Background
According to the Visual Studio ALM Rangers, there are two major approaches to sharing resources (e.g. common libraries which are used in many separate products) in TFS 2012:
Workspace mapping, setting up workspaces so that they point to the appropriate version of each required library and product.
Shared folders, using branch/merge to get and update the shared resource
At a glance, shared folders seems like the way to go, but a client that I am working with has experienced a lot of problems with that approach in Starteam, and is reluctant to try it again in TFS. I am currently in the process of assisting the client migrating from Starteam to TFS.
I have listed pros and cons with each approach, but I am uncertain if I have missed something.
Workspace mapping:
Simple to setup and understand
Easy to test a library change in several products
Easy to get latest changes in a library, and to submit changes to a library
No tracability, or at least less tracability, e.g. if a change in a library was introduced in Product A, how to track that change in Product B
Changes in libraries may affect products in an uncontrolled manner
Build gets more complicated
Each user must set up his/her workspace individually (but there are workspace templates in TFS 2012 Power Tools)
Folder mapping:
Everything that is needed is configured in a given branch
Isolation between products and branches
Builds are simplified
More control of changes
Requires more disk space
Requires more administration in the form of branching/merging and setup of branches
One particular problem is how to test library changes in several products. As I understand that would require testing in product A, then reverse integrate to library and forward integrate to product B, then test that product and so on.
Conclusion, and final question
The client has successfully used something similar to workspace mapping in Starteam for 10 years, and plan to continue to use that approach in TFS. Although they have the problem to keep track of library changes that affects several products.
They are afraid that folder sharing will get messy and complicated.
My question is, have I missed something in my list above? Are there more reasons for why an organisation not should use workspace mapping, or for why they should use folder sharing.
Related
We are investigate how to using modern code version systems for a legacy systems.
However, there are some difficulties that we do not really know how to solve. How do you think we should going to organize this?
Some parts of the code we have responsibility for, and some parts the customer are responsibility for.
The code that is in our development system is the master, and all changes are made there and then sent out to the customer. So far is it like a common project. But for some directories with a few hundred files, the client is responsible for the code. Customer can change these files as they want, and do not have inform us. When we do an update for them, we need to get a copy of these files and merge our changes before we give the new version to the customer.
We have released 4 base versions of the system over the 25 years the system have lived. Our six customers, however, use different versions of the system, which means that we have to consider it. In addition, each customer has their own requirements, and we have made major adjustments for each customer.
We have several parallel projects related to the same files. Today we have major problems that the various projects interfere with each other, so we will benefit greatly from having changes in different branches. But how do we organize this?
Some projects affect only one customer, other projects involving several customers. And some projects may be introduced into a new future base version.
So my question is how to organize this in our new version system.
I'm currently in charge of migrating our asp.net applications from source safe to TFS. We have three or four very similar apps (let us say e-commerce) that currently share a core library (services, business logic, entities, data access etc).
The applications are similar but not identical so one app might get a feature set the others won't get etc.
I want to stop the sharing of code and instead set up branches (if that fits) so if I change something in Application A:s core library I will need to merge the changes with the other branches instead of them getting the changes automatically. This to avoid surprises when you update from your trunk and suddenly the core has changed for another project and this project breaks in some way.
Any suggestions on how I should set this up in TFS? Should I have a "main" Core that is not directly used in any project that is the parent of all the other cores so I can push changes up to that one from one core and then distribute it to the other cores? Does that make sense and would it be easy to set up in TFS?
In response to your comment, I'd suggest you to read up on Feature branches on the CodePlex website.
Scenario 4 – Branch for Feature
In this scenario, you create a
development branch, perform work in
that branch, and then merge your work
back into your main source tree. You
organize your development branches
based on product features. The
following is a physical view showing
branching for feature development:
My Team Project
Development -> Isolated development branch container
Feature A -> Feature branch
Source
Feature B -> Feature branch
Source
Feature C -> Feature branch
Source
Main -> Main Integration branch
Source
We are alos moving from SS to TFS in the near future.
As I perceive it, we are going to keep our SS repository online and start fresh over in TFS. Our framework probably will get its own project in TFS. Project specific shared units will need to get merged from time to time.
