/* Returns 1 on error */
static l_int32
test_mem_gif(const char *fname,
l_int32 index)
{
l_uint8 *data = NULL;
l_int32 same;
size_t size = 0;
PIX *pixs;
PIX *pixd = NULL;
if ((pixs = pixRead(fname)) == NULL) {
fprintf(stderr, "Failure to read gif file: %s\n", fname);
return 1;
}
if (pixWriteMem(&data, &size, pixs, IFF_GIF)) {
fprintf(stderr, "Mem gif write fail on image %d\n", index);
return 1;
}
if ((pixd = pixReadMem(data, size)) == NULL) {
fprintf(stderr, "Mem gif read fail on image %d\n", index);
lept_free(data);
return 1;
}
pixEqual(pixs, pixd, &same);
pixDestroy(&pixs);
pixDestroy(&pixd);
lept_free(data);
if (!same && index < 6) {
fprintf(stderr, "Mem gif write/read fail for file %s\n", fname);
return 1;
}
else
return 0;
}
/* Returns 1 on error */
static l_int32
test_mem_png(const char *fname)
{
l_uint8 *data = NULL;
l_int32 same;
size_t size = 0;
PIX *pixs;
PIX *pixd = NULL;
if ((pixs = pixRead(fname)) == NULL) {
fprintf(stderr, "Failure to read %s\n", fname);
return 1;
}
if (pixWriteMem(&data, &size, pixs, IFF_PNG)) {
fprintf(stderr, "Mem write fail for png\n");
return 1;
}
if ((pixd = pixReadMem(data, size)) == NULL) {
fprintf(stderr, "Mem read fail for png\n");
lept_free(data);
return 1;
}
pixEqual(pixs, pixd, &same);
if (!same)
fprintf(stderr, "Mem write/read fail for file %s\n", fname);
pixDestroy(&pixs);
pixDestroy(&pixd);
lept_free(data);
return (!same);
}
// Returns the Pix image for the image_data. Must be pixDestroyed after use.
Pix* ImageData::GetPixInternal(const GenericVector<char>& image_data) {
Pix* pix = NULL;
if (!image_data.empty()) {
// Convert the array to an image.
const unsigned char* u_data =
reinterpret_cast<const unsigned char*>(&image_data[0]);
pix = pixReadMem(u_data, image_data.size());
}
return pix;
}
/* Returns 1 on error */
static l_int32
test_writemem(PIX *pixs,
l_int32 format,
char *psfile)
{
l_uint8 *data = NULL;
l_int32 same;
size_t size = 0;
PIX *pixd = NULL;
if (format == IFF_PS) {
pixWriteMemPS(&data, &size, pixs, NULL, 0, 1.0);
l_binaryWrite(psfile, "w", data, size);
lept_free(data);
return 0;
}
/* Fail silently if library is not available */
#if !HAVE_LIBJPEG
if (format == IFF_JFIF_JPEG)
return 0;
#endif /* !HAVE_LIBJPEG */
#if !HAVE_LIBPNG
if (format == IFF_PNG)
return 0;
#endif /* !HAVE_LIBPNG */
#if !HAVE_LIBTIFF
if (format == IFF_TIFF)
return 0;
#endif /* !HAVE_LIBTIFF */
if (pixWriteMem(&data, &size, pixs, format)) {
fprintf(stderr, "Mem write fail for format %d\n", format);
return 1;
}
if ((pixd = pixReadMem(data, size)) == NULL) {
fprintf(stderr, "Mem read fail for format %d\n", format);
lept_free(data);
return 1;
}
if (format == IFF_JFIF_JPEG) {
fprintf(stderr, "jpeg size = %ld\n", size);
pixDisplayWrite(pixd, 1);
same = TRUE;
}
else {
pixEqual(pixs, pixd, &same);
if (!same)
fprintf(stderr, "Mem write/read fail for format %d\n", format);
}
pixDestroy(&pixd);
lept_free(data);
return (!same);
}
jlong Java_com_googlecode_leptonica_android_ReadFile_nativeReadMem(JNIEnv *env, jclass clazz,
jbyteArray image, jint length) {
jbyte *image_buffer = env->GetByteArrayElements(image, NULL);
int buffer_length = env->GetArrayLength(image);
PIX *pix = pixReadMem((const l_uint8 *) image_buffer, buffer_length);
env->ReleaseByteArrayElements(image, image_buffer, JNI_ABORT);
return (jlong) pix;
}
/*!
