Following SO thread shows Managing DLL references in multiple projects across different solutions. Additionally I want to know how to manage the version control for these dependency DLLs ?
Should the DLL (which is published to other external projects) be committed too, every time code change happens ?
Team Development with TFS Guide (Final Release)
This guide has proven to be invaluable to my team. They describe several scenarios and what the pros and cons are of each. For anyone managing a TFS environment this is a must read.
With regards to the DLL's from external projects. We will keep a copy of the source in TFS if we can get access to the source some times, use your best judgement. We keep a copy of the DLL in source control always. The DLL goes into a "SharedBinaries" folder next to source code so that they can be branched together.
It is critical that you be able to branch these DLL's along with source. It is also critical that they be in TFS so that you can do an automated build with little or no build machine configuration. My own personal goal while managing TFS is to be ready for a new developer to join the team and with a single get of source code be able to execute a successful build for local debugging.
EDIT: Different department builds DLL
Like all good IT answers I have to start with "it depends". If the other department is truly segregated from your department and you have little or no knowledge of what they are working or when they will be working on it. If they just occasionally tell you that they have done some things and you should now incorporate the changes then I would lean towards the DLL being committed to the repository every time that the department consuming it wants to change it.
If on the other hand we are really just talking about different teams in the same department where there is lots of cross talk and water cooler communication then I would expect that you could making something else work with just some project references.
It sounds to me like it is the former and not the latter situation that you find yourself in today. I would try to get the department that is creating the shared code to "release" the shared code like Microsoft releases the .NET Framework. Get them to just build the API and give you some DLL's and some documentation. Then the groups that are incorporating those DLL's into there products can check them in separately into a repository of there own control and isolate themselves from code churn while the department working on the these reused DLL's can work on the next version of them.
You should take this all with a grain of salt. This is just one guy rambling on about what might be a good idea. There are many more ways to solve these problems and they are all correct given different circumstances. If you are asking 5 people and you get 5 different responses I wouldn't be surprised.
Related
It's unclear how the repositories are connected and tied together. Each git repo has a different package of cs files, with their own unique features and design schemas. Rather than a hierarchical design structure, where in my understanding or example:
Abp-base (free git repo), would have core essentials
AbpZero-paid (membership subscription to premium git repo), would add additional features into their design that would encourage others to pay for.
AbpZeroTemplate-demo (download from website, not git), would be a sample utilizing the paid content to push and drive sales to unlock the tool to it's full potential.
I downloaded and even paid for the different tiers, but all of the contents are all over the place. When I tried to combine them so that I could take advantage of a complete product, nothing lined up properly, and I get breakage everywhere.
How am I supposed to change the crucial schema design flaws when half of the .proj links to .dll that expect the class to be structured a certain way?
Example:
User is structured: "Name", "Surname". But i want it to be "Firstname", "Lastname." Some may ask, what's the difference? Just like how "Maiden name" is different from "Last name" or "Family name." Those little nuances are'nt procedural thinking. If you went to a stranger and asked "what's your name?" and they said "Chuck Jones," you would then ask to clarify, "What's your first name?"
Each git repository is running off of a different versioning number, that doesn't match with their counterpart. AbpZero is on a stable release version 5.x, and you go to GitHub for Abp and it's saying version 3.x is stable and green. And one of them is using User<TTenant,TUser> as a class, and the other is using User<TUser>. So not only am I getting mixed signals on how they're supposed to work (because git repos point to each other, and rely on .dll for continuity) but they're also not even sharing the same coding structure.
Can anybody help me merge the old and new together, or tell me the direction on how the design is being built?
Disclaimer: I'm a contributor to ASP.NET Boilerplate
ASP.NET Core is a web framework developed by Microsoft, as you know. There are plenty different applications based on it.
ABP is an application framework on top of ASP.NET Core (and MVC 5.x but it's a different story - I skip). It tries to help real life applications on top of ASP.NET Core. It has many features like advanced authorization, background jobs, javascript proxies, DDD infrastructure and so on.. This is completely free and developed on Github: https://github.com/aspnetboilerplate/aspnetboilerplate Thousands of developers are actively using it to build their applications.
