mirror/tesseract
1
0
Fork 0
mirror of https://github.com/tesseract-ocr/tesseract.git synced 2025-06-07 18:02:40 +08:00
Code Issues Actions 10 Packages Projects Releases Wiki Activity

Convert ALTO XML ID generation to conditional code based on the current pge

Browse Source 
This will ensure, validated ALTO XML output is generated while keeping IDs for the first page consistent as before.
This commit is contained in:
Alex Jank 2025-01-26 10:30:03 +01:00
parent c702b488f4
commit c6b7e0523d
No known key found for this signature in database
1 changed files with 20 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
src/api/altorenderer.cpp
View File

@ -51,6 +51,20 @@ static void AddBoxToAlto(const ResultIterator *it, PageIteratorLevel level,
}
}
static std::string GetID(const char *prefix, int page_number, int counter) {
std::stringstream idstr;
// IDs will only have the counter for the first page to keep them consistent
// with the IDs assigned before this change was made.
// From the second page on, IDs will also contain the page number to make them unique.
if (page_number == 0) {
idstr << prefix << "_" << counter;
} else {
idstr << prefix << "_" << page_number << "_" << counter;
}
return idstr.str();
}
///
/// Append the ALTO XML for the beginning of the document
///
@ -168,7 +182,7 @@ char *TessBaseAPI::GetAltoText(ETEXT_DESC *monitor, int page_number) {
case PT_PULLOUT_IMAGE: {
// Handle all kinds of images.
// TODO: optionally add TYPE, for example TYPE="photo".
alto_str << "\t\t\t\t<Illustration ID=\"cblock_" << page_number << "_" << bcnt++ << "\"";
alto_str << "\t\t\t\t<Illustration ID=\"" << GetID("cblock", page_number, bcnt++) << "\"";
AddBoxToAlto(res_it.get(), RIL_BLOCK, alto_str);
alto_str << "</Illustration>\n";
res_it->Next(RIL_BLOCK);
@ -177,7 +191,7 @@ char *TessBaseAPI::GetAltoText(ETEXT_DESC *monitor, int page_number) {
case PT_HORZ_LINE:
case PT_VERT_LINE:
// Handle horizontal and vertical lines.
alto_str << "\t\t\t\t<GraphicalElement ID=\"cblock_" << page_number << "_" << bcnt++ << "\"";
alto_str << "\t\t\t\t<GraphicalElement ID=\"" << GetID("cblock", page_number, bcnt++) << "\"";
AddBoxToAlto(res_it.get(), RIL_BLOCK, alto_str);
alto_str << "</GraphicalElement >\n";
res_it->Next(RIL_BLOCK);
@ -190,24 +204,24 @@ char *TessBaseAPI::GetAltoText(ETEXT_DESC *monitor, int page_number) {
}
if (res_it->IsAtBeginningOf(RIL_BLOCK)) {
alto_str << "\t\t\t\t<ComposedBlock ID=\"cblock_" << page_number << "_" << bcnt << "\"";
alto_str << "\t\t\t\t<ComposedBlock ID=\"" << GetID("cblock", page_number, bcnt) << "\"";
AddBoxToAlto(res_it.get(), RIL_BLOCK, alto_str);
alto_str << "\n";
}
if (res_it->IsAtBeginningOf(RIL_PARA)) {
alto_str << "\t\t\t\t\t<TextBlock ID=\"block_" << page_number << "_" << tcnt << "\"";
alto_str << "\t\t\t\t\t<TextBlock ID=\"" << GetID("block", page_number, tcnt) << "\"";
AddBoxToAlto(res_it.get(), RIL_PARA, alto_str);
alto_str << "\n";
}
if (res_it->IsAtBeginningOf(RIL_TEXTLINE)) {
alto_str << "\t\t\t\t\t\t<TextLine ID=\"line_" << page_number << "_" << lcnt << "\"";
alto_str << "\t\t\t\t\t\t<TextLine ID=\"" << GetID("line", page_number, lcnt) << "\"";
AddBoxToAlto(res_it.get(), RIL_TEXTLINE, alto_str);
alto_str << "\n";
}
alto_str << "\t\t\t\t\t\t\t<String ID=\"string_" << page_number << "_" << wcnt << "\"";
alto_str << "\t\t\t\t\t\t\t<String ID=\"" << GetID("string", page_number, wcnt) << "\"";
AddBoxToAlto(res_it.get(), RIL_WORD, alto_str);
alto_str << " CONTENT=\"";