The way you structure your repository depends on your specific situation. Every branch scenario has its specific advantages and drawbacks.
How many projects
How many developers
Are the developers dedicated
Do you need concurrent hot fixes
Do you need service packs
Take a look at the CodePlex branching guide for all the information you need to make an informed decision about your TFS structure. Print out the cheat sheets and pin them to your wall for quick reference.
Before executing on your branch plan,
pay attention to this cautionary
message - every branch you create does
have a cost so make sure you get some
value from it. The mechanics of
branching in TFS are simplified to a
single right click branch command.
However, the total cost of branching
is paid by reduced code velocity to
main, merge conflicts and additional
testing can be expensive.
I am assuming you have already investigated whether you truly need to make your "copies" seperate team projects. Remember the TFS concept of a "Team Project" is a VERY LARGE high level container. It is not the same thing as what most IT shops consider a "Project". Think of "Microsoft Vista" or "Office 2007" as a project, not, say "A new release of Company XYZ's Accounts Receivable System" as a project in the Team Project sense.
I have a client that decided on one single Team Project for TFS. There is nothing wrong with this - and it is truly the best scenario in many circumstances.
If you truly need a very strong isolation between your copies of the application (perhaps they are seperate clients and you need very strong security seperation) and must have seperate team projects.
That said - you still - as you've stated need to share code between instances of your application. The first thing I would strongly recommend is to get away from "Cut and Paste" sharing. I would truly try to isolate the shared code into a seperate Solution and generate binaries for that (perhaps you've already done this!)
This is covered in the Codeplex TFS: http://tfsguide.codeplex.com/
Another approach I've done for several clients - is to have a Team Project that contains the shared code. The "Build" creates the binaries for the shared code - and the "Deploy" simply copies those to a "known location" (ie UNC share on the build machine)
For the applications that are "Consumers" of the "Framework" we simply used the "AdditionalReferencesPath" Item group to include the location of that known location.
Furthermore - this tool: http://tfsdepreplicator.codeplex.com/ can be helpful. This would allow you to have builds automatically triggered for your "Consumer" Projects whenever the "Framework" solution is built.
My brief answer is that you should only setup one 'TFS project' and simply organize your different projects, i.e. your individual applications, and each shared library, as separate folders under that one TFS project. The alternative is to include specific (binary) builds of the shared libraries in each individual application – if you do that then you can organize each application into it's own TFS project, tho you can't merge changes or branch those projects without using the TFS command line (and some non-obvious commands to boot).
I was trying to determine the same information, this guide on codeplex is perfect
http://vsarbranchingguide.codeplex.com/releases
Includes terminology and different branching workflow approaches as well as cheat sheets.
Ok, when hired on to my current company a year ago, I was tasked with migrating our development teams from VSS. They already had it in their minds that they wanted Subversion, and since I had experience using and setting up subversion, I was a good candidate. I first tried to sell TFS because it woul dhave solved the problem I am in right now, but since money is tight, and Subversion is free... well you get it. Anyway, I have finalized the propsal and the only thing standing in the way is the following.
I proposed that we store only our source code in SVN, and all documentation, release builds, and other project artifracts be stored in our SharePoint portal, so we don't have to give non developer stakeholders access to SVN. When I presented the proposal, all was excepted but the question arose about how to manage the syncronization between the artifacts (Ex: How to is document x version 3.1.2 associated with release 4.5.2). My initial reaction is to create a section in the SharePoint porject page for each new release that will hold the artificats (and keep track of changes too). Is there a better way of doing this? Does anyone know of anyone doing this? Or any integration packages to sync SVN with SharePoint?
Here is some info on the companies development environment. All of our software is for internal use, we sell none of it, so our customers are all in-house. We have 2 types of developers: 1. those who take care of maintainance and customization of third party software, and 2. those who write proprieatry software (which is where I fall). Our software we write is mostly .NET, but the 3rd party software is all over the board (COBAL, C, FORTRAN, Other crap that no ones cares about anymore).
Please advise, as I need to get this submitted soon. I HATE VSS!!!!!!!!!! and I need relief!
What we do internally is putting all docs under our version control system, I think it's much easier. Then, of course, you have to give access to not-developers.
