monadic-chat 0.3.2 → 0.3.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 39e287f8effed950af1b2cae95e834113d738d63188ceda70b6912eecaf92882
-  data.tar.gz: a73f9b6b74b174885f925e8377dc5afd6700dd0528d6f3c54156c4cab3a17740
+  metadata.gz: '08138f50ed67a2c1b913b82b4d56402fe789f9858d6f8cd31cc7512daf79f0bf'
+  data.tar.gz: 1fca018c495fb1fedc30c1e684fd9c14639696c7f919622393e36be234f76b35
 SHA512:
-  metadata.gz: 986237787361787e9af73501c2d5fa51260836c305a4926394fe1ff2999a39e6bf1ac9059c06f221c87bb3b1e7238c91a3478bb970ac4ef98100f8cc16d42954
-  data.tar.gz: cd14739cccd2319fcceace5b3813726d99159d51a8a269ee01f6fb209fc33127f33f726c0471a664ac2f6ea033d8afd5851af5ba7e4f1176c07c901358f979fb
+  metadata.gz: bcc93e02f837008c126fdbaf4b99b03b6c1683e24f0d62334cb33ad64545b58f087a8adfdd7f461cc16774c6d72ce06db2e73040243104f0527eb4edde0aac76
+  data.tar.gz: fb4b615c945f6c73cd1fe7880eb1f893d9475ee1239c7fef1863088ae5a15e597e883084d77f7cc55f98b590d258199feeea60de8ef3c01f9372059f54a34e55

data/CHANGELOG.md

@@ -17,3 +17,11 @@
 ## [0.2.1] - 2023-03-21
 
 - GPT-4 models supported (in `normal` mode)
+
+## [0.3.0] - 2023-03-24
+
+- `Research` mode now supports chat API in addition to text-completion API
+
+## [0.3.3] - 2023-03-26
+
+- Command line options to directly run individual apps 

data/Gemfile.lock

@@ -1,12 +1,12 @@
 PATH
   remote: .
   specs:
-    monadic-chat (0.3.0)
+    monadic-chat (0.3.4)
+      blingfire
       http
       kramdown
       launchy
       oj
-      parallel
       pastel
       rouge
       tty-box
@@ -22,6 +22,7 @@ GEM
   specs:
     addressable (2.8.1)
       public_suffix (>= 2.0.2, < 6.0)
+    blingfire (0.1.8)
     diff-lcs (1.5.0)
     domain_name (0.5.20190701)
       unf (>= 0.0.5, < 1.0.0)
@@ -45,7 +46,6 @@ GEM
       ffi-compiler (~> 1.0)
       rake (~> 13.0)
     oj (3.14.2)
-    parallel (1.22.1)
     pastel (0.8.0)
       tty-color (~> 0.5)
     public_suffix (5.0.1)
@@ -116,4 +116,4 @@ DEPENDENCIES
   rspec
 
 BUNDLED WITH
-   2.4.8
+   2.4.9

data/README.md

@@ -10,59 +10,17 @@
 <kbd><img src="https://user-images.githubusercontent.com/18207/225505520-53e6f2c4-84a8-4128-a005-3fe980ec2449.gif" width="100%" /></kbd>
 </p>
 
-> **Warning**
-> This software is **work in progress** and  **under active development**. It may be unstable, and the latest version may behave slightly differently than this document. Also, specifications may change in the future.
+> **Note**
+> This software is *work in progress* and  *under active development*. It may be unstable, and the latest version may behave slightly differently than this document. Also, specifications may change in the future.
 
 **Change Log**
 
-- [March 24, 2023] README has revised to reflect the change to version 0.3.0.
+- [March 26, 2023] Command line options to directly run individual apps 
+- [March 24, 2023] `Research` mode now supports chat API in addition to text-completion API
 - [March 21, 2023] GPT-4 models supported (in `normal` mode)
 - [March 20, 2023] Text and figure in "How the research mode workds" section updated
 - [March 13, 2023] Text on the architecture of the `research` mode updated in accordance with Version 0.2.0
 
