install tessdata from github repository

/***************************************************************************
 *
 * compile:
 *  g++ get_tessdata.cpp -o get_tessdata -std=c++0x -lcurl -ltesseract
 * run:
 *  ./get_tessdata
 *
 * You may opt to use, copy, modify, merge, publish, distribute and/or sell
 * copies of the Software, and permit persons to whom the Software is
 * furnished to do so, under the terms of the COPYING file.
 *
 * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
 * KIND, either express or implied.
 *
 ***************************************************************************/

#include <cmath>
#include <cstring>

#define CURL_STATICLIB
#include <curl/curl.h>
#include <tesseract/baseapi.h>

// source: http://stackoverflow.com/questions/1637587/c-libcurl-console-progress-bar
int progress_func(void* ptr, double TotalToDownload, double NowDownloaded,
                  double TotalToUpload, double NowUploaded) {
    // ensure that the file to be downloaded is not empty
    // because that would cause a division by zero error later on
    if (TotalToDownload <= 0.0) {
        return 0;
    }

    // how wide you want the progress meter to be
    int totaldotz = 40;
    double fractiondownloaded = NowDownloaded / TotalToDownload;
    // part of the progressmeter that's already "full"
    int dotz = std::round(fractiondownloaded * totaldotz);

    // create the "meter"
    int ii = 0;
    printf("%3.0f%% [", fractiondownloaded * 100);
    // part that's full already
    for ( ; ii < dotz; ii++) {
        printf("=");
    }
    // remaining part (spaces)
    for ( ; ii < totaldotz; ii++) {
        printf(" ");
    }
    // and back to line begin - do not forget the fflush to avoid output buffering problems!
    printf("]\r");
    fflush(stdout);
    // if you don't return 0, the transfer will be aborted - see the documentation
    return 0;
}

size_t write_data(void *ptr, size_t size, size_t nmemb, FILE *stream) {
    size_t written;
    written = fwrite(ptr, size, nmemb, stream);
    return written;
}

void receive_file(const char *file_name, char *download_url) {
    CURL *curl;
    FILE *fp;
    CURLcode res;

    curl = curl_easy_init();
    if (curl) {
        fp = fopen(file_name,"wb");
        if (!fp) {
            fprintf(stderr, "Can not open output file! Quitting...\n");
        }

        curl_easy_setopt(curl, CURLOPT_URL, download_url);
        curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, write_data);
        curl_easy_setopt(curl, CURLOPT_WRITEDATA, fp);
        // Internal CURL progressmeter must be disabled if we provide our own callback
        curl_easy_setopt(curl, CURLOPT_NOPROGRESS, false);
        // Install the callback function
        curl_easy_setopt(curl, CURLOPT_PROGRESSFUNCTION, progress_func);

        fprintf(stdout, "Starting download of %s:\n", download_url);
        res = curl_easy_perform(curl);

        if(res != CURLE_OK) {
            fprintf(stderr, "\nThere were errors during download:\n%s\n",
                    curl_easy_strerror(res));
        } else {
            fprintf(stdout, "\nDownload finished!\n");
        }

        /* now extract transfer info */
        double speed, download_time;
        curl_easy_getinfo(curl, CURLINFO_SPEED_DOWNLOAD, &speed);
        curl_easy_getinfo(curl, CURLINFO_TOTAL_TIME, &download_time);

        fprintf(stdout, "Speed: %.3f bytes/sec during %.3f seconds\n",
                speed, download_time);

        /* always cleanup */
        curl_easy_cleanup(curl);
        fclose(fp);
    } else {
        fprintf(stdout, "Can not init curl... Closing.\n");
    }
}

int main(int argc, char* argv[]) {
    const char* file_name = "eng.traineddata";
    const char* base_url = "https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/";
    bool force = 0;

    if (argc < 1) {
        fprintf(stderr, "Usage is:\n\t%s -f <traineddata> [-b url] [--force]\n\nFor example:"
                "\n\t%s -f eng.%s -b %s\n", argv[0], argv[0], file_name, base_url);
        return(1);
    } else {
        for (int i = 1; i < argc; i++) {
            if (i + 1 < argc) {
                if (strcmp(argv[i], "-f") == 0) {
                    file_name = argv[i + 1];
                } else if (strcmp(argv[i], "-b") == 0) {
                    base_url = argv[i + 1];
                }
            } else if (strcmp(argv[i], "--force") == 0) {
                // force will trigger download even file is already downloaded
                force = 1;
            }
        }
    }

    char url[strlen(base_url) + strlen(file_name) + 1];
    strcpy(url, base_url);
    strcat(url, file_name);

    // install path
    tesseract::TessBaseAPI *api = new tesseract::TessBaseAPI();
    if (api->Init(NULL, "eng")) {
        fprintf(stderr, "Could not initialize tesseract.\n");
        return 1;
    }
    const char *datapath = api->GetDatapath();
    fprintf(stdout, "datapath is %s\n", datapath);

    char destination[strlen(datapath) + strlen(file_name) + 1];
    strcpy(destination, datapath);
    strcat(destination, file_name);

    // Check if file is not downloaded already e.g. installation failed
    if (FILE *fp = fopen(destination, "r")) {
        fseek(fp, 0, SEEK_END);
        int size = ftell(fp);
        fclose(fp);
        if (size > 0 && !force) {
            fprintf(stdout, "%s is already downloaded.\n", destination);
        } else {
            receive_file(destination, url);
        }

    } else {
        receive_file(destination, url);
    }

    return 0;
}