AI Instructions

## Data access: ARX.SDK

This project uses ARX.SDK 1.0.0-rc.1 with MongoDB (package `ARX.SDK.MongoDb`, service `ArxMongoDbService`).

- Use the `arx-sdk` skill for any code that reads or writes the database. Its references for this provider are in
  `.claude/skills/arx-sdk/references/mongodb/`: read `arx-mongodb-guide.md` and `arx-mongodb-types.md` first.
- Use only the ARX types and members listed in `arx-mongodb-types.md`. If something isn't there, say ARX doesn't
  offer it; don't invent a name.
- Start from the closest verified sample in the operation's file (`arx-mongodb-create.md` and the others).
- After changing ARX code: `dotnet build`, run the tests, and check every ARX call against the types file.
- When the ARX.SDK package version changes, download the skill again and update the version above.

## Data access: ARX.SDK

This project uses ARX.SDK 1.0.0-rc.1 with MongoDB (package `ARX.SDK.MongoDb`, service `ArxMongoDbService`). Its knowledge
files, generated from the library and its verified samples, are in `docs/arx-sdk/references/mongodb/`; the files named
below are there. Read `arx-mongodb-guide.md` and `arx-mongodb-types.md` before writing data access code,
and the operation's file (`arx-mongodb-create.md`, `-create-multiple.md`, `-update.md`, `-update-multiple.md`,
`-delete.md`, `-delete-multiple.md`, `-retrieve.md`, `-retrieve-multiple.md`, `-execute.md`) for its samples.

You help a developer write C# code that uses ARX.SDK with MongoDB (package `ARX.SDK.MongoDb`). Follow these rules.

1. **Use ARX's types for database work.** Create, update, delete and read documents, and run transactions through
   `ArxMongoDbService`, `ArxMongoConnection` and `ArxTransaction`. Don't write driver code (`IMongoCollection`,
   `BsonDocument`, `FilterDefinition`) for work ARX does.
2. **Don't invent API.** Use only the types, methods, properties and parameters listed in `arx-mongodb-types.md`. If
   something isn't there, say ARX doesn't offer it rather than guessing a name.
3. **A document is an `ArxEntity`.** `var account = new ArxEntity("account");` then values by field name:
   `account["name"] = "Northwind Traders";`. The entity's name is the collection. Field names keep their case.
4. **Declare every collection's id once, at startup,** in an `ArxSchemaCatalog`, and give it to the service:
   `ArxSchemaCatalog.Create().Table("account", ArxIdColumn.Generated<string>("account_id")).Build()`, then
   `new ArxMongoDbService(connectionString, schema)`; the connection string names the database
   (`mongodb://host:27017/crm`). The declared id is the document's `_id`: a generated string id is an ObjectId ARX makes
   (returned as its 24-character text), a generated Guid one ARX makes, `ArxIdColumn.Supplied<string>("code")` an id the
   application gives. A generated number id is ARX8001 (no sequences) and `NoIdColumn` is ARX1001 (every document has
   an `_id`). Add business keys with `.Key("account", "account_number")` and back them with unique indexes.
5. **Lookups are declared**, since MongoDB has no foreign keys: `.Lookup("contact", "account_id", "account")`. Write one
   with `new ArxEntityReference("account", id)`; it reads back as an `ArxEntityReference` with the account's display
   name (`.DisplayName("account", "name")`).
6. **Every call is asynchronous** (`CreateAsync`, `CreateMultipleAsync`, `UpdateAsync`, `UpdateMultipleAsync`,
   `DeleteAsync`, `DeleteMultipleAsync`, `RetrieveAsync`, `RetrieveMultipleAsync`); await it, and pass the caller's
   `CancellationToken`. Make one service for the application: it shares one MongoDB client per connection string.
7. **Writes of several documents are all or nothing** (CreateMultiple, UpdateMultiple, DeleteMultiple with checks), in
   a transaction: MongoDB needs a replica set or a sharded cluster for that (one replica set node is enough). On a
   standalone server they are ARX8001.
8. **Update and Delete find documents by the id, a declared key, or fields you name:**
   `new ArxUpdateOptions { Match = ArxMatch.Key("account_number") }`, `ArxMatch.Columns(...)`. An update writes only the
   attributes that are set (`$set`); null stores null. A document's `_id` can't change (ARX8001).
   **To change or delete every document a condition matches** (from 0.1.0-preview.5): an entity with only the values to set
   and an `ArxFilterExpression`, `int changed = await service.UpdateMultipleAsync(values, filter);`, or
   `await service.DeleteMultipleAsync("account", filter)`; one `updateMany` or `deleteMany` in a transaction (a replica set), all or nothing, returning the count. The
   entity carries no id (ARX1001 with a filter); an empty filter is ARX1001 unless
   `new ArxFilterWriteOptions { AllRows = true }`. For a large insert whose ids aren't needed:
   `CreateMultipleAsync(entities, new ArxCreateMultipleOptions { ReturnIds = false })` (every id `ArxRecordId.None`).
9. **Reads.** `RetrieveAsync` returns `null` when nothing matches. `RetrieveMultipleAsync` takes an
   `ArxQueryExpression` (criteria, columns, order, paging, joins as link entities, aggregates) and returns the
   documents in `.Entities`. `Null` matches a stored null and a missing field; NotEqual never matches either. Page by
   `PageInfo.PageNumber`, or pass each page's `PagingCookie` to the next query. Execute is ARX8001: MongoDB has no
   stored procedures.
   **Query cheat sheet** (the exact names; there are no others, such as `AddGroupBy` or `MoreRecords`):
   `var query = new ArxQueryExpression("book");` · columns: `query.ColumnSet = new ArxColumnSet("title", "year");`
   (empty: every column) · conditions: `query.Criteria.AddCondition("year", ArxConditionOperator.Between, 2000, 2010);`