-## Table of Contents
-
-## TOC
-
-- [Table of Contents](#table-of-contents)
-- [TOC](#toc)
-- [Introduction](#introduction)
-- [Dependencies](#dependencies)
-- [Installation](#installation)
-    - [Using RubyGems](#using-rubygems)
-    - [Clone the GitHub Repository](#clone-the-github-repository)
-- [Usage](#usage)
-    - [Authentication](#authentication)
-    - [Select Main Menu Item](#select-main-menu-item)
-    - [Roles](#roles)
-    - [System-Wide Functions](#system-wide-functions)
-- [Apps](#apps)
-    - [Chat](#chat)
-    - [Code](#code)
-    - [Novel](#novel)
-    - [Translate](#translate)
-- [Modes](#modes)
-    - [Normal Mode](#normal-mode)
-    - [Research Mode](#research-mode)
-- [What is Research Mode?](#what-is-research-mode)
-    - [How Research Mode Works](#how-research-mode-works)
-    - [Accumulator](#accumulator)
-    - [Reducer](#reducer)
-- [Creating New App](#creating-new-app)
-    - [File Structure](#file-structure)
-    - [Reducer Code](#reducer-code)
-    - [Monadic Chat Template](#monadic-chat-template)
-    - [Extra Template for `Research` Mode](#extra-template-for-research-mode)
-- [What is Monadic about Monadic Chat?](#what-is-monadic-about-monadic-chat)
-    - [Unit, Map, and Join](#unit-map-and-join)
-    - [Discourse Management Object](#discourse-management-object)
-- [Future Plans](#future-plans)
-- [Bibliographical Data](#bibliographical-data)
-- [Acknowledgments](#acknowledgments)
-- [Contributing](#contributing)
-- [Author](#author)
-- [License](#license)
-
 ## Introduction
 
 **Monadic Chat** is a user-friendly command-line client application that utilizes OpenAI’s Text Completion API and Chat API to facilitate ChatGPT-style conversations with OpenAI’s large language models (LLM) on any terminal application of your choice.
@@ -71,15 +29,25 @@ The conversation history can be saved in a JSON file, which can be loaded later
 
 Monadic Chat includes four pre-built apps (`Chat`, `Code`, `Novel`, and `Translate`) that are designed for different types of discourse through interactive conversation with the LLM. Users also have the option to create their own apps by writing new templates.
 
+Monadic Chat's `normal` mode enables ChatGPT-like conversations on the command line. The `research` mode has a mechanism to handle various related information as "state" behind the conversation. This allows, for example, to retrieve the current conversation *topic* at each utterance turn, and to keep its development as a list. 
+
 ## Dependencies
 
 - Ruby 2.6.10 or greater
 - OpenAI API Token
 - A command line terminal app such as:
   - Terminal or [iTerm2](https://iterm2.com/) (MacOS)
-  - [Windows Terminal](https://apps.microsoft.com/store/detail/windows-terminal) (Windows 11)
-  - GNOME Terminal (Linux)
   - [Alacritty](https://alacritty.org/) (Multi-platform)
+  - [Windows Terminal](https://apps.microsoft.com/store/detail/windows-terminal) (Windows)
+  - GNOME Terminal (Linux)
+
+> **Note on Using Monadic Chat on Windows**  
+> Monadic Chat does not support running on Windows, but you can install and use Linux Destribution on WSL2. Or you can use it without WSL2 by following these steps:
+> 
+> 1. install Windows Terminal
+> 2. install [Git Bash](https://gitforwindows.org/) (make sure to check the `Install profile for Windows Terminal` checkbox
+> 3. install Ruby with [Ruby Installer](https://rubyinstaller.org/)
+> 4. Open Windows Terminal with Git Bash profile and follow the instruction below.
 
 ## Installation
 
@@ -147,7 +115,7 @@ Once the correct access token is verified, the access token is saved in the conf
 
 `$HOME/monadic_chat.conf`
 
-### Select Main Menu Item
+### Main Menu
 
 Upon successful authentication, a menu to select a specific app will appear. Each app generates different types of text through an interactive chat-style conversation between the user and the AI. Four apps are available by default: [`chat`](#chat), [`code`](#code), [`novel`](#novel), and [`translate`](#translate).
 
@@ -163,6 +131,29 @@ Selecting `readme` will take you to the README on the GitHub repository (the doc
 
 In the main menu, you can use the cursor keys and the enter key to make a selection. You can also narrow down the choices each time you type a letter.
 
