mirror/tesseract
1
0
Fork 0
mirror of https://github.com/tesseract-ocr/tesseract.git synced 2024-11-30 23:49:05 +08:00
Code Issues Actions 10 Packages Projects Releases Wiki Activity

Only generate dir for HOCR when needed - fixes #208

Browse Source 
Takes advantage of inheritance and dir="ltr" default to:
 - only generate paragraph dirs which are not ltr
 - only generate word dirs which don't match enclosing paragraph

Tested against LTR, RTL, and mixed direction files. Files for the
latter two cases are in a separate commit on the ltr-test-files branch.
This commit is contained in:
Tom Morris 2016-02-14 16:51:29 -05:00 committed by Zdenko Podobný
parent c3ad0de69b
commit 381b3a56c6
1 changed files with 10 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
api/baseapi.cpp
View File

@ -1432,6 +1432,7 @@ char* TessBaseAPI::GetHOCRText(int page_number) {
int lcnt = 1, bcnt = 1, pcnt = 1, wcnt = 1; int lcnt = 1, bcnt = 1, pcnt = 1, wcnt = 1;
int page_id = page_number + 1; // hOCR uses 1-based page numbers. int page_id = page_number + 1; // hOCR uses 1-based page numbers.
bool para_is_ltr = true; // Default direction is LTR
bool font_info = false; bool font_info = false;
GetBoolVariable("hocr_font_info", &font_info); GetBoolVariable("hocr_font_info", &font_info);
@ -1481,15 +1482,15 @@ char* TessBaseAPI::GetHOCRText(int page_number) {
// Open any new block/paragraph/textline. // Open any new block/paragraph/textline.
if (res_it->IsAtBeginningOf(RIL_BLOCK)) { if (res_it->IsAtBeginningOf(RIL_BLOCK)) {
para_is_ltr = true; // reset to default direction
hocr_str += " <div class='ocr_carea'"; hocr_str += " <div class='ocr_carea'";
AddIdTohOCR(&hocr_str, "block", page_id, bcnt); AddIdTohOCR(&hocr_str, "block", page_id, bcnt);
AddBoxTohOCR(res_it, RIL_BLOCK, &hocr_str); AddBoxTohOCR(res_it, RIL_BLOCK, &hocr_str);
} }
if (res_it->IsAtBeginningOf(RIL_PARA)) { if (res_it->IsAtBeginningOf(RIL_PARA)) {
hocr_str += "\n <p class='ocr_par'"; hocr_str += "\n <p class='ocr_par'";
if (res_it->ParagraphIsLtr()) { para_is_ltr = res_it->ParagraphIsLtr();
hocr_str += " dir='ltr'"; if (!para_is_ltr) {
} else {
hocr_str += " dir='rtl'"; hocr_str += " dir='rtl'";
} }
AddIdTohOCR(&hocr_str, "par", page_id, pcnt); AddIdTohOCR(&hocr_str, "par", page_id, pcnt);
@ -1531,8 +1532,11 @@ char* TessBaseAPI::GetHOCRText(int page_number) {
hocr_str += "'"; hocr_str += "'";
} }
switch (res_it->WordDirection()) { switch (res_it->WordDirection()) {
case DIR_LEFT_TO_RIGHT: hocr_str += " dir='ltr'"; break; // Only emit direction if different from current paragraph direction
case DIR_RIGHT_TO_LEFT: hocr_str += " dir='rtl'"; break; case DIR_LEFT_TO_RIGHT: if (!para_is_ltr) hocr_str += " dir='ltr'"; break;
case DIR_RIGHT_TO_LEFT: if (para_is_ltr) hocr_str += " dir='rtl'"; break;
case DIR_MIX:
case DIR_NEUTRAL:
default: // Do nothing. default: // Do nothing.
break; break;
} }
@ -1562,6 +1566,7 @@ char* TessBaseAPI::GetHOCRText(int page_number) {
if (last_word_in_para) { if (last_word_in_para) {
hocr_str += "\n </p>\n"; hocr_str += "\n </p>\n";
pcnt++; pcnt++;
para_is_ltr = true; // back to default direction
} }
if (last_word_in_block) { if (last_word_in_block) {
hocr_str += " </div>\n"; hocr_str += " </div>\n";