only_ever_generator 0.3.1 → 0.3.2

package/dist/bootstrap/app.js

@@ -13,8 +13,8 @@ exports.OnlyEverGenerator = void 0;
 const generate_cards_1 = require("../card_gen/generate_cards");
 const parse_source_content_1 = require("../parse/parse_source_content");
 const open_ai_service_1 = require("../services/open_ai_service");
-const card_gen_prompt_1 = require("../constants/prompts/card_gen_prompt");
-const typology_prompt_1 = require("../constants/prompts/typology_prompt");
+// import { returnCardGenPrompt } from "../constants/prompts/card_gen_prompt";
+// import { returnTypologyPrompt } from "../constants/prompts/typology_prompt";
 const generate_typology_1 = require("../typology_gen/generate_typology");
 const generate_args_1 = require("../utils/generate_args");
 const calculate_gap_fill_1 = require("../gap_fill/calculate_gap_fill");
@@ -41,8 +41,8 @@ class OnlyEverGenerator {
             // parsedData.type == 'cards' ? this.typologyResponse = parsedData.taxonomy :  this.typologyResponse = null;
             this.typologyResponse = generationContent.content.taxonomy;
         this.expectedFields = generationContent.content.fields; //returnFields();
-        this.promptForTypology = (0, typology_prompt_1.returnTypologyPrompt)(generationContent.prompt.typology);
-        this.promptForCardGen = (0, card_gen_prompt_1.returnCardGenPrompt)(generationContent.prompt.card_generation);
+        this.promptForTypology = generationContent.prompt.typology;
+        this.promptForCardGen = generationContent.prompt.card_generation;
     }
     generate() {
         return __awaiter(this, arguments, void 0, function* (generate_typology = false, generate_card = false) {
@@ -94,7 +94,8 @@ class OnlyEverGenerator {
     }
     generateCard(prompt, additionalContent, isGapFill) {
         return __awaiter(this, void 0, void 0, function* () {
-            let generateCardsResp = yield new generate_cards_1.GenerateCards(this.openAiService).generateCards(prompt !== null && prompt !== void 0 ? prompt : "", this.parsedContent + additionalContent, isGapFill);
+            var _a;
+            let generateCardsResp = yield new generate_cards_1.GenerateCards(this.openAiService).generateCards(prompt !== null && prompt !== void 0 ? prompt : "", JSON.stringify(this.parsedContent) + additionalContent, isGapFill, (_a = this.parsedContent.headings) !== null && _a !== void 0 ? _a : []);
             // let response =  await this.openAiService?.sendRequest(prompt,this.parsedContent);
             // response['type'] = 'card_gen';
             return generateCardsResp;

package/dist/card_gen/generate_cards.js

@@ -15,7 +15,7 @@ class GenerateCards {
     constructor(openAiService) {
         this.openAiService = openAiService;
     }
-    generateCards(prompt, parsedContent, isGapFill) {
+    generateCards(prompt, parsedContent, isGapFill, headings) {
         return __awaiter(this, void 0, void 0, function* () {
             var _a, _b, _c;
             let response = yield ((_a = this.openAiService) === null || _a === void 0 ? void 0 : _a.sendRequest(prompt, parsedContent));
@@ -31,7 +31,7 @@ class GenerateCards {
             if (response.status_code == 200) {
                 response.metadata.status = "completed";
                 //return response;
-                return this.parse(response, isGapFill);
+                return this.parse(response, isGapFill, headings);
             }
             else {
                 response.metadata.status = "failed";
@@ -39,7 +39,7 @@ class GenerateCards {
             }
         });
     }
-    parse(generatedData, isGapFill) {
+    parse(generatedData, isGapFill, headings) {
         try {
             const cardData = [];
             let usage_data = generatedData.metadata;
@@ -50,6 +50,11 @@ class GenerateCards {
             const type = generatedData.type;
             if (unparsedTestCards !== undefined && unparsedTestCards.length != 0) {
                 for (let elem of unparsedTestCards) {
+                    if (headings.includes(elem.card_reference)) {
+                    }
+                    else {
+                        elem.card_reference = '';
+                    }
                     if (elem.type == "flash") {
                         cardData.push(this.parseFlashCard(elem));
                     }

package/dist/constants/prompt_data.js

@@ -1,6 +1,8 @@
 "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.returnPromptData = void 0;
+const card_gen_prompt_1 = require("./prompts/card_gen_prompt");
+const typology_prompt_1 = require("./prompts/typology_prompt");
 const promptData = {
     typology: {
         role: `
@@ -293,6 +295,9 @@ const promptData = {
     }
 };
 function returnPromptData() {
-    return promptData;
+    return {
+        "typology": (0, typology_prompt_1.returnTypologyPrompt)(),
+        "card_generation": (0, card_gen_prompt_1.returnCardGenPrompt)(''),
+    };
 }
 exports.returnPromptData = returnPromptData;

