Comments
Sort by recent activity
You might also consider a snippet in SQL Prompt if you don't want something for every single object. This one will create a stub proc if it doesn't exist and then an ALTER will work. You could also change the logic to handle tables or use DROP IF EXISTS IF NOT EXISTS( SELECT name FROM sys.objects o WHERE o.name = '$myproc$')
CREATE PROCEDURE $myproc$
as begin select 1 end / comments
You might also consider a snippet in SQL Prompt if you don't want something for every single object. This one will create a stub proc if it doesn't exist and then an ALTER will work. You could also...
Kendra has a view on some of this here: https://www.youtube.com/watch?v=20xJTUokUxM&t=1s You can work with the objects, database by database, and certainly commit from each one. The challenge with deployment is understanding where the dependencies exist and which order in which to deploy changes. For builds, you likely want all databases o the build server, and then deploy to a new name for each to test the changes. / comments
Kendra has a view on some of this here: https://www.youtube.com/watch?v=20xJTUokUxM&t=1sYou can work with the objects, database by database, and certainly commit from each one. The challenge with d...
Not directly. The tools work by looking for changes, and there isn't a way to detect that this is a rename v a drop and add, at least not easily. When you import the migration script, that is the time to change this. I'd take this script: [image] And edit it to be this: [image] As a side note, renaming objects is an anti-pattern that you should avoid where possible. That being said, it is necessary sometimes. While SQL Prompt has a smart rename to find references and ensure things are renamed, for tables/columns, you need to manually fix the migration script.
/ comments
Not directly. The tools work by looking for changes, and there isn't a way to detect that this is a rename v a drop and add, at least not easily. When you import the migration script, that is the t...
Is this SQL Change Automation project in VS/SSMS or are you starting in a SQL Source Control project? / comments
Is this SQL Change Automation project in VS/SSMS or are you starting in a SQL Source Control project?
Second vote for David's strategy. With either the state (SQL Source Control) or migrations ( SQL Change Automation in VS or SSMS), there isn't an easy way to break out the features and ensure things work. This is really a VCS task, where you branch off with just certain changes and verify they will compile and deploy. / comments
Second vote for David's strategy. With either the state (SQL Source Control) or migrations ( SQL Change Automation in VS or SSMS), there isn't an easy way to break out the features and ensure thing...
In what way? Are you trying to tell the build system what migrations should be included? Or are you trying to add metadata lost build of what's been built? / comments
In what way? Are you trying to tell the build system what migrations should be included? Or are you trying to add metadata lost build of what's been built?
What's large? This is intended for lookup type datasets that tend to be hundreds or thousands of rows. When getting to GB of data, this is not the type of technology to use. / comments
What's large? This is intended for lookup type datasets that tend to be hundreds or thousands of rows. When getting to GB of data, this is not the type of technology to use.
OK, here's me Azure DevOps build step I use in most demos. I do have the agent running internally and just use AzDevOps for orchestration. [image] If I specify a server, then builds occur there. Do this is a point of contention for parallel builds.Keep this in mind if you have lots of parallel builds. Here is a demo build running right now. It's the dlmau_____ one.
[image]
I don't specify a db (leave blank), which means the SCA cmdlet will create a randomly named db and perform the build there. This is the preferred way, because if I need to run multiple builds, I can. If I specify a db name, then this database will be used. This is where parallel builds from different developers can collide. In this case, I set a name and started a build. Here's the config. [image] This db doesn't exist, so when I run the build, I see this.
[image]
Does this make sense?
/ comments
OK, here's me Azure DevOps build step I use in most demos. I do have the agent running internally and just use AzDevOps for orchestration. If I specify a server, then builds occur there. Do this is...
Be careful of using a specific database as this can cause issues with parallel builds / comments
Be careful of using a specific database as this can cause issues with parallel builds
Are thinking non-prod environments like an integration/qa/staging/etc environment? Or a dev environment? Part of what I would do for the former, if needed, is orchestrate this in the pipeline to clean out or drop the database. / comments
Are thinking non-prod environments like an integration/qa/staging/etc environment? Or a dev environment? Part of what I would do for the former, if needed, is orchestrate this in the pipeline to cl...