mirror/tesseract
1
0
Fork 0
mirror of https://github.com/tesseract-ocr/tesseract.git synced 2025-06-06 17:32:41 +08:00
Code Issues Actions 10 Packages Projects Releases Wiki Activity

Result of clang tidy on recent merge

Browse Source
This commit is contained in:
Ray Smith 2016-11-07 10:46:33 -08:00
parent 8972c0e916
commit 2c837dffc3
139 changed files with 2488 additions and 2947 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
COPYING
View File

@ -17,5 +17,5 @@ in this distribution is now licensed under the Apache License:
Other Dependencies and Licenses:
================================
Tesseract uses Leptonica library (http://leptonica.com/) with a very weakly
restricted copyright license (http://leptonica.com/about-the-license.html)
Tesseract uses Leptonica library (http://leptonica.com/) which essentially
uses a BSD 2-clause license. (http://leptonica.com/about-the-license.html)

5
android/jni/Android.mk
View File

@ -4,7 +4,7 @@ include $(CLEAR_VARS)
LOCAL_MODULE := tesseract-$(APP_ABI)
LOCAL_STATIC_LIBRARIES := \
mobile_base \
base \
leptonica-$(APP_ABI)
LOCAL_C_INCLUDES := $(APP_C_INCLUDES)
@ -51,7 +51,6 @@ LOCAL_CFLAGS := -DANDROID_BUILD -DNO_CUBE_BUILD -DGRAPHICS_DISABLED
include $(BUILD_SHARED_LIBRARY)
$(call import-module,mobile/base)
$(call import-module,mobile/base)
$(call import-module,base/port)
$(call import-module,mobile/util/hash)
$(call import-module,third_party/leptonica/android/jni)

