tiktoken_ruby 0.0.6-x86_64-linux → 0.0.8-x86_64-linux

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: df3386d523b807be99acc4d52ba9c82278b3eb23b85c7b4c0e2007827978be19
-  data.tar.gz: e96e54eec674ffc0d895e1c60fa081d8b32d5f72c242246585916d92cf3da0ae
+  metadata.gz: a3a7f91d1a21f0a8062c03988d79393676a8f6a727c58efe094f27d274c27f9f
+  data.tar.gz: 15a0056ddedcf37c6a907445027411a9a43d41a4f067fa1fab81df7ba4432adb
 SHA512:
-  metadata.gz: f7ec03086d285833ab22acea40b0c5841c4417e502612a032a7e919bec6dfae25853b5cc81ac78af154b80f248895d46e199f7e090d7d5c98333384f777716e4
-  data.tar.gz: 4364eccf5dc41de7dfe8afee572caa305307a34543e8271dcdc4ec8efb71713103b73b6c8f05d6f0413ecff41ee2ec651ffbd9388fe73b0f576e116d4fe31d95
+  metadata.gz: 59eb286c3a2084951ee7ca6a9ba7ea585e27747fcfbd9b79271a81cfe5c978168915d48486829f6d25d43a567136b422db6da8a1b925327747fbec1a013fbbb0
+  data.tar.gz: 310063bff0976e90055dd3685a64d6c27e0cde1e5ff21ca7891f875065b3b1e7984c8adcf0f7fde3d854674fe6e85d53377b7cca634206c9675163339bb2a87a

data/Gemfile

@@ -2,15 +2,11 @@
 
 source "https://rubygems.org"
 
-# Specify your gem's dependencies in tiktoken_ruby.gemspec
 gemspec
 
-gem "rake", "~> 13.0"
-
+gem "rake"
 gem "rake-compiler"
-
-gem "rspec", "~> 3.0"
-
-gem "standard", "~> 1.3"
-
-gem "yard-doctest", "~> 0.1.17"
+gem "rspec"
+gem "standard"
+gem "yard-doctest"
+gem "racc"

data/Gemfile.lock

@@ -1,64 +1,74 @@
 PATH
   remote: .
   specs:
-    tiktoken_ruby (0.0.6)
-      rb_sys (~> 0.9.68)
+    tiktoken_ruby (0.0.8)
+      rb_sys (= 0.9.87)
 
 GEM
   remote: https://rubygems.org/
   specs:
     ast (2.4.2)
     diff-lcs (1.5.0)
-    json (2.6.3)
+    json (2.7.1)
     language_server-protocol (3.17.0.3)
-    minitest (5.18.0)
-    parallel (1.22.1)
-    parser (3.2.1.1)
+    lint_roller (1.1.0)
+    minitest (5.21.2)
+    parallel (1.24.0)
+    parser (3.3.0.4)
       ast (~> 2.4.1)
+      racc
+    racc (1.7.3)
     rainbow (3.1.1)
-    rake (13.0.6)
-    rake-compiler (1.2.1)
+    rake (13.1.0)
+    rake-compiler (1.2.5)
       rake
-    rb_sys (0.9.68)
-    regexp_parser (2.7.0)
-    rexml (3.2.5)
+    rb_sys (0.9.87)
+    regexp_parser (2.9.0)
+    rexml (3.2.6)
     rspec (3.12.0)
       rspec-core (~> 3.12.0)
       rspec-expectations (~> 3.12.0)
       rspec-mocks (~> 3.12.0)
-    rspec-core (3.12.1)
+    rspec-core (3.12.2)
       rspec-support (~> 3.12.0)
-    rspec-expectations (3.12.2)
+    rspec-expectations (3.12.3)
       diff-lcs (>= 1.2.0, < 2.0)
       rspec-support (~> 3.12.0)
-    rspec-mocks (3.12.4)
+    rspec-mocks (3.12.6)
       diff-lcs (>= 1.2.0, < 2.0)
       rspec-support (~> 3.12.0)
-    rspec-support (3.12.0)
-    rubocop (1.48.1)
+    rspec-support (3.12.1)
+    rubocop (1.59.0)
       json (~> 2.3)
+      language_server-protocol (>= 3.17.0)
       parallel (~> 1.10)
-      parser (>= 3.2.0.0)
+      parser (>= 3.2.2.4)
       rainbow (>= 2.2.2, < 4.0)
       regexp_parser (>= 1.8, < 3.0)
       rexml (>= 3.2.5, < 4.0)
-      rubocop-ast (>= 1.26.0, < 2.0)
+      rubocop-ast (>= 1.30.0, < 2.0)
       ruby-progressbar (~> 1.7)
       unicode-display_width (>= 2.4.0, < 3.0)
-    rubocop-ast (1.27.0)
+    rubocop-ast (1.30.0)
       parser (>= 3.2.1.0)
-    rubocop-performance (1.16.0)
-      rubocop (>= 1.7.0, < 2.0)
-      rubocop-ast (>= 0.4.0)
+    rubocop-performance (1.20.2)
+      rubocop (>= 1.48.1, < 2.0)
+      rubocop-ast (>= 1.30.0, < 2.0)
     ruby-progressbar (1.13.0)
-    standard (1.25.1)
+    standard (1.33.0)
       language_server-protocol (~> 3.17.0.2)
-      rubocop (= 1.48.1)
-      rubocop-performance (= 1.16.0)
-    unicode-display_width (2.4.2)
-    webrick (1.7.0)
-    yard (0.9.28)
-      webrick (~> 1.7.0)
+      lint_roller (~> 1.0)
+      rubocop (~> 1.59.0)
+      standard-custom (~> 1.0.0)
+      standard-performance (~> 1.3)
+    standard-custom (1.0.2)
+      lint_roller (~> 1.0)
+      rubocop (~> 1.50)
+    standard-performance (1.3.1)
+      lint_roller (~> 1.1)
+      rubocop-performance (~> 1.20.2)
+    unicode-display_width (2.5.0)
+    yard (0.9.34)
     yard-doctest (0.1.17)
       minitest
       yard
@@ -70,12 +80,13 @@ PLATFORMS
   x86_64-linux
 
