Repository files navigation

RepoDB is an open-source .NET ORM library that bridges the gaps of micro-ORMs and full-ORMs. It helps you simplify the switch-over of when to use the BASIC and ADVANCE operations during the development.

To get the latest updates about this library, follow us onTwitter!

Like with any other ORMs, RepoDB does provide the preliminarymethods needed for your basic operations. The good thing is, it also does provide the operations that is needed to cater your edge-cases like2nd-Layer Cache,Tracing,Repositories,Property Handlers andBatch/Bulk Operations.

If you are to useRepoDB, yourdevelopment experience is as simple as opening a connection and calling the advance operations with a very minimal code. It is the reason that makes this library the simpliest ORM to use.

When you do thebulk operations, the generated value of theidentity columns will be set back to the data models, just right after your execution. It is an important use-case that is/may needed by you and/or most developers, and both theBulkInsert andBulkMerge operations addressed this need.

RepoDB also does support the different way-of-executions (theatomic, thebatch and thebulk). Through this, it is easy for you to establish your repository that can process the smallest-to-the-largest datasets without even affecting the efficiency and the performance of your application.

Please click any of the link below to fast-track your learnings.

• SqlServer
• SqLite
• MySql
• PostgreSql

Or, learn a specific feature.

• Batch Operations
• Bulk Operations
• Caching
• Class Handlers
• Class Mapping
• Dynamics
• Connection Persistency
• Enumeration
• Expression Trees
• Hints
• Implicit Mapping
• Multiple Query
• Property Handlers
• Repositories
• Targeted Operations
• Tracing
• Transaction
• Type Mapping

Otherwise, please visit ourdocumentation page to learn more.

The execute methods below support all the RDBMS data providers.

• ExecuteQuery
• ExecuteNonQuery
• ExecuteScalar
• ExecuteReader
• ExecuteQueryMultiple

Whereas the fluent methods below only support theSQL Server,SQLite,MySQL andPostgreSQL RDBMS data providers.

• Query
• Insert
• Merge
• Delete
• Update

Clickhere to see all the operations.

By default, .NET is auto-resolving the references, however, we strongly recommend that you always explicitly reference theRepoDb core library. The rationale behind this is that, theRepoDb core library is a fast-moving package in which all the alpha/beta releases, hotfixes and/or even the actual releases could happen without affecting the extension libraries.

Please note that we are releasing an actual next released-version if the changes are having minimal impact but is important for the other users.

You can always target the version when installing the library, even it is on a semantic release.

>Install-Package RepoDb-version1.x.x-betaX

By default, RepoDB does not do the automatic .NET CLR Type conversion during the serialization and deserialization process. The coercion support is completely dependent to the ADO.NET coercion capability.

It is in purpose to strictly notify you (as a library user) the design and/or the implementation problem of the entity model if being compared to its corresponding database table/view.

If you wish to have an automatic conversion, simply set theConverter.ConversionType property toAutomatic.

RepoDb.Converter.ConversionType=ConversionType.Automatic;

Note: The exception that is being thrown is dependent to what the underlying ADO.NET coercion exception. If theAutomatic conversion is used, the extracted value will always be evaluated and an additional conversion logic will be used (if needed). The conversion logic is through the AOT compilation ofSystem.Linq.Expressions.Expression.Convert and/orSystem.Convert.

As the compiler exception is a bit low-level and is not descriptive for the native language, therefore, when compiling the process of hydration from/to the database and the application, a customized exception is being thrown to provide a detailed exception messages to the callers.

On the other hand, as part of the standard when writing code in RepoDB (i.e.: respect the default exception handling of .NET, ensure an unharmonized exception when bubbling up the exception messages), RepoDB does not contain a single line of code that catches and rethrowing any exception (try-catch statement). Any exception happens within the library whether it is an ADO.NET exception and/or whatever will be bubble up natively back to the callers.