ABP framework does not provide a User Interface itself, but provides the infrastructure. So, you should create an empty ASP.NET Core application, add nuget packages, configure the framework and modules, create a UI layout and some basic pages to start your development. This is also a time consuming work. That's why we have created "STARTUP TEMPLATES". There are two startup template:
(A) A free & open source startup template can be downloaded here: https://aspnetboilerplate.com/Templates (It just gets the code from https://github.com/aspnetboilerplate/module-zero-core-template and renames solution for you). This is a simple startup template (but more powerful then ASP.NET Core's standard template when you make File > New Project).
(B) A PAID startup template, which becomes the product on https://aspnetzero.com/. This template has much much more features on top of the framework. You can see all on its web site. Since it's a paid product, the Github repository is private and only available to paid customers.
You either start with template (A) or template (B). You CAN NOT combine them. I believe it's very obvious. For instance. Microsoft provides different startup templates for ASP.NET Core: One of them is Angular, one of them is Razor Pages, one of them is empty... and so on.. So, you select one of them based on your needs.
Both of (A) and (B) templates work on the ABP framework and adds via nuget packages (it's already combined/integrated).
So, after this explanation, I believe no one would even think how to combine them.
How am i supposed to change the crucial schema design flaws when half of the .proj links to .dll that expect the class to be structured a certain way?
When you use a framework or a pre-build solution, you have significant advantages. However, there is nothing in the World that has only advantages. Every solution/framework/library will come with its own limitations, you like or not.
With that in mind.. Despite of other vendors, we provide the FULL SOURCE CODE! User class or another schema class is not inside a CLOSED/SECRET DLL. They are on nuget packages and are open source as I described before. They are framework code. If you want to change the framework code it's always possible. Just download the ABP framework on Github repository, include into your solution, remove nuget packages and add project references. Then you can freely change the User class. Source code is open & here:
https://github.com/aspnetboilerplate/aspnetboilerplate/blob/dev/src/Abp.ZeroCore/Authorization/Users/AbpUser.cs
https://github.com/aspnetboilerplate/aspnetboilerplate/blob/dev/src/Abp.Zero.Common/Authorization/Users/AbpUserBase.cs
Each git repository is running off of a different versioning number
Why we follow the same version numbers for different products those are released and developed separately? Does Microsoft have same version number for all products?
I wish i had enough money as the next guy to afford throwing around $1600 a year just to resolve issues with poorly documented code, but sadly i do not
There are thousands of customers paid it and they are happy about our product and they are renewing their licenses to be able to access to latest source code after 1 year. If anybody wants a refund, we always provide a money back without any question. Refund request ratio is smaller than 0.5% because developers love our framework and product.
My membership expired while i was resolving conflict issues with the code, and i immediately lost the ability to use git repo connected to that service
You can contact info#aspnetzero.com and request the latest version of the code (when the time your license expired). We always help to our customers about that. It would be better if you had contact with us.
I have a few (3) core projects I want to share across many solutions (12+).
So, say I have 12 websites and they use some shared back end core code (in this case I'm not talking about shared js, css or views - I'm talking about business objects, entity stuff, etc.).
I need to be able to identify which site has which version of the shared code in dev, test, prod, etc. so a developer can get the website code and get the right version of the shared code to develop or patch the website.
And then the MS build server needs to know which version of the shared code to get for the deployment.
To solve this, I'm seeing people branch that core code - which seems absurd to do 12+ times. (I do expect to branch the core code sometimes for things like hot fixes and long running projects.)
I'm also seeing people copy DLLs of the core code and check those in.
I would think I would list the dependencies for my solutions based on TFS label names somewhere so developers can easily get the apps running with the right code and given a tfs label the build server can get the code for the website and the proper version of the core code. I'm using TFS & VS 2013 at the moment too, so there's that.
So, is there a way to do this that's straightforward, supportable/scale-able and intuitive? Thanks - Peter
Labels in TFS is very limited. For example once the label created you couldn't change and update it. If one of your core projects updated, did you need to create a new label for it. If you did and use the new label for one of your solution. However you found there are some bugs in this update, you need a newer update of your core project to fix the bug. Then a newer label created, you need to manually maintain the dependencies which seems not to be an easy job.
Moreover how to list the dependencies for your solutions based on TFS label names? TFS don't have this built-in option, seems the only way is store it in a txt or someother files and check in the source control. Every time the developer open a website application need to check it first and get label from server to their workspace and work on it.