* pixaGenerateFontFromString()
*
* Input: fontsize (4, 6, 8, ... , 20, in pts at 300 ppi)
* &bl1 (<return> baseline of row 1)
* &bl2 (<return> baseline of row 2)
* &bl3 (<return> baseline of row 3)
* Return: pixa of font bitmaps for 95 characters, or null on error
*
* Notes:
* (1) See pixaGenerateFontFromFile() for details.
*/
PIXA *
pixaGenerateFontFromString(l_int32 fontsize,
l_int32 *pbl0,
l_int32 *pbl1,
l_int32 *pbl2)
{
l_uint8 *data;
l_int32 redsize, nbytes;
PIX *pix;
PIXA *pixa;
PROCNAME("pixaGenerateFontFromString");
if (!pbl0 || !pbl1 || !pbl2)
return (PIXA *)ERROR_PTR("&bl not all defined", procName, NULL);
*pbl0 = *pbl1 = *pbl2 = 0;
redsize = (fontsize / 2) - 2;
if (redsize < 0 || redsize > NUM_FONTS)
return (PIXA *)ERROR_PTR("invalid font size", procName, NULL);
if (fontsize == 4) {
data = decodeBase64(fontdata_4, strlen(fontdata_4), &nbytes);
} else if (fontsize == 6) {
data = decodeBase64(fontdata_6, strlen(fontdata_6), &nbytes);
} else if (fontsize == 8) {
data = decodeBase64(fontdata_8, strlen(fontdata_8), &nbytes);
} else if (fontsize == 10) {
data = decodeBase64(fontdata_10, strlen(fontdata_10), &nbytes);
} else if (fontsize == 12) {
data = decodeBase64(fontdata_12, strlen(fontdata_12), &nbytes);
} else if (fontsize == 14) {
data = decodeBase64(fontdata_14, strlen(fontdata_14), &nbytes);
} else if (fontsize == 16) {
data = decodeBase64(fontdata_16, strlen(fontdata_16), &nbytes);
} else if (fontsize == 18) {
data = decodeBase64(fontdata_18, strlen(fontdata_18), &nbytes);
} else { /* fontsize == 20 */
data = decodeBase64(fontdata_20, strlen(fontdata_20), &nbytes);
}
if (!data)
return (PIXA *)ERROR_PTR("data not made", procName, NULL);
pix = pixReadMem(data, nbytes);
FREE(data);
if (!pix)
return (PIXA *)ERROR_PTR("pix not made", procName, NULL);
pixa = pixaGenerateFont(pix, fontsize, pbl0, pbl1, pbl2);
pixDestroy(&pix);
return pixa;
}
int main(int argc,
char **argv)
{
char buf[512];
char *pathname, *datastr, *formstr;
l_uint8 *data1, *data2;
l_int32 i, bl1, bl2, bl3, sbytes, formbytes, fontsize, rbytes;
size_t nbytes;
PIX *pix1, *pix2, *pixd;
PIXA *pixa;
L_REGPARAMS *rp;
if (regTestSetup(argc, argv, &rp))
return 1;
/* ------------ Generate pixa char bitmap files from file ----------- */
lept_rmdir("filefonts");
lept_mkdir("filefonts");
for (i = 0; i < 9; i++) {
pixaSaveFont("fonts", "/tmp/filefonts", sizes[i]);
pathname = genPathname("/tmp/filefonts", outputfonts[i]);
pixa = pixaRead(pathname);
if (rp->display) {
fprintf(stderr, "Found %d chars in font size %d\n",
pixaGetCount(pixa), sizes[i]);
}
pixd = pixaDisplayTiled(pixa, 1500, 0, 15);
regTestWritePixAndCheck(rp, pixd, IFF_PNG); /* 0 - 8 */
if (i == 2) pixDisplayWithTitle(pixd, 100, 0, NULL, rp->display);
pixDestroy(&pixd);
pixaDestroy(&pixa);
lept_free(pathname);
}
lept_rmdir("filefonts");
/* ---------- Generate pixa char bitmap files from string --------- */
lept_rmdir("strfonts");
lept_mkdir("strfonts");
for (i = 0; i < 9; i++) {
pixaSaveFont(NULL, "/tmp/strfonts", sizes[i]);
pathname = genPathname("/tmp/strfonts", outputfonts[i]);
