simon987/sist2
1
0
Fork 0
mirror of https://github.com/simon987/sist2.git synced 2025-04-19 18:26:43 +00:00
Code Issues Projects Releases Wiki Activity

fix #228: Handle multiple languages for tesseract OCR

Browse Source
This commit is contained in:
Yatao Li 2022-01-09 03:09:46 +08:00
parent fe1aa6dd4c
commit d10c13d46d
1 changed files with 26 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

29
src/cli.c
View File

@ -77,6 +77,15 @@ void exec_args_destroy(exec_args_t *args) {
free(args); free(args);
} }
__always_inline
static const char* find_tessdata(const char* p_lang, size_t len) {
char filename[128];
memcpy(filename, p_lang, len);
filename[len] = '\0';
strcat(filename, ".traineddata");
return find_file_in_paths(TESS_DATAPATHS, filename);
}
int scan_args_validate(scan_args_t *args, int argc, const char **argv) { int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
if (argc < 2) { if (argc < 2) {
fprintf(stderr, "Required positional argument: PATH.\n"); fprintf(stderr, "Required positional argument: PATH.\n");
@ -171,16 +180,28 @@ int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
if (args->tesseract_lang != NULL) { if (args->tesseract_lang != NULL) {
TessBaseAPI *api = TessBaseAPICreate(); TessBaseAPI *api = TessBaseAPICreate();
char filename[128]; const char *p_lang = args->tesseract_lang;
sprintf(filename, "%s.traineddata", args->tesseract_lang); const char *e_lang = p_lang + strlen(p_lang);
const char *path = find_file_in_paths(TESS_DATAPATHS, filename); const char *path = NULL;
if (path == NULL) { while (p_lang < e_lang) {
const char *p_plus = strstr(p_lang, "+");
if (p_plus == NULL) {
p_plus = e_lang;
}
const char* _path = find_tessdata(p_lang, p_plus - p_lang);
if (_path == NULL) {
LOG_FATAL("cli.c", "Could not find tesseract language file!"); LOG_FATAL("cli.c", "Could not find tesseract language file!");
} else if (path != NULL && path != _path) {
LOG_FATAL("cli.c", "Multiple tesseract language files are not in the same directory!");
}
path = _path;
p_lang = p_plus + 1;
} }
ret = TessBaseAPIInit3(api, path, args->tesseract_lang); ret = TessBaseAPIInit3(api, path, args->tesseract_lang);
if (ret != 0) { if (ret != 0) {
fprintf(stderr, "Could not initialize tesseract with lang '%s'\n", args->tesseract_lang); fprintf(stderr, "Could not initialize tesseract with lang '%s'\n", args->tesseract_lang);
TessBaseAPIDelete(api);
return 1; return 1;
} }
TessBaseAPIEnd(api); TessBaseAPIEnd(api);