   (Equal, NotEqual, GreaterThan, GreaterEqual, LessThan, LessEqual, Between, In, NotIn, Like, Contains, BeginsWith,
   EndsWith, Null, NotNull and their Not/DoesNot forms) · an OR group: `var either = query.Criteria.AddFilter(ArxLogicalOperator.Or);`
   then `either.AddCondition(…)` · order: `query.AddOrder("title", ArxOrderType.Ascending);`
   · a join: `var author = query.AddLink("author", "author_id", "id", ArxJoinOperator.Inner); author.EntityAlias = "a";
   author.ColumnSet = new ArxColumnSet("name");`, read as `row["a.name"]` · paging:
   `query.PageInfo = new ArxPagingInfo { Count = 20, PageNumber = 1, ReturnTotalRecordCount = true };`, then
   `result.Entities`, `result.HasNextPage`, `result.TotalRecordCount` (or page by `result.PagingCookie`) · grouped counts:
   `query.AggregateInfo = new ArxAggregateInfo(); query.AggregateInfo.AddAggregateColumn("books", "id", ArxAggregateOperator.Count);
   query.AggregateInfo.AddGroupByColumn("author_id");` (Count reads back as a `long`). A foreign key column reads back as
   an `ArxEntityReference` (with the referenced row's name) only when the catalog declares it:
   `.Lookup("book", "author_id", "author")` and `.DisplayName("author", "name")`; otherwise it is the plain id.
10. **Values are never run as queries:** pass them as .NET values. A name with `$` or SQL syntax is ARX1017, so names
    that carry operators are refused, but a valid name is used as given and reaches any collection or field the database
    user can read: map a name from a user to one the application allows (an allow list), or preload the schema and set
    `Strict = true`. Search text is escaped.
11. **Dates:** a BSON date holds milliseconds, in UTC: a `DateTime` is stored as UTC (Local converted, Unspecified taken
    as UTC), a `DateTimeOffset` as its instant (the offset isn't kept), a `DateOnly` as midnight UTC. Never
    `default(DateTime)` (ARX1013: use `null`).
12. **Errors** are `ArxException`s with a stable `Code`: catch by code
    (`catch (ArxException error) when (error.Code == ArxErrorCode.UniqueViolation)`: catch the base class, since the class tells where an error was found and the code what happened) and retry only when
    `error.IsTransient` is true. `arx-mongodb-error-codes.md` lists every code. Log `error.Message`, not the driver's:
    MongoDB's duplicate key message quotes the value.
13. **Transactions:** `await using var connection = new ArxMongoConnection(service);`,
    `await using var transaction = await connection.BeginTransactionAsync();`, pass `transaction` to each call, then
    `await transaction.CommitAsync();`. Disposing without committing rolls back. Code written against `IArxDataService` opens one without the provider type: `await using var transaction = await service.BeginTransactionAsync();` (from 0.1.0-preview.5; the transaction owns its connection). MongoDB has no savepoints: a call that
    fails at the database aborts the transaction, and batch scopes have no savepoints: leave `Savepoints` unset (`true` is ARX8001), and a failed call rolls the scope back.
14. **Base answers on the samples** in the operation files: they are verified code. Say which sample a pattern comes
    from when that helps the developer find more.
15. **Reading values and namespaces.** `entity["col"]` is the value or null; a misspelt or unread attribute
   reads null too, so take column names from the table. `GetAttributeValue<T>` of NULL is `default(T)` (0 for `int`:
   ask for `int?` to tell NULL from 0); it converts numbers when nothing is lost (a Count is always a `long`, Sum and
   Avg are decimal or double by provider: read them as `long` and `decimal`), and reads a declared lookup (an
   `ArxEntityReference`, also in a grouped query) as its id, and a UTC or Local `DateTime` as a `DateTimeOffset` (and
   back, as UTC); a number that doesn't fit is ARX4001 (1.5 as `int`), any other mismatch is ARX4004, an `ArxException`.
   Through a link a lookup column is its plain id, not an `ArxEntityReference`.
   Namespaces: `ArxEntity`, `ArxEntityReference`, `ArxColumnSet`, `ArxSchemaCatalog` and `ArxMatch` are in
   `ARX.SDK.Services`; `ArxQueryExpression`, `ArxConditionOperator`, `ArxJoinOperator`, `ArxAggregateOperator` and
   `ArxOrderType` in `ARX.SDK.Query`; `ArxException` and `ArxErrorCode` in `ARX.SDK`; `ArxMongoDbService` and `ArxMongoConnection` in `ARX.SDK.MongoDb.Services`. Write every `using` a file needs.
   Names that differ from the Dynamics XRM SDK: the join operators are `ArxJoinOperator.Inner` and `ArxJoinOperator.Left` (not XRM's `LeftOuter`);
   a link's columns are `link.ColumnSet`, its alias `link.EntityAlias`, its conditions `link.LinkCriteria`; the referenced
   table of an `ArxEntityReference` is `EntityName` and its id `RecordId`. `UpdateAsync` throws ARX2002 RecordNotFound
   when no record matches (`DeleteAsync` returns false or 0 instead). A page's `TotalRecordCount` is an `int`; an aggregate
   Count is a `long`.
16. **Ids and linked columns are plain values.** `entity.RecordId` is an `ArxRecordId`: read it with
    `AsInt64()`, `AsGuid()` or `AsString()`, pass it where an id goes (`RetrieveAsync("account", id)`,
    `new ArxEntityReference("account", entity.RecordId!)`), or pass the plain value instead. `ArxRecordId` is a class, not
    a nullable struct: `RecordId` is null when no id is known, and `.Value` is the id as an `object` (pass the
    `ArxRecordId` itself, or `AsInt64()`, where a typed id goes, never `.Value`). A linked table's column reads back as `row["alias.column"]`, the plain value (no `AliasedValue`); without an alias the prefix is the table's name (`row["account.name"]`). Table names
    are matched exactly (`"Account"` isn't `account`: ARX1016); column names ignoring case.