If you are working with this package, you are required to bootstrap the connection object on the startup.

vardbSetting=newSqlServerDbSetting();DbSettingMapper.Add<System.Data.SqlClient.SqlConnection>(dbSetting,true);DbHelperMapper.Add<System.Data.SqlClient.SqlConnection>(newSqlServerDbHelper(),true);StatementBuilderMapper.Add<System.Data.SqlClient.SqlConnection>(newSqlServerStatementBuilder(dbSetting),true);

Or, you can replicate the actualSqlServerBootstrap class implementation and attach it to your solution. Then, call the local class initializer method explicitly.

ForRepoDb.SqlServer package, starting the versionv1.1.5-beta4, theMicrosoft.Data.SqlClient v4.0.0 is used, however, it seems to require a TLS 1.2 when connecting to the database via Integrated Security.

In most cases, an exception below is thrown if the mentioned security chain is not enabled.

Aconnection was successfully establishedwith the server,but then an error occurred during the login process.(provider: SSL Provider,error:0- The certificate chain was issued by an authority thatis not trusted.)--->System.ComponentModel.Win32Exception:The certificatechain wasissued byan authoritythatis nottrusted..

The issue above can be rectified by simply enabling the TLS 1.2. Alternatively, the argumentTrustServerCertificate can be used on the connection string.

Note: By enabling theTrustServerCertificate argument, as per Microsoft,the transport layer will use SSL to encrypt the channel and bypass walking the certificate chain to validate trust. Therefore, only enable this flag if needed.

It is very important for you and to the community of .NET to learn the things the library is capable and is-not capable of doing, so please spend time reading thelimitation page before using the library.

The benchmark result shown on this section is the result of the community-approved ORM bencher tool, theRawDataAccessBencher tool.

Below is the actual recent official executionresult.

RepoDB shows an impressive performance and memory-efficiency if being compared with other ORMs available in the .NET ecosystem. It has positioned itself just right behind the logic-less hand-coded materializer if being benchmark with .NET Core and .NET Framework. However, RepoDB is the fastest and the most-efficient ORM if being benchmark with .NET 5, even beating the hand-coded materializer.

The AOT compilation (IL/Expression) has some degree of performance impact, even just for milliseconds, therefore, if you are to materialize RepoDB, it is highly recommended to always eliminate the first execution.

To avoid the bias, you as well should exclude the first execution of the other ORMs during the benchmarking.

We would like to make RepoDB the mainstream hybrid-ORM library for .NET technology. Please help us build and realize the solution.

To contribute, please find afor-grabs item and issue a PR. Otherwise, you may create anew issue for us to look-at and discuss.

If you wish to contribute to the documentation site, it is hosted in theRepoDb.NET repository. Your expertise is needed to correct the forms, if needed.

Your biggest contribution is to utilize and share this library to the other developers.

• Blog it
• Discuss it
• Document it
• Share it
• Use it

Or, show your support by simply giving a ⭐ on this project.

Please get in touch with us via:

• GitHub - for any issues, requests and problems.
• StackOverflow - for any technical questions.
• Twitter - for the latest news.
• Gitter Chat - for direct and live Q&A.

Ensure to visit ourSupport Policy to get more details about our policies when handling the operational support for this library.

• Building the Solutions - let us build your copies.
• Coding Standards - let us be uniformed.
• Issuing a Pull-Request - let us be aligned and notified.
• Reporting an Issue - let us be organized for easy tracking and fixing.

Thanks to all thecontributors of this project, and toScott Hanselman forExploring the .NET open source hybrid ORM library RepoDB.

And also, thanks to these awesome OSS projects.

• AppVeyor - for the builds and test-executions.
• GitHub - for hosting this project.
• Gitter - for the community engagements.
• Jekyll - for powering our website.
• Just-the-Docs - for being the awesome library documentation template.
• Moq - for being the tests mocking framework.
• Nuget - for the package deliveries.
• RawDataAccessBencher - for measuring the performance and efficiency.
• SharpLab - for helping us on our IL coding.
• Shields - for the awesome badges.
• StackEdit - for being the markdown file editor.
• Microsoft.Data.Sqlite,System.Data.SQLite.Core,MySql.Data,MySqlConnector,Npgsql - for being the extended DB provider drivers.

Apache-2.0 - Copyright © 2019 -Michael Camara Pendon