package/dist/constants/prompts/card_gen_prompt.js

@@ -23,7 +23,7 @@ exports.returnCardGenPrompt = void 0;
 //      {
 //          "type": "flash" | "mcq" | "cloze" | "match",
 //          "card_content": { "front": "...", "back": "..." | "prompt": "...", "choices": [ ... ] | "prompt": "...", "options": [ ... ] | "...": "...", "....": "..." },
-//          "card_reference": "source_title#heading",
+//          "card_reference": "heading",
 //          "concepts": ["Concept1", "Concept2", "..."],
 //          "facts": ["Fact1", "Fact2", "..."],
 //          "bloom_level": <1-5>
@@ -80,7 +80,7 @@ exports.returnCardGenPrompt = void 0;
 //      "front": "<content for the front>",
 //      "back": "<content for the back>"
 //  },
-//  "card_reference": "source_title#heading",
+//  "card_reference": "heading",
 //  "concepts": ["Concept1", "Concept2", "..."],
 //  "facts": ["Fact1", "Fact2", "..."],
 //  "bloom_level": <1-5>
@@ -98,7 +98,7 @@ exports.returnCardGenPrompt = void 0;
 //          "... up to 8 choices"
 //      ]
 //  },
-//  "card_reference": "source_title#heading",
+//  "card_reference": "heading",
 //  "concepts": ["Concept1", "Concept2", "..."],
 //  "facts": ["Fact1", "Fact2", "..."],
 //  "bloom_level": <1-5>
@@ -122,7 +122,7 @@ exports.returnCardGenPrompt = void 0;
 //          "... up to 8 choices"
 //      ]
 //  },
-//  "card_reference": "source_title#heading",
+//  "card_reference": "heading",
 //  "concepts": ["Concept1", "Concept2", "..."],
 //  "facts": ["Fact1", "Fact2", "..."],
 //  "bloom_level": <1-5>
@@ -141,14 +141,14 @@ exports.returnCardGenPrompt = void 0;
 //      "left_choice 2": "right_choice 2",
 //      "... up to 8 total pairs"
 //  },
-//  "card_reference": "source_title#heading",
+//  "card_reference": "heading",
 //  "concepts": ["Concept1", "Concept2", "..."],
 //  "facts": ["Fact1", "Fact2", "..."],
 //  "bloom_level": <1-5>
 // }
 // •	Maximum character length for each item in a pair: 42
 //     `,
-//     reference : `Each test card needs a reference. A reference can either be the entire source or a specific heading in the source. Whenever possible, pick a main heading to direct the user to the most relevant part of the source material. The reference schema is as follows: source_title#main_heading, where #main_heading is optional.`,
+//     reference : `Each test card needs a reference. A reference can either be the entire source or a specific heading in the source. Whenever possible, pick a main heading to direct the user to the most relevant part of the source material. The reference schema is as follows: ""main_heading, where #main_heading is optional.`,
 //     checkcoverage: `Once you are done generating the test cards. Go back and evaulate the full list of concepts and fact that include any of the missing concepts or facts along with the list that was provided as the input. 
 // Are there any concept or fact that don't have a test card yet? If yes, go back and create one.
 // Once you are done creating come back to this step again to check if you have full coverage of all the concepts and facts in the source. You can stop generating test questions once you achieve full coverage.
@@ -184,7 +184,7 @@ json
         {
             "type": "flash" | "mcq" | "cloze" | "match",
             "card_content": { "front": "...", "back": "..." | "prompt": "...", "choices": [ ... ] | "prompt": "...", "options": [ ... ] | "right_choice 1": "...", "left_choice 1": "..." },
-            "card_reference": "source_title#heading",
+            "card_reference": "heading",
             "concepts": ["Concept1", "Concept2", "..."],
             "facts": ["Fact1", "Fact2", "..."]
         }
@@ -248,7 +248,7 @@ json
         "front": "<content for the front>",
         "back": "<content for the back>"
     },
-    "card_reference": "source_title#heading",
+    "card_reference": "heading",
     "concepts": ["Concept1", "Concept2", "..."],
     "facts": ["Fact1", "Fact2", "..."],
     "bloom_level": <1-5>
@@ -267,7 +267,7 @@ json
             "... up to 8 choices"
         ]
     },
-    "card_reference": "source_title#heading",
+    "card_reference": "heading",
     "concepts": ["Concept1", "Concept2", "..."],
     "facts": ["Fact1", "Fact2", "..."],
     "bloom_level": <1-5>
@@ -293,7 +293,7 @@ json
             "... up to 8 choices"
         ]
     },
-    "card_reference": "source_title#heading",
+    "card_reference": "heading",
     "concepts": ["Concept1", "Concept2", "..."],
     "facts": ["Fact1", "Fact2", "..."],
     "bloom_level": <1-5>
@@ -313,14 +313,14 @@ json
         "right_choice 2": "left_choice 2",
         "... up to 8 total pairs"
     },
-    "card_reference": "source_title#heading",
+    "card_reference": "heading",
     "concepts": ["Concept1", "Concept2", "..."],
     "facts": ["Fact1", "Fact2", "..."],
     "bloom_level": <1-5>
 }
 •	Maximum character length for each item in a pair: 42
 
-Each test card needs a reference. A reference can either be the entire source or a specific heading in the source. Whenever possible, pick a main heading to direct the user to the most relevant part of the source material. The reference schema is as follows: source_title#main_heading, where #main_heading is optional.
+Each test card needs a reference. A reference can either be the entire source or a specific heading in the source. Whenever possible, pick a main heading to direct the user to the most relevant part of the source material. The reference schema is as follows: ""main_heading, where #main_heading is optional.
 Once you are done generating the test cards. Go back and evaulate the full list of concepts and fact that include any of the missing concepts or facts along with the list that was provided as the input. 
 
 Are there any concept or fact that don't have a test card yet? If yes, go back and create one.