pixa = pixaRead(pathname);
if (rp->display) {
fprintf(stderr, "Found %d chars in font size %d\n",
pixaGetCount(pixa), sizes[i]);
}
pixd = pixaDisplayTiled(pixa, 1500, 0, 15);
regTestWritePixAndCheck(rp, pixd, IFF_PNG); /* 9 - 17 */
if (i == 2) pixDisplayWithTitle(pixd, 100, 150, NULL, rp->display);
pixDestroy(&pixd);
pixaDestroy(&pixa);
lept_free(pathname);
}
/* ----- Use pixaGetFont() and write the result out -----*/
lept_rmdir("pafonts");
lept_mkdir("pafonts");
for (i = 0; i < 9; i++) {
pixa = pixaGetFont("/tmp/strfonts", sizes[i], &bl1, &bl2, &bl3);
fprintf(stderr, "Baselines are at: %d, %d, %d\n", bl1, bl2, bl3);
snprintf(buf, sizeof(buf), "/tmp/pafonts/chars-%d.pa", sizes[i]);
pixaWrite(buf, pixa);
if (i == 2) {
pixd = pixaDisplayTiled(pixa, 1500, 0, 15);
pixDisplayWithTitle(pixd, 100, 300, NULL, rp->display);
pixDestroy(&pixd);
}
pixaDestroy(&pixa);
}
lept_rmdir("pafonts");
/* ------- Generate 4/3 encoded ascii strings from tiff files ------ */
lept_rmdir("fontencode");
lept_mkdir("fontencode");
for (i = 0; i < 9; i++) {
fontsize = 2 * i + 4;
pathname = genPathname("fonts", inputfonts[i]);
data1 = l_binaryRead(pathname, &nbytes);
datastr = encodeBase64(data1, nbytes, &sbytes);
if (rp->display)
fprintf(stderr, "nbytes = %lu, sbytes = %d\n",
(unsigned long)nbytes, sbytes);
formstr = reformatPacked64(datastr, sbytes, 4, 72, 1, &formbytes);
snprintf(buf, sizeof(buf), "/tmp/fontencode/formstr_%d.txt", fontsize);
l_binaryWrite(buf, "w", formstr, formbytes);
regTestCheckFile(rp, buf); /* 18-26 */
if (i == 8)
pix1 = pixReadMem(data1, nbytes); /* original */
FREE(data1);
data2 = decodeBase64(datastr, sbytes, &rbytes);
snprintf(buf, sizeof(buf), "/tmp/fontencode/image_%d.tif", fontsize);
l_binaryWrite(buf, "w", data2, rbytes);
if (i == 8) {
pix2 = pixReadMem(data2, rbytes); /* encode/decode */
regTestComparePix(rp, pix1, pix2); /* 27 */
pixDestroy(&pix1);
pixDestroy(&pix2);
}
FREE(data2);
FREE(pathname);
FREE(datastr);
FREE(formstr);
}
/* ------------ Get timing for font generation ----------- */
startTimer();
for (i = 0; i < 100; i++) {
pixa = pixaGenerateFontFromString(sizes[5], &bl1, &bl2, &bl3);
pixaDestroy(&pixa);
}
fprintf(stderr, "Time for font gen = %7.4f sec\n", stopTimer() / 100.0);
return regTestCleanup(rp);
}
char *internalDoScanningOperation(char *uuid, char *lang) {
int request_resolution = 0;
int docid;
int current_page = 0;
int total_requested_pages;
double totbytes = 0;
SANE_Status status;
SANE_Handle *openDeviceHandle;
SANE_Byte *raw_image;
SANE_Parameters pars;
char *docid_s;
char *total_requested_pages_s;
char *devName;
char *outFilename;
char *raw_image_format;
char *header;
o_log(DEBUGM, "doScanningOperation: sane initialized uuid(%s)",(char *)uuid);
updateScanProgress(uuid, SCAN_WAITING_ON_SCANNER, 0);
// Open the device
devName = getScanParam(uuid, SCAN_PARAM_DEVNAME);
o_log(DEBUGM, "sane_open of \"%s\"",devName);
status = sane_open ((SANE_String_Const) devName, (SANE_Handle)&openDeviceHandle);
if(status != SANE_STATUS_GOOD) {
handleSaneErrors("Cannot open device ", devName, status, 0);