In your case, using SVN, why don't you put everything inside and then use the webinterface to give access to the stakeholders? It's easy enough for them :-P
I would use SVN for both documents and source code.
Advantages:
You can synchronize versions of
documents with versions of source
code.
You have everything in one place, so
no two repositories to administrate.
Disadvantages:
You'd probably need to manage the
access rights for some stakeholders
to some parts of the folder
structures.
SVN is not the most appropriate tool
for document management
In order to solve the possible concurrent changes to the same document, you can use SVN property svn:needs-lock for these items, to make them editable by one person, who locks the item.
As pablo said, you can access the documents (at least for reading them) through the web interface.
You could expose the svn repo via the web interface and link to that in sharepoint. That way people who need to edit the documents would need access to subversion but anyone could easily access the documents "read only".
In our organization, we have docs/artifacts, code everything in SVN and have given access to non-technical stakeholders as well who use tortoise client.
however you can look at the following option
Option 1 : create a ASP.Net interface for non-technical users
You can build a simple web interface in ASP.net, configure that with a single user so you would not have to create separate users for all the nontechnical stakeholders and they would get access to the docs with proper version control, etc. you could look at sharpsvn for the implementation aspect. the disadvantage of this approach would be that you might have to invest some time in developing this app
Option 2 : ofcourse, create separate users for each non-developer stakeholder
This answer is probably too late for you implementation, but the simplest integration path may be to store the docs in SVN and then publish to Sharepoint with an svn-hook.
Build artifacts could be programatically published the same way from you build scripts.
You can upload docs to SharePoint using a simple POST
i.e.
http://blogs.msdn.com/rohitpuri/archive/2007/04/10/upload-download-file-to-from-wss-document-library-using-dav.aspx
Probably a little late, too, but I would avoid putting the documents in SVN if you have a SharePoint system setup. Though SVN does a fantastic job for source code, for document management it doesn't provide the ease-of-use of SharePoint. If you have it already setup and you are a primarily MS based network, SharePoint makes a lot of sense and can handle revision control for the MS based documentation much better than SVN.
Yes, you can manage access to SVN documents with a needs-lock, but chances are at some point you'll have a non-developer needing to access the documents. Explaining SVN to a non-developer, non-techie is not an easy thing.
Our small development shop is looking to migrate our projects from VSS to TFS, and we're evaluating TFS vs. others (haven't pulled the trigger quite yet). The nature of our software shop is such that we have 100+ projects in VSS ranging from small one-man-show projects to massive enterprise-wide applications.
We are trying to determine how to structure our projects in the transition and have, for the most part, decided on putting everything into one project site/system with each project having a subfolder off the root.
With this type of setup, we are concerned that we will lose a lot of the functionality that TFS provides (bug tracking, scrum burndowns, reporting, document storage etc.) because all the projects will be in the same portal/project space and it will be difficult to separate out individual project tickets/items.
Does anyone have experience with this? What was your solution? Did you stick with TFS?
The answer to this question requires some planning on your part: how you intend to use TFS, and which of those capabilities has inherent limitations in the product. I would summarize my advice as:
You will need [at least] 1 team project per process template. That is, if two teams want to adopt / customize different processes, they will need to be separated.
Once condition #1 is satisfied, you probably don't need as many separate Team Projects as you think. 90% of TFS features & settings are hierarchical in nature, allowing you to scope them as broadly or narrowly as each of your projects requires.
For complete details, see:
http://www.codeplex.com/BranchingGuidance/Wiki/View.aspx?title=Guidance%20for%20Structuring%20Team%20Projects
http://blogs.msdn.com/richardb/archive/2007/05/01/tfs-team-project-whitepaper.aspx
http://msdn2.microsoft.com/en-us/library/aa974183(vs.80).aspx
The approach I've taken was to have a TFS project for each logical grouping of assemblies -- So we've a framework project that contains assemblies common to all our applicaitons, we then have a separate project for our quotations system, another for the costing system and so forth. Whilst the workspace mappings get a bit "interesting", it does allow different design methodologies for different projects, and at different timescales -- so one team might be half way through a sprint (Most projects use Scrum for Team System), at the same time as another is just starting...