Before finishing: `dotnet build`, run the tests, and check every ARX call against `arx-mongodb-types.md`. When the
ARX.SDK package version changes, replace the knowledge files with the new version's and update this section.

---
applyTo: "**/*.cs"
---

# ARX.SDK data access

This project uses ARX.SDK 1.0.0-rc.1 with MongoDB (package `ARX.SDK.MongoDb`, service `ArxMongoDbService`). Its knowledge
files are in `docs/arx-sdk/references/mongodb/`; the files named below are there.

You help a developer write C# code that uses ARX.SDK with MongoDB (package `ARX.SDK.MongoDb`). Follow these rules.

1. **Use ARX's types for database work.** Create, update, delete and read documents, and run transactions through
   `ArxMongoDbService`, `ArxMongoConnection` and `ArxTransaction`. Don't write driver code (`IMongoCollection`,
   `BsonDocument`, `FilterDefinition`) for work ARX does.
2. **Don't invent API.** Use only the types, methods, properties and parameters listed in `arx-mongodb-types.md`. If
   something isn't there, say ARX doesn't offer it rather than guessing a name.
3. **A document is an `ArxEntity`.** `var account = new ArxEntity("account");` then values by field name:
   `account["name"] = "Northwind Traders";`. The entity's name is the collection. Field names keep their case.
4. **Declare every collection's id once, at startup,** in an `ArxSchemaCatalog`, and give it to the service:
   `ArxSchemaCatalog.Create().Table("account", ArxIdColumn.Generated<string>("account_id")).Build()`, then
   `new ArxMongoDbService(connectionString, schema)`; the connection string names the database
   (`mongodb://host:27017/crm`). The declared id is the document's `_id`: a generated string id is an ObjectId ARX makes
   (returned as its 24-character text), a generated Guid one ARX makes, `ArxIdColumn.Supplied<string>("code")` an id the
   application gives. A generated number id is ARX8001 (no sequences) and `NoIdColumn` is ARX1001 (every document has
   an `_id`). Add business keys with `.Key("account", "account_number")` and back them with unique indexes.
5. **Lookups are declared**, since MongoDB has no foreign keys: `.Lookup("contact", "account_id", "account")`. Write one
   with `new ArxEntityReference("account", id)`; it reads back as an `ArxEntityReference` with the account's display
   name (`.DisplayName("account", "name")`).
6. **Every call is asynchronous** (`CreateAsync`, `CreateMultipleAsync`, `UpdateAsync`, `UpdateMultipleAsync`,
   `DeleteAsync`, `DeleteMultipleAsync`, `RetrieveAsync`, `RetrieveMultipleAsync`); await it, and pass the caller's
   `CancellationToken`. Make one service for the application: it shares one MongoDB client per connection string.
7. **Writes of several documents are all or nothing** (CreateMultiple, UpdateMultiple, DeleteMultiple with checks), in
   a transaction: MongoDB needs a replica set or a sharded cluster for that (one replica set node is enough). On a
   standalone server they are ARX8001.
8. **Update and Delete find documents by the id, a declared key, or fields you name:**
   `new ArxUpdateOptions { Match = ArxMatch.Key("account_number") }`, `ArxMatch.Columns(...)`. An update writes only the
   attributes that are set (`$set`); null stores null. A document's `_id` can't change (ARX8001).
   **To change or delete every document a condition matches** (from 0.1.0-preview.5): an entity with only the values to set
   and an `ArxFilterExpression`, `int changed = await service.UpdateMultipleAsync(values, filter);`, or
   `await service.DeleteMultipleAsync("account", filter)`; one `updateMany` or `deleteMany` in a transaction (a replica set), all or nothing, returning the count. The
   entity carries no id (ARX1001 with a filter); an empty filter is ARX1001 unless
   `new ArxFilterWriteOptions { AllRows = true }`. For a large insert whose ids aren't needed:
   `CreateMultipleAsync(entities, new ArxCreateMultipleOptions { ReturnIds = false })` (every id `ArxRecordId.None`).
9. **Reads.** `RetrieveAsync` returns `null` when nothing matches. `RetrieveMultipleAsync` takes an
   `ArxQueryExpression` (criteria, columns, order, paging, joins as link entities, aggregates) and returns the
   documents in `.Entities`. `Null` matches a stored null and a missing field; NotEqual never matches either. Page by
   `PageInfo.PageNumber`, or pass each page's `PagingCookie` to the next query. Execute is ARX8001: MongoDB has no
   stored procedures.
   **Query cheat sheet** (the exact names; there are no others, such as `AddGroupBy` or `MoreRecords`):
   `var query = new ArxQueryExpression("book");` · columns: `query.ColumnSet = new ArxColumnSet("title", "year");`
   (empty: every column) · conditions: `query.Criteria.AddCondition("year", ArxConditionOperator.Between, 2000, 2010);`
   (Equal, NotEqual, GreaterThan, GreaterEqual, LessThan, LessEqual, Between, In, NotIn, Like, Contains, BeginsWith,
   EndsWith, Null, NotNull and their Not/DoesNot forms) · an OR group: `var either = query.Criteria.AddFilter(ArxLogicalOperator.Or);`
   then `either.AddCondition(…)` · order: `query.AddOrder("title", ArxOrderType.Ascending);`
   · a join: `var author = query.AddLink("author", "author_id", "id", ArxJoinOperator.Inner); author.EntityAlias = "a";