132
api/baseapi.cpp
View File

@ -809,9 +809,7 @@ int CubeAPITest(Boxa* boxa_blocks, Pixa* pixa_blocks,
* has not been subjected to a call of Init, SetImage, Recognize, Clear, End
* DetectOS, or anything else that changes the internal PAGE_RES.
*/
PageIterator* TessBaseAPI::AnalyseLayout() {
return AnalyseLayout(false);
}
PageIterator* TessBaseAPI::AnalyseLayout() { return AnalyseLayout(false); }
PageIterator* TessBaseAPI::AnalyseLayout(bool merge_similar_words) {
if (FindLines() == 0) {
@ -1379,8 +1377,9 @@ static void AddBaselineCoordsTohOCR(const PageIterator *it,
hocr_str->add_str_double(" ", round(p0 * 1000.0) / 1000.0);
}
static void AddIdTohOCR(STRING* hocr_str, const std::string base, int num1, int num2) {
const unsigned long BUFSIZE = 64;
static void AddIdTohOCR(STRING* hocr_str, const std::string base, int num1,
int num2) {
const size_t BUFSIZE = 64;
char id_buffer[BUFSIZE];
if (num2 >= 0) {
snprintf(id_buffer, BUFSIZE - 1, "%s_%d_%d", base.c_str(), num1, num2);
@ -1393,8 +1392,7 @@ static void AddIdTohOCR(STRING* hocr_str, const std::string base, int num1, int
*hocr_str += "'";
}
static void AddBoxTohOCR(const ResultIterator *it,
PageIteratorLevel level,
static void AddBoxTohOCR(const ResultIterator* it, PageIteratorLevel level,
STRING* hocr_str) {
int left, top, right, bottom;
it->BoundingBox(level, &left, &top, &right, &bottom);
@ -1410,7 +1408,7 @@ static void AddBoxTohOCR(const ResultIterator *it,
// add custom height measures
float row_height, descenders, ascenders; // row attributes
it->RowAttributes(&row_height, &descenders, &ascenders);
// TODO: Do we want to limit these to a single decimal place?
// TODO(rays): Do we want to limit these to a single decimal place?
hocr_str->add_str_double("; x_size ", row_height);
hocr_str->add_str_double("; x_descenders ", descenders * -1);
hocr_str->add_str_double("; x_ascenders ", ascenders);
@ -1418,9 +1416,8 @@ static void AddBoxTohOCR(const ResultIterator *it,
*hocr_str += "\">";
}
static void AddBoxToTSV(const PageIterator *it,
PageIteratorLevel level,
STRING* hocr_str) {
static void AddBoxToTSV(const PageIterator* it, PageIteratorLevel level,
STRING* hocr_str) {
int left, top, right, bottom;
it->BoundingBox(level, &left, &top, &right, &bottom);
hocr_str->add_str_int("\t", left);
@ -1429,8 +1426,6 @@ static void AddBoxToTSV(const PageIterator *it,
hocr_str->add_str_int("\t", bottom - top);
}
/**
* Make a HTML-formatted string with hOCR markup from the internal
* data structures.
@ -1440,7 +1435,7 @@ static void AddBoxToTSV(const PageIterator *it,
* STL removed from original patch submission and refactored by rays.
*/
char* TessBaseAPI::GetHOCRText(int page_number) {
return GetHOCRText(NULL,page_number);
return GetHOCRText(NULL, page_number);
}
/**
@ -1452,13 +1447,12 @@ char* TessBaseAPI::GetHOCRText(int page_number) {
* STL removed from original patch submission and refactored by rays.
*/
char* TessBaseAPI::GetHOCRText(ETEXT_DESC* monitor, int page_number) {
if (tesseract_ == NULL ||
(page_res_ == NULL && Recognize(monitor) < 0))
if (tesseract_ == NULL || (page_res_ == NULL && Recognize(monitor) < 0))
return NULL;
int lcnt = 1, bcnt = 1, pcnt = 1, wcnt = 1;
int page_id = page_number + 1; // hOCR uses 1-based page numbers.
bool para_is_ltr = true; // Default direction is LTR
bool para_is_ltr = true; // Default direction is LTR
const char* paragraph_lang = NULL;
bool font_info = false;
GetBoolVariable("hocr_font_info", &font_info);
@ -1470,13 +1464,13 @@ char* TessBaseAPI::GetHOCRText(ETEXT_DESC* monitor, int page_number) {
#ifdef _WIN32
// convert input name from ANSI encoding to utf-8
int str16_len = MultiByteToWideChar(CP_ACP, 0, input_file_->string(), -1,
NULL, 0);
int str16_len =
MultiByteToWideChar(CP_ACP, 0, input_file_->string(), -1, NULL, 0);
wchar_t *uni16_str = new WCHAR[str16_len];
str16_len = MultiByteToWideChar(CP_ACP, 0, input_file_->string(), -1,
uni16_str, str16_len);
int utf8_len = WideCharToMultiByte(CP_UTF8, 0, uni16_str, str16_len, NULL,
0, NULL, NULL);
int utf8_len = WideCharToMultiByte(CP_UTF8, 0, uni16_str, str16_len, NULL, 0,
NULL, NULL);
char *utf8_str = new char[utf8_len];
WideCharToMultiByte(CP_UTF8, 0, uni16_str, str16_len, utf8_str,
utf8_len, NULL, NULL);
@ -1509,7 +1503,7 @@ char* TessBaseAPI::GetHOCRText(ETEXT_DESC* monitor, int page_number) {
// Open any new block/paragraph/textline.
if (res_it->IsAtBeginningOf(RIL_BLOCK)) {
para_is_ltr = true; // reset to default direction
para_is_ltr = true; // reset to default direction
hocr_str += " <div class='ocr_carea'";
AddIdTohOCR(&hocr_str, "block", page_id, bcnt);
AddBoxTohOCR(res_it, RIL_BLOCK, &hocr_str);
@ -1523,9 +1517,9 @@ char* TessBaseAPI::GetHOCRText(ETEXT_DESC* monitor, int page_number) {
AddIdTohOCR(&hocr_str, "par", page_id, pcnt);
paragraph_lang = res_it->WordRecognitionLanguage();
if (paragraph_lang) {
hocr_str += " lang='";
hocr_str += paragraph_lang;
hocr_str += "'";
hocr_str += " lang='";
hocr_str += paragraph_lang;
hocr_str += "'";
}
AddBoxTohOCR(res_it, RIL_PARA, &hocr_str);
}
@ -1567,8 +1561,12 @@ char* TessBaseAPI::GetHOCRText(ETEXT_DESC* monitor, int page_number) {
}
switch (res_it->WordDirection()) {
// Only emit direction if different from current paragraph direction
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_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.
@ -1600,7 +1598,7 @@ char* TessBaseAPI::GetHOCRText(ETEXT_DESC* monitor, int page_number) {
if (last_word_in_para) {
hocr_str += "\n </p>\n";
pcnt++;
para_is_ltr = true; // back to default direction
para_is_ltr = true; // back to default direction
}
if (last_word_in_block) {
hocr_str += " </div>\n";
@ -1620,8 +1618,7 @@ char* TessBaseAPI::GetHOCRText(ETEXT_DESC* monitor, int page_number) {
* page_number is 0-based but will appear in the output as 1-based.
*/
char* TessBaseAPI::GetTSVText(int page_number) {
if (tesseract_ == NULL ||
(page_res_ == NULL && Recognize(NULL) < 0))
if (tesseract_ == NULL || (page_res_ == NULL && Recognize(NULL) < 0))