package/dist/constants/prompts/typology_prompt.js

@@ -1,108 +1,30 @@
 "use strict";
-// const typologyPromptString = `
-// As a dedicated assistant at a learning company, your role involves analyzing educational content to categorize and summarize it. You will process content (in JSON format) that represents text and images from diverse sources such as PDFs, book chapters, videos, and websites. Follow these steps:
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.returnTypologyPrompt = void 0;
-// 1. Classify the content into one to three predefined fields of knowledge.
-// 2. Identify key concepts within the content.
-// 3. Detect concrete facts that are empirically verified and relevant to the subject.
-// 4. Decide whether the identified concepts and facts should be used to generate learning materials like flashcards based on their educational value.
-// 5. If the generate_cards is true then summarize the content using a series of summary cards.
-// Please format your findings in this JSON schema:
-// json
-// {
-//     "field": ["primary_field", "secondary_field", "tertiary_field"],
-//     "concepts": ["concept1", "concept2", "concept3", "..."],
-//     "facts": ["fact1", "fact2", "fact3", "..."],
-//     "generate_cards": [
-//         state: true or false,
-//         false_reason: "reason for marking the source as false. Leave empty for true."
-//     ],
-//     "summary_cards": ["summary_card1", "summary_card2", "summary_card3", "..."]
-// }
-// Further instruction on how to perform these tasks are below.
-// Every source must be placed under a field. This is the broadest category of knowledge. A source should belong to at least one and at most 3 fields. Only include fields that a source is strongly associated with. The field names in your response must exactly match the names of 18 fields listed below.
-// 1.	Sciences: Focus on Biology, Chemistry, Physics, Astronomy, Mathematics, and Computer Science.
-// 2.	Technology & Engineering: Emphasize Information Technology, Engineering disciplines, AI, and Robotics.
-// 3.	Humanities & Cultural Studies: Highlight History, Literature, Languages, Arts, Philosophy, and Anthropological Studies.
-// 4.	Social Sciences & Global Studies: Include Sociology, Psychology, Economics, Political Science, Anthropology, and International Relations.
-// 5.	Business & Management: Encompass Entrepreneurship, Marketing, Finance, Leadership, and Ethics.
-// 6.	Health & Medicine: Cover Medical Sciences, Public Health, Nutrition, Wellness, and Mental Health.
-// 7.	Environmental Studies & Earth Sciences: Discuss Ecology, Climate Science, Geology, and Environmental Policy.
-// 8.	Education, Learning & Personal Development: Talk about Educational Theories, Teaching Methods, and Personal Skills.
-// 9.	Creative & Performing Arts: Include Visual Arts, Music, Theater, Dance, and Design Principles.
-// 10.	Law, Governance & Ethics: Focus on Legal Studies, Public Administration, Policy Analysis, and Ethical Decision-Making.
-// 11.	Recreation, Lifestyle & Practical Skills: Highlight Hobbies, Sports, Travel, Lifestyle Choices, and Practical Skills.
-// 12.	Technology & Media Literacy: Discuss Digital Literacy, Media Studies, and the Impact of Digital Media.
-// 13.	Philosophy & Critical Thinking: Emphasize Moral Philosophy, Ethical Frameworks, and Critical Thinking.
-// 14.	Space & Astronomical Sciences: Focus on Space Exploration, Astronomy, and Astrophysics.
-// 15.	Agriculture & Food Sciences: Discuss Sustainable Farming, Food Technology, and Nutrition.
-// 16.	Trades & Craftsmanship: Cover Hands-on Skills in Trades and Crafts.
-// 17.	Reference & Indexing: Include Summaries, Timelines, Directories, Glossaries, Bibliographies, and other Reference Material.
-// 18.	Other: Use for content that doesn’t fit into the above categories.
-// Identify key concepts within the content after classifying the field. Concepts are the fundamental ideas or categories that form the basis of knowledge within any discipline. They help organize and explain information, making it accessible and relatable.
-// 1. **Definition of a Concept**: Concepts should be significant ideas that recur within the content and are essential for understanding the main themes.
-// 2. **Inclusion Criteria**: Include a concept only if it is discussed in detail, meaning it is explained thoroughly, tied to specific examples, or highlighted as a critical part of the subject matter.
-// List the concepts in the following JSON format:
-// json
-// {
-//     "concepts": ["concept1", "concept2", "concept3", "..."]
-// }
-// After classifying the content and identifying key concepts, proceed to extract and list verifiable facts. Facts are objective statements that must be supported by empirical evidence or observation, such as specific data on events, people, numbers, dates, or well-established ideas.
-// 1. **Definition of a Fact**: Ensure each fact is a standalone piece of information that is concrete and can be independently verified.
-// 2. **Selection Criteria**: Choose facts based on their significance to the content's main themes or concepts, their educational value, or their foundational role in the subject.
-// Record the facts in the following JSON format:
-// json
-// {
-//     "facts": ["fact1", "fact2", "fact3", "..."]
-// }
-//     After analyzing the content, classifying its field, and identifying key concepts, and facts, assess whether the discovered elements warrant the creation of testing materials. Consider if these elements provide significant educational value to an average learner by enhancing understanding, offering practical applications, or supporting crucial educational goals. If you decide that testing cards don't need to be generated then please provide a reason in less than 90 characters.
-// 1. **Value Assessment**: Determine if the concepts and facts are essential for understanding the broader topic, are likely to be used in practical scenarios, or help in achieving educational benchmarks.
-// 2. **Criteria for Material Generation**: Generate testing materials if the concepts and facts are central to the content, have broad applicability, and are likely to reinforce or expand the learner’s knowledge significantly.
-// Make your decision using this criterion and reflect it in the JSON format as follows:
-// json
-// "generate_cards": 
-//     {   state: true or false,
-//         false_reason: "reason for marking the source as false. Leave empty for true."
-//     }
-//     After analyzing the content, identifying key concepts, and facts, summarize the material using a series of engaging and informative cards. These cards should capture the essence of the content while highlighting the critical concepts and facts that you previously identified.
-// 1. **Inclusion Criteria**: The generate_cards should be true. Return an empty array if the generate_cards is false.
-// 2. **Summarization Objective**: Each card is a step in a journey through the content. The series should collectively summarize the source while emphasizing important learning points.
-// 3. **Character Limit**: Maintain a limit of 320 characters per card to ensure each message is concise yet informative.
-// 4. **Engagement and Flow**: Write in an engaging style that maintains the user’s interest. Arrange the cards in a logical order that reflects the flow of the original content.
-// Format your output in JSON as follows:
-// json
-// {
-//     "summary_cards": ["summary_card1", "summary_card2", "summary_card3", "..."]
-// }
-// `;
-const typologyPromptString = {
-    role: `
+const typologyPromptString = `
 As a dedicated assistant at a learning company, your role involves analyzing educational content to categorize and summarize it. You will process content (in JSON format) that represents text and images from diverse sources such as PDFs, book chapters, videos, and websites. Follow these steps:
 
 1. Classify the content into one to three predefined fields of knowledge.
-2. Extract key concepts within the content.
-3. Extract concrete facts that are relevant to the subject and referenced in the content.
+2. Identify key concepts within the content.
+3. Detect concrete facts that are empirically verified and relevant to the subject.
 4. Decide whether the identified concepts and facts should be used to generate learning materials like flashcards based on their educational value.
 5. If the generate_cards is true then summarize the content using a series of summary cards.
 