updateScanProgress(uuid, SCAN_ERRO_FROM_SCANNER, status);
free(devName);
return 0;
}
free(devName);
/* ========================================================== */
if ( ! setOptions( (char *)uuid, openDeviceHandle, &request_resolution ) )
return 0;
o_log(DEBUGM, "sane_start: setOptions returned request_resolution %d\n",request_resolution);
int timeout = 5;
while( 0 < timeout ) {
status = sane_start (openDeviceHandle);
if(status == SANE_STATUS_GOOD) {
break;
}
else {
if(status == SANE_STATUS_DEVICE_BUSY ) {
// BUSY signal could be the scanner just having a
// bit of lag - specially network connected devices
timeout--;
if ( timeout == 0 ) {
handleSaneErrors("Cannot start scanning", "even after trying several time", status, 0);
updateScanProgress(uuid, SCAN_ERRO_FROM_SCANNER, status);
return 0;
}
else {
o_log(WARNING, "Device reports not ready to 'start', waiting 500ms. Will try another %d times", timeout);
usleep(500 * 1000); // 500ms or 0.5sec
}
}
else {
handleSaneErrors("Cannot start scanning", "", status, 0);
updateScanProgress(uuid, SCAN_ERRO_FROM_SCANNER, status);
return 0;
}
}
}
// Get scanning params (from the scanner)
if( request_resolution == 0 ) {
o_log(DEBUGM, "Resolution did not get set in scanner setup.");
updateScanProgress(uuid, SCAN_INTERNAL_ERROR, 10004);
return 0;
}
o_log(DEBUGM, "Get scanning params");
status = sane_get_parameters (openDeviceHandle, &pars);
o_log(INFORMATION, "Scanner Parm : stat=%s form=%d,lf=%d,bpl=%d,pixpl=%d,lin=%d,dep=%d",
sane_strstatus (status),
pars.format, pars.last_frame,
pars.bytes_per_line, pars.pixels_per_line,
pars.lines, pars.depth);
switch (pars.format) {
case SANE_FRAME_GRAY:
o_log(DEBUGM, "Expecting Gray data (1 channel only).");
raw_image_format = o_strdup( "P5" );
break;
case SANE_FRAME_RGB:
o_log(DEBUGM, "Expecting RGB data (3 channels).");
raw_image_format = o_strdup( "P6" );
break;
default:
o_log(DEBUGM, "backend returns three frames speratly. We do not currently support this.");
updateScanProgress(uuid, SCAN_INTERNAL_ERROR, 10003);
return 0;
break;
}
header = o_printf ("%s\n# SANE data follows\n%d %d\n%d\n",
raw_image_format, pars.pixels_per_line, pars.lines,
(pars.depth <= 8) ? 255 : 65535);
free( raw_image_format );
// Save Record
//
docid_s = getScanParam(uuid, SCAN_PARAM_DOCID);
total_requested_pages_s = getScanParam(uuid, SCAN_PARAM_REQUESTED_PAGES);
total_requested_pages = atoi(total_requested_pages_s);
free(total_requested_pages_s);
if( docid_s == NULL ) {
o_log(DEBUGM, "Saving record");
updateScanProgress(uuid, SCAN_DB_WORKING, 0);
docid_s = addNewScannedDoc(pars.lines, pars.pixels_per_line, request_resolution, total_requested_pages);
setScanParam(uuid, SCAN_PARAM_DOCID, docid_s);
setScanParam(uuid, SCAN_PARAM_ON_PAGE, "1");
current_page = 1;
}
else {
char *current_page_s = getScanParam(uuid, SCAN_PARAM_ON_PAGE);
current_page = atoi(current_page_s);
free(current_page_s);
current_page++;
current_page_s = itoa(current_page, 10);
setScanParam(uuid, SCAN_PARAM_ON_PAGE, current_page_s);
free(current_page_s);
}
docid = atoi(docid_s);
