/* array = '[' [ value { ',' value } ] ']' */
static int parse_array(struct frozen *f) {
int i = 0, current_path_len;
char buf[20];
CALL_BACK(f, JSON_TYPE_ARRAY_START, NULL, 0);
TRY(test_and_skip(f, '['));
{
{
SET_STATE(f, f->cur - 1, "", 0);
while (cur(f) != ']') {
snprintf(buf, sizeof(buf), "[%d]", i);
i++;
current_path_len = append_to_path(f, buf, strlen(buf));
f->cur_name =
f->path + strlen(f->path) - strlen(buf) + 1 /*opening brace*/;
f->cur_name_len = strlen(buf) - 2 /*braces*/;
TRY(parse_value(f));
truncate_path(f, current_path_len);
if (cur(f) == ',') f->cur++;
}
TRY(test_and_skip(f, ']'));
truncate_path(f, fstate.path_len);
CALL_BACK(f, JSON_TYPE_ARRAY_END, fstate.ptr, f->cur - fstate.ptr);
}
}
return 0;
}
/* number = [ '-' ] digit+ [ '.' digit+ ] [ ['e'|'E'] ['+'|'-'] digit+ ] */
static int parse_number(struct frozen *f) {
int ch = cur(f);
SET_STATE(f, f->cur, "", 0);
if (ch == '-') f->cur++;
EXPECT(f->cur < f->end, JSON_STRING_INCOMPLETE);
EXPECT(is_digit(f->cur[0]), JSON_STRING_INVALID);
while (f->cur < f->end && is_digit(f->cur[0])) f->cur++;
if (f->cur < f->end && f->cur[0] == '.') {
f->cur++;
EXPECT(f->cur < f->end, JSON_STRING_INCOMPLETE);
EXPECT(is_digit(f->cur[0]), JSON_STRING_INVALID);
while (f->cur < f->end && is_digit(f->cur[0])) f->cur++;
}
if (f->cur < f->end && (f->cur[0] == 'e' || f->cur[0] == 'E')) {
f->cur++;
EXPECT(f->cur < f->end, JSON_STRING_INCOMPLETE);
if ((f->cur[0] == '+' || f->cur[0] == '-')) f->cur++;
EXPECT(f->cur < f->end, JSON_STRING_INCOMPLETE);
EXPECT(is_digit(f->cur[0]), JSON_STRING_INVALID);
while (f->cur < f->end && is_digit(f->cur[0])) f->cur++;
}
truncate_path(f, fstate.path_len);
CALL_BACK(f, JSON_TYPE_NUMBER, fstate.ptr, f->cur - fstate.ptr);
return 0;
}
/* identifier = letter { letter | digit | '_' } */
static int parse_identifier(struct frozen *f) {
EXPECT(is_alpha(cur(f)), JSON_STRING_INVALID);
{
SET_STATE(f, f->cur, "", 0);
while (f->cur < f->end &&
(*f->cur == '_' || is_alpha(*f->cur) || is_digit(*f->cur))) {
f->cur++;
}
truncate_path(f, fstate.path_len);
CALL_BACK(f, JSON_TYPE_STRING, fstate.ptr, f->cur - fstate.ptr);
}
return 0;
}
/* pair = key ':' value */
static int parse_pair(struct frozen *f) {
int current_path_len;
const char *tok;
skip_whitespaces(f);
tok = f->cur;
TRY(parse_key(f));
current_path_len =
append_to_path(f, *tok == '"' ? tok + 1 : tok,
*tok == '"' ? f->cur - tok - 2 : f->cur - tok);
TRY(test_and_skip(f, ':'));
TRY(parse_value(f));
truncate_path(f, current_path_len);
return 0;
}
/* object = '{' pair { ',' pair } '}' */
static int parse_object(struct frozen *f) {
CALL_BACK(f, JSON_TYPE_OBJECT_START, NULL, 0);
TRY(test_and_skip(f, '{'));
{
SET_STATE(f, f->cur - 1, ".", 1);
while (cur(f) != '}') {
TRY(parse_pair(f));
if (cur(f) == ',') f->cur++;
}
TRY(test_and_skip(f, '}'));
truncate_path(f, fstate.path_len);
CALL_BACK(f, JSON_TYPE_OBJECT_END, fstate.ptr, f->cur - fstate.ptr);
}
return 0;
}
static int expect(struct frozen *f, const char *s, int len,
enum json_token_type tok_type) {
int i, n = left(f);
SET_STATE(f, f->cur, "", 0);
for (i = 0; i < len; i++) {
if (i >= n) return JSON_STRING_INCOMPLETE;
if (f->cur[i] != s[i]) return JSON_STRING_INVALID;
}
f->cur += len;
truncate_path(f, fstate.path_len);
CALL_BACK(f, tok_type, fstate.ptr, f->cur - fstate.ptr);
return 0;
}
/* array = '[' [ value { ',' value } ] ']' */
static int parse_array(struct frozen *f) {
int i = 0, current_path_len;
char buf[20];
TRY(test_and_skip(f, '['));