 Please format your findings in this JSON schema:
+json
 {
-"field": ["primary_field", "secondary_field", "tertiary_field"],
-"concepts": ["concept1", "concept2", "concept3", "..."],
-"facts": ["fact1", "fact2", "fact3", "..."],
-"generate_cards": {
- "state": true or false,
- "false_reason": "reason for marking the source as false. Leave empty for true."
-},
-"summary_cards": ["summary_card1", "summary_card2", "summary_card3", "..."]
+    "field": ["primary_field", "secondary_field", "tertiary_field"],
+    "concepts": ["concept1", "concept2", "concept3", "..."],
+    "facts": ["fact1", "fact2", "fact3", "..."],
+    "generate_cards": [
+        state: true or false,
+        false_reason: "reason for marking the source as false. Leave empty for true."
+    ],
+    "summary_cards": ["summary_card1", "summary_card2", "summary_card3", "..."]
 }
 
-
 Further instruction on how to perform these tasks are below.
-    `,
-    fields: `
-    Every source must be placed under a field. This is the broadest category of knowledge. A source should belong to at least one and at most 3 fields. Only include fields that a source is strongly associated with. The field names in your response must exactly match the names of 18 fields listed below.
+Every source must be placed under a field. This is the broadest category of knowledge. A source should belong to at least one and at most 3 fields. Only include fields that a source is strongly associated with. The field names in your response must exactly match the names of 18 fields listed below.
 
 1.	Sciences: Focus on Biology, Chemistry, Physics, Astronomy, Mathematics, and Computer Science.
 2.	Technology & Engineering: Emphasize Information Technology, Engineering disciplines, AI, and Robotics.
@@ -122,62 +44,54 @@ Further instruction on how to perform these tasks are below.
 16.	Trades & Craftsmanship: Cover Hands-on Skills in Trades and Crafts.
 17.	Reference & Indexing: Include Summaries, Timelines, Directories, Glossaries, Bibliographies, and other Reference Material.
 18.	Other: Use for content that doesn’t fit into the above categories.
-    `,
-    concepts: `
-    Extract key concepts within the content after classifying the field. Please be as exhaustive as possible.
 
-Definition of a Concept: Concepts are fundamental ideas that form the basis of knowledge in any discipline. They help organize and explain information, making it accessible and relatable.
-Inclusion Criteria: Include a concept only if it is discussed in detail, meaning it is explained thoroughly, tied to specific examples, or highlighted as a critical part of the subject matter.
-List the concepts in the following JSON format:
+Identify key concepts within the content after classifying the field. Concepts are the fundamental ideas or categories that form the basis of knowledge within any discipline. They help organize and explain information, making it accessible and relatable.
+
+1. **Definition of a Concept**: Concepts should be significant ideas that recur within the content and are essential for understanding the main themes.
+2. **Inclusion Criteria**: Include a concept only if it is discussed in detail, meaning it is explained thoroughly, tied to specific examples, or highlighted as a critical part of the subject matter.
 
+List the concepts in the following JSON format:
+json
 {
- "concepts": 
- [
-     "concept1", 
-     "concept2", 
-     "concept3", 
-     "..."
- ]
+    "concepts": ["concept1", "concept2", "concept3", "..."]
 }
-    `,
-    facts: `
-    After classifying the content and identifying key concepts, proceed to extract and list verifiable facts.
+After classifying the content and identifying key concepts, proceed to extract and list verifiable facts. Facts are objective statements that must be supported by empirical evidence or observation, such as specific data on events, people, numbers, dates, or well-established ideas.
 
-Definition of a Fact: Ensure each fact is a standalone piece of information that is concrete and can be independently verified.
-Selection Criteria: Choose facts based on their significance to the content's main themes or concepts, their educational value, or their foundational role in the subject.
-Record the facts in the following JSON format:
+1. **Definition of a Fact**: Ensure each fact is a standalone piece of information that is concrete and can be independently verified.
+2. **Selection Criteria**: Choose facts based on their significance to the content's main themes or concepts, their educational value, or their foundational role in the subject.
 
+Record the facts in the following JSON format:
+json
 {
- "facts": ["fact1", "fact2", "fact3", "..."]
+    "facts": ["fact1", "fact2", "fact3", "..."]
 }
-    `,
-    generate: `
     After analyzing the content, classifying its field, and identifying key concepts, and facts, assess whether the discovered elements warrant the creation of testing materials. Consider if these elements provide significant educational value to an average learner by enhancing understanding, offering practical applications, or supporting crucial educational goals. If you decide that testing cards don't need to be generated then please provide a reason in less than 90 characters.
 
-Value Assessment: Determine if the concepts and facts are essential for understanding the broader topic, are likely to be used in practical scenarios, or help in achieving educational benchmarks.
-Criteria for Material Generation: Generate testing materials if the concepts and facts are central to the content, have broad applicability, and are likely to reinforce or expand the learner’s knowledge significantly.
-Make your decision using this criterion and reflect it in the JSON format as follows:
+1. **Value Assessment**: Determine if the concepts and facts are essential for understanding the broader topic, are likely to be used in practical scenarios, or help in achieving educational benchmarks.
+2. **Criteria for Material Generation**: Generate testing materials if the concepts and facts are central to the content, have broad applicability, and are likely to reinforce or expand the learner’s knowledge significantly.
 
+Make your decision using this criterion and reflect it in the JSON format as follows:
+json
 "generate_cards": 
- {   state: true or false,
-     false_reason: "reason for marking the source as false. Leave empty for true."
- }
-    `,
-    summarize: `
+    {   state: true or false,
+        false_reason: "reason for marking the source as false. Leave empty for true."
+    }
+
     After analyzing the content, identifying key concepts, and facts, summarize the material using a series of engaging and informative cards. These cards should capture the essence of the content while highlighting the critical concepts and facts that you previously identified.
 
-Inclusion Criteria: The generate_cards should be true. Return an empty array if the generate_cards is false.
-Summarization Objective: Each card is a step in a journey through the content. The series should collectively summarize the source while emphasizing important learning points.
-Character Limit: Maintain a limit of 320 characters per card to ensure each message is concise yet informative.
-Engagement and Flow: Write in an engaging style that maintains the user’s interest. Arrange the cards in a logical order that reflects the flow of the original content.
-Format your output in JSON as follows:
 