 DEPENDENCIES
-  rake (~> 13.0)
+  racc
+  rake
   rake-compiler
-  rspec (~> 3.0)
-  standard (~> 1.3)
+  rspec
+  standard
   tiktoken_ruby!
-  yard-doctest (~> 0.1.17)
+  yard-doctest
 
 BUNDLED WITH
-   2.4.6
+   2.4.4

data/README.md

@@ -1,8 +1,14 @@
 [![Gem Version](https://badge.fury.io/rb/tiktoken_ruby.svg)](https://badge.fury.io/rb/tiktoken_ruby)
+
 # tiktoken_ruby
 
 [Tiktoken](https://github.com/openai/tiktoken) is BPE tokenizer from OpenAI used with their GPT models.
-This is a wrapper around it aimed primarily at enabling accurate counts of GPT model tokens used. 
+This is a wrapper around it aimed primarily at enabling accurate counts of GPT model tokens used.
+
+## Request for maintainers
+
+I can't really put substantial time into maintaining this. Probably nothing more than a couple hours every few months. If you have experience maintaining ruby gems and would like to
+lend a hand please send me an email or reply to this [issue](https://github.com/IAPark/tiktoken_ruby/issues/26)
 
 ## Installation
 
@@ -15,17 +21,19 @@ If bundler is not being used to manage dependencies, install the gem by executin
     $ gem install tiktoken_ruby
 
 ## Usage
+
 Usage should be very similar to the python library. Here's a simple example
 
 Encode and decode text
+
 ```ruby
 require 'tiktoken_ruby'
-
 enc = Tiktoken.get_encoding("cl100k_base")
 enc.decode(enc.encode("hello world")) #=> "hello world"
 ```
 
 Encoders can also be retrieved by model name
+
 ```ruby
 require 'tiktoken_ruby'
 
@@ -53,7 +61,6 @@ bundle exec rake compile
 bundle exec rake spec
 ```
 
-
 ## License
 
 The gem is available as open source under the terms of the [MIT License](https://opensource.org/licenses/MIT).

data/lib/tiktoken_ruby/encoding.rb

@@ -1,6 +1,8 @@
 # frozen_string_literal: true
 
 class Tiktoken::Encoding
+  CACHE_MUTEX = Mutex.new
+
   attr_reader :name
 
   # This returns a new Tiktoken::Encoding instance for the requested encoding
@@ -15,8 +17,10 @@ class Tiktoken::Encoding
   # @param encoding [Symbol] The name of the encoding to load
   # @return [Tiktoken::Encoding] The encoding instance
   def self.for_name_cached(encoding)
-    @encodings ||= {}
-    @encodings[encoding.to_sym] ||= Tiktoken::Encoding.for_name(encoding)
+    CACHE_MUTEX.synchronize do
+      @encodings ||= {}
+      @encodings[encoding.to_sym] ||= Tiktoken::Encoding.for_name(encoding)
+    end
   end
 
   # Encodes the text as a list of integer tokens. This encoding will encode special non text tokens

data/lib/tiktoken_ruby/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Tiktoken
-  VERSION = "0.0.6"
+  VERSION = "0.0.8"
 end

data/lib/tiktoken_ruby.rb

@@ -28,22 +28,22 @@ module Tiktoken
 
     # Gets the encoding for an OpenAI model
     # @param model_name [Symbol|String] The name of the model to get the encoding for
-    # @return [Tiktoken::Encoding] The encoding instance
+    # @return [Tiktoken::Encoding, nil] The encoding instance, or nil if no encoding is found
     # @example Count tokens for text
     #   enc = Tiktoken.encoding_for_model("gpt-4")
     #   enc.encode("hello world").length #=> 2
     def encoding_for_model(model_name)
-      PREFIX_MODELS.each do |prefix|
-        if model_name.to_s.start_with?("#{prefix}-")
-          model_name = prefix
-          break
-        end
+      if MODEL_TO_ENCODING_NAME.key?(model_name.to_sym)
+        return get_encoding(MODEL_TO_ENCODING_NAME[model_name.to_sym])
       end
 
