filesparser.c/PPTExtractor.cpp

#include "PPTExtractor.h"
#include <zip.h>
#include <pugixml.hpp>
#include <iostream>
#include <sstream>

PPTExtractor::PPTExtractor(const std::string& filePath) : filePath(filePath) {}

PPTExtractor::~PPTExtractor() {}

bool PPTExtractor::isPPTX() {
    return filePath.substr(filePath.find_last_of(".") + 1) == "pptx";
}

bool PPTExtractor::extractText(std::vector<std::string>& slidesText) {
    if (isPPTX()) {
        return extractTextFromPPTX(slidesText);
    }
    else {
        std::cerr << "Only PPTX files are supported." << std::endl;
        return false;
    }
}

bool PPTExtractor::extractTextFromPPTX(std::vector<std::string>& slidesText) {
    int err = 0;
    zip_t* zip = zip_open(filePath.c_str(), 0, &err);
    if (!zip) {
        std::cerr << "Failed to open PPTX file." << std::endl;
        return false;
    }

    for (int i = 1; ; ++i) {
        std::ostringstream oss;
        oss << "ppt/slides/slide" << i << ".xml";
        std::string slidePath = oss.str();

        struct zip_stat st;
        if (zip_stat(zip, slidePath.c_str(), 0, &st) != 0) {
            break; // No more slides
        }

        zip_file_t* slideFile = zip_fopen(zip, slidePath.c_str(), 0);
        if (!slideFile) {
            std::cerr << "Failed to open slide: " << slidePath << std::endl;
            continue;
        }

        char* buffer = new char[st.size];
        zip_fread(slideFile, buffer, st.size);
        zip_fclose(slideFile);

        pugi::xml_document doc;
        pugi::xml_parse_result result = doc.load_buffer(buffer, st.size);
        delete[] buffer;

        if (!result) {
            std::cerr << "Failed to parse slide XML: " << slidePath << std::endl;
            continue;
        }

        std::string slideText;
        for (pugi::xml_node node : doc.select_nodes("//a:t")) {
            slideText += node.child_value();
            slideText += "\n";
        }

        slidesText.push_back(slideText);
    }

    zip_close(zip);
    return true;
}