   author.ColumnSet = new ArxColumnSet("name");`, read as `row["a.name"]` · paging:
   `query.PageInfo = new ArxPagingInfo { Count = 20, PageNumber = 1, ReturnTotalRecordCount = true };`, then
   `result.Entities`, `result.HasNextPage`, `result.TotalRecordCount` (or page by `result.PagingCookie`) · grouped counts:
   `query.AggregateInfo = new ArxAggregateInfo(); query.AggregateInfo.AddAggregateColumn("books", "id", ArxAggregateOperator.Count);
   query.AggregateInfo.AddGroupByColumn("author_id");` (Count reads back as a `long`). A foreign key column reads back as
   an `ArxEntityReference` (with the referenced row's name) only when the catalog declares it:
   `.Lookup("book", "author_id", "author")` and `.DisplayName("author", "name")`; otherwise it is the plain id.
10. **Values are never run as queries:** pass them as .NET values. A name with `$` or SQL syntax is ARX1017, so names
    that carry operators are refused, but a valid name is used as given and reaches any collection or field the database
    user can read: map a name from a user to one the application allows (an allow list), or preload the schema and set
    `Strict = true`. Search text is escaped.
11. **Dates:** a BSON date holds milliseconds, in UTC: a `DateTime` is stored as UTC (Local converted, Unspecified taken
    as UTC), a `DateTimeOffset` as its instant (the offset isn't kept), a `DateOnly` as midnight UTC. Never
    `default(DateTime)` (ARX1013: use `null`).
12. **Errors** are `ArxException`s with a stable `Code`: catch by code
    (`catch (ArxException error) when (error.Code == ArxErrorCode.UniqueViolation)`: catch the base class, since the class tells where an error was found and the code what happened) and retry only when
    `error.IsTransient` is true. `arx-mongodb-error-codes.md` lists every code. Log `error.Message`, not the driver's:
    MongoDB's duplicate key message quotes the value.
13. **Transactions:** `await using var connection = new ArxMongoConnection(service);`,
    `await using var transaction = await connection.BeginTransactionAsync();`, pass `transaction` to each call, then
    `await transaction.CommitAsync();`. Disposing without committing rolls back. Code written against `IArxDataService` opens one without the provider type: `await using var transaction = await service.BeginTransactionAsync();` (from 0.1.0-preview.5; the transaction owns its connection). MongoDB has no savepoints: a call that
    fails at the database aborts the transaction, and batch scopes have no savepoints: leave `Savepoints` unset (`true` is ARX8001), and a failed call rolls the scope back.
14. **Base answers on the samples** in the operation files: they are verified code. Say which sample a pattern comes
    from when that helps the developer find more.
15. **Reading values and namespaces.** `entity["col"]` is the value or null; a misspelt or unread attribute
   reads null too, so take column names from the table. `GetAttributeValue<T>` of NULL is `default(T)` (0 for `int`:
   ask for `int?` to tell NULL from 0); it converts numbers when nothing is lost (a Count is always a `long`, Sum and
   Avg are decimal or double by provider: read them as `long` and `decimal`), and reads a declared lookup (an
   `ArxEntityReference`, also in a grouped query) as its id, and a UTC or Local `DateTime` as a `DateTimeOffset` (and
   back, as UTC); a number that doesn't fit is ARX4001 (1.5 as `int`), any other mismatch is ARX4004, an `ArxException`.
   Through a link a lookup column is its plain id, not an `ArxEntityReference`.
   Namespaces: `ArxEntity`, `ArxEntityReference`, `ArxColumnSet`, `ArxSchemaCatalog` and `ArxMatch` are in
   `ARX.SDK.Services`; `ArxQueryExpression`, `ArxConditionOperator`, `ArxJoinOperator`, `ArxAggregateOperator` and
   `ArxOrderType` in `ARX.SDK.Query`; `ArxException` and `ArxErrorCode` in `ARX.SDK`; `ArxMongoDbService` and `ArxMongoConnection` in `ARX.SDK.MongoDb.Services`. Write every `using` a file needs.
   Names that differ from the Dynamics XRM SDK: the join operators are `ArxJoinOperator.Inner` and `ArxJoinOperator.Left` (not XRM's `LeftOuter`);
   a link's columns are `link.ColumnSet`, its alias `link.EntityAlias`, its conditions `link.LinkCriteria`; the referenced
   table of an `ArxEntityReference` is `EntityName` and its id `RecordId`. `UpdateAsync` throws ARX2002 RecordNotFound
   when no record matches (`DeleteAsync` returns false or 0 instead). A page's `TotalRecordCount` is an `int`; an aggregate
   Count is a `long`.
16. **Ids and linked columns are plain values.** `entity.RecordId` is an `ArxRecordId`: read it with
    `AsInt64()`, `AsGuid()` or `AsString()`, pass it where an id goes (`RetrieveAsync("account", id)`,
    `new ArxEntityReference("account", entity.RecordId!)`), or pass the plain value instead. `ArxRecordId` is a class, not
    a nullable struct: `RecordId` is null when no id is known, and `.Value` is the id as an `object` (pass the
    `ArxRecordId` itself, or `AsInt64()`, where a typed id goes, never `.Value`). A linked table's column reads back as `row["alias.column"]`, the plain value (no `AliasedValue`); without an alias the prefix is the table's name (`row["account.name"]`). Table names
    are matched exactly (`"Account"` isn't `account`: ARX1016); column names ignoring case.