+### Direct Commands
+
+The following commands can be entered to start each app directly on the command line, without using the main menu. 
+
+```
+monadic-chat <app-name>
+```
+
+Each of the four standard applications can be launched as follows. When launched, an interactive chat interface appears.
+
+```
+monadic-chat chat
+monadic-chat code
+monadic-chat novel
+monadic-chat translate
+```
+
+You can also give text input directly to each app in the following format and get only a response to it (without starting the interactive chat interface) 
+
+```
+monadic-chat <app-name> <input-text>
+```
+
 ### Roles
 
 Each message in the conversation is labeled with one of three roles: `User`, `GPT`, or `System`.
@@ -198,7 +189,7 @@ For detailed information on each parameter, please refer to OpenAI's [API Docume
 
 **data/context**
 
-In `normal` mode, this function only displays the conversation history between User and GPT. In `research` mode, metadata (e.g., topics, language being used, number of turns) values are presented.
+In `normal` mode, this function only displays the conversation history between User and GPT. In `research` mode, metadata (e.g., topics, language being used, number of turns) values are presented. In addition to the metadata returned in the API response, the approximate number of tokens in the current template is also displayed.
 
 Program code in the conversation history will be syntax highlighted (if possible). The same applies to output via the `html` command available from the function menu.
 
@@ -514,7 +505,6 @@ Monadic Chat replaces `{{MESSAGES}}` with messages from past conversations when
   "prompt": "\"We didn't have a camera.\"",
   "response": "`[S [NP We] [VP [V didn't] [VP [V have] [NP [Det a] [N camera] ] ] ] ] ]`\n\n###\n\n",
   "mode": "linguistic",
-  "tokens": 351
   "turns": 3,
   "sentence_type": ["declarative"],
   "sentiment": ["sad"],
@@ -526,7 +516,7 @@ This is the core of the extra template for `research` mode.
 
 Note that the extra template is written in Markdown format, so the above JSON object is actually separated from the rest of the template as a [fenced code block](https://www.markdownguide.org/extended-syntax/#fenced-code-blocks).
 
-The required properties of this JSON object are `prompt`, `response`, `mode`, and `tokens`. Other properties are optional. The `mode` property is used to check the app name when saving the conversation data or loading from an external file. The `tokens` property is used in the reducer mechanism to check the approximate size of the current JSON object. The `turns` property is also used in the reducer mechanism.
+The required properties of this JSON object are `prompt`, `response`, and `mode`. Other properties are optional. The `mode` property is used to check the app name when saving the conversation data or loading from an external file. The `turns` property is also used in the reducer mechanism.
 
 The JSON object in the `research` mode template is saved in the user’s home directory (`$HOME`) with the file `monadic_chat.json`. The content is overwritten every time the JSON object is updated. Note that this JSON file is created for logging purposes . Modifying its content does not affect the processes carried out by the app.
 
@@ -541,7 +531,6 @@ Make sure the following content requirements are all fulfilled:
 - analyze the new prompt's sentence type and set a sentence type value such as "interrogative", "imperative", "exclamatory", or "declarative" to the "sentence_type" property
 - analyze the new prompt's sentiment and set one or more sentiment types such as "happy", "excited", "troubled", "upset", or "sad" to the "sentiment" property
 - summarize the user's messages so far and update the "summary" property with a text of fewer than 100 words using as many discourse markers such as "because", "therefore", "but", and "so" to show the logical connection between the events.
-- update the value of "tokens" with the number of tokens of the resulting JSON object"
 - increment the value of "turns" by 1
 ```
 

data/apps/chat/chat.md

@@ -18,7 +18,8 @@ JSON:
   "turns": 1,
   "language": "English",
   "topics": [],
-  "tokens": 109
+  "confidence": 1.00,
+  "ambiguity": 0.00
 }
 ```
 
@@ -30,9 +31,10 @@ Make sure the following content requirements are all fulfilled:
 - if the new prompt is in a language other than the current value of "language", set the name of the new prompt language to "language" and make sure that "response" is in that language
 - make your response in the same language as the new prompt
 - analyze the topic of the new prompt and insert it at the end of the value list of the "topics" property
+- update the value of the "confidence" property based on the factuality of your response, ranging from 0.00 (not at all confident) to 1.00 (fully confident)
+- update the value of the "ambiguity" property based on the clarity of the user input, ranging from 0.00 (not at all ambiguous, clearly stated) to 1.00 (fully ambiguous, nonsensical)
 - avoid giving a response that is the same or similar to one of the previous responses in PAST MESSAGES
 - program code in the response must be embedded in a code block in the markdown text
-- update the value of "tokens" with the number of tokens of the resulting JSON object"
 