It is true that to garner all of the benefits of TFS, it is best to use separate projects, but those benefits should be weighed against the administrative overhead associated with managing many projects. Years ago, I used Visual Source Safe...After I left Microsoft, I switched to Subversion. After returning to Microsoft, I am using TFS and so far I am very happy with it.
The process guidance, the reports, the integrated bugtracking, and the tight IDE integration serve my needs perfectly. Plus, the TFS SDK allows for some interesting extensibiilty scenarios.
I've used several SCC providers and we've settled on TFS for all of the features it has that others don't. Bug correlation, CI and automated testing certainly topped the list of benefits.
As for whether you use multiple project or not, I'd say it depends on if the projects share any common code. We tend to use a TFS project for all "related" code assets, so if we have several different solutions that do similar things and share a lot of code, we use a single TFS project. If they have nothing in common, then they become separate projects.
I am not sure if this was fixed in 2008 but in 2005 when you built a project that was a subfolder of a root project, MSBuild will pull the entire source tree of the root project - even files that are not part of your subfolder.
Depending on how much source you are managing this can greatly increase your build times.
I realize this article is old, but TFS 2010 now supports a wonderful feature call Team Project Collections which is simply another level of indirection or grouping on top of Projects.
This makes it much easier to create Team Projects without clogging up your namespace and encourages better organization!
Great Link talking more about Collections
http://blogs.msdn.com/b/bharry/archive/2009/04/19/team-foundation-server-2010-key-concepts.aspx
I a not a sharepoint user but I hear its very similar concept to Sharepoint collections :)
My organization has begun slowly repurposing itself to a less product-oriented business model and more contract-oriented business model over the last year or two. During the past year, I was shifted into the new contracting business to help put out fires and fill orders. While the year as a whole was profitable (and therefore, by at least one measure, successful, we had a couple projects that really dinged our numbers for the year back around June.
I was talking with my manager before the Christmas holiday, and he mentioned that, while he doesn't like the term "post-mortem" (I have no idea what's wrong with the term, any business folks or managers out there know?), he did want to hold a meeting sometime mid-January where the entire contract group would review the year and try to figure out what went right, what went wrong, and what initiatives we can perform to try to improve profitability.
For various reasons (I'll go into more detail if it's requested), I believe that one thing our team, and indeed the organization as a whole, would benefit from is some form of organized code-sharing. The same things get done again and again by different people and they end up getting done (and broken) in different ways. I'd like to at least establish a repository where people can grab code that performs a certain task and include (or, realistically, copy/paste) that code in their own projects.
What should I propose as a workable common source repository for a team of at least 10-12 full-time devs, plus anywhere from 5-50 (very) part time developers who are temporarily loaned to the contract group for specialized work?
The answer required some cultural information for any chance at a reasonable answer, so I'll provide it here, along with some of my thoughts on the topic:
Developers will not be forced to use this repository. The barrier to
entry must be as low as possible to
encourage participation, or it will
be ignored. Sadly, this means
that anything which requires an
additional software client to be
installed and run will likely fail.
ClickOnce deployment's about as
close as we can get, and that's awfully iffy.
We are a risk-averse, Microsoft shop. I may be able to sell open-source solutions, but they'll be looked upon with suspicion. All devs have VSS, the corporate director has declared that VSTS is not viable going forward. If it isn't too difficult a setup and the license is liberal, I could still try to ninja a VSTS server into the lab.
Some of my fellow devs care about writing quality, reliable software, some don't. I'd like to protect any shared code written by those who care from those who don't. Common configuration management practices (like checking out code while it's being worked on) are completely ignored by at least a fifth of my colleagues on the contract team.
We're better at writing processes than following them. I will pretty much have to have some form of written process to be able to sell this to my manager. I believe it will have to be lightweight, flexible, and enforced by the tools to be remotely relevant because my manager is the only person who will ever read it.
Don't assume best practices. I would very much like to include things like mandatory code reviews to enforce use of static analysis tools (FxCop, StyleCop) on common code. This raises the bar, however, because no such practices are currently performed in a consistent manner.
I will be happy to provide any additional requested information. :)
EDIT: (Responsing to questions)
Perhaps contracting isn't the correct term. We absolutely own our own code assets. A significant part of the business model on paper (though not, yet, in practice) is that we own the code/projects we write and we can re-sell them to other customers. Our projects typically take the form of adding some special functionality to one of the company's many existing software products.