Check every ARX call against `arx-mongodb-types.md`, and build before finishing.

---
description: Writing or changing data access code that uses ARX.SDK (ARX.SDK.MongoDb)
globs: **/*.cs
alwaysApply: false
---

# ARX.SDK data access

This project uses ARX.SDK 1.0.0-rc.1 with MongoDB (package `ARX.SDK.MongoDb`, service `ArxMongoDbService`). Its knowledge
files are in `docs/arx-sdk/references/mongodb/`; the files named below are there. Read `arx-mongodb-guide.md` and
`arx-mongodb-types.md` before writing data access code.

You help a developer write C# code that uses ARX.SDK with MongoDB (package `ARX.SDK.MongoDb`). Follow these rules.

1. **Use ARX's types for database work.** Create, update, delete and read documents, and run transactions through
   `ArxMongoDbService`, `ArxMongoConnection` and `ArxTransaction`. Don't write driver code (`IMongoCollection`,
   `BsonDocument`, `FilterDefinition`) for work ARX does.
2. **Don't invent API.** Use only the types, methods, properties and parameters listed in `arx-mongodb-types.md`. If
   something isn't there, say ARX doesn't offer it rather than guessing a name.
3. **A document is an `ArxEntity`.** `var account = new ArxEntity("account");` then values by field name:
   `account["name"] = "Northwind Traders";`. The entity's name is the collection. Field names keep their case.
4. **Declare every collection's id once, at startup,** in an `ArxSchemaCatalog`, and give it to the service:
   `ArxSchemaCatalog.Create().Table("account", ArxIdColumn.Generated<string>("account_id")).Build()`, then
   `new ArxMongoDbService(connectionString, schema)`; the connection string names the database
   (`mongodb://host:27017/crm`). The declared id is the document's `_id`: a generated string id is an ObjectId ARX makes
   (returned as its 24-character text), a generated Guid one ARX makes, `ArxIdColumn.Supplied<string>("code")` an id the
   application gives. A generated number id is ARX8001 (no sequences) and `NoIdColumn` is ARX1001 (every document has
   an `_id`). Add business keys with `.Key("account", "account_number")` and back them with unique indexes.
5. **Lookups are declared**, since MongoDB has no foreign keys: `.Lookup("contact", "account_id", "account")`. Write one
   with `new ArxEntityReference("account", id)`; it reads back as an `ArxEntityReference` with the account's display
   name (`.DisplayName("account", "name")`).
6. **Every call is asynchronous** (`CreateAsync`, `CreateMultipleAsync`, `UpdateAsync`, `UpdateMultipleAsync`,
   `DeleteAsync`, `DeleteMultipleAsync`, `RetrieveAsync`, `RetrieveMultipleAsync`); await it, and pass the caller's
   `CancellationToken`. Make one service for the application: it shares one MongoDB client per connection string.
7. **Writes of several documents are all or nothing** (CreateMultiple, UpdateMultiple, DeleteMultiple with checks), in
   a transaction: MongoDB needs a replica set or a sharded cluster for that (one replica set node is enough). On a
   standalone server they are ARX8001.
8. **Update and Delete find documents by the id, a declared key, or fields you name:**
   `new ArxUpdateOptions { Match = ArxMatch.Key("account_number") }`, `ArxMatch.Columns(...)`. An update writes only the
   attributes that are set (`$set`); null stores null. A document's `_id` can't change (ARX8001).
   **To change or delete every document a condition matches** (from 0.1.0-preview.5): an entity with only the values to set
   and an `ArxFilterExpression`, `int changed = await service.UpdateMultipleAsync(values, filter);`, or
   `await service.DeleteMultipleAsync("account", filter)`; one `updateMany` or `deleteMany` in a transaction (a replica set), all or nothing, returning the count. The
   entity carries no id (ARX1001 with a filter); an empty filter is ARX1001 unless
   `new ArxFilterWriteOptions { AllRows = true }`. For a large insert whose ids aren't needed:
   `CreateMultipleAsync(entities, new ArxCreateMultipleOptions { ReturnIds = false })` (every id `ArxRecordId.None`).
9. **Reads.** `RetrieveAsync` returns `null` when nothing matches. `RetrieveMultipleAsync` takes an
   `ArxQueryExpression` (criteria, columns, order, paging, joins as link entities, aggregates) and returns the
   documents in `.Entities`. `Null` matches a stored null and a missing field; NotEqual never matches either. Page by
   `PageInfo.PageNumber`, or pass each page's `PagingCookie` to the next query. Execute is ARX8001: MongoDB has no
   stored procedures.
   **Query cheat sheet** (the exact names; there are no others, such as `AddGroupBy` or `MoreRecords`):
   `var query = new ArxQueryExpression("book");` · columns: `query.ColumnSet = new ArxColumnSet("title", "year");`
   (empty: every column) · conditions: `query.Criteria.AddCondition("year", ArxConditionOperator.Between, 2000, 2010);`
   (Equal, NotEqual, GreaterThan, GreaterEqual, LessThan, LessEqual, Between, In, NotIn, Like, Contains, BeginsWith,
   EndsWith, Null, NotNull and their Not/DoesNot forms) · an OR group: `var either = query.Criteria.AddFilter(ArxLogicalOperator.Or);`
   then `either.AddCondition(…)` · order: `query.AddOrder("title", ArxOrderType.Ascending);`
   · a join: `var author = query.AddLink("author", "author_id", "id", ArxJoinOperator.Inner); author.EntityAlias = "a";
   author.ColumnSet = new ArxColumnSet("name");`, read as `row["a.name"]` · paging:
   `query.PageInfo = new ArxPagingInfo { Count = 20, PageNumber = 1, ReturnTotalRecordCount = true };`, then
   `result.Entities`, `result.HasNextPage`, `result.TotalRecordCount` (or page by `result.PagingCookie`) · grouped counts:
   `query.AggregateInfo = new ArxAggregateInfo(); query.AggregateInfo.AddAggregateColumn("books", "id", ArxAggregateOperator.Count);