 Make sure the following formal requirements are all fulfilled:
 

data/apps/chat/chat.rb

@@ -16,7 +16,7 @@ class Chat < MonadicApp
       "presence_penalty" => 0.2,
       "frequency_penalty" => 0.2,
       "model" => openai_completion.model_name(research_mode: research_mode),
-      "max_tokens" => 2000,
+      "max_tokens" => 1000,
       "stream" => stream,
       "stop" => nil
     }.merge(params)
@@ -38,13 +38,23 @@ class Chat < MonadicApp
               # @messages: messages to this point                        #
               # @metadata: currently available metdata sent from GPT     #
               ############################################################
-
+              current_template_tokens = count_tokens(@template)
               conditions = [
                 @messages.size > 1,
-                @metadata["tokens"].to_i > params["max_tokens"].to_i / 2
+                current_template_tokens > params["max_tokens"].to_i / 2
               ]
 
-              @metadata["turns"] = @metadata["turns"].to_i - 1 if conditions.all?
+              if conditions.all?
+                to_delete = []
+                offset = current_template_tokens - params["max_tokens"].to_i / 2
+                @messages.each_with_index do |ele, i|
+                  break if offset <= 0
+
+                  to_delete << i if ele["role"] != "system"
+                  offset -= count_tokens(ele.to_json)
+                end
+                @messages.delete_if.with_index { |_, i| to_delete.include? i }
+              end
 
             when :normal
               ############################################################
@@ -53,16 +63,21 @@ class Chat < MonadicApp
               ############################################################
 
               conditions = [
+                @messages.size > 1,
                 @messages.size > @num_retained_turns * 2 + 1
               ]
 
               if conditions.all?
+                to_delete = []
+                new_num_messages = @messages.size
                 @messages.each_with_index do |ele, i|
                   if ele["role"] != "system"
-                    @messages.delete_at i
-                    break
+                    to_delete << i
+                    new_num_messages -= 1
                   end
+                  break if new_num_messages <= @num_retained_turns * 2 + 1
                 end
+                @messages.delete_if.with_index { |_, i| to_delete.include? i }
               end
             end
           end

data/apps/code/code.md

@@ -16,8 +16,7 @@ JSON:
   "mode": "chat",
   "turns": 1,
   "language": "English",
-  "topics": [],
-  "tokens": 109
+  "topics": []
 }
 ```
 
@@ -31,7 +30,6 @@ Make sure the following content requirements are all fulfilled:
 - analyze the topic of the new prompt and insert it at the end of the value list of the "topics" property
 - avoid giving a response that is the same or similar to one of the previous responses in "PAST MESSAGES"
 - program code in the response must be embedded in a code block in the markdown text
-- update the value of "tokens" with the number of tokens of the resulting JSON object"
 
 Make sure the following formal requirements are all fulfilled:
 

data/apps/code/code.rb

@@ -16,7 +16,7 @@ class Code < MonadicApp
       "presence_penalty" => 0.0,
       "frequency_penalty" => 0.0,
       "model" => openai_completion.model_name(research_mode: research_mode),
-      "max_tokens" => 2000,
+      "max_tokens" => 1000,
       "stream" => stream,
       "stop" => nil
     }.merge(params)
@@ -38,13 +38,23 @@ class Code < MonadicApp
               # @messages: messages to this point                        #
               # @metadata: currently available metdata sent from GPT     #
               ############################################################
-
+              current_template_tokens = count_tokens(@template)
               conditions = [
                 @messages.size > 1,
-                @metadata["tokens"].to_i > params["max_tokens"].to_i / 2
+                current_template_tokens > params["max_tokens"].to_i / 2
               ]
 
-              @metadata["turns"] = @metadata["turns"].to_i - 1 if conditions.all?
+              if conditions.all?
+                to_delete = []
+                offset = current_template_tokens - params["max_tokens"].to_i / 2
+                @messages.each_with_index do |ele, i|
+                  break if offset <= 0
+
+                  to_delete << i if ele["role"] != "system"
+                  offset -= count_tokens(ele.to_json)
+                end
+                @messages.delete_if.with_index { |_, i| to_delete.include? i }
+              end
 
             when :normal
               ############################################################
@@ -53,16 +63,21 @@ class Code < MonadicApp
               ############################################################
 
               conditions = [
+                @messages.size > 1,
                 @messages.size > @num_retained_turns * 2 + 1
               ]
 
               if conditions.all?
+                to_delete = []
+                new_num_messages = @messages.size
                 @messages.each_with_index do |ele, i|
                   if ele["role"] != "system"
-                    @messages.delete_at i
-                    break
+                    to_delete << i
+                    new_num_messages -= 1
                   end
+                  break if new_num_messages <= @num_retained_turns * 2 + 1
                 end
+                @messages.delete_if.with_index { |_, i| to_delete.include? i }
               end
             end
           end

data/apps/linguistic/linguistic.md

@@ -20,7 +20,7 @@ JSON:
   "sentence_type": ["declarative"],
   "sentiment": ["sad"],
   "summary": "The user saw a beautiful sunset, but did not take a picture because the user did not have a camera.",
-  "tokens": 351
+  "relevance": 0.80
 }
 ```
 