From the sounds of it you have a opportunity during the "post-mortem"to present some solutions. I would create a presentation outlining your ideas and present them at this meeting. Before that I would recommend that you set up some solutions and demonstrate it during your presentation. Some things to do -
Evangelize component based programming (A good read is Programming .NET Components - Jubal Lowy). Advocate the DRY (Don't Repeat Yourself) principle of coding.
Set up a central common location in you repository for all your re-usable code libraries. This should have the reference implementation of your re-usable code library.
Make it easy for people to use your code libraries by providing project templates for common scenarios with the code libraries already baked in. This way your colleagues will have a consistent template to work from. You can leverage the VS.NET project template capabilities to this - check out the following links VSX Project System (VS.Net 2008), Code Project article on creating Project Templates
Use a build automation tool like MSBuild (which is bundled in VS2005 and up) to copy over just the components needed for a particular project. Make this part of your build setup in the IDE (VS.NET 2005 and up have nifty ways to set up pre-compile and post-compile tasks using MSBuild)
I know there is resistance for open source solutions but I would still recommend setting up and using a continuous automation system like CruiseControl.NET so that you can leverage it to compile and test your projects on a regular basis from a central repository where the re-usable code library is maintained. This way any changes to the code library can be quickly checked to make sure it does not break anything, It also helps bring out version issues with the various projects.
If you can set this up on a machine and show it during your post-mortem as part of the steps that can be taken to improve, you should get better buy since you are showing something already working that can be scaled up easily.
Hope this helps and best of luck with your evangelism :-)
I came across this set of frameworks recently called the Chuck Norris Frameworks - They are available on NuGet at http://nuget.org/packages/chucknorris . You should definitely check them out, as they have some nice templates for your ASP.NET projects. Also definitely checkout Nuget.
organize by topic, require unit tests (feature-level) for check-in/acceptance into library; add a wiki to explain what/why and for searching
One question: You say this is a consulting group. What code assets do you have? I would think most of your teams' coding efforts would be owned by your clients as part of your work-for-hire contract. If you are going to do this you need to make absolutely certain that your contracts grant you rights to your employees' work.
Maven has solved code reuse in the Java community - you should go check it out.
I have a .NET developer that's devised something similar for our internal use for .NET assemblies. Because there's no comparable .NET Internet community, this tool will just access an internal repository in our corporate network. Otherwise will work rather much the way Maven does.
Maven could really be used to manage .NET assemblies directly (we use it with our Flex .swf and .swc code modules) is just .NET folk would have to get over using a Java tool and would probably have to write a Maven plugin to drive msbuild.
First of all for code organization check out Microsoft Framework Design Guidelines at http://msdn.microsoft.com/en-us/library/ms229042.aspx and then create a central Location source control for the new framework that your going to create. Set up some default namespaces, assemblies for cleaner seperation and make sure everyone gets a daily build.
Just an additional point, since we have "shared code" in my shop as well.
We found out this is very much a packaging issue:
Whatever code your are producing or tool you are using, what you should have is a common build tool able to package your sources into a "delivery component", with everything used to actually execute the code, but also the documentation (compressed), and the source (compressed).
The main interest into having a such a "delivery package unit" is to have as less files to deploy as possible, in order to ease the download of those units.
The build process can very well be managed by Maven or any other (ant/nant) tool you want.
When some audit team want to examine all our projects, we just deploy on their post the same packages we deploy on a production machine, except they will un-compressed the source files and do their work.
Since our source files also includes whatever files are needed to compile them (like for instance eclipse files), they even can re-compile those projects in their development environment).
That way:
Developers will not be forced to use this repository. The barrier to entry must be as low as possible to encourage participation, or it will be ignored: it is just a script to execute to get the "delivery module" with everything in it they need (a maven repository can be used for that too)
We are a risk-averse, Microsoft shop: you can use any repository you want
Some of my fellow devs care about writing quality, reliable software, some don't: this has nothing to do with the quality of code written in these packages modules
We're better at writing processes than following them: the only process involved in this is the packaging process, and it can be fairly automated
Don't assume best practices: you are not forced to apply any kind of static code analysis before packaging executable and source files.