   query.AggregateInfo.AddGroupByColumn("author_id");` (Count reads back as a `long`). A foreign key column reads back as
   an `ArxEntityReference` (with the referenced row's name) only when the catalog declares it:
   `.Lookup("book", "author_id", "author")` and `.DisplayName("author", "name")`; otherwise it is the plain id.
10. **Values are never run as queries:** pass them as .NET values. A name with `$` or SQL syntax is ARX1017, so names
    that carry operators are refused, but a valid name is used as given and reaches any collection or field the database
    user can read: map a name from a user to one the application allows (an allow list), or preload the schema and set
    `Strict = true`. Search text is escaped.
11. **Dates:** a BSON date holds milliseconds, in UTC: a `DateTime` is stored as UTC (Local converted, Unspecified taken
    as UTC), a `DateTimeOffset` as its instant (the offset isn't kept), a `DateOnly` as midnight UTC. Never
    `default(DateTime)` (ARX1013: use `null`).
12. **Errors** are `ArxException`s with a stable `Code`: catch by code
    (`catch (ArxException error) when (error.Code == ArxErrorCode.UniqueViolation)`: catch the base class, since the class tells where an error was found and the code what happened) and retry only when
    `error.IsTransient` is true. `arx-mongodb-error-codes.md` lists every code. Log `error.Message`, not the driver's:
    MongoDB's duplicate key message quotes the value.
13. **Transactions:** `await using var connection = new ArxMongoConnection(service);`,
    `await using var transaction = await connection.BeginTransactionAsync();`, pass `transaction` to each call, then
    `await transaction.CommitAsync();`. Disposing without committing rolls back. Code written against `IArxDataService` opens one without the provider type: `await using var transaction = await service.BeginTransactionAsync();` (from 0.1.0-preview.5; the transaction owns its connection). MongoDB has no savepoints: a call that
    fails at the database aborts the transaction, and batch scopes have no savepoints: leave `Savepoints` unset (`true` is ARX8001), and a failed call rolls the scope back.
14. **Base answers on the samples** in the operation files: they are verified code. Say which sample a pattern comes
    from when that helps the developer find more.
15. **Reading values and namespaces.** `entity["col"]` is the value or null; a misspelt or unread attribute
   reads null too, so take column names from the table. `GetAttributeValue<T>` of NULL is `default(T)` (0 for `int`:
   ask for `int?` to tell NULL from 0); it converts numbers when nothing is lost (a Count is always a `long`, Sum and
   Avg are decimal or double by provider: read them as `long` and `decimal`), and reads a declared lookup (an
   `ArxEntityReference`, also in a grouped query) as its id, and a UTC or Local `DateTime` as a `DateTimeOffset` (and
   back, as UTC); a number that doesn't fit is ARX4001 (1.5 as `int`), any other mismatch is ARX4004, an `ArxException`.
   Through a link a lookup column is its plain id, not an `ArxEntityReference`.
   Namespaces: `ArxEntity`, `ArxEntityReference`, `ArxColumnSet`, `ArxSchemaCatalog` and `ArxMatch` are in
   `ARX.SDK.Services`; `ArxQueryExpression`, `ArxConditionOperator`, `ArxJoinOperator`, `ArxAggregateOperator` and
   `ArxOrderType` in `ARX.SDK.Query`; `ArxException` and `ArxErrorCode` in `ARX.SDK`; `ArxMongoDbService` and `ArxMongoConnection` in `ARX.SDK.MongoDb.Services`. Write every `using` a file needs.
   Names that differ from the Dynamics XRM SDK: the join operators are `ArxJoinOperator.Inner` and `ArxJoinOperator.Left` (not XRM's `LeftOuter`);
   a link's columns are `link.ColumnSet`, its alias `link.EntityAlias`, its conditions `link.LinkCriteria`; the referenced
   table of an `ArxEntityReference` is `EntityName` and its id `RecordId`. `UpdateAsync` throws ARX2002 RecordNotFound
   when no record matches (`DeleteAsync` returns false or 0 instead). A page's `TotalRecordCount` is an `int`; an aggregate
   Count is a `long`.
16. **Ids and linked columns are plain values.** `entity.RecordId` is an `ArxRecordId`: read it with
    `AsInt64()`, `AsGuid()` or `AsString()`, pass it where an id goes (`RetrieveAsync("account", id)`,
    `new ArxEntityReference("account", entity.RecordId!)`), or pass the plain value instead. `ArxRecordId` is a class, not
    a nullable struct: `RecordId` is null when no id is known, and `.Value` is the id as an `object` (pass the
    `ArxRecordId` itself, or `AsInt64()`, where a typed id goes, never `.Value`). A linked table's column reads back as `row["alias.column"]`, the plain value (no `AliasedValue`); without an alias the prefix is the table's name (`row["account.name"]`). Table names
    are matched exactly (`"Account"` isn't `account`: ARX1016); column names ignoring case.

Check every ARX call against `arx-mongodb-types.md`, and build before finishing.

## Data access: ARX.SDK

This project uses ARX.SDK 1.0.0-rc.1 with MongoDB (package `ARX.SDK.MongoDb`, service `ArxMongoDbService`). Its knowledge
files, generated from the library and its verified samples, are in `docs/arx-sdk/references/mongodb/`; the files named
below are there. Read `arx-mongodb-guide.md` and `arx-mongodb-types.md` before writing data access code,
and the operation's file (`arx-mongodb-create.md`, `-create-multiple.md`, `-update.md`, `-update-multiple.md`,
`-delete.md`, `-delete-multiple.md`, `-retrieve.md`, `-retrieve-multiple.md`, `-execute.md`) for its samples.

You help a developer write C# code that uses ARX.SDK with MongoDB (package `ARX.SDK.MongoDb`). Follow these rules.

