[officialasishkumar]

Fixes #39546

GetTextReader, GetFieldValue<TextReader> and GetChars wrap GetStream in a StreamReader that decodes UTF-8. GetStream exposes text in the database encoding, because neither sqlite3_blob_read nor sqlite3_column_blob converts it. In a database created with PRAGMA encoding = 'UTF-16le' or 'UTF-16be', these methods therefore return text like t\0�\0s\0t\0 without an error, while GetString returns the right value.

Fix

SqliteDataRecord now creates its text reader in one place, used by GetTextReader and GetChars:

• For a TEXT value in a database that isn't UTF-8, it reads the value with GetString, because sqlite3_column_text always converts to UTF-8. I chose this over a StreamReader with the database encoding because the bytes of a value that isn't read through a blob depend on earlier calls: once GetString has read a value, sqlite3_column_blob returns it as UTF-8, so the database encoding wouldn't reliably describe them.
• In every other case nothing changes. UTF-8 databases keep streaming (GetTextReader_works and GetTextReader_works_streaming are unchanged), and BLOB values are still decoded as UTF-8.

The encoding comes from PRAGMA encoding, which runs at most once per result set and only when a TEXT value is read this way. All databases on a connection share one encoding, because SQLite rejects attaching a database with a different one.

GetStream and GetBytes still return the raw bytes.

Note that on UTF-16 databases, these methods now hold the whole value in memory rather than streaming it.

Tests (SqliteDataReaderTest)

• GetTextReader_works_for_database_encoding and GetChars_works_for_database_encoding: UTF-16 LE and BE, with a rowid (blob path), without one, and from an expression, plus UTF-8 as a guard.
• GetTextReader_works_for_parameter_when_UTF16_database and GetFieldValue_of_TextReader_works_when_UTF16_database.
• GetTextReader_reads_blob_as_UTF8_when_UTF16_database: guards that BLOB values keep their behavior.

Without the source change, 12 of these test cases fail. The UTF-8 and BLOB guards pass either way. With it, the Microsoft.Data.Sqlite.Tests project passes on macOS arm64 (829 passed, 1 skipped). I only ran that test project locally, not the full repository suite.

• I've read the guidelines for contributing and seen the walkthrough
• I've posted a comment on an issue with a detailed description of how I am planning to contribute and got approval from a member of the team
• The code builds and tests pass locally (also verified by our automated build checks)
• Commit messages follow this format:

        Summary of the changes
        - Detail 1
        - Detail 2

        Fixes #bugnumber

• Tests for the changes have been added (for bug fixes / features)
• Code follows the same patterns and style as existing code in this repo

[Copilot]

🟢 Approval recommended

The UTF-16 text-decoding fix has no active review issues.

No open issues 🎉

What changed in this PR

Fixes #39546 by correcting text decoding in Microsoft.Data.Sqlite for UTF-16 databases across GetTextReader, GetFieldValue<TextReader>, and GetChars.

Changes:

• Centralizes text-reader creation, using GetString for non-UTF-8 TEXT values and caching the database encoding per result set.
• Preserves UTF-8 streaming and BLOB decoding; UTF-16 text is materialized in memory, while raw-byte APIs remain unchanged.
• Adds regression tests for database encodings, query shapes, parameters, and BLOB behavior.

File	Description
test/​Microsoft.Data.Sqlite.Tests/​SqliteDataReaderTest.cs	Adds encoding regression tests and UTF-8/BLOB compatibility guards.
src/​Microsoft.Data.Sqlite.Core/​SqliteDataRecord.cs	Shares encoding-aware text-reader creation between text and character accessors.

_{Review effort: Balanced}