Dave Slusher
Welcome to our new blog! The paint is still wet and there may be construction dust in odd places so if you find some let me know. We will discuss in the coming weeks and months a little about how we built it using a combination of ServiceNow infrastructure and the Hugo static site generator. Some functionality does not yet exist, such as commenting and liking that existed on the previous Community-site-hosted blog.
It is that time again in the ServiceNow year, new release Early Access season! The Kingston release has just become available to the Personal Developer Instances. If you are eagerly awaiting the new features, you can now upgrade and take them for a test drive.
In another post, josh.nerius covered one of the primary features of interest to developers in this release: Flow Designer and IntegrationHub. I am going to provide a survey of some of the other topics.
After Your Pull Request is Accepted
Last week we posted information on how you can use ServiceNow projects to participate in Hacktoberfest. This involves some work server-side for the maintainer to be able to emulate the merging of GitHub pull requests. Let’s say you participated, submitted a pull request and it was accepted and merged into the main repository. Now what?
To reiterate slightly, you will have your own fork of the repository, and your ServiceNow instance is connected to your copy of the repository.
Source control integration was added as a feature to ServiceNow in the Geneva release. That increased by a wide margin the quality of development tools available to the ServiceNow developer. One could save code, easily move from instance to instance, backup personal developer instances, etc. One of the details of the implementation is that under the hood it is committing update sets. This complicates standard collaboration tools. GitHub pull requests and normal patch files assume that they are working on the text of code so when that text is really an XML payload, that presents a big challenge for diffing and merging.
In September 2017, a group of us convened in San Diego for a CreatorCon on the Road event. We spent a day at the Paradise Point resort with Pat Casey, Joe Davis and Jonathan Sparks reprising their keynotes from Orlando earlier this year. After that we had a full day of breakout sessions and workshops, along with exhibits and pods showing off various projects. We now have a set of the breakout talks posted on our Developer Program YouTube channel.
Last spring several of us with the developer program had a trip to India to put on some Developer Days events. We returned again in August to bring the CreatorCon on the Road experience to the country.
One of the main drivers of the timing was the fact that the new ServiceNow office in Hyderabad was opening on August 9th so there would be a puja ceremony and festivities around that.
In case you were not aware, there is a collection of Technical Best Practices that are published on the Developer Portal. This material is also available via navigation under the Learn > Documentation header in the top navigation. This is important information that needs to have a good home where it is accessible by developers at any time. We often have sessions covering this material at Knowledge (I co-presented on in 2014, for example) but it can be difficult to know where to access it on a daily basis.
Last week at Slack Frontiers, a big announcement was made about a new partnership between Slack and ServiceNow. Allan Leinwand (our CTO) joined April Underwood of Slack to discuss and demonstrate the upcoming prebuilt integration between the two services to be released in Kingston. The video is here and the ServiceNow announcement begins around 14:30. Allan also has a blog post discussing the partnership.
If you watch the video and/or read the announcement, you might be wondering exactly what this means for ServiceNow developers.
One of the ongoing pain points that ServiceNow developer run into is staying abreast on information about the various server side and client side APIs. It can take some thought to remember which API is occuring in the browser vs the server. That can be compounded even more by the complication of which server side APIs are global vs scoped, and then which version they were introduced or their scope access was altered.
In an interesting coincidence, I recently had to solve this issue in my own Travel Tracker app (as seen on a number of Live Coding Happy Hour episodes). Just today, someone in the Developer Program asked the same thing so I thought I would document it for posterity. What happens when an incoming connection you don’t control (such as a webhook) isn’t using application/json or application/xml as the Content-Type on the HTTP transaction?
