0.0.5

0.0.5

• Minimal: OPT: override FindFirstCharacterToEncodeUtf8. Use SearchValues for net8+.
• Minimal: OPT: FindFirstCharacterToEncode: detect and skip legal surrogates instead of making encoder handle all surrogates.
• Minimal: OPT: After SearchValues, if the first disallowed char is ascii, we can return immediately (we know this is a char that needs escaping)

Full Changelog: 0.0.4...0.0.5

Nuget: https://www.nuget.org/packages/ExtremeJsonEncoders/0.0.5

0.0.4

0.0.4

• Use upper case hex as default (Shared and ctors). It is defacto standard, it seems.
• Minimal: add ctor options: shortEscapes, lowerCaseHex, extraAsciiEscapeChars
• Maximal: add ctor options: shortEscapes, lowerCaseHex
• FindFirstCharacterToEncode: use SearchValues on net8+ to skip leading non-escaped ascii chars faster

Full Changelog: 0.0.3...0.0.4

Nuget: https://www.nuget.org/packages/ExtremeJsonEncoders/0.0.4

0.0.3

0.0.3

• MinimalJsonEncoder: increase performance by used code from dotnet runtime.
• Both: add bool parameter in ctor, lowerCaseHex. Default is true (also true via the Shared-instances).
• Handle invalid chars (partial surrogates). They will now be replaced with Rune.ReplacementChar, just like the other encoders do. This is a side effect of using code from dotnet runtime.

Full Changelog: 0.0.2...0.0.3

Nuget: https://www.nuget.org/packages/ExtremeJsonEncoders/0.0.3

0.0.2

0.0.2

• FIX: MinimalJsonEncoder encoded \n,\t,\r,etc. (control characters that has a two-character sequence) fully escaped (\u000#)
• FIX: MaximalJsonEncoder encoded \n incorrectly (\u000: instead of \u000a)

Full Changelog: 0.0.1...0.0.2

Nuget: https://www.nuget.org/packages/ExtremeJsonEncoders/0.0.2

0.0.1

0.0.1

• First release.

Full Changelog: https://github.com/osexpert/ExtremeJsonEncoders/commits/0.0.1

Nuget: https://www.nuget.org/packages/ExtremeJsonEncoders/0.0.1