{
SET_STATE(f, f->cur - 1, JSON_TYPE_ARRAY, "", 0);
while (cur(f) != ']') {
snprintf(buf, sizeof(buf), "[%d]", i);
i++;
current_path_len = append_to_path(f, buf, strlen(buf));
TRY(parse_value(f));
truncate_path(f, current_path_len);
if (cur(f) == ',') f->cur++;
}
TRY(test_and_skip(f, ']'));
CALL_BACK(f);
}
return 0;
}
/* string = '"' { quoted_printable_chars } '"' */
static int parse_string(struct frozen *f) {
int n, ch = 0, len = 0;
TRY(test_and_skip(f, '"'));
{
SET_STATE(f, f->cur, "", 0);
for (; f->cur < f->end; f->cur += len) {
ch = *(unsigned char *) f->cur;
len = get_utf8_char_len((unsigned char) ch);
EXPECT(ch >= 32 && len > 0, JSON_STRING_INVALID); /* No control chars */
EXPECT(len <= left(f), JSON_STRING_INCOMPLETE);
if (ch == '\\') {
EXPECT((n = get_escape_len(f->cur + 1, left(f))) > 0, n);
len += n;
} else if (ch == '"') {
truncate_path(f, fstate.path_len);
CALL_BACK(f, JSON_TYPE_STRING, fstate.ptr, f->cur - fstate.ptr);
f->cur++;
break;
};
}
}
return ch == '"' ? 0 : JSON_STRING_INCOMPLETE;
}
static void format_filelist_filename(LPWSTR file, LPWSTR out)
{
LPWSTR pos_basename;
LPWSTR truncpos1, truncpos2;
WCHAR myDocs[MAX_PATH];
SHGetFolderPathW(NULL, CSIDL_PERSONAL, NULL, SHGFP_TYPE_CURRENT, myDocs);
pos_basename = file_basename(file);
truncpos1 = NULL;
truncpos2 = NULL;
*(pos_basename-1) = 0;
if(!lstrcmpiW(file, myDocs) || (lstrlenW(pos_basename) > FILELIST_ENTRY_LENGTH))
{
truncpos1 = pos_basename;
*(pos_basename-1) = '\\';
} else
{
LPWSTR pos;
BOOL morespace = FALSE;
*(pos_basename-1) = '\\';
for(pos = file; pos < pos_basename; pos++)
{
if(*pos == '\\' || *pos == '/')
{
if(truncpos1)
{
if((pos - file + lstrlenW(pos_basename)) > FILELIST_ENTRY_LENGTH)
break;
truncpos1 = pos;
morespace = TRUE;
break;
}
if((pos - file + lstrlenW(pos_basename)) > FILELIST_ENTRY_LENGTH)
break;
truncpos1 = pos;
}
}
if(morespace)
{
for(pos = pos_basename; pos >= truncpos1; pos--)
{
if(*pos == '\\' || *pos == '/')
{
if((truncpos1 - file + lstrlenW(pos_basename) + pos_basename - pos) > FILELIST_ENTRY_LENGTH)
break;
truncpos2 = pos;
}
}
}
}
if(truncpos1 == pos_basename)
lstrcatW(out, pos_basename);
else if(truncpos1 == truncpos2 || !truncpos2)
lstrcatW(out, file);
else
truncate_path(file, out, truncpos1, truncpos2);
}
int main(int argc, char **argv) {
#ifdef USING_GETTEXT
setlocale (LC_ALL, "");
bindtextdomain (PACKAGE, LOCALEDIR);
textdomain (PACKAGE);
#endif
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);
exit(0);
}
tesseract::TessBaseAPI api;
int rc = api.Init(argv[0], NULL);
if (rc) {
fprintf(stderr, "Could not initialize tesseract.\n");
exit(1);
}
if (argc == 2 && strcmp(argv[1], "--list-langs") == 0) {
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.Clear();
exit(0);
}
api.End();
// 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;
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], "-psm") == 0 && arg + 1 < argc) {
pagesegmode = static_cast<tesseract::PageSegMode>(atoi(argv[arg + 1]));
++arg;
} else if (image == NULL) {
image = argv[arg];
} else if (output == NULL) {
output = argv[arg];
}
++arg;
}
if (output == NULL) {
fprintf(stderr, _("Usage:%s imagename outputbase [-l lang] "
"[-psm pagesegmode] [configfile...]\n\n"), argv[0]);
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"));
fprintf(stderr, _("-l lang and/or -psm pagesegmode must occur before any"
"configfile.\n\n"));
fprintf(stderr, _("Single options:\n"));
fprintf(stderr, _(" -v --version: version info\n"));
