video2document/services/modules/replace_speaker/replaceSpeaker.js

const fs = require('fs');
const path = require('path');

const outputDir = path.join(__dirname, "../../../storage/documents"); // path for output directory

if (!fs.existsSync(outputDir)) {
    fs.mkdirSync(outputDir, { recursive: true }); // Create output directory if it doesn't exist
}

const module_exports = {
    name: "replace_speaker",
    type: "processor",
    displayname: "Speaker Name Replacer",
    description: "Replaces speaker placeholder names with actual names based on a mapping in HTML files",

    async function(parameter) {
        return new Promise(async (resolve, reject) => {
            try {
                // console.log("Speaker replacer module invoked with parameters:", parameter);

                resolve(await this.replaceNames(
                    parameter.inputHtmlPath,      // Path to input HTML file
                    parameter.speakerMappingPath  // Path to speaker mapping file (JSON)
                ));

            } catch (error) {
                // console.error("Error in speaker replacer module:", error);
                reject(error)
            }
        })
    },

    replaceNames: async function(inputHtmlPath, speakerMappingPath) {
        return new Promise(async(resolve, reject) => {
            try {
                const htmlContent = await fs.promises.readFile(inputHtmlPath, "utf-8"); // read HTML file
                const mappingData = await fs.promises.readFile(speakerMappingPath, "utf-8"); // read mapping file
                
                // Parse mapping - supports JSON or simple format
                let speakerMap = {};
                try {
                    speakerMap = JSON.parse(mappingData); // Try to parse as JSON
                } catch (e) {
                    // If not JSON, try simple format: "Speaker A,Mike\nSpeaker B,Stefan"
                    const lines = mappingData.trim().split('\n');
                    lines.forEach(line => {
                        const [placeholder, realName] = line.split(',').map(s => s.trim());
                        if (placeholder && realName) {
                            speakerMap[placeholder] = realName;
                        }
                    });
                }

                // Replace all speaker names in HTML content
                let outputContent = htmlContent;
                Object.entries(speakerMap).forEach(([placeholder, realName]) => {
                    // Create regex to replace all occurrences (case-sensitive)
                    const regex = new RegExp(`\\b${placeholder}\\b`, 'g');
                    outputContent = outputContent.replace(regex, realName);
                });

                // Generate output file path based on input file name
                const inputFileName = path.basename(inputHtmlPath, path.extname(inputHtmlPath));
                const outPath = path.join(outputDir, `${inputFileName}_replaced.html`);
                
                // Write output to file
                fs.writeFileSync(outPath, outputContent, "utf8");

                // console.log("Replaced HTML file written to:", outPath);
                resolve(outPath)

            } catch (error) {
                // console.error("Error replacing speaker names:", error);
                reject(error)
            }
        })
    }
};

module.exports = module_exports;

// CLI Mode: Allow direct execution
if (require.main === module) {
    (async () => {
        const args = process.argv.slice(2);
        
        if (args.length < 2) {
            console.error("Usage: node string-replacer.js <inputHtmlPath> <speakerMappingPath>");
            console.error("Example: node string-replacer.js ./document.html ./speaker_mapping.json");
            console.error("\nMapping file formats:");
            console.error("JSON: {\"Speaker A\": \"Mike\", \"Speaker B\": \"Stefan\"}");
            console.error("or simple: Speaker A,Mike\\nSpeaker B,Stefan");
            process.exit(1);
        }

        const [inputHtmlPath, speakerMappingPath] = args;

        // Check if files exist
        if (!fs.existsSync(inputHtmlPath)) {
            console.error(`ERROR: HTML file not found: ${inputHtmlPath}`);
            process.exit(1);
        }

        if (!fs.existsSync(speakerMappingPath)) {
            console.error(`ERROR: Speaker mapping file not found: ${speakerMappingPath}`);
            process.exit(1);
        }

        console.log("Starting speaker name replacement...");
        console.log(`HTML file: ${inputHtmlPath}`);
        console.log(`Mapping file: ${speakerMappingPath}`);
        
        await module_exports.replaceNames(
            inputHtmlPath,
            speakerMappingPath
        );
        
        console.log("Done!");
    })();
}