Last Monday 26th we started the Sprint number 163. It’s been already a while since we started using scrum as you can see. We’ve run bi-weekly sprints for years now.
We’d like to share some of the things we’re working on so the Plastic and SemanticMerge users can get a glimpse of what we are doing.
[Update]: my friend Rami Shaft, from Israel, just sent me an email to share some special properties of the number 163:
- 163 - has a very special property in number theory, there is no explanation for this.
- 163 is a strong prime in the sense that it is greater than the arithmetic mean of its two neighboring primes.
- 163 is a lucky prime and a fortunate number.
- 163 is a strictly non-palindromic number. Given 163, the Mertens function returns 0.
- 163 is a Heegner number.
- The square root of 163 occurs in several interesting pieces of mathematics.
- and more and more...
Revamped method hist
Have you ever heard of our “method hist”? Maybe not, but it is something we launched almost three years ago and lets you right click on a method on your code and find the history of this specific method.
The motivation behind methodhist (as we call it) is that 9 out of 10 times you look at a file’s history you’re really looking for the history of a given method.
So methodhist will:
- Focus on the method you’re looking for (showing you just the history of the body of the method and not the full file).
- Locate the method even when it was moved to a different position in the file.
Current Plastic SCM users can access this feature through the Visual Studio Package (2005 onwards) and we also developed a version for SVN users: http://www.methodhistory.com.
Why are we redoing the whole thing then?
Well, now we’re plugin the “semantic merge” (www.semanticmerge.com) technology inside method hist wich means:
- New methodhist will not only locate the method when it has been moved, it will also do it if the method was renamed, moved to a different class (inside the same file) and so on.
- It has full language support: the old version had trouble with C# properties, some constructors. The new version is based on a vastly improved parser so it doesn’t have this sort of issues.
- It works for Java, C# and Vb.net (and soon for all other languages that semantic merge will support).
- The GUI has been deeply modified and improved so that it works much better than before, as you can see here:
We’re also polishing the last bits inside the VStudio package.
This new feature will be made available real soon, inside Plastic SCM 4.2 (more on naming later) so I expect you’ll be able to give it a try in no more than a week or two.
One of the big goals for the sprint is to continue improving our Eclipse support. There’s nothing wrong about it right now but we’re working on a few great improvements.
The first line of work is related to work with huge projects (where facing projects bigger than 500k files in a workspace), performance related, polishing some existing Eclipse perpectives and so on. New branch and changeset views (customizable with queries) will be ready at the end of this sprint.
The second and really exciting new feature is the Branch Explorer within Eclipse! Yes, finally it is going to be there, and here you can see a very initial preview:
Cloaked files improvements
We’re doing some improvements on the “cloaked” files feature. Basically being able to handle exceptions correctly. Something like saying: cloak everything unless *.c with much simpler rules than today.
The goal here is to improve the way in which workspaces can be configured, introduced a more flexible way. It is the first step towards fully configurable workspaces.
We will merge this feature on the 4.2 branch and it will be also available next week.
We can’t unveil this one yet, but we’re integrating with a well-known specific-purpose IDE. It has been under development for months already but we’re getting close to production.
Azure cloud server
Yes, you guessed it right. The goal for this sprint is to have a new iteration of our cloud server initiative up and running. The initial versions have been running for months already but with some restrictions.
Now we’re focusing on adapting the server code to use the cloud caching features (required to have multiple “workers” (Azure’s jargon) cooperating and to achieve true scalability on cloud operations) and running a good number of performance tests.
As you might guess the final goal is to launch a cloud service based on Plastic SCM and specifically tailored for commercial teams (since we don’t think there’s much room for another open source initiative :P).
It has been under development for months and tested internally but we’re approaching public alpha.
Plastic SCM 5.0 launch
Yes, we’re about to launch Plastic SCM 5.0. We’re going to rename 4.2 (which has been under development for more than one year, and available under the “labs” section on the website) to 5.0 to make clearer it is a new version.
There are many new things on 5.0 (as I said, available on the “labs section”) including performance, security, new features but the first thing you will notice is the new icons in the GUI:
Plus the two new I3 themes, the black one and a “blue one” that we think is pretty cool:
Now the themes are fully configurable with config files which will allow our designer to come up with new variations.
Did you notice the new logo? :)
The semanticmerge squad has also quite a few goals for this sprint. We reviewed all the answers we got from the SemanticMerge User Survey and focused on improving the most demanded ones. The number one was startup speed and it was already finished and launched last week.
Now we’re working on:
- Usability fixes.
- External parser support: basically opening up the parser system so that developers can add their own languages.
- C support: progressing here, about to be in alpha mode.
- Mac OS X: we won’t work on it this sprint but we will continue the development the next one.