Usually the purpose of sharing code between projects is reducing maintenance. There’s two main code sharing paths: source and binary. The difference between them you could take a look at this blog: Code Sharing in Team Foundation Server
Sharing code between products is a primary cause of quality erosion and elevated bug counts. I would recommend you to build separately and sharing binary output through NuGet which use preferable.
Also take a look below similar questions:
Sharing code between solutions in TFS
TFS 2010 Branch Across Team Projects - Best Practices
We have Team Foundation Server 2008 deployed as our source control management system. A team that is responsible for multiple products is asking for all their products to be put under a single TFS project. Their reason is because the products are all in a similar domain.
Here are my reasons against:
The workspace mappings will get weird, since projects will be mapped to subfolders
Continuous Integration may be a problem, since a single project can't be referenced
Tracking history of source control activity could be problematic
This just feels like an overall bad idea, but I would like some concrete reasons against it. If I'm completely off-base and this is a good approach to take, I'd like to hear that as well.
What are the pros/cons?
I have experience storing multiple Visual Studio Solutions (seperate products) under one TFS Team Project in both TFS2008 and TFS2010. Here is my take.
In both versions we create a folder for the Product, then a folder for the branches (Main, etc.) This makes it easy to see what product we are working on, and we can see the history of the product seperate from other products. Continuous integration works just fine with multiple build definitions, one for each product. We only create one workspace mapping for the entire TFS Team Project.
The shortfall in TFS2008 is that it can be difficult to manage work items for each Product. In TFS2008 the work items apply to the entire Team Project and it is not as easy as it should be to figure out which work item belongs to which product.
In TFS2010 the work items have an Areas and Iterations section. We use the Area to define the Product. So each Work Item gets an Area that matches the Product name. This has worked very well for us.
If you are not using work items heavily in TFS2008 than I don't think you should avoid putting multiple Products in one TFS Team Project, certinally not for the reasons you listed above.
Using one Team Project does haves some advantages:
1. There is ony one Team Project to manage and there is only one Share Point site.
2. You can see history across the entire Team Project easily.
My thoughts:
If there are assemblies shared amongst the projects, it makes sense to lump them together, otherwise you will run into the same problems that many people have discussed here, on how to handle shared assemblies.
You shouldn't encounter any problems with workspace mappings. Within our organization, we simply map $/ to a folder and go from there. Otherwise you could very easily map individual source control folders to different areas on disk. The only recommendation I would have is to put that mapping in a batch file, so that new members can run the batch and be consistent.
The only thing that you might lose out on a bit by lumping these all together is quick and easy reporting. If everything is in its own Team Project, the built-in reporting works "out of the box." If you put things together, you'll need to set up additional areas and iterations in order to do the reporting and tracking.
In our organization we have upward of 15 separate team projects, but every single one of them has more than one "product" underneath. We've been running this way for two years and really haven't had any problem with it, with the exception of the reporting.
Using a single Team Project for more than one software is a perfectly acceptable solution if you don't use separate templates for them. Martin Hinshelwood has a detailed blog post on the subject.
http://blog.hinshelwood.com/when-should-i-use-areas-in-tfs-instead-of-team-projects-in-team-foundation-server-2010/
This is a follow-on post to our previous post (Help Structuring VS2010 Solutions/Projects and TFS2010).
We have a few questions regarding how to structure our VS2010 solutions and projects for best organization, as well as to save in and use TFS2010.
Currently, our code is structured something like:
/OverallAppName
OverallAppName.sln
-/Client
- -/WindowsFormsProject1
WindowsFormsProject1.sln
- -/WindowsFormsProject2
WindowsFormsProject2.sln
-/Components
- -/ClassLibrary1 (common library referenced by other projects)
ClassLibrary1.sln
- -/ClassLibrary2
ClassLibrary2.sln
- -/ClassLibrary3
ClassLibrary3.sln
- -/ClassLibrary4
ClassLibrary4.sln
- -/ClassLibrary5
ClassLibrary5.sln
-/Server
- -/WindowsServiceProject1
WindowsServiceProject1.sln
- -/WindowsServiceProject2
WindowsServiceProject2.sln
- -/WebProject1
WebProject1.sln
- -/WebProject2
WebProject2.sln
Since, right now, we’re in the process of moving from VSS to TFS2010, we’re wanting to structure all our solutions/projects to be most efficient, most logical, easiest to maintain, easiest to reference, and easiest to use with and build in TFS2010, and we’re needing some advice on the “best” way to structure everything with a partitioned solution model.
