I am planning to create a web application using ASP.NET MVC4 using the code first approach.
I am using Visual Studio 2012 and I have SQL Server 2012. What I would like to do is setup the solution in a way that I can use SQL Server instead of the default being SQL Express or localDB. I would like to have the solution regenerate the database if any changes are made to the models and seed the database with some test data when that occurs.
I have gone through a bunch of tutorials and they all seem to be using the SQL Express or LocalDB database. I was able to get the solution to generate the database on SQL Server the first time it was run, but after I make changes to the model such as alter the schema name from 'dbo' to lets say 'test' or change the table name, it doesn't seem to pick up those changes.
Can anyone give me some insight on how I might be able to accomplish this?
Thanks
All you need to do is have your connection string in the web.config pointing to your SQL server for it to use that server instance.
As for having it update on model change, have a look into DropCreateDatabaseIfModelChanges.
In your Global.asax you can set your database initializer like this
Database.SetInitializer(new DropCreateDatabaseIfModelChanges<DbContext>());
You will need a using statement too
using System.Data.Entity;
There is a free video guide to MVC4 by Scott Allen which can give you more info on Migrations and Code First (which is what you are asking about). The 4th section of the video guide has a 13 minute segment on Data Migrations which is a full walk through. If you can spare the time, watch that for a full detailed guide.http://www.asp.net/mvc/videos/pluralsight-building-applications-with-aspnet-mvc-4
I might point out this is only useful in development and there is no way to make it work for production as you would lose your data. If you need something that is suitable for a production environment look into SSDT (SQL Server Data Tools). You can manage your database as a SQL project in your solution and deploy updates using a DACPAC file. The file manages all the updates in production and can be set to protect against data loss. You can change and deploy your Db from your project and still use EF with database first and EF models etc. All you do is make your changes in the Db project and hit debug to update your test Db. When publishing you can choose to export the DACPAC for updating live.
Related
I have created a functioning LightSwitch (Visual studio 2012 Pro) application...
Now it's time to publish! The only problem is that I can't seem to figure out how to change the schema name to something other than dbo.
That is, I don't want the output tables to be dbo.XXXX, I want them to be my_schema.XXXX instead.
Also, note that the database is on SQL Server 2005.
Does anyone have suggestions on how to achieve this?
Thanks!
Update: Sorry for not being specific. I don't want to change the schema for existing tables in my database. I want my LightSwitch application to use the new schema. I don't mind publishing it and then modifying the schema on the tables, if I can then somehow also modify the LightSwitch application to use the new schema. But ideally, there would be a way to tell LightSwitch to use a different schema when it publishes.
For those who don't know what LightSwitch is, it is a Microsoft .Net Rapid Application Development tool for SilverLight business applications, and is built into Visual Studio 2012. My application is written in VB.Net
You can't change the schema name for an intrinsic database (the database that is created with LightSwitch). Maybe you'll be able to in V3, but you can't currently do it in V1 or V2. To achieve this, you'll have to create the database in something like SSMS (SQL Server Management Studio, then attach to it as an external data source.
Seeing as you've already done all the creating in LightSwitch, you could publish it as is, then attach to it as I mentioned above. I hope you don't have screens created, because unless you want to manually edit all of the lsml, you won't be able to use the attached database in any already created screens.
Any time you need to do something "out of the ordinary" with LightSwitch's database, you need to create it outside of LightSwitch, then attach to it. Why? LightSwitch was originally created with the idea in mind that the users didn't need to know anything about SQL Server.
I've started working on my database desktop application. The goal is to have one centralized database, so that each employee could connect.
My goals:
Centralized database on SQL Server 2008 Express
Use Sql to Linq
First run of application should create structure of database on server.
Now my question: how to do the last point?
I know that I can generate sql script via SQL Managment Studio and then execute it, but maybe there is some better way?
Like the comments said, I'm not sure why you would want to do this, but if you have one centralized database machine and you want to create a new database per user upon the first connection, then you could create one canonical/template database and the clone it via scripting on the first connect.
Your UI would say something like "Please wait while we create your new database..." and you'd run CREATE scripts for the database/tables/stored procedures/views/etc.
Just remember it might be a major pain to update the schemas of these databases once the cat is out of the bag.
Here's task: We have an sql server database. which is hosted at our server. What we need to do is: we need to create a non-techy-users interface (basically insert/edit forms) and let these non-techy-users to install this database locally, since they are located in the areas without internet connection. Then when they're done using the database we get the data from them and inster it in our database.
The biggest concern is that it is not trivial for non-it people to install sql server. Can you please advise me what solution should I choose? Simple Access should work fine, but i really do not want to mess with it and have data conversion back and forth between engines.