fprintf(stderr, _(" --list-langs: list available languages for tesseract engine\n"));
exit(1);
}
api.SetOutputName(output);
STRING tessdata_dir;
truncate_path(argv[0], &tessdata_dir);
rc = api.Init(tessdata_dir.string(), lang, tesseract::OEM_DEFAULT,
&(argv[arg]), argc - arg, NULL, NULL, false);
if (rc) {
fprintf(stderr, "Could not initialize tesseract.\n");
exit(1);
}
// 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);
tprintf("Tesseract Open Source OCR Engine v%s with Leptonica\n",
tesseract::TessBaseAPI::Version());
FILE* fin = fopen(image, "rb");
if (fin == NULL) {
printf("Cannot open input file: %s\n", image);
exit(2);
}
fclose(fin);
PIX *pixs;
if ((pixs = pixRead(image)) == NULL) {
printf("Unsupported image type.\n");
exit(3);
}
pixDestroy(&pixs);
STRING text_out;
if (!api.ProcessPages(image, NULL, 0, &text_out)) {
printf("Error during processing.\n");
}
bool output_hocr = false;
api.GetBoolVariable("tessedit_create_hocr", &output_hocr);
bool output_box = false;
api.GetBoolVariable("tessedit_create_boxfile", &output_box);
STRING outfile = output;
outfile += output_hocr ? ".html" : output_box ? ".box" : ".txt";
FILE* fout = fopen(outfile.string(), "wb");
if (fout == NULL) {
printf("Cannot create output file %s\n", outfile.string());
exit(1);
}
fwrite(text_out.string(), 1, text_out.length(), fout);
fclose(fout);
return 0; // Normal exit
}
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);
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;
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], "-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:%s imagename outputbase|stdout [-l lang] "
"[-psm pagesegmode] [-c configvar=value] "
"[configfile...]\n\n", argv[0]);
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");
fprintf(stderr, "multiple -c arguments are allowed.\n");
fprintf(stderr, "-l lang, -psm pagesegmode and any -c options must occur"
"before any configfile.\n\n");
fprintf(stderr, "Single options:\n");
fprintf(stderr, " -v --version: version info\n");
fprintf(stderr, " --list-langs: list available languages for tesseract "
"engine\n");
fprintf(stderr, " --print-parameters: print tesseract parameters to the "
"stdout\n");
exit(1);
}
tesseract::TessBaseAPI api;
STRING tessdata_dir;
truncate_path(argv[0], &tessdata_dir);
api.SetOutputName(output);
int rc = api.Init(tessdata_dir.string(), 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);
tprintf("Tesseract Open Source OCR Engine v%s with Leptonica\n",
tesseract::TessBaseAPI::Version());
FILE* fin = fopen(image, "rb");
if (fin == NULL) {
fprintf(stderr, "Cannot open input file: %s\n", image);
exit(2);
}
fclose(fin);
bool output_hocr = false;
api.GetBoolVariable("tessedit_create_hocr", &output_hocr);
bool output_box = false;
api.GetBoolVariable("tessedit_create_boxfile", &output_box);
FILE* fout = stdout;
if (strcmp(output, "-") && strcmp(output, "stdout")) {
STRING outfile = output;
outfile += output_hocr ? ".html" : output_box ? ".box" : ".txt";
fout = fopen(outfile.string(), "wb");
if (fout == NULL) {
fprintf(stderr, "Cannot create output file %s\n", outfile.string());
exit(1);
}
}
STRING text_out;
if (!api.ProcessPages(image, NULL, 0, &text_out)) {
fprintf(stderr, "Error during processing.\n");
if (fout != stdout)
fclose(fout);
exit(1);
}
fwrite(text_out.string(), 1, text_out.length(), fout);
if (fout != stdout)
fclose(fout);
else
clearerr(fout);
return 0; // Normal exit
}