free(docid_s);
totbytes = (double)((pars.bytes_per_line * pars.lines));
/* ========================================================== */
raw_image = collectData( (char *)uuid, openDeviceHandle, totbytes, pars.bytes_per_line, header );
o_log(INFORMATION, "Scanning done.");
o_log(DEBUGM, "sane_cancel");
sane_cancel(openDeviceHandle);
o_log(DEBUGM, "sane_close");
sane_close(openDeviceHandle);
// Convert Raw into JPEG
//
updateScanProgress(uuid, SCAN_CONVERTING_FORMAT, 0);
PIX *pix;
if ( ( pix = pixReadMem( raw_image, (pars.bytes_per_line*pars.lines)+strlen(header) ) ) == NULL) {
o_log(ERROR, "Could not load the image data into a PIX");
}
updateScanProgress(uuid, SCAN_CONVERTING_FORMAT, 55);
o_log(INFORMATION, "Convertion process: Loaded (depth: %d)", pixGetDepth(pix));
free(raw_image);
free(header);
outFilename = o_printf("%s/scans/%d_%d.jpg", BASE_DIR, docid, current_page);
pixWrite(outFilename, pix, IFF_JFIF_JPEG);
free(outFilename);
updateScanProgress(uuid, SCAN_CONVERTING_FORMAT, 100);
o_log(INFORMATION, "Conversion process: Complete");
// Do OCR - on this page
// - OCR libs just wants the raw data and not the image header
ocrImage( uuid, docid, current_page, request_resolution, pix, lang );
#ifdef CAN_PHASH
// Calulate the pHash, so we can compare images later
if( current_page == 1 ) {
updateScanProgress(uuid, SCAN_CALULATING_PHASH, 0);
unsigned long long hash = getImagePhash_px( pix );
savePhash( docid, hash );
}
#endif /* CAN_PHASH */
pixDestroy( &pix );
// cleaup && What should we do next
//
o_log(DEBUGM, "mostly done.");
if(current_page >= total_requested_pages)
updateScanProgress(uuid, SCAN_FINISHED, docid);
else
updateScanProgress(uuid, SCAN_WAITING_ON_NEW_PAGE, ++current_page);
o_log(DEBUGM, "Page scan done.");
return o_strdup("OK");
}
int main(int argc, char **argv) {
if ((argc == 2 && strcmp(argv[1], "-v") == 0) ||
(argc == 2 && strcmp(argv[1], "--version") == 0)) {
char *versionStrP;
fprintf(stderr, "tesseract %s\n", tesseract::TessBaseAPI::Version());
versionStrP = getLeptonicaVersion();
fprintf(stderr, " %s\n", versionStrP);
lept_free(versionStrP);
versionStrP = getImagelibVersions();
fprintf(stderr, " %s\n", versionStrP);
lept_free(versionStrP);
#ifdef USE_OPENCL
cl_platform_id platform;
cl_uint num_platforms;
cl_device_id devices[2];
cl_uint num_devices;
char info[256];
int i;
fprintf(stderr, " OpenCL info:\n");
clGetPlatformIDs(1, &platform, &num_platforms);
fprintf(stderr, " Found %d platforms.\n", num_platforms);
clGetPlatformInfo(platform, CL_PLATFORM_NAME, 256, info, 0);
fprintf(stderr, " Platform name: %s.\n", info);
clGetPlatformInfo(platform, CL_PLATFORM_VERSION, 256, info, 0);
fprintf(stderr, " Version: %s.\n", info);
clGetDeviceIDs(platform, CL_DEVICE_TYPE_ALL, 2, devices, &num_devices);
fprintf(stderr, " Found %d devices.\n", num_devices);
for (i = 0; i < num_devices; ++i) {
clGetDeviceInfo(devices[i], CL_DEVICE_NAME, 256, info, 0);
fprintf(stderr, " Device %d name: %s.\n", i+1, info);
}
#endif
exit(0);
}
// Make the order of args a bit more forgiving than it used to be.