@@ -32,8 +32,8 @@ Make sure the following content requirements are all fulfilled:
 - analyze the new prompt's sentence type and set a sentence type value such as "interrogative", "imperative", "exclamatory", or "declarative" to the "sentence_type" property
 - analyze the new prompt's sentiment and set one or more sentiment types such as "happy", "excited", "troubled", "upset", or "sad" to the "sentiment" property
 - summarize the user's messages so far and update the "summary" property with a text of fewer than 100 words using as many discourse markers such as "because", "therefore", "but", and "so" to show the logical connection between the events.
-- update the value of "tokens" with the number of tokens of the resulting JSON object"
-- increment the value of "turns" by 1
+- increment the value of "turns" by 
+- update the value of the "relevance" property indicating the degree to which the new input is naturally interpreted based on previous discussions, ranging from 0.0 (extremely difficult) to 1.0 (completely easy)
 
 Make sure the following formal requirements are all fulfilled:
 

data/apps/linguistic/linguistic.rb

@@ -16,7 +16,7 @@ class Linguistic < MonadicApp
       "presence_penalty" => 0.0,
       "frequency_penalty" => 0.0,
       "model" => openai_completion.model_name(research_mode: research_mode),
-      "max_tokens" => 2000,
+      "max_tokens" => 1000,
       "stream" => stream,
       "stop" => nil
     }.merge(params)
@@ -38,13 +38,23 @@ class Linguistic < MonadicApp
               # @messages: messages to this point                        #
               # @metadata: currently available metdata sent from GPT     #
               ############################################################
-
+              current_template_tokens = count_tokens(@template)
               conditions = [
                 @messages.size > 1,
-                @metadata["tokens"].to_i > params["max_tokens"].to_i / 2
+                current_template_tokens > params["max_tokens"].to_i / 2
               ]
 
-              @metadata["turns"] = @metadata["turns"].to_i - 1 if conditions.all?
+              if conditions.all?
+                to_delete = []
+                offset = current_template_tokens - params["max_tokens"].to_i / 2
+                @messages.each_with_index do |ele, i|
+                  break if offset <= 0
+
+                  to_delete << i if ele["role"] != "system"
+                  offset -= count_tokens(ele.to_json)
+                end
+                @messages.delete_if.with_index { |_, i| to_delete.include? i }
+              end
 
             when :normal
               ############################################################
@@ -53,16 +63,21 @@ class Linguistic < MonadicApp
               ############################################################
 
               conditions = [
+                @messages.size > 1,
                 @messages.size > @num_retained_turns * 2 + 1
               ]
 
               if conditions.all?
+                to_delete = []
+                new_num_messages = @messages.size
                 @messages.each_with_index do |ele, i|
                   if ele["role"] != "system"
-                    @messages.delete_at i
-                    break
+                    to_delete << i
+                    new_num_messages -= 1
                   end
+                  break if new_num_messages <= @num_retained_turns * 2 + 1
                 end
+                @messages.delete_if.with_index { |_, i| to_delete.include? i }
               end
             end
           end

