@thecodingwhale/cv-processor/dist/AICVProcessorX.js

CV Processor to extract structured data from PDF resumes using TypeScript

"use strict";
var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
    if (k2 === undefined) k2 = k;
    var desc = Object.getOwnPropertyDescriptor(m, k);
    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
      desc = { enumerable: true, get: function() { return m[k]; } };
    }
    Object.defineProperty(o, k2, desc);
}) : (function(o, m, k, k2) {
    if (k2 === undefined) k2 = k;
    o[k2] = m[k];
}));
var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
    Object.defineProperty(o, "default", { enumerable: true, value: v });
}) : function(o, v) {
    o["default"] = v;
});
var __importStar = (this && this.__importStar) || (function () {
    var ownKeys = function(o) {
        ownKeys = Object.getOwnPropertyNames || function (o) {
            var ar = [];
            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
            return ar;
        };
        return ownKeys(o);
    };
    return function (mod) {
        if (mod && mod.__esModule) return mod;
        var result = {};
        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
        __setModuleDefault(result, mod);
        return result;
    };
})();
Object.defineProperty(exports, "__esModule", { value: true });
exports.AICVProcessorX = void 0;
const fs = __importStar(require("fs"));
const pdfParse = __importStar(require("pdf-parse"));
class AICVProcessorX {
    constructor(aiProvider, verbose = false) {
        this.tokenUsage = {
            promptTokens: 0,
            completionTokens: 0,
            totalTokens: 0,
            estimatedCost: 0,
        };
        this.aiProvider = aiProvider;
        this.verbose = verbose;
    }
    async processCv(pdfPath) {
        console.log(`Processing CV: ${pdfPath}`);
        this.resetTokenUsage();
        // :one: Extract text from PDF locally
        const pdfBuffer = fs.readFileSync(pdfPath);
        const parsed = await pdfParse(pdfBuffer);
        let extractedText = parsed.text;
        if (this.verbose) {
            console.log(':white_check_mark: PDF text extracted (local)');
            console.log(`Length: ${extractedText.length} characters`);
        }
        // :two: Pre-clean text → remove irrelevant sections
        extractedText = this.cleanExtractedText(extractedText);
        if (this.verbose) {
            console.log(':white_check_mark: Cleaned text ready for AI processing');
        }
        // :three: Send cleaned text directly to AI with a structured prompt
        const aiPrompt = this.buildAIPrompt(extractedText);
        const aiResponse = await this.aiProvider.complete({
            prompt: aiPrompt,
            maxTokens: 1500, // adjust as needed
            temperature: 0.1,
        });
        this.addTokenUsageFromResponse(aiResponse.usage);
        if (this.verbose) {
            console.log(':white_check_mark: AI completed pattern extraction');
        }
        const cvData = JSON.parse(aiResponse.text);
        return cvData;
    }
    cleanExtractedText(text) {
        // Very basic cleaning — you can enhance regex based on format
        return text
            .replace(/PROFILE[\s\S]*?(?=\n\d{4})/i, '') // remove profile section up to first year
            .replace(/NOTES[\s\S]*$/i, '') // remove notes section at end
            .trim();
    }
    buildAIPrompt(text) {
        return `
You are an AI data extractor for an actor's resume system. I will provide you resume text. Extract credits and convert them into JSON matching this schema:

{
  "resume": [
    {
      "category": "<Category>",
      "category_id": "<UUIDv4>",
      "credits": [
        { "id": "<UUIDv4>", "year": "YYYY", "title": "<Title>", "role": "<Role>", "director": "<Director>", "attached_media": [] }
      ]
    }
  ],
  "resume_show_years": true
}

:white_check_mark: Official categories:
["Commercial","Film","Television","Theatre","Print / Fashion","Training","Voice","Stunt","Corporate","MC/Presenting","Extras","Other"]

Categorization rules:
- Only use these categories.
- Map synonyms and related words logically to closest category.
- If unsure → assign to "Other".
- If director missing → "director": ""
- Each category + credit must have unique UUIDv4
- Ignore profile, skills, notes, memberships.

Resume text:
${text}
    `;
    }
    addTokenUsageFromResponse(usage) {
        if (usage) {
            this.tokenUsage.promptTokens += usage.prompt_tokens || 0;
            this.tokenUsage.completionTokens += usage.completion_tokens || 0;
            this.tokenUsage.totalTokens += usage.total_tokens || 0;
            this.tokenUsage.estimatedCost += ((usage.total_tokens || 0) * 0.03) / 1000; // example pricing
        }
    }
    resetTokenUsage() {
        this.tokenUsage = {
            promptTokens: 0,
            completionTokens: 0,
            totalTokens: 0,
            estimatedCost: 0,
        };
    }
}
exports.AICVProcessorX = AICVProcessorX;