+1. **Inclusion Criteria**: The generate_cards should be true. Return an empty array if the generate_cards is false.
+2. **Summarization Objective**: Each card is a step in a journey through the content. The series should collectively summarize the source while emphasizing important learning points.
+3. **Character Limit**: Maintain a limit of 320 characters per card to ensure each message is concise yet informative.
+4. **Engagement and Flow**: Write in an engaging style that maintains the user’s interest. Arrange the cards in a logical order that reflects the flow of the original content.
+
+Format your output in JSON as follows:
+json
 {
- "summary_cards": ["summary_card1", "summary_card2", "summary_card3", "..."]
+    "summary_cards": ["summary_card1", "summary_card2", "summary_card3", "..."]
 }
-    `,
-};
-function returnTypologyPrompt(typologyPrompt) {
-    return typologyPrompt;
+`;
+function returnTypologyPrompt() {
+    return typologyPromptString;
 }
 exports.returnTypologyPrompt = returnTypologyPrompt;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "only_ever_generator",
-  "version": "0.3.1",
+  "version": "0.3.2",
   "main": "dist/index.js",
   "scripts": {
     "start": "npm run build && nodemon dist/index.js",

package/src/bootstrap/app.ts

@@ -1,8 +1,8 @@
 import { GenerateCards } from "../card_gen/generate_cards";
 import { ParseSourceContent } from "../parse/parse_source_content";
 import { OpenAiService } from "../services/open_ai_service";
-import { returnCardGenPrompt } from "../constants/prompts/card_gen_prompt";
-import { returnTypologyPrompt } from "../constants/prompts/typology_prompt";
+// import { returnCardGenPrompt } from "../constants/prompts/card_gen_prompt";
+// import { returnTypologyPrompt } from "../constants/prompts/typology_prompt";
 import { GenerateTypology } from "../typology_gen/generate_typology";
 import { GenerateArgs } from "../utils/generate_args";
 import { gapFilling } from "../gap_fill/calculate_gap_fill";
@@ -48,8 +48,8 @@ export class OnlyEverGenerator {
     this.typologyResponse = generationContent.content.taxonomy
     
     this.expectedFields =  generationContent.content.fields; //returnFields();
-    this.promptForTypology = returnTypologyPrompt(generationContent.prompt.typology);
-    this.promptForCardGen = returnCardGenPrompt(generationContent.prompt.card_generation);
+    this.promptForTypology = generationContent.prompt.typology;
+    this.promptForCardGen =generationContent.prompt.card_generation;
   }
 
 
@@ -124,8 +124,9 @@ export class OnlyEverGenerator {
   async generateCard(prompt: string, additionalContent: string, isGapFill: boolean) {
     let generateCardsResp = await new GenerateCards(this.openAiService).generateCards(
       prompt ?? "",
-      this.parsedContent + additionalContent,
-      isGapFill
+      JSON.stringify(this.parsedContent) + additionalContent,
+      isGapFill,
+      this.parsedContent.headings ?? [],
     );
   
     // let response =  await this.openAiService?.sendRequest(prompt,this.parsedContent);

package/src/card_gen/generate_cards.ts

@@ -7,7 +7,7 @@ export class GenerateCards {
     this.openAiService = openAiService;
   }
 
-  async generateCards(prompt: string, parsedContent: string, isGapFill: boolean) {
+  async generateCards(prompt: string, parsedContent: string,isGapFill: boolean,headings: Array<any>) {
     let response = await this.openAiService?.sendRequest(prompt, parsedContent);
    // console.log("response to card generation ", response);
     response["type"] = isGapFill ? "gap_fill":"card_gen";
@@ -21,14 +21,14 @@ export class GenerateCards {
       if(response.status_code == 200){
         response.metadata.status = "completed";
         //return response;
-       return this.parse(response, isGapFill);
+       return this.parse(response, isGapFill,headings);
       } else {
         response.metadata.status = "failed";
         return response;
       }
   }
 
-  parse(generatedData: any, isGapFill: boolean) {
+  parse(generatedData: any, isGapFill: boolean, headings:Array<any>) {
     try{
     const cardData = [];
     let usage_data = generatedData.metadata;
@@ -39,6 +39,11 @@ export class GenerateCards {
     const type = generatedData.type;
 if(unparsedTestCards !== undefined && unparsedTestCards.length != 0) {
   for (let elem of unparsedTestCards) {
+    if(headings.includes(elem.card_reference)){
+
+    }else{
+      elem.card_reference = '';
+    }
     if (elem.type == "flash") {
       cardData.push(this.parseFlashCard(elem));
     } else if (elem.type == "mcq") {
@@ -249,3 +254,5 @@ return question;
     return displayTitle;
   }
 }
+
+

package/src/constants/prompt_data.ts

@@ -1,3 +1,6 @@
+import { returnCardGenPrompt } from "./prompts/card_gen_prompt";
+import { returnTypologyPrompt } from "./prompts/typology_prompt";
+
 const promptData: any = {
     typology: {
         role: `
@@ -293,5 +296,8 @@ const promptData: any = {
 }
 
 export function returnPromptData(){
-    return promptData;
+    return {
+        "typology": returnTypologyPrompt(),
+        "card_generation": returnCardGenPrompt(''),
+    };
 }

package/src/constants/prompts/card_gen_prompt.ts

@@ -25,7 +25,7 @@
 //      {
 //          "type": "flash" | "mcq" | "cloze" | "match",
 //          "card_content": { "front": "...", "back": "..." | "prompt": "...", "choices": [ ... ] | "prompt": "...", "options": [ ... ] | "...": "...", "....": "..." },
-//          "card_reference": "source_title#heading",
+//          "card_reference": "heading",
 //          "concepts": ["Concept1", "Concept2", "..."],
 //          "facts": ["Fact1", "Fact2", "..."],
 //          "bloom_level": <1-5>
@@ -94,7 +94,7 @@
 //      "front": "<content for the front>",
 //      "back": "<content for the back>"
 //  },
-//  "card_reference": "source_title#heading",
+//  "card_reference": "heading",
 //  "concepts": ["Concept1", "Concept2", "..."],
 //  "facts": ["Fact1", "Fact2", "..."],
 //  "bloom_level": <1-5>
@@ -114,7 +114,7 @@
 //          "... up to 8 choices"
 //      ]
 //  },
-//  "card_reference": "source_title#heading",
+//  "card_reference": "heading",
 //  "concepts": ["Concept1", "Concept2", "..."],
 //  "facts": ["Fact1", "Fact2", "..."],
 //  "bloom_level": <1-5>
@@ -141,7 +141,7 @@
 //          "... up to 8 choices"
 //      ]
 //  },
-//  "card_reference": "source_title#heading",
+//  "card_reference": "heading",
 //  "concepts": ["Concept1", "Concept2", "..."],
 //  "facts": ["Fact1", "Fact2", "..."],
 //  "bloom_level": <1-5>
@@ -163,7 +163,7 @@
 //      "left_choice 2": "right_choice 2",
 //      "... up to 8 total pairs"
 //  },
-//  "card_reference": "source_title#heading",
+//  "card_reference": "heading",
 //  "concepts": ["Concept1", "Concept2", "..."],
 //  "facts": ["Fact1", "Fact2", "..."],
 //  "bloom_level": <1-5>
@@ -171,7 +171,7 @@
 
