mirror/tesseract
1
0
Fork 0
mirror of https://github.com/tesseract-ocr/tesseract.git synced 2024-11-27 12:49:35 +08:00
Code Issues Actions 10 Packages Projects Releases Wiki Activity

Remove trailing whitespace from PAGE XML <Unicode> output

Browse Source 
Fixes: 577e8a8b93 ("Add PAGE XML renderer / export (#4214)")
Signed-off-by: Stefan Weil <sw@weilnetz.de>
This commit is contained in:
Stefan Weil 2024-05-03 07:44:57 +02:00
parent d461867812
commit 4a49e0340f
1 changed files with 6 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
src/api/pagerenderer.cpp
View File

@ -873,7 +873,11 @@ char *TessBaseAPI::GetPAGEText(ETEXT_DESC *monitor, int page_number) {
if (res_it->IsAtBeginningOf(RIL_TEXTLINE)) {
// writing_direction_before = writing_direction;
line_conf = ((res_it->Confidence(RIL_TEXTLINE)) / 100.);
line_content << HOcrEscape(res_it->GetUTF8Text(RIL_TEXTLINE)).c_str();
std::string textline = res_it->GetUTF8Text(RIL_TEXTLINE);
if (textline.back() == '\n') {
textline.erase(textline.length() - 1);
}
line_content << HOcrEscape(textline.c_str());
line_str << "\t\t\t<TextLine id=\"r" << rcnt << "l" << lcnt << "\" ";
if (writing_direction != 0 &&
writing_direction != writing_direction_block) {
@ -1078,12 +1082,10 @@ char *TessBaseAPI::GetPAGEText(ETEXT_DESC *monitor, int page_number) {
region_content << line_content.str();
line_content.str("");
if (!last_word_in_cblock) {
region_content << "\n\t\t\t\t\t";
region_content << '\n';
}
lcnt++;
wcnt = 0;
} else {
line_content << " ";
}
// Write region information to the output