with them.
Recently I've been looking at automating more installation tasks via PowerShell, in order to try and standardise the processes developers use when setting up new instances for work. I've been digging out helpful bits of script from across the internet, and I plan to write a few posts to document some of the approaches I've been experimenting with. The first of these posts covers info on automating host names and their bindings:
Quick bit of script this week. Ages back I wrote up the process for working out how to run unattended installations of Sitecore using its .exe installer and then converted that to PowerShell. Recently I've been doing a bit more work on install automations in PowerShell. (Which I plan to write up in future posts) But while doing this, I realised that the flip side of automating the install is that it's also helpful to be able to remove an instance of Sitecore automatically too...
Spending a lot of time developing websites with Sitecore means I've reported my fair share of bugs to support. One particular issue that I came across in 6.6 a while back stands out as my "favourite" issue I've reported – just because the error stops a lot of developers in their tracks thinking "how on earth can this happen?". If you wrote this code, would you think it could crash?
Well it can crash. And to a less experienced developer it can be a very confusing error:
Recently a client of mine came up with some requirements that involved submitting data to the Web Forms for Marketers database via code. Having done a bit of Googling, I came across a Stack Overflow answer on the same subject which seemed to offer a solution. Implementing this code did indeed allow submitting data – however it didn't trigger Save Actions. So while the data will go to the configured WFfM data store, emails or CRM integrations won't get triggered.
That was an issue for my client, who wanted to have email notifications as well. So starting from the Stack Overflow responses, I looked into how the WFfM save actions can be correctly triggered via code.