 // •	Maximum character length for each item in a pair: 42
 //     `,
-//     reference : `Each test card needs a reference. A reference can either be the entire source or a specific heading in the source. Whenever possible, pick a main heading to direct the user to the most relevant part of the source material. The reference schema is as follows: source_title#main_heading, where #main_heading is optional.`,
+//     reference : `Each test card needs a reference. A reference can either be the entire source or a specific heading in the source. Whenever possible, pick a main heading to direct the user to the most relevant part of the source material. The reference schema is as follows: ""main_heading, where #main_heading is optional.`,
 //     checkcoverage: `Once you are done generating the test cards. Go back and evaulate the full list of concepts and fact that include any of the missing concepts or facts along with the list that was provided as the input. 
 
 // Are there any concept or fact that don't have a test card yet? If yes, go back and create one.
@@ -213,7 +213,7 @@ json
         {
             "type": "flash" | "mcq" | "cloze" | "match",
             "card_content": { "front": "...", "back": "..." | "prompt": "...", "choices": [ ... ] | "prompt": "...", "options": [ ... ] | "right_choice 1": "...", "left_choice 1": "..." },
-            "card_reference": "source_title#heading",
+            "card_reference": "heading",
             "concepts": ["Concept1", "Concept2", "..."],
             "facts": ["Fact1", "Fact2", "..."]
         }
@@ -277,7 +277,7 @@ json
         "front": "<content for the front>",
         "back": "<content for the back>"
     },
-    "card_reference": "source_title#heading",
+    "card_reference": "heading",
     "concepts": ["Concept1", "Concept2", "..."],
     "facts": ["Fact1", "Fact2", "..."],
     "bloom_level": <1-5>
@@ -296,7 +296,7 @@ json
             "... up to 8 choices"
         ]
     },
-    "card_reference": "source_title#heading",
+    "card_reference": "heading",
     "concepts": ["Concept1", "Concept2", "..."],
     "facts": ["Fact1", "Fact2", "..."],
     "bloom_level": <1-5>
@@ -322,7 +322,7 @@ json
             "... up to 8 choices"
         ]
     },
-    "card_reference": "source_title#heading",
+    "card_reference": "heading",
     "concepts": ["Concept1", "Concept2", "..."],
     "facts": ["Fact1", "Fact2", "..."],
     "bloom_level": <1-5>
@@ -342,14 +342,14 @@ json
         "right_choice 2": "left_choice 2",
         "... up to 8 total pairs"
     },
-    "card_reference": "source_title#heading",
+    "card_reference": "heading",
     "concepts": ["Concept1", "Concept2", "..."],
     "facts": ["Fact1", "Fact2", "..."],
     "bloom_level": <1-5>
 }
 •	Maximum character length for each item in a pair: 42
 
-Each test card needs a reference. A reference can either be the entire source or a specific heading in the source. Whenever possible, pick a main heading to direct the user to the most relevant part of the source material. The reference schema is as follows: source_title#main_heading, where #main_heading is optional.
+Each test card needs a reference. A reference can either be the entire source or a specific heading in the source. Whenever possible, pick a main heading to direct the user to the most relevant part of the source material. The reference schema is as follows: ""main_heading, where #main_heading is optional.
 Once you are done generating the test cards. Go back and evaulate the full list of concepts and fact that include any of the missing concepts or facts along with the list that was provided as the input. 
 
 Are there any concept or fact that don't have a test card yet? If yes, go back and create one.