-      encoding_name = MODEL_TO_ENCODING_NAME[model_name.to_sym]
-      return nil unless encoding_name
+      _prefix, encoding = MODEL_PREFIX_TO_ENCODING.find do |prefix, _encoding|
+        model_name.start_with?(prefix.to_s)
+      end
 
-      get_encoding(encoding_name)
+      if encoding
+        get_encoding(encoding)
+      end
     end
 
     # Lists all the encodings that are supported
@@ -67,12 +67,22 @@ module Tiktoken
       :cl100k_base
     ]
 
-    # taken from the python library here https://github.com/openai/tiktoken/blob/main/tiktoken/model.py#L13-L53
+    # taken from the python library here https://github.com/openai/tiktoken/blob/main/tiktoken/model.py
     # that is also MIT licensed but by OpenAI
     MODEL_TO_ENCODING_NAME = {
+      # chat
       "gpt-4": "cl100k_base",
       "gpt-3.5-turbo": "cl100k_base",
-      # text
+      "gpt-35-turbo": "cl100k_base",  # Azure deployment name
+      # base
+      "davinci-002": "cl100k_base",
+      "babbage-002": "cl100k_base",
+      # embeddings
+      "text-embedding-ada-002": "cl100k_base",
+      "text-embedding-3-small": "cl100k_base",
+      "text-embedding-3-large": "cl100k_base",
+      # DEPRECATED MODELS
+      # text (DEPRECATED)
       "text-davinci-003": "p50k_base",
       "text-davinci-002": "p50k_base",
       "text-davinci-001": "r50k_base",
@@ -83,19 +93,17 @@ module Tiktoken
       curie: "r50k_base",
       babbage: "r50k_base",
       ada: "r50k_base",
-      # code
+      # code (DEPRECATED)
       "code-davinci-002": "p50k_base",
       "code-davinci-001": "p50k_base",
       "code-cushman-002": "p50k_base",
       "code-cushman-001": "p50k_base",
       "davinci-codex": "p50k_base",
       "cushman-codex": "p50k_base",
-      # edit
+      # edit (DEPRECATED)
       "text-davinci-edit-001": "p50k_edit",
       "code-davinci-edit-001": "p50k_edit",
-      # embeddings
-      "text-embedding-ada-002": "cl100k_base",
-      # old embeddings
+      # old embeddings (DEPRECATED)
       "text-similarity-davinci-001": "r50k_base",
       "text-similarity-curie-001": "r50k_base",
       "text-similarity-babbage-001": "r50k_base",
@@ -105,10 +113,21 @@ module Tiktoken
       "text-search-babbage-doc-001": "r50k_base",
       "text-search-ada-doc-001": "r50k_base",
       "code-search-babbage-code-001": "r50k_base",
-      "code-search-ada-code-001": "r50k_base"
+      "code-search-ada-code-001": "r50k_base",
+      # open source
+      gpt2: "gpt2"
     }
 
-    # these are models that have a versioned models that are otherwise identical
-    PREFIX_MODELS = ["gpt-4", "gpt-3.5-turbo"]
+    MODEL_PREFIX_TO_ENCODING = {
+      # chat
+      "gpt-4-": "cl100k_base",  # e.g., gpt-4-0314, etc., plus gpt-4-32k
+      "gpt-3.5-turbo-": "cl100k_base",  # e.g, gpt-3.5-turbo-0301, -0401, etc.
+      "gpt-35-turbo-": "cl100k_base",  # Azure deployment name
+      # fine-tuned
+      "ft:gpt-4": "cl100k_base",
+      "ft:gpt-3.5-turbo": "cl100k_base",
+      "ft:davinci-002": "cl100k_base",
+      "ft:babbage-002": "cl100k_base"
+    }
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: tiktoken_ruby
 version: !ruby/object:Gem::Version
-  version: 0.0.6
+  version: 0.0.8
 platform: x86_64-linux
 authors:
 - IAPark
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2023-09-24 00:00:00.000000000 Z
+date: 2024-04-04 00:00:00.000000000 Z
 dependencies: []
 description: An unofficial Ruby wrapper for Tiktoken, a BPE tokenizer written by and
   used by OpenAI. It can be used to count the number of tokens in text before sending
@@ -28,9 +28,9 @@ files:
 - Rakefile
 - doctest_helper.rb
 - lib/tiktoken_ruby.rb
-- lib/tiktoken_ruby/3.0/tiktoken_ruby.so
 - lib/tiktoken_ruby/3.1/tiktoken_ruby.so
 - lib/tiktoken_ruby/3.2/tiktoken_ruby.so
+- lib/tiktoken_ruby/3.3/tiktoken_ruby.so
 - lib/tiktoken_ruby/encoding.rb
 - lib/tiktoken_ruby/version.rb
 - sig/tiktoken_ruby.rbs
@@ -49,10 +49,10 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: '3.0'
+      version: '3.1'
   - - "<"
     - !ruby/object:Gem::Version
-      version: 3.3.dev
+      version: 3.4.dev
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="