const char* lang = "eng";
const char* image = NULL;
const char* output = NULL;
const char* datapath = NULL;
bool noocr = false;
bool list_langs = false;
bool print_parameters = false;
tesseract::PageSegMode pagesegmode = tesseract::PSM_AUTO;
int arg = 1;
while (arg < argc && (output == NULL || argv[arg][0] == '-')) {
if (strcmp(argv[arg], "-l") == 0 && arg + 1 < argc) {
lang = argv[arg + 1];
++arg;
} else if (strcmp(argv[arg], "--tessdata-dir") == 0 && arg + 1 < argc) {
datapath = argv[arg + 1];
++arg;
} else if (strcmp(argv[arg], "--list-langs") == 0) {
noocr = true;
list_langs = true;
} else if (strcmp(argv[arg], "-psm") == 0 && arg + 1 < argc) {
pagesegmode = static_cast<tesseract::PageSegMode>(atoi(argv[arg + 1]));
++arg;
} else if (strcmp(argv[arg], "--print-parameters") == 0) {
noocr = true;
print_parameters = true;
} else if (strcmp(argv[arg], "-c") == 0 && arg + 1 < argc) {
// handled properly after api init
++arg;
} else if (image == NULL) {
image = argv[arg];
} else if (output == NULL) {
output = argv[arg];
}
++arg;
}
if (argc == 2 && strcmp(argv[1], "--list-langs") == 0) {
list_langs = true;
noocr = true;
}
if (output == NULL && noocr == false) {
fprintf(stderr, "Usage:\n %s imagename|stdin outputbase|stdout "
"[options...] [configfile...]\n\n", argv[0]);
fprintf(stderr, "OCR options:\n");
fprintf(stderr, " --tessdata-dir /path\tspecify location of tessdata"
" path\n");
fprintf(stderr, " -l lang[+lang]\tspecify language(s) used for OCR\n");
fprintf(stderr, " -c configvar=value\tset value for control parameter.\n"
"\t\t\tMultiple -c arguments are allowed.\n");
fprintf(stderr, " -psm pagesegmode\tspecify page segmentation mode.\n");
fprintf(stderr, "These options must occur before any configfile.\n\n");
fprintf(stderr,
"pagesegmode values are:\n"
" 0 = Orientation and script detection (OSD) only.\n"
" 1 = Automatic page segmentation with OSD.\n"
" 2 = Automatic page segmentation, but no OSD, or OCR\n"
" 3 = Fully automatic page segmentation, but no OSD. (Default)\n"
" 4 = Assume a single column of text of variable sizes.\n"
" 5 = Assume a single uniform block of vertically aligned text.\n"
" 6 = Assume a single uniform block of text.\n"
" 7 = Treat the image as a single text line.\n"
" 8 = Treat the image as a single word.\n"
" 9 = Treat the image as a single word in a circle.\n"
" 10 = Treat the image as a single character.\n\n");
fprintf(stderr, "Single options:\n");
fprintf(stderr, " -v --version: version info\n");
fprintf(stderr, " --list-langs: list available languages for tesseract "
"engine. Can be used with --tessdata-dir.\n");
fprintf(stderr, " --print-parameters: print tesseract parameters to the "
"stdout.\n");
exit(1);
}
if (output != NULL && strcmp(output, "-") && strcmp(output, "stdout")) {
tprintf("Tesseract Open Source OCR Engine v%s with Leptonica\n",
tesseract::TessBaseAPI::Version());
}
PERF_COUNT_START("Tesseract:main")
tesseract::TessBaseAPI api;
api.SetOutputName(output);
int rc = api.Init(datapath, lang, tesseract::OEM_DEFAULT,
&(argv[arg]), argc - arg, NULL, NULL, false);
if (rc) {
fprintf(stderr, "Could not initialize tesseract.\n");
exit(1);
}
char opt1[255], opt2[255];
for (arg = 0; arg < argc; arg++) {
if (strcmp(argv[arg], "-c") == 0 && arg + 1 < argc) {
strncpy(opt1, argv[arg + 1], 255);
*(strchr(opt1, '=')) = 0;
strncpy(opt2, strchr(argv[arg + 1], '=') + 1, 255);
opt2[254] = 0;
++arg;
if (!api.SetVariable(opt1, opt2)) {
fprintf(stderr, "Could not set option: %s=%s\n", opt1, opt2);
}
}
}
if (list_langs) {
GenericVector<STRING> languages;
api.GetAvailableLanguagesAsVector(&languages);
fprintf(stderr, "List of available languages (%d):\n",
languages.size());
for (int index = 0; index < languages.size(); ++index) {
STRING& string = languages[index];
fprintf(stderr, "%s\n", string.string());
}
api.End();
exit(0);
}
if (print_parameters) {
FILE* fout = stdout;
fprintf(stdout, "Tesseract parameters:\n");
api.PrintVariables(fout);
api.End();
exit(0);
}
// We have 2 possible sources of pagesegmode: a config file and
// the command line. For backwards compatability reasons, the
// default in tesseract is tesseract::PSM_SINGLE_BLOCK, but the
// default for this program is tesseract::PSM_AUTO. We will let
// the config file take priority, so the command-line default
// can take priority over the tesseract default, so we use the
// value from the command line only if the retrieved mode
// is still tesseract::PSM_SINGLE_BLOCK, indicating no change
// in any config file. Therefore the only way to force
// tesseract::PSM_SINGLE_BLOCK is from the command line.