package/src/constants/prompts/typology_prompt.ts

@@ -1,121 +1,27 @@
-// const typologyPromptString = `
-// As a dedicated assistant at a learning company, your role involves analyzing educational content to categorize and summarize it. You will process content (in JSON format) that represents text and images from diverse sources such as PDFs, book chapters, videos, and websites. Follow these steps:
-
-// 1. Classify the content into one to three predefined fields of knowledge.
-// 2. Identify key concepts within the content.
-// 3. Detect concrete facts that are empirically verified and relevant to the subject.
-// 4. Decide whether the identified concepts and facts should be used to generate learning materials like flashcards based on their educational value.
-// 5. If the generate_cards is true then summarize the content using a series of summary cards.
-
-// Please format your findings in this JSON schema:
-// json
-// {
-//     "field": ["primary_field", "secondary_field", "tertiary_field"],
-//     "concepts": ["concept1", "concept2", "concept3", "..."],
-//     "facts": ["fact1", "fact2", "fact3", "..."],
-//     "generate_cards": [
-//         state: true or false,
-//         false_reason: "reason for marking the source as false. Leave empty for true."
-//     ],
-//     "summary_cards": ["summary_card1", "summary_card2", "summary_card3", "..."]
-// }
-
-// Further instruction on how to perform these tasks are below.
-// Every source must be placed under a field. This is the broadest category of knowledge. A source should belong to at least one and at most 3 fields. Only include fields that a source is strongly associated with. The field names in your response must exactly match the names of 18 fields listed below.
-
-// 1.	Sciences: Focus on Biology, Chemistry, Physics, Astronomy, Mathematics, and Computer Science.
-// 2.	Technology & Engineering: Emphasize Information Technology, Engineering disciplines, AI, and Robotics.
-// 3.	Humanities & Cultural Studies: Highlight History, Literature, Languages, Arts, Philosophy, and Anthropological Studies.
-// 4.	Social Sciences & Global Studies: Include Sociology, Psychology, Economics, Political Science, Anthropology, and International Relations.
-// 5.	Business & Management: Encompass Entrepreneurship, Marketing, Finance, Leadership, and Ethics.
-// 6.	Health & Medicine: Cover Medical Sciences, Public Health, Nutrition, Wellness, and Mental Health.
-// 7.	Environmental Studies & Earth Sciences: Discuss Ecology, Climate Science, Geology, and Environmental Policy.
-// 8.	Education, Learning & Personal Development: Talk about Educational Theories, Teaching Methods, and Personal Skills.
-// 9.	Creative & Performing Arts: Include Visual Arts, Music, Theater, Dance, and Design Principles.
-// 10.	Law, Governance & Ethics: Focus on Legal Studies, Public Administration, Policy Analysis, and Ethical Decision-Making.
-// 11.	Recreation, Lifestyle & Practical Skills: Highlight Hobbies, Sports, Travel, Lifestyle Choices, and Practical Skills.
-// 12.	Technology & Media Literacy: Discuss Digital Literacy, Media Studies, and the Impact of Digital Media.
-// 13.	Philosophy & Critical Thinking: Emphasize Moral Philosophy, Ethical Frameworks, and Critical Thinking.
-// 14.	Space & Astronomical Sciences: Focus on Space Exploration, Astronomy, and Astrophysics.
-// 15.	Agriculture & Food Sciences: Discuss Sustainable Farming, Food Technology, and Nutrition.
-// 16.	Trades & Craftsmanship: Cover Hands-on Skills in Trades and Crafts.
-// 17.	Reference & Indexing: Include Summaries, Timelines, Directories, Glossaries, Bibliographies, and other Reference Material.
-// 18.	Other: Use for content that doesn’t fit into the above categories.
-
-// Identify key concepts within the content after classifying the field. Concepts are the fundamental ideas or categories that form the basis of knowledge within any discipline. They help organize and explain information, making it accessible and relatable.
-
-// 1. **Definition of a Concept**: Concepts should be significant ideas that recur within the content and are essential for understanding the main themes.
-// 2. **Inclusion Criteria**: Include a concept only if it is discussed in detail, meaning it is explained thoroughly, tied to specific examples, or highlighted as a critical part of the subject matter.
-
-// List the concepts in the following JSON format:
-// json
-// {
-//     "concepts": ["concept1", "concept2", "concept3", "..."]
-// }
-// After classifying the content and identifying key concepts, proceed to extract and list verifiable facts. Facts are objective statements that must be supported by empirical evidence or observation, such as specific data on events, people, numbers, dates, or well-established ideas.
-
-// 1. **Definition of a Fact**: Ensure each fact is a standalone piece of information that is concrete and can be independently verified.
-// 2. **Selection Criteria**: Choose facts based on their significance to the content's main themes or concepts, their educational value, or their foundational role in the subject.
-
-// Record the facts in the following JSON format:
-// json
-// {
-//     "facts": ["fact1", "fact2", "fact3", "..."]
-// }
-//     After analyzing the content, classifying its field, and identifying key concepts, and facts, assess whether the discovered elements warrant the creation of testing materials. Consider if these elements provide significant educational value to an average learner by enhancing understanding, offering practical applications, or supporting crucial educational goals. If you decide that testing cards don't need to be generated then please provide a reason in less than 90 characters.
-
-// 1. **Value Assessment**: Determine if the concepts and facts are essential for understanding the broader topic, are likely to be used in practical scenarios, or help in achieving educational benchmarks.
-// 2. **Criteria for Material Generation**: Generate testing materials if the concepts and facts are central to the content, have broad applicability, and are likely to reinforce or expand the learner’s knowledge significantly.
-
-// Make your decision using this criterion and reflect it in the JSON format as follows:
-// json
-// "generate_cards": 
-//     {   state: true or false,
-//         false_reason: "reason for marking the source as false. Leave empty for true."
-//     }
-
-//     After analyzing the content, identifying key concepts, and facts, summarize the material using a series of engaging and informative cards. These cards should capture the essence of the content while highlighting the critical concepts and facts that you previously identified.
-
-
-// 1. **Inclusion Criteria**: The generate_cards should be true. Return an empty array if the generate_cards is false.
-// 2. **Summarization Objective**: Each card is a step in a journey through the content. The series should collectively summarize the source while emphasizing important learning points.
-// 3. **Character Limit**: Maintain a limit of 320 characters per card to ensure each message is concise yet informative.
-// 4. **Engagement and Flow**: Write in an engaging style that maintains the user’s interest. Arrange the cards in a logical order that reflects the flow of the original content.
-
-// Format your output in JSON as follows:
-// json
-// {
-//     "summary_cards": ["summary_card1", "summary_card2", "summary_card3", "..."]
-// }
-// `;
-
-const typologyPromptString : any = {
-    role: `
+const typologyPromptString = `
 As a dedicated assistant at a learning company, your role involves analyzing educational content to categorize and summarize it. You will process content (in JSON format) that represents text and images from diverse sources such as PDFs, book chapters, videos, and websites. Follow these steps:
 
 1. Classify the content into one to three predefined fields of knowledge.
-2. Extract key concepts within the content.
-3. Extract concrete facts that are relevant to the subject and referenced in the content.
+2. Identify key concepts within the content.
+3. Detect concrete facts that are empirically verified and relevant to the subject.
 4. Decide whether the identified concepts and facts should be used to generate learning materials like flashcards based on their educational value.
 5. If the generate_cards is true then summarize the content using a series of summary cards.
 
 Please format your findings in this JSON schema:
+json
 {
-"field": ["primary_field", "secondary_field", "tertiary_field"],
-"concepts": ["concept1", "concept2", "concept3", "..."],
-"facts": ["fact1", "fact2", "fact3", "..."],
-"generate_cards": {
- "state": true or false,
- "false_reason": "reason for marking the source as false. Leave empty for true."
-},
-"summary_cards": ["summary_card1", "summary_card2", "summary_card3", "..."]
+    "field": ["primary_field", "secondary_field", "tertiary_field"],
+    "concepts": ["concept1", "concept2", "concept3", "..."],
+    "facts": ["fact1", "fact2", "fact3", "..."],
+    "generate_cards": [
+        state: true or false,
+        false_reason: "reason for marking the source as false. Leave empty for true."
+    ],
+    "summary_cards": ["summary_card1", "summary_card2", "summary_card3", "..."]
 }
 
-
 Further instruction on how to perform these tasks are below.
-    `,
-    fields: `
-    Every source must be placed under a field. This is the broadest category of knowledge. A source should belong to at least one and at most 3 fields. Only include fields that a source is strongly associated with. The field names in your response must exactly match the names of 18 fields listed below.
+Every source must be placed under a field. This is the broadest category of knowledge. A source should belong to at least one and at most 3 fields. Only include fields that a source is strongly associated with. The field names in your response must exactly match the names of 18 fields listed below.
 