1. **Use ARX's types for database work.** Create, update, delete and read documents, and run transactions through
   `ArxMongoDbService`, `ArxMongoConnection` and `ArxTransaction`. Don't write driver code (`IMongoCollection`,
   `BsonDocument`, `FilterDefinition`) for work ARX does.
2. **Don't invent API.** Use only the types, methods, properties and parameters listed in `arx-mongodb-types.md`. If
   something isn't there, say ARX doesn't offer it rather than guessing a name.
3. **A document is an `ArxEntity`.** `var account = new ArxEntity("account");` then values by field name:
   `account["name"] = "Northwind Traders";`. The entity's name is the collection. Field names keep their case.
4. **Declare every collection's id once, at startup,** in an `ArxSchemaCatalog`, and give it to the service:
   `ArxSchemaCatalog.Create().Table("account", ArxIdColumn.Generated<string>("account_id")).Build()`, then
   `new ArxMongoDbService(connectionString, schema)`; the connection string names the database
   (`mongodb://host:27017/crm`). The declared id is the document's `_id`: a generated string id is an ObjectId ARX makes
   (returned as its 24-character text), a generated Guid one ARX makes, `ArxIdColumn.Supplied<string>("code")` an id the
   application gives. A generated number id is ARX8001 (no sequences) and `NoIdColumn` is ARX1001 (every document has
   an `_id`). Add business keys with `.Key("account", "account_number")` and back them with unique indexes.
5. **Lookups are declared**, since MongoDB has no foreign keys: `.Lookup("contact", "account_id", "account")`. Write one
   with `new ArxEntityReference("account", id)`; it reads back as an `ArxEntityReference` with the account's display
   name (`.DisplayName("account", "name")`).
6. **Every call is asynchronous** (`CreateAsync`, `CreateMultipleAsync`, `UpdateAsync`, `UpdateMultipleAsync`,
   `DeleteAsync`, `DeleteMultipleAsync`, `RetrieveAsync`, `RetrieveMultipleAsync`); await it, and pass the caller's
   `CancellationToken`. Make one service for the application: it shares one MongoDB client per connection string.
7. **Writes of several documents are all or nothing** (CreateMultiple, UpdateMultiple, DeleteMultiple with checks), in
   a transaction: MongoDB needs a replica set or a sharded cluster for that (one replica set node is enough). On a
   standalone server they are ARX8001.
8. **Update and Delete find documents by the id, a declared key, or fields you name:**
   `new ArxUpdateOptions { Match = ArxMatch.Key("account_number") }`, `ArxMatch.Columns(...)`. An update writes only the
   attributes that are set (`$set`); null stores null. A document's `_id` can't change (ARX8001).
   **To change or delete every document a condition matches** (from 0.1.0-preview.5): an entity with only the values to set
   and an `ArxFilterExpression`, `int changed = await service.UpdateMultipleAsync(values, filter);`, or
   `await service.DeleteMultipleAsync("account", filter)`; one `updateMany` or `deleteMany` in a transaction (a replica set), all or nothing, returning the count. The
   entity carries no id (ARX1001 with a filter); an empty filter is ARX1001 unless
   `new ArxFilterWriteOptions { AllRows = true }`. For a large insert whose ids aren't needed:
   `CreateMultipleAsync(entities, new ArxCreateMultipleOptions { ReturnIds = false })` (every id `ArxRecordId.None`).
9. **Reads.** `RetrieveAsync` returns `null` when nothing matches. `RetrieveMultipleAsync` takes an
   `ArxQueryExpression` (criteria, columns, order, paging, joins as link entities, aggregates) and returns the
   documents in `.Entities`. `Null` matches a stored null and a missing field; NotEqual never matches either. Page by
   `PageInfo.PageNumber`, or pass each page's `PagingCookie` to the next query. Execute is ARX8001: MongoDB has no
   stored procedures.
   **Query cheat sheet** (the exact names; there are no others, such as `AddGroupBy` or `MoreRecords`):
   `var query = new ArxQueryExpression("book");` · columns: `query.ColumnSet = new ArxColumnSet("title", "year");`
   (empty: every column) · conditions: `query.Criteria.AddCondition("year", ArxConditionOperator.Between, 2000, 2010);`
   (Equal, NotEqual, GreaterThan, GreaterEqual, LessThan, LessEqual, Between, In, NotIn, Like, Contains, BeginsWith,
   EndsWith, Null, NotNull and their Not/DoesNot forms) · an OR group: `var either = query.Criteria.AddFilter(ArxLogicalOperator.Or);`
   then `either.AddCondition(…)` · order: `query.AddOrder("title", ArxOrderType.Ascending);`
   · a join: `var author = query.AddLink("author", "author_id", "id", ArxJoinOperator.Inner); author.EntityAlias = "a";
   author.ColumnSet = new ArxColumnSet("name");`, read as `row["a.name"]` · paging:
   `query.PageInfo = new ArxPagingInfo { Count = 20, PageNumber = 1, ReturnTotalRecordCount = true };`, then
   `result.Entities`, `result.HasNextPage`, `result.TotalRecordCount` (or page by `result.PagingCookie`) · grouped counts:
   `query.AggregateInfo = new ArxAggregateInfo(); query.AggregateInfo.AddAggregateColumn("books", "id", ArxAggregateOperator.Count);
   query.AggregateInfo.AddGroupByColumn("author_id");` (Count reads back as a `long`). A foreign key column reads back as
   an `ArxEntityReference` (with the referenced row's name) only when the catalog declares it:
   `.Lookup("book", "author_id", "author")` and `.DisplayName("author", "name")`; otherwise it is the plain id.
10. **Values are never run as queries:** pass them as .NET values. A name with `$` or SQL syntax is ARX1017, so names
    that carry operators are refused, but a valid name is used as given and reaches any collection or field the database
    user can read: map a name from a user to one the application allows (an allow list), or preload the schema and set
    `Strict = true`. Search text is escaped.
11. **Dates:** a BSON date holds milliseconds, in UTC: a `DateTime` is stored as UTC (Local converted, Unspecified taken
    as UTC), a `DateTimeOffset` as its instant (the offset isn't kept), a `DateOnly` as midnight UTC. Never
    `default(DateTime)` (ARX1013: use `null`).
12. **Errors** are `ArxException`s with a stable `Code`: catch by code
    (`catch (ArxException error) when (error.Code == ArxErrorCode.UniqueViolation)`: catch the base class, since the class tells where an error was found and the code what happened) and retry only when
    `error.IsTransient` is true. `arx-mongodb-error-codes.md` lists every code. Log `error.Message`, not the driver's:
    MongoDB's duplicate key message quotes the value.
13. **Transactions:** `await using var connection = new ArxMongoConnection(service);`,
    `await using var transaction = await connection.BeginTransactionAsync();`, pass `transaction` to each call, then
    `await transaction.CommitAsync();`. Disposing without committing rolls back. Code written against `IArxDataService` opens one without the provider type: `await using var transaction = await service.BeginTransactionAsync();` (from 0.1.0-preview.5; the transaction owns its connection). MongoDB has no savepoints: a call that
    fails at the database aborts the transaction, and batch scopes have no savepoints: leave `Savepoints` unset (`true` is ARX8001), and a failed call rolls the scope back.
14. **Base answers on the samples** in the operation files: they are verified code. Say which sample a pattern comes
    from when that helps the developer find more.
15. **Reading values and namespaces.** `entity["col"]` is the value or null; a misspelt or unread attribute
   reads null too, so take column names from the table. `GetAttributeValue<T>` of NULL is `default(T)` (0 for `int`:
   ask for `int?` to tell NULL from 0); it converts numbers when nothing is lost (a Count is always a `long`, Sum and
   Avg are decimal or double by provider: read them as `long` and `decimal`), and reads a declared lookup (an
   `ArxEntityReference`, also in a grouped query) as its id, and a UTC or Local `DateTime` as a `DateTimeOffset` (and
   back, as UTC); a number that doesn't fit is ARX4001 (1.5 as `int`), any other mismatch is ARX4004, an `ArxException`.
   Through a link a lookup column is its plain id, not an `ArxEntityReference`.
   Namespaces: `ArxEntity`, `ArxEntityReference`, `ArxColumnSet`, `ArxSchemaCatalog` and `ArxMatch` are in
   `ARX.SDK.Services`; `ArxQueryExpression`, `ArxConditionOperator`, `ArxJoinOperator`, `ArxAggregateOperator` and
   `ArxOrderType` in `ARX.SDK.Query`; `ArxException` and `ArxErrorCode` in `ARX.SDK`; `ArxMongoDbService` and `ArxMongoConnection` in `ARX.SDK.MongoDb.Services`. Write every `using` a file needs.
   Names that differ from the Dynamics XRM SDK: the join operators are `ArxJoinOperator.Inner` and `ArxJoinOperator.Left` (not XRM's `LeftOuter`);
   a link's columns are `link.ColumnSet`, its alias `link.EntityAlias`, its conditions `link.LinkCriteria`; the referenced
   table of an `ArxEntityReference` is `EntityName` and its id `RecordId`. `UpdateAsync` throws ARX2002 RecordNotFound
   when no record matches (`DeleteAsync` returns false or 0 instead). A page's `TotalRecordCount` is an `int`; an aggregate
   Count is a `long`.
16. **Ids and linked columns are plain values.** `entity.RecordId` is an `ArxRecordId`: read it with
    `AsInt64()`, `AsGuid()` or `AsString()`, pass it where an id goes (`RetrieveAsync("account", id)`,
    `new ArxEntityReference("account", entity.RecordId!)`), or pass the plain value instead. `ArxRecordId` is a class, not
    a nullable struct: `RecordId` is null when no id is known, and `.Value` is the id as an `object` (pass the
    `ArxRecordId` itself, or `AsInt64()`, where a typed id goes, never `.Value`). A linked table's column reads back as `row["alias.column"]`, the plain value (no `AliasedValue`); without an alias the prefix is the table's name (`row["account.name"]`). Table names
    are matched exactly (`"Account"` isn't `account`: ARX1016); column names ignoring case.