return NULL;
int lcnt = 1, bcnt = 1, pcnt = 1, wcnt = 1;
@ -1629,9 +1626,10 @@ char* TessBaseAPI::GetTSVText(int page_number) {
STRING tsv_str("");
int page_num = page_id, block_num = 0, par_num = 0, line_num = 0, word_num = 0;
int page_num = page_id, block_num = 0, par_num = 0, line_num = 0,
word_num = 0;
tsv_str.add_str_int("1\t", page_num); // level 1 - page
tsv_str.add_str_int("1\t", page_num); // level 1 - page
tsv_str.add_str_int("\t", block_num);
tsv_str.add_str_int("\t", par_num);
tsv_str.add_str_int("\t", line_num);
@ -1642,7 +1640,7 @@ char* TessBaseAPI::GetTSVText(int page_number) {
tsv_str.add_str_int("\t", rect_height_);
tsv_str += "\t-1\t\n";
ResultIterator *res_it = GetIterator();
ResultIterator* res_it = GetIterator();
while (!res_it->Empty(RIL_BLOCK)) {
if (res_it->Empty(RIL_WORD)) {
res_it->Next(RIL_WORD);
@ -1652,46 +1650,46 @@ char* TessBaseAPI::GetTSVText(int page_number) {
// Add rows for any new block/paragraph/textline.
if (res_it->IsAtBeginningOf(RIL_BLOCK)) {
block_num++, par_num = 0, line_num = 0, word_num = 0;
tsv_str.add_str_int("2\t", page_num); // level 2 - block
tsv_str.add_str_int("2\t", page_num); // level 2 - block
tsv_str.add_str_int("\t", block_num);
tsv_str.add_str_int("\t", par_num);
tsv_str.add_str_int("\t", line_num);
tsv_str.add_str_int("\t", word_num);
AddBoxToTSV(res_it, RIL_BLOCK, &tsv_str);
tsv_str += "\t-1\t\n"; // end of row for block
tsv_str += "\t-1\t\n"; // end of row for block
}
if (res_it->IsAtBeginningOf(RIL_PARA)) {
par_num++, line_num = 0, word_num = 0;
tsv_str.add_str_int("3\t", page_num); // level 3 - paragraph
tsv_str.add_str_int("3\t", page_num); // level 3 - paragraph
tsv_str.add_str_int("\t", block_num);
tsv_str.add_str_int("\t", par_num);
tsv_str.add_str_int("\t", line_num);
tsv_str.add_str_int("\t", word_num);
AddBoxToTSV(res_it, RIL_PARA, &tsv_str);
tsv_str += "\t-1\t\n"; // end of row for para
tsv_str += "\t-1\t\n"; // end of row for para
}
if (res_it->IsAtBeginningOf(RIL_TEXTLINE)) {
line_num++, word_num = 0;
tsv_str.add_str_int("4\t", page_num); // level 4 - line
tsv_str.add_str_int("4\t", page_num); // level 4 - line
tsv_str.add_str_int("\t", block_num);
tsv_str.add_str_int("\t", par_num);
tsv_str.add_str_int("\t", line_num);
tsv_str.add_str_int("\t", word_num);
AddBoxToTSV(res_it, RIL_TEXTLINE, &tsv_str);
tsv_str += "\t-1\t\n"; // end of row for line
tsv_str += "\t-1\t\n"; // end of row for line
}
// Now, process the word...
int left, top, right, bottom;
bool bold, italic, underlined, monospace, serif, smallcaps;
int pointsize, font_id;
const char *font_name;
const char* font_name;
res_it->BoundingBox(RIL_WORD, &left, &top, &right, &bottom);
font_name = res_it->WordFontAttributes(&bold, &italic, &underlined,
&monospace, &serif, &smallcaps,
&pointsize, &font_id);
font_name =
res_it->WordFontAttributes(&bold, &italic, &underlined, &monospace,
&serif, &smallcaps, &pointsize, &font_id);
word_num++;
tsv_str.add_str_int("5\t", page_num); // level 5 - word
tsv_str.add_str_int("5\t", page_num); // level 5 - word
tsv_str.add_str_int("\t", block_num);
tsv_str.add_str_int("\t", par_num);
tsv_str.add_str_int("\t", line_num);
@ -1712,11 +1710,11 @@ char* TessBaseAPI::GetTSVText(int page_number) {
tsv_str += res_it->GetUTF8Text(RIL_SYMBOL);
res_it->Next(RIL_SYMBOL);
} while (!res_it->Empty(RIL_BLOCK) && !res_it->IsAtBeginningOf(RIL_WORD));
tsv_str += "\n"; // end of row
tsv_str += "\n"; // end of row
wcnt++;
}
char *ret = new char[tsv_str.length() + 1];
char* ret = new char[tsv_str.length() + 1];
strcpy(ret, tsv_str.string());
delete res_it;
return ret;
@ -1760,7 +1758,7 @@ char* TessBaseAPI::GetBoxText(int page_number) {
int total_length = blob_count * kBytesPerBoxFileLine + utf8_length +
kMaxBytesPerLine;
char* result = new char[total_length];
strcpy(result, "\0");
result[0] = '\0';
int output_length = 0;
LTRResultIterator* it = GetLTRIterator();
do {
@ -1907,17 +1905,17 @@ char* TessBaseAPI::GetUNLVText() {
return result;
}
/**
* The recognized text is returned as a char* which is coded
* as UTF8 and must be freed with the delete [] operator.
* page_number is a 0-based page index that will appear in the osd file.
*/
/**
* The recognized text is returned as a char* which is coded
* as UTF8 and must be freed with the delete [] operator.
* page_number is a 0-based page index that will appear in the osd file.
*/
char* TessBaseAPI::GetOsdText(int page_number) {
OSResults osr;
bool osd = DetectOS(&osr);
if (!osd) {
return NULL;
return NULL;
}
int orient_id = osr.best_result.orientation_id;
@ -1931,19 +1929,19 @@ char* TessBaseAPI::GetOsdText(int page_number) {
int orient_deg = orient_id * 90;
// clockwise rotation needed to make the page upright
int rotate = OrientationIdToValue(orient_id);
int rotate = OrientationIdToValue(orient_id);
char* osd_buf = new char[255];
snprintf(osd_buf, 255,
"Page number: %d\n"
"Orientation in degrees: %d\n"
"Rotate: %d\n"
"Orientation confidence: %.2f\n"
"Script: %s\n"
"Script confidence: %.2f\n",
page_number,
orient_deg, rotate, orient_conf,
script_name, script_conf);
const int kOsdBufsize = 255;
char* osd_buf = new char[kOsdBufsize];
snprintf(osd_buf, kOsdBufsize,
"Page number: %d\n"
"Orientation in degrees: %d\n"
"Rotate: %d\n"
"Orientation confidence: %.2f\n"
"Script: %s\n"
"Script confidence: %.2f\n",
page_number, orient_deg, rotate, orient_conf, script_name,
script_conf);
return osd_buf;
}
@ -2063,7 +2061,7 @@ void TessBaseAPI::Clear() {
if (thresholder_ != NULL)
thresholder_->Clear();
ClearResults();
SetInputImage(NULL);
if (tesseract_ != NULL) SetInputImage(NULL);
}
/**
@ -2767,7 +2765,7 @@ void TessBaseAPI::GetFeaturesForBlob(TBLOB* blob,
INT_FX_RESULT_STRUCT fx_info;
tesseract_->ExtractFeatures(*blob, false, &bl_features,
&cn_features, &fx_info, &outline_counts);
if (cn_features.size() == 0 || cn_features.size() > MAX_NUM_INT_FEATURES) {
if (cn_features.empty() || cn_features.size() > MAX_NUM_INT_FEATURES) {
*num_features = 0;
return; // Feature extraction failed.
}