Sync Framework for SQL Server: your application uses a lite weight, embedded SQL Server CE (no installation, just a couple of DLLs deployed along with your app) and the sync framework manages the synchronization with the 'mother ship' SQL Server.
Out of interest, why do they need their own installation? Can't you create a new database on your existing instance?
If you're looking for an easy way to create insert/edit forms on your database, have you considered looking at Microsoft's new LightSwitch product (currently in Beta) or Microsoft's Dynamic Data?
i'm currently working on a asp.mvc application with entity framework as db backend which will be running on ms azure platform.
on my development machine i'm running a sqlexpress instance which hosts my development database (like i said may app connects via entity framework to it).
deploying the database schema from my local sqlexpress to sql azure is pretty easy via the entity framework database generation wizard (for whatever reasons the wizard always wants to create some clustered indexes which i correct manually in the generated sql file).
but i can't figure out a way to keep my data! the auto generated sql script always dumps all my tables and creates new ones...thats ok for initial setup, but not to upgrade a existing database.
there must be a nice way to perform a schema update without dataloss...please help! i have already tried sql management studio (r2) and SQLAzureMW (available on codeplex)...but they don't do the job :(
please don't tell me i have to code my own tool to do that!
looking for your help
thx
Edit: here's how i do it now
I use the SQL scripts created by the EF migration wizard for my local SQL Express and modify them myself to be compatible with SQL Azure. Is less work that one might think and works perfect :)
Have you tried to use BCP to export and then import your data? I believe you should be able to use BCP to export data from your SQL Express instance into a file, and then import data from the file into SQL Azure. The SQL Azure team has a blog posting that describes using BCP with SQL Azure - Link.
Additionally, in the future, the upcoming release of SQL Azure Data Sync CTP2 might be able to help you out.
Out of curiosity, what problems where you having with SQLAzureMW?
RedGate is famous for SQL server synchronization and data synchronization for deployment purposes.
They have a v9 beta out of their DataCompare and SchemaCompare products that are compatible with SQL Azure. I've been using both successfully and like them alot.
They are pricey when purchased, if you are a small business.
http://www.red-gate.com/Azure
This sounds vary similar to this question, the only answer to which is "wait for the next version of entity framework". I'll admit that I don't use EF, but I make my changes to tables the old school way with SQL Server Management Studio and scripts. ALTER TABLE FTW.
EDIT: If you don't have access to SQL Server Management studio, the new version of the SQL Azure portal includes a Silverlight based application that allows you to run queries and scripts
hope that everybody here is OK.
We are using VS 2008 as development tool, TFS 2008 as version control as well as build automation. Some of our developer use dbpro for databases changes and some use SQL Server management studio.
I am trying to automate build for Web Application built using C# and VB.Net.
Our scenario is such that we have a central database to which our web application connects.
Whenever we supply our clients with a new functionality or a bug fix, we supply them incremental builds.
The SQL script is checked into source control for every incremental build when they have made and tested there changes on our central DB Server.
I want to generate Differential script that can be run at the client as an incremental update script. Now to come about it is a problem. Sometimes our developers tend to forget the database change-sets and the script in the source control is missing an SP or a two.
Also, sometimes we need to insert default data into some of the tables that have strict stringent values and not test values. Like a table that contains Services provided by the panel, we add a new service name, signature, credentials and service address, etc etc in the ServiceTable. Besides this many other tables may have test data that may not be needed.
If we use DataCompare, it will generate changeset for required data (important for client to enable certain services) and our test data that was added to the database as a result of our testing of the functionality or bug fix.
Currently i am using SQLSchemaCompareTask (from Visual Studio 2008 Team Database Professional Power Tools API) in the TFSBuild.proj file of the build definition for TFS 2008.
Using SQLSchemaCompareTask, the script generated contains database names like [dbo]. etc which are not desired as the script fails when run against SQL Server 2000 databses (Some of our client still use SQL Server 2000) databases as teh backend of the application.
Also default data can't be generated by this process.
To overcome this problem, i have to come up with a solution that can compare databases and generate script automatically that does not have to be manually reviewed again before being sent to the client.
Please suggest effective methodology of such SQL script generation and suggest whether two different databases may be used or something ? Is there any toolkit or api that can enable build automation for SQL Server databases?
Thank you all.
Regards
Steve
Try to use SQL Examiner Suite for this:
http://www.sqlaccessories.com/SQL_Examiner_Suite/
The tool compares both schema and data and produces synchronization scrips (or differentials scripts). You can automate script creation with supplied command-line tool.
Rather than collating many individual change set scripts (and therefore occasionally missing objects out), why not use schema compare and data compare to create a single script from your database project using a database equivalent to your client's on the target? This should create a script tailored to their requirements.
In data compare you can exclude test data records that you don't want pushed to your client by unchecking them in the lower grid.