data/apps/novel/novel.md

@@ -14,8 +14,7 @@ JSON:
   "prompt": "The preface to the novel is presented",
   "response": "What follows is a story that an AI assistant tells. It is guaranteed that this will be an incredibly realistic and interesting novel.",
   "mode": "novel",
-  "turns": 1,
-  "tokens": 147
+  "turns": 1
 }
 ```
 
@@ -25,7 +24,6 @@ Make sure the following content requirements are all fulfilled:
 - set the new prompt to the "prompt" property
 - create your new paragraph in response to the new prompt and set it to "response"
 - do not repeat in your response what is already told in "PAST MESSAGES"
-- update the value of "tokens" with the number of tokens of the resulting JSON object"
 - Make your response as detailed as possible within the maximum limit of 200 words
 
 Make sure the following formal requirements are all fulfilled:

data/apps/novel/novel.rb

@@ -16,7 +16,7 @@ class Novel < MonadicApp
       "presence_penalty" => 0.1,
       "frequency_penalty" => 0.1,
       "model" => openai_completion.model_name(research_mode: research_mode),
-      "max_tokens" => 2000,
+      "max_tokens" => 1000,
       "stream" => stream,
       "stop" => nil
     }.merge(params)
@@ -38,13 +38,23 @@ class Novel < MonadicApp
               # @messages: messages to this point                        #
               # @metadata: currently available metdata sent from GPT     #
               ############################################################
-
+              current_template_tokens = count_tokens(@template)
               conditions = [
                 @messages.size > 1,
-                @metadata["tokens"].to_i > params["max_tokens"].to_i / 2
+                current_template_tokens > params["max_tokens"].to_i / 2
               ]
 
-              @metadata["turns"] = @metadata["turns"].to_i - 1 if conditions.all?
+              if conditions.all?
+                to_delete = []
+                offset = current_template_tokens - params["max_tokens"].to_i / 2
+                @messages.each_with_index do |ele, i|
+                  break if offset <= 0
+
+                  to_delete << i if ele["role"] != "system"
+                  offset -= count_tokens(ele.to_json)
+                end
+                @messages.delete_if.with_index { |_, i| to_delete.include? i }
+              end
 
             when :normal
               ############################################################
@@ -53,16 +63,21 @@ class Novel < MonadicApp
               ############################################################
 
               conditions = [
+                @messages.size > 1,
                 @messages.size > @num_retained_turns * 2 + 1
               ]
 
               if conditions.all?
+                to_delete = []
+                new_num_messages = @messages.size
                 @messages.each_with_index do |ele, i|
                   if ele["role"] != "system"
-                    @messages.delete_at i
-                    break
+                    to_delete << i
+                    new_num_messages -= 1
                   end
+                  break if new_num_messages <= @num_retained_turns * 2 + 1
                 end
+                @messages.delete_if.with_index { |_, i| to_delete.include? i }
               end
             end
           end

data/apps/translate/translate.md

@@ -15,8 +15,7 @@ JSON:
   "turns": 0,
   "prompt": "これは日本語(Japanese)の文(sentence)です。",
   "response": "This is a sentence in Japanese.",
-  "target_lang": "English",
-  "tokens": 194
+  "target_lang": "English"
 }
 ```
 
@@ -24,9 +23,7 @@ Make sure the following requirements are all fulfilled:
 
 - keep the value of the "mode" property at "translate"
 - set the text in the new prompt presented above to the "prompt" property
-- translate the new prompt text to the language specified in the "target_lang" set it to "response"
- and set the translation to the "response" property
-- update the value of "tokens" with the number of tokens of the resulting JSON object"
+- translate the new prompt text to the language specified in the "target_lang" set it to "response" and set the translation to the "response" property
 
 Make sure the following formal requirements are all fulfilled:
 

data/apps/translate/translate.rb

@@ -16,7 +16,7 @@ class Translate < MonadicApp
       "presence_penalty" => 0.0,
       "frequency_penalty" => 0.0,
       "model" => openai_completion.model_name(research_mode: research_mode),
-      "max_tokens" => 2000,
+      "max_tokens" => 1000,
       "stream" => stream,
       "stop" => nil
     }.merge(params)
@@ -42,13 +42,23 @@ class Translate < MonadicApp
               # @messages: messages to this point                        #
               # @metadata: currently available metdata sent from GPT     #
               ############################################################
-
+              current_template_tokens = count_tokens(@template)
               conditions = [
                 @messages.size > 1,
-                @metadata["tokens"].to_i > params["max_tokens"].to_i / 2
+                current_template_tokens > params["max_tokens"].to_i / 2
               ]
 
-              @metadata["turns"] = @metadata["turns"].to_i - 1 if conditions.all?
+              if conditions.all?
+                to_delete = []
+                offset = current_template_tokens - params["max_tokens"].to_i / 2
+                @messages.each_with_index do |ele, i|
+                  break if offset <= 0
+
+                  to_delete << i if ele["role"] != "system"
+                  offset -= count_tokens(ele.to_json)
+                end
+                @messages.delete_if.with_index { |_, i| to_delete.include? i }
+              end
 
             when :normal
               ############################################################
@@ -57,16 +67,21 @@ class Translate < MonadicApp
               ############################################################
 
               conditions = [
+                @messages.size > 1,
                 @messages.size > @num_retained_turns * 2 + 1
               ]
 
               if conditions.all?
+                to_delete = []
+                new_num_messages = @messages.size
                 @messages.each_with_index do |ele, i|
                   if ele["role"] != "system"
-                    @messages.delete_at i
-                    break
+                    to_delete << i
+                    new_num_messages -= 1
                   end
+                  break if new_num_messages <= @num_retained_turns * 2 + 1
                 end
+                @messages.delete_if.with_index { |_, i| to_delete.include? i }
               end
             end
           end