// It would be simpler if we could set the value before Init,
// but that doesn't work.
if (api.GetPageSegMode() == tesseract::PSM_SINGLE_BLOCK)
api.SetPageSegMode(pagesegmode);
bool stdInput = !strcmp(image, "stdin") || !strcmp(image, "-");
Pix* pixs = NULL;
if (stdInput) {
char byt;
GenericVector<l_uint8> ch_data;
std::istream file(std::cin.rdbuf());
#ifdef WIN32
if (_setmode(_fileno(stdin), _O_BINARY) == -1)
tprintf("ERROR: cin to binary: %s", strerror(errno));
#endif // WIN32
while (file.get(byt)) {
ch_data.push_back(byt);
}
std::cin.ignore(std::cin.rdbuf()->in_avail() + 1);
pixs = pixReadMem(&ch_data[0], ch_data.size());
}
if (pagesegmode == tesseract::PSM_AUTO_ONLY ||
pagesegmode == tesseract::PSM_OSD_ONLY) {
int ret_val = 0;
if (!pixs)
pixs = pixRead(image);
if (!pixs) {
fprintf(stderr, "Cannot open input file: %s\n", image);
exit(2);
}
api.SetImage(pixs);
if (pagesegmode == tesseract::PSM_OSD_ONLY) {
OSResults osr;
if (api.DetectOS(&osr)) {
int orient = osr.best_result.orientation_id;
int script_id = osr.get_best_script(orient);
float orient_oco = osr.best_result.oconfidence;
float orient_sco = osr.best_result.sconfidence;
tprintf("Orientation: %d\nOrientation in degrees: %d\n" \
"Orientation confidence: %.2f\n" \
"Script: %d\nScript confidence: %.2f\n",
orient, OrientationIdToValue(orient), orient_oco,
script_id, orient_sco);
} else {
ret_val = 1;
}
} else {
tesseract::Orientation orientation;
tesseract::WritingDirection direction;
tesseract::TextlineOrder order;
float deskew_angle;
tesseract::PageIterator* it = api.AnalyseLayout();
if (it) {
it->Orientation(&orientation, &direction, &order, &deskew_angle);
tprintf("Orientation: %d\nWritingDirection: %d\nTextlineOrder: %d\n" \
"Deskew angle: %.4f\n",
orientation, direction, order, deskew_angle);
} else {
ret_val = 1;
}
delete it;
}
pixDestroy(&pixs);
exit(ret_val);
}
tesseract::TessResultRenderer* renderer = NULL;
bool b;
api.GetBoolVariable("tessedit_create_hocr", &b);
if (b && renderer == NULL) renderer = new tesseract::TessHOcrRenderer();
api.GetBoolVariable("tessedit_create_pdf", &b);
if (b && renderer == NULL)
renderer = new tesseract::TessPDFRenderer(api.GetDatapath());
api.GetBoolVariable("tessedit_create_boxfile", &b);
if (b && renderer == NULL) renderer = new tesseract::TessBoxTextRenderer();
if (renderer == NULL) renderer = new tesseract::TessTextRenderer();
if (pixs) {
api.ProcessPage(pixs, 0, NULL, NULL, 0, renderer);
pixDestroy(&pixs);
} else {
FILE* fin = fopen(image, "rb");
if (fin == NULL) {
fprintf(stderr, "Cannot open input file: %s\n", image);
exit(2);
}
fclose(fin);
if (!api.ProcessPages(image, NULL, 0, renderer)) {
fprintf(stderr, "Error during processing.\n");
exit(1);
}
}
FILE* fout = stdout;
if (strcmp(output, "-") && strcmp(output, "stdout")) {
STRING outfile = STRING(output)
+ STRING(".")
+ STRING(renderer->file_extension());
fout = fopen(outfile.string(), "wb");
if (fout == NULL) {
fprintf(stderr, "Cannot create output file %s\n", outfile.string());
exit(1);
}
}
const char* data;
inT32 data_len;
if (renderer->GetOutput(&data, &data_len)) {
fwrite(data, 1, data_len, fout);
if (fout != stdout)
fclose(fout);
else
clearerr(fout);
}
PERF_COUNT_END
return 0; // Normal exit
}