Before finishing: `dotnet build`, run the tests, and check every ARX call against `arx-mongodb-types.md`. When the
ARX.SDK package version changes, replace the knowledge files with the new version's and update this section.

You help a developer write C# code that uses ARX.SDK with MongoDB (package `ARX.SDK.MongoDb`). Follow these rules.

1. **Use ARX's types for database work.** Create, update, delete and read documents, and run transactions through
   `ArxMongoDbService`, `ArxMongoConnection` and `ArxTransaction`. Don't write driver code (`IMongoCollection`,
   `BsonDocument`, `FilterDefinition`) for work ARX does.
2. **Don't invent API.** Use only the types, methods, properties and parameters listed in `arx-mongodb-types.md`. If
   something isn't there, say ARX doesn't offer it rather than guessing a name.
3. **A document is an `ArxEntity`.** `var account = new ArxEntity("account");` then values by field name:
   `account["name"] = "Northwind Traders";`. The entity's name is the collection. Field names keep their case.
4. **Declare every collection's id once, at startup,** in an `ArxSchemaCatalog`, and give it to the service:
   `ArxSchemaCatalog.Create().Table("account", ArxIdColumn.Generated<string>("account_id")).Build()`, then
   `new ArxMongoDbService(connectionString, schema)`; the connection string names the database
   (`mongodb://host:27017/crm`). The declared id is the document's `_id`: a generated string id is an ObjectId ARX makes
   (returned as its 24-character text), a generated Guid one ARX makes, `ArxIdColumn.Supplied<string>("code")` an id the
   application gives. A generated number id is ARX8001 (no sequences) and `NoIdColumn` is ARX1001 (every document has
   an `_id`). Add business keys with `.Key("account", "account_number")` and back them with unique indexes.
5. **Lookups are declared**, since MongoDB has no foreign keys: `.Lookup("contact", "account_id", "account")`. Write one
   with `new ArxEntityReference("account", id)`; it reads back as an `ArxEntityReference` with the account's display
   name (`.DisplayName("account", "name")`).
6. **Every call is asynchronous** (`CreateAsync`, `CreateMultipleAsync`, `UpdateAsync`, `UpdateMultipleAsync`,
   `DeleteAsync`, `DeleteMultipleAsync`, `RetrieveAsync`, `RetrieveMultipleAsync`); await it, and pass the caller's
   `CancellationToken`. Make one service for the application: it shares one MongoDB client per connection string.
7. **Writes of several documents are all or nothing** (CreateMultiple, UpdateMultiple, DeleteMultiple with checks), in
   a transaction: MongoDB needs a replica set or a sharded cluster for that (one replica set node is enough). On a
   standalone server they are ARX8001.
8. **Update and Delete find documents by the id, a declared key, or fields you name:**
   `new ArxUpdateOptions { Match = ArxMatch.Key("account_number") }`, `ArxMatch.Columns(...)`. An update writes only the
   attributes that are set (`$set`); null stores null. A document's `_id` can't change (ARX8001).
   **To change or delete every document a condition matches** (from 0.1.0-preview.5): an entity with only the values to set
   and an `ArxFilterExpression`, `int changed = await service.UpdateMultipleAsync(values, filter);`, or
   `await service.DeleteMultipleAsync("account", filter)`; one `updateMany` or `deleteMany` in a transaction (a replica set), all or nothing, returning the count. The
   entity carries no id (ARX1001 with a filter); an empty filter is ARX1001 unless
   `new ArxFilterWriteOptions { AllRows = true }`. For a large insert whose ids aren't needed:
   `CreateMultipleAsync(entities, new ArxCreateMultipleOptions { ReturnIds = false })` (every id `ArxRecordId.None`).
9. **Reads.** `RetrieveAsync` returns `null` when nothing matches. `RetrieveMultipleAsync` takes an
   `ArxQueryExpression` (criteria, columns, order, paging, joins as link entities, aggregates) and returns the
   documents in `.Entities`. `Null` matches a stored null and a missing field; NotEqual never matches either. Page by
   `PageInfo.PageNumber`, or pass each page's `PagingCookie` to the next query. Execute is ARX8001: MongoDB has no
   stored procedures.
   **Query cheat sheet** (the exact names; there are no others, such as `AddGroupBy` or `MoreRecords`):
   `var query = new ArxQueryExpression("book");` · columns: `query.ColumnSet = new ArxColumnSet("title", "year");`
   (empty: every column) · conditions: `query.Criteria.AddCondition("year", ArxConditionOperator.Between, 2000, 2010);`
   (Equal, NotEqual, GreaterThan, GreaterEqual, LessThan, LessEqual, Between, In, NotIn, Like, Contains, BeginsWith,
   EndsWith, Null, NotNull and their Not/DoesNot forms) · an OR group: `var either = query.Criteria.AddFilter(ArxLogicalOperator.Or);`
   then `either.AddCondition(…)` · order: `query.AddOrder("title", ArxOrderType.Ascending);`
   · a join: `var author = query.AddLink("author", "author_id", "id", ArxJoinOperator.Inner); author.EntityAlias = "a";
   author.ColumnSet = new ArxColumnSet("name");`, read as `row["a.name"]` · paging:
   `query.PageInfo = new ArxPagingInfo { Count = 20, PageNumber = 1, ReturnTotalRecordCount = true };`, then
   `result.Entities`, `result.HasNextPage`, `result.TotalRecordCount` (or page by `result.PagingCookie`) · grouped counts:
   `query.AggregateInfo = new ArxAggregateInfo(); query.AggregateInfo.AddAggregateColumn("books", "id", ArxAggregateOperator.Count);
   query.AggregateInfo.AddGroupByColumn("author_id");` (Count reads back as a `long`). A foreign key column reads back as
   an `ArxEntityReference` (with the referenced row's name) only when the catalog declares it:
   `.Lookup("book", "author_id", "author")` and `.DisplayName("author", "name")`; otherwise it is the plain id.
10. **Values are never run as queries:** pass them as .NET values. A name with `$` or SQL syntax is ARX1017, so names
    that carry operators are refused, but a valid name is used as given and reaches any collection or field the database
    user can read: map a name from a user to one the application allows (an allow list), or preload the schema and set
    `Strict = true`. Search text is escaped.
11. **Dates:** a BSON date holds milliseconds, in UTC: a `DateTime` is stored as UTC (Local converted, Unspecified taken
    as UTC), a `DateTimeOffset` as its instant (the offset isn't kept), a `DateOnly` as midnight UTC. Never
    `default(DateTime)` (ARX1013: use `null`).
12. **Errors** are `ArxException`s with a stable `Code`: catch by code
    (`catch (ArxException error) when (error.Code == ArxErrorCode.UniqueViolation)`: catch the base class, since the class tells where an error was found and the code what happened) and retry only when
    `error.IsTransient` is true. `arx-mongodb-error-codes.md` lists every code. Log `error.Message`, not the driver's:
    MongoDB's duplicate key message quotes the value.
13. **Transactions:** `await using var connection = new ArxMongoConnection(service);`,
    `await using var transaction = await connection.BeginTransactionAsync();`, pass `transaction` to each call, then
    `await transaction.CommitAsync();`. Disposing without committing rolls back. Code written against `IArxDataService` opens one without the provider type: `await using var transaction = await service.BeginTransactionAsync();` (from 0.1.0-preview.5; the transaction owns its connection). MongoDB has no savepoints: a call that
    fails at the database aborts the transaction, and batch scopes have no savepoints: leave `Savepoints` unset (`true` is ARX8001), and a failed call rolls the scope back.
14. **Base answers on the samples** in the operation files: they are verified code. Say which sample a pattern comes
    from when that helps the developer find more.
15. **Reading values and namespaces.** `entity["col"]` is the value or null; a misspelt or unread attribute
   reads null too, so take column names from the table. `GetAttributeValue<T>` of NULL is `default(T)` (0 for `int`:
   ask for `int?` to tell NULL from 0); it converts numbers when nothing is lost (a Count is always a `long`, Sum and
   Avg are decimal or double by provider: read them as `long` and `decimal`), and reads a declared lookup (an
   `ArxEntityReference`, also in a grouped query) as its id, and a UTC or Local `DateTime` as a `DateTimeOffset` (and
   back, as UTC); a number that doesn't fit is ARX4001 (1.5 as `int`), any other mismatch is ARX4004, an `ArxException`.
   Through a link a lookup column is its plain id, not an `ArxEntityReference`.
   Namespaces: `ArxEntity`, `ArxEntityReference`, `ArxColumnSet`, `ArxSchemaCatalog` and `ArxMatch` are in
   `ARX.SDK.Services`; `ArxQueryExpression`, `ArxConditionOperator`, `ArxJoinOperator`, `ArxAggregateOperator` and
   `ArxOrderType` in `ARX.SDK.Query`; `ArxException` and `ArxErrorCode` in `ARX.SDK`; `ArxMongoDbService` and `ArxMongoConnection` in `ARX.SDK.MongoDb.Services`. Write every `using` a file needs.
   Names that differ from the Dynamics XRM SDK: the join operators are `ArxJoinOperator.Inner` and `ArxJoinOperator.Left` (not XRM's `LeftOuter`);
   a link's columns are `link.ColumnSet`, its alias `link.EntityAlias`, its conditions `link.LinkCriteria`; the referenced
   table of an `ArxEntityReference` is `EntityName` and its id `RecordId`. `UpdateAsync` throws ARX2002 RecordNotFound
   when no record matches (`DeleteAsync` returns false or 0 instead). A page's `TotalRecordCount` is an `int`; an aggregate
   Count is a `long`.
16. **Ids and linked columns are plain values.** `entity.RecordId` is an `ArxRecordId`: read it with
    `AsInt64()`, `AsGuid()` or `AsString()`, pass it where an id goes (`RetrieveAsync("account", id)`,
    `new ArxEntityReference("account", entity.RecordId!)`), or pass the plain value instead. `ArxRecordId` is a class, not
    a nullable struct: `RecordId` is null when no id is known, and `.Value` is the id as an `object` (pass the
    `ArxRecordId` itself, or `AsInt64()`, where a typed id goes, never `.Value`). A linked table's column reads back as `row["alias.column"]`, the plain value (no `AliasedValue`); without an alias the prefix is the table's name (`row["account.name"]`). Table names
    are matched exactly (`"Account"` isn't `account`: ARX1016); column names ignoring case.