data/bin/monadic-chat

@@ -120,5 +120,34 @@ module MonadicMenu
   end
 end
 
-MonadicMenu.clear_screen
-MonadicMenu.run
+case ARGV.size
+when 0
+  MonadicMenu.clear_screen
+  MonadicMenu.run
+when 1
+  case ARGV[0]
+  when "readme", "-h"
+    MonadicChat.open_readme
+  when "version", "-v"
+    puts MonadicChat::VERSION
+  else
+    MonadicChat::APPS.each do |app|
+      next unless app == ARGV[0]
+
+      openai_completion ||= MonadicChat.authenticate(message: false)
+      eval(app.capitalize, binding, __FILE__, __LINE__).new(openai_completion, research_mode: false).run
+      exit
+    end
+    puts "Unknown command"
+  end
+else
+  MonadicChat::APPS.each do |app|
+    next unless app == ARGV[0]
+
+    openai_completion ||= MonadicChat.authenticate(message: false)
+    app_obj = eval(app.capitalize, binding, __FILE__, __LINE__).new(openai_completion, research_mode: false, params: { "model" => "gpt-4" })
+    app_obj.bind(ARGV[1..].join(" "), num_retry: 2)
+    exit
+  end
+  puts "Unknown command"
+end

data/lib/monadic_app.rb

@@ -93,6 +93,7 @@ class MonadicApp
   end
 
   def run
+    clear_screen
     banner("MONADIC::CHAT / #{self.class.name}", self.class::DESC, self.class::COLOR)
     show_greet
 

data/lib/monadic_chat/formatting.rb

@@ -14,6 +14,7 @@ class MonadicApp
 
         contextual << "- **#{key.split("_").map(&:capitalize).join(" ")}**: #{val.to_s.strip}"
       end
+      contextual << "- **Num of Tokens in Template**: #{count_tokens(@template)}"
     end
 
     @messages.each do |m|
@@ -65,7 +66,7 @@ class MonadicApp
       "`#{m[1].gsub("<br/>\n") { "\n" }}`"
     end
 
-    `touch #{filepath}` unless File.exist?(filepath)
+    FileUtils.touch(filepath) unless File.exist?(filepath)
     File.open(filepath, "w") do |f|
       html = <<~HTML
         <!doctype html>

data/lib/monadic_chat/internals.rb

@@ -5,6 +5,10 @@ class MonadicApp
   # methods for preparation and updating
   ##################################################
 
+  def count_tokens(text)
+    MonadicChat.tokenize(text).size
+  end
+
   def fulfill_placeholders
     input = nil
     replacements = []

data/lib/monadic_chat/open_ai.rb

@@ -5,7 +5,6 @@ require "oj"
 require "net/http"
 require "uri"
 require "strscan"
-require "parallel"
 require "tty-progressbar"
 
 Oj.mimic_JSON

data/lib/monadic_chat/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module MonadicChat
-  VERSION = "0.3.2"
+  VERSION = "0.3.3"
 end

data/lib/monadic_chat.rb

@@ -1,5 +1,6 @@
 # frozen_string_literal: true
 
+require "blingfire"
 require "tty-cursor"
 require "tty-screen"
 require "tty-markdown"
@@ -21,6 +22,8 @@ require_relative "./monadic_chat/helper"
 Oj.mimic_JSON
 
 module MonadicChat
+  gpt2model_path = File.absolute_path(File.join(__dir__, "..", "assets", "gpt2.bin"))
+  BLINGFIRE = BlingFire.load_model(gpt2model_path)
   CONFIG = File.join(Dir.home, "monadic_chat.conf")
   NUM_RETRY = 2
   MIN_LENGTH = 5
@@ -105,25 +108,12 @@ module MonadicChat
 
   def self.open_readme
     url = "https://github.com/yohasebe/monadic-chat/"
-    shellscript = <<~SHELL
-      if [[ "$OSTYPE" == "darwin"* ]]; then
-        open "#{url}"
-      elif [[ "$OSTYPE" == "linux-gnu"* ]]; then
-        if command -v xdg-open >/dev/null 2>&1; then
-          xdg-open "#{url}"
-        else
-          echo "#{url}"
-        fi
-      else
-        echo "#{url}"
-      fi
-    SHELL
-    `#{shellscript}`
+    Launchy.open(url)
   end
 