 1.	Sciences: Focus on Biology, Chemistry, Physics, Astronomy, Mathematics, and Computer Science.
 2.	Technology & Engineering: Emphasize Information Technology, Engineering disciplines, AI, and Robotics.
@@ -135,63 +41,58 @@ Further instruction on how to perform these tasks are below.
 16.	Trades & Craftsmanship: Cover Hands-on Skills in Trades and Crafts.
 17.	Reference & Indexing: Include Summaries, Timelines, Directories, Glossaries, Bibliographies, and other Reference Material.
 18.	Other: Use for content that doesn’t fit into the above categories.
-    `,
-    concepts: `
-    Extract key concepts within the content after classifying the field. Please be as exhaustive as possible.
 
-Definition of a Concept: Concepts are fundamental ideas that form the basis of knowledge in any discipline. They help organize and explain information, making it accessible and relatable.
-Inclusion Criteria: Include a concept only if it is discussed in detail, meaning it is explained thoroughly, tied to specific examples, or highlighted as a critical part of the subject matter.
-List the concepts in the following JSON format:
+Identify key concepts within the content after classifying the field. Concepts are the fundamental ideas or categories that form the basis of knowledge within any discipline. They help organize and explain information, making it accessible and relatable.
 
+1. **Definition of a Concept**: Concepts should be significant ideas that recur within the content and are essential for understanding the main themes.
+2. **Inclusion Criteria**: Include a concept only if it is discussed in detail, meaning it is explained thoroughly, tied to specific examples, or highlighted as a critical part of the subject matter.
+
+List the concepts in the following JSON format:
+json
 {
- "concepts": 
- [
-     "concept1", 
-     "concept2", 
-     "concept3", 
-     "..."
- ]
+    "concepts": ["concept1", "concept2", "concept3", "..."]
 }
-    `,
-    facts: `
-    After classifying the content and identifying key concepts, proceed to extract and list verifiable facts.
+After classifying the content and identifying key concepts, proceed to extract and list verifiable facts. Facts are objective statements that must be supported by empirical evidence or observation, such as specific data on events, people, numbers, dates, or well-established ideas.
 
-Definition of a Fact: Ensure each fact is a standalone piece of information that is concrete and can be independently verified.
-Selection Criteria: Choose facts based on their significance to the content's main themes or concepts, their educational value, or their foundational role in the subject.
-Record the facts in the following JSON format:
+1. **Definition of a Fact**: Ensure each fact is a standalone piece of information that is concrete and can be independently verified.
+2. **Selection Criteria**: Choose facts based on their significance to the content's main themes or concepts, their educational value, or their foundational role in the subject.
 
+Record the facts in the following JSON format:
+json
 {
- "facts": ["fact1", "fact2", "fact3", "..."]
+    "facts": ["fact1", "fact2", "fact3", "..."]
 }
-    `,
-    generate: `
     After analyzing the content, classifying its field, and identifying key concepts, and facts, assess whether the discovered elements warrant the creation of testing materials. Consider if these elements provide significant educational value to an average learner by enhancing understanding, offering practical applications, or supporting crucial educational goals. If you decide that testing cards don't need to be generated then please provide a reason in less than 90 characters.
 
-Value Assessment: Determine if the concepts and facts are essential for understanding the broader topic, are likely to be used in practical scenarios, or help in achieving educational benchmarks.
-Criteria for Material Generation: Generate testing materials if the concepts and facts are central to the content, have broad applicability, and are likely to reinforce or expand the learner’s knowledge significantly.
-Make your decision using this criterion and reflect it in the JSON format as follows:
+1. **Value Assessment**: Determine if the concepts and facts are essential for understanding the broader topic, are likely to be used in practical scenarios, or help in achieving educational benchmarks.
+2. **Criteria for Material Generation**: Generate testing materials if the concepts and facts are central to the content, have broad applicability, and are likely to reinforce or expand the learner’s knowledge significantly.
 
+Make your decision using this criterion and reflect it in the JSON format as follows:
+json
 "generate_cards": 
- {   state: true or false,
-     false_reason: "reason for marking the source as false. Leave empty for true."
- }
-    `,
-    summarize:`
+    {   state: true or false,
+        false_reason: "reason for marking the source as false. Leave empty for true."
+    }
+
     After analyzing the content, identifying key concepts, and facts, summarize the material using a series of engaging and informative cards. These cards should capture the essence of the content while highlighting the critical concepts and facts that you previously identified.
 
-Inclusion Criteria: The generate_cards should be true. Return an empty array if the generate_cards is false.
-Summarization Objective: Each card is a step in a journey through the content. The series should collectively summarize the source while emphasizing important learning points.
-Character Limit: Maintain a limit of 320 characters per card to ensure each message is concise yet informative.
-Engagement and Flow: Write in an engaging style that maintains the user’s interest. Arrange the cards in a logical order that reflects the flow of the original content.
-Format your output in JSON as follows:
 
+1. **Inclusion Criteria**: The generate_cards should be true. Return an empty array if the generate_cards is false.
+2. **Summarization Objective**: Each card is a step in a journey through the content. The series should collectively summarize the source while emphasizing important learning points.
+3. **Character Limit**: Maintain a limit of 320 characters per card to ensure each message is concise yet informative.
+4. **Engagement and Flow**: Write in an engaging style that maintains the user’s interest. Arrange the cards in a logical order that reflects the flow of the original content.
+
+Format your output in JSON as follows:
+json
 {
- "summary_cards": ["summary_card1", "summary_card2", "summary_card3", "..."]
+    "summary_cards": ["summary_card1", "summary_card2", "summary_card3", "..."]
 }
-    `,
-    
-  }
+`;
+
+
+
+
 
-export function returnTypologyPrompt(typologyPrompt: any){
-    return typologyPrompt;
+export function returnTypologyPrompt(){
+    return typologyPromptString;
 }