mirror/tesseract
1
0
Fork 0
mirror of https://github.com/tesseract-ocr/tesseract.git synced 2025-01-22 18:13:42 +08:00
Code Issues Actions 11 Packages Projects Releases Wiki Activity

Merge pull request #2705 from stweil/title

Browse Source 
Add new parameter "document_title" to set the title in OCR output files
This commit is contained in:
Egor Pugin 2019-10-11 02:23:36 +03:00 committed by GitHub
commit 247cd0edc4
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 5 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
src/api/altorenderer.cpp
View File

@ -20,6 +20,7 @@
# include "host.h" // windows.h for MultiByteToWideChar, ... # include "host.h" // windows.h for MultiByteToWideChar, ...
#endif #endif
#include "renderer.h" #include "renderer.h"
#include "strngs.h" // for STRING
namespace tesseract { namespace tesseract {
@ -68,7 +69,7 @@ bool TessAltoRenderer::BeginDocumentHandler() {
AppendString(title()); AppendString(title());
AppendString( AppendString(
"\t\t\t</fileName>\n" "</fileName>\n"
"\t\t</sourceImageInformation>\n" "\t\t</sourceImageInformation>\n"
"\t\t<OCRProcessing ID=\"OCR_0\">\n" "\t\t<OCRProcessing ID=\"OCR_0\">\n"
"\t\t\t<ocrProcessingStep>\n" "\t\t\t<ocrProcessingStep>\n"

5
src/api/baseapi.cpp
View File

@ -96,6 +96,7 @@
#include "werd.h" // for WERD, WERD_IT, W_FUZZY_NON, W_FUZZY_SP #include "werd.h" // for WERD, WERD_IT, W_FUZZY_NON, W_FUZZY_SP
static BOOL_VAR(stream_filelist, false, "Stream a filelist from stdin"); static BOOL_VAR(stream_filelist, false, "Stream a filelist from stdin");
static STRING_VAR(document_title, "", "Title of output document (used for hOCR and PDF output)");
namespace tesseract { namespace tesseract {
@ -993,7 +994,7 @@ bool TessBaseAPI::ProcessPagesFileList(FILE *flist,
} }
// Begin producing output // Begin producing output
if (renderer && !renderer->BeginDocument(unknown_title_)) { if (renderer && !renderer->BeginDocument(document_title.c_str())) {
return false; return false;
} }
@ -1208,7 +1209,7 @@ bool TessBaseAPI::ProcessPagesInternal(const char* filename,
} }
// Begin the output // Begin the output
if (renderer && !renderer->BeginDocument(unknown_title_)) { if (renderer && !renderer->BeginDocument(document_title.c_str())) {
pixDestroy(&pix); pixDestroy(&pix);
return false; return false;
} }

6
src/api/baseapi.h
View File

@ -923,12 +923,6 @@ class TESS_API TessBaseAPI {
int timeout_millisec, int timeout_millisec,
TessResultRenderer* renderer, TessResultRenderer* renderer,
int tessedit_page_number); int tessedit_page_number);
// There's currently no way to pass a document title from the
// Tesseract command line, and we have multiple places that choose
// to set the title to an empty string. Using a single named
// variable will hopefully reduce confusion if the situation changes
// in the future.
const char *unknown_title_ = "";
}; // class TessBaseAPI. }; // class TessBaseAPI.
/** Escape a char string - remove &<>"' with HTML codes. */ /** Escape a char string - remove &<>"' with HTML codes. */