Any suggestions????? How can we structure all these different types of VS2010 projects into a logical structure that separate groups can work on individual pieces (not the entire solution), we can still have project references, we can stored in TFS2010 and build and branch in there, and follow “recommended best practices”?
Thanks.
(Sorry, I'm not sure the formatting came out very good.)
While I admire your commitment to trying to keep everything as one large solution, I think this is going to defeat some of the best features TFS has to offer in the realm of automated builds by sticking to this.
The reason I say that is because you can use builds triggered by check-in to immediately build the code to prove it works (or better yet, use a Gated check-in). The usefulness of these builds are inversely proportional to the time they take to run. So if you have a massive solution that takes 20 minutes to build then it's going to take away from the advantages of those types of builds. If however you had several smaller solutions that took about 5 minutes each then you'll only get the modified solutions building on check-in and know the results sooner.
From what you've listed above I'd be inclined to have a solution for each set of artefacts that can be released separately. In your example that's probably one for each of the clients, one for each of the web applications and one for all of the common libraries.
Folder structure wise it'll not be much different to what you have above (assuming I'm interpreting it correctly)
/OverallApplication
/Clients
/Client1
-Client1.sln
/Client1Project1
-Client1Project1.csproj
/Client1Project2
-Client1Project1.csproj
...
...
/Components
-Components.sln
/ClassLibrary1
-ClassLibrary1.csproj
/ClassLibrary2
-ClassLibrary2.csproj
...
/Server
/WebApp1
-WebApp1.sln
/WebApp1Project1
-WebApp1Project1.csproj
/WebApp1Project2
-WebApp1Project1.csproj
...
...
/CodeSigningKey
-KeyPair.snk
/ReferencedAssemblies
/Manufacturer1
-Manufacturer1Assembly1.dll
...
...
The common libraries can still be added as project references in the server and client solutions. I've introduced a few new folders for common items such as the code signing key and 3rd party assemblies that are referenced (such as the Enterprise Library).
On top of that you'll want to employ a branching strategy of some kind to keep Main, Dev and Release code branches separate. I recommend a little light reading of the ALM Rangers branching guide on codeplex for that.
http://vsarbranchingguide.codeplex.com/releases
We are working on a big application, comprising around 100 projects (40 views, 40 controllers/models, 20 Utilities libraries). We have outsourced the bulk of the work and the deliverables come in fairly randomly.
When we get a deliverable (a project), we need to run FxCop, StyleCop, the associated unit-tests, etc, etc. before committing it to source control. To make this easier, we have mandated that every project has a solution file. This allows us to simply run an automated script on the solution file which tests it before checking it in.
My question is "Can you think of any drawbacks to having 1 solution for each project?".
Drawbacks we have already discussed include:
Additional maintenance required by developers. This doesn't bother us as we have outsourced development on a fixed-price contract.
SourceSafe bindings in solution file. This could have been a huge issue, but luckily we migrated to TFS about a year ago.
We're in a similar boat with about 200 projects, many of them common use, accross our various solutions of varying sizes.
While a disadvantage is load time, one advantage is debugging - i.e. if your code is calling into other assemblies, then it's nice to have everything in the same solution.
Also, we do keep our unit and integration tests along with our core project (DLL or EXE) all in the same solution, so even on a bare bones template we have at least three per solution.
Ultimately I'd say the largest advantage of a common solution boils down to cross-project debugging, IMO. - But I would never just toss them together unless there was at least this, or some other compelling reason.
On a side note - we do not allow a project without a solution for the reasons you noted above (running FxCop, etc.) plus Continuous Integration. One disadvantage of a large solution with several projects is build time - but it does help to know if messing with a component ended up breaking an unrelated solution/project.
We did discover one drawback in the end. All our projects are in Visual Studio 2008. If we want to upgrade one project to Visual Studio 2010, we realised we would almost certainly need to upgrade every other project (as they are all, directly or indirectly, either dependent on, or depended on, each other).