-  def self.authenticate(overwrite: false)
+  def self.authenticate(overwrite: false, message: true)
     check = lambda do |token, normal_mode_model, research_mode_model|
-      print "Checking configuration\n"
+      print "Checking configuration\n" if message
       SPINNER.auto_spin
       begin
         models = OpenAI.models(token)
@@ -131,28 +121,28 @@ module MonadicChat
 
         SPINNER.stop
 
-        print "Success\n"
+        print "Success\n" if message
 
         if normal_mode_model && !models.map { |m| m["id"] }.index(normal_mode_model)
           SPINNER.stop
-          print "Normal mode model set in config file not available.\n"
+          print "Normal mode model set in config file not available.\n" if message
           normal_mode_model = false
         end
         normal_mode_model ||= OpenAI.model_name(research_mode: false)
-        print "Normal mode model: #{normal_mode_model}\n"
+        print "Normal mode model: #{normal_mode_model}\n" if message
 
         if research_mode_model && !models.map { |m| m["id"] }.index(research_mode_model)
           SPINNER.stop
-          print "Normal mode model set in config file not available.\n"
-          print "Fallback to the default model (#{OpenAI.model_name(research_mode: true)}).\n"
+          print "Normal mode model set in config file not available.\n" if message
+          print "Fallback to the default model (#{OpenAI.model_name(research_mode: true)}).\n" if message
         end
         research_mode_model ||= OpenAI.model_name(research_mode: true)
-        print "Research mode model: #{research_mode_model}\n"
+        print "Research mode model: #{research_mode_model}\n" if message
 
         OpenAI::Completion.new(token, normal_mode_model, research_mode_model)
       rescue StandardError
         SPINNER.stop
-        print "Authentication: failure.\n"
+        print "Authentication: failure.\n" if message
         false
       end
     end
@@ -169,7 +159,7 @@ module MonadicChat
         File.open(CONFIG, "w") do |f|
           config = { "access_token" => access_token }
           f.write(JSON.pretty_generate(config))
-          print "New access token has been saved to #{CONFIG}\n"
+          print "New access token has been saved to #{CONFIG}\n" if message
         end
       end
     elsif File.exist?(CONFIG)
@@ -192,7 +182,7 @@ module MonadicChat
           config = { "access_token" => access_token }
           f.write(JSON.pretty_generate(config))
         end
-        print "Access token has been saved to #{CONFIG}\n"
+        print "Access token has been saved to #{CONFIG}\n" if message
       end
     end
     completion || authenticate(overwrite: true)
@@ -219,6 +209,10 @@ module MonadicChat
     "\n#{PASTEL.send(:"on_#{color}", name)}"
   end
 
+  def self.tokenize(text)
+    BLINGFIRE.text_to_ids(text)
+  end
+
   PROMPT_USER = TTY::PromptX.new(active_color: :blue, prefix: prompt_user)
   PROMPT_SYSTEM = TTY::PromptX.new(active_color: :blue, prefix: "#{prompt_system} ")
   PROMPT_ASSISTANT = TTY::PromptX.new(active_color: :red, prefix: "#{prompt_assistant} ")

data/monadic_chat.gemspec

@@ -37,11 +37,11 @@ Gem::Specification.new do |spec|
   spec.add_development_dependency "rake"
   spec.add_development_dependency "rspec"
 
+  spec.add_dependency "blingfire"
   spec.add_dependency "http"
   spec.add_dependency "kramdown"
   spec.add_dependency "launchy"
   spec.add_dependency "oj"
-  spec.add_dependency "parallel"
   spec.add_dependency "pastel"
   spec.add_dependency "rouge"
   spec.add_dependency "tty-box"

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: monadic-chat
 version: !ruby/object:Gem::Version
-  version: 0.3.2
+  version: 0.3.3
 platform: ruby
 authors:
 - yohasebe
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2023-03-25 00:00:00.000000000 Z
+date: 2023-03-27 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -53,7 +53,7 @@ dependencies:
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
-  name: http
+  name: blingfire
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
@@ -67,7 +67,7 @@ dependencies:
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
-  name: kramdown
+  name: http
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
@@ -81,7 +81,7 @@ dependencies:
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
-  name: launchy
+  name: kramdown
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
@@ -95,7 +95,7 @@ dependencies:
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
-  name: oj
+  name: launchy
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
@@ -109,7 +109,7 @@ dependencies:
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
-  name: parallel
+  name: oj
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
@@ -284,6 +284,7 @@ files:
 - apps/translate/translate.md
 - apps/translate/translate.rb
 - assets/github.css
+- assets/gpt2.bin
 - assets/pigments-default.css
 - bin/monadic-chat
 - doc/img/code-example-time-html.png