14
api/baseapi.h
View File

@ -588,8 +588,8 @@ class TESS_API TessBaseAPI {
* data structures.
* page_number is 0-based but will appear in the output as 1-based.
* monitor can be used to
* cancel the recognition
* receive progress callbacks
* cancel the recognition
* receive progress callbacks
*/
char* GetHOCRText(ETEXT_DESC* monitor, int page_number);
@ -750,13 +750,9 @@ class TESS_API TessBaseAPI {
*/
static void NormalizeTBLOB(TBLOB *tblob, ROW *row, bool numeric_mode);
Tesseract* tesseract() const {
return tesseract_;
}
Tesseract* tesseract() const { return tesseract_; }
OcrEngineMode oem() const {
return last_oem_requested_;
}
OcrEngineMode oem() const { return last_oem_requested_; }
void InitTruthCallback(TruthCallback *cb) { truth_cb_ = cb; }
@ -898,7 +894,7 @@ class TESS_API TessBaseAPI {
const char* retry_config, int timeout_millisec,
TessResultRenderer* renderer,
int tessedit_page_number);
// TIFF supports multipage so gets special consideration
// TIFF supports multipage so gets special consideration.
bool ProcessPagesMultipageTiff(const unsigned char *data,
size_t size,
const char* filename,

11
api/pdfrenderer.cpp
View File

@ -20,12 +20,12 @@
#include "config_auto.h"
#endif
#include "allheaders.h"
#include "baseapi.h"
#include "renderer.h"
#include "math.h"
#include "renderer.h"
#include "strngs.h"
#include "tprintf.h"
#include "allheaders.h"
#ifdef _MSC_VER
#include "mathfix.h"
@ -441,8 +441,8 @@ char* TessPDFRenderer::GetPDFTextObjects(TessBaseAPI* api,
int code = unicodes[i];
// Convert to UTF-16BE https://en.wikipedia.org/wiki/UTF-16
if ((code > 0xD7FF && code < 0xE000) || code > 0x10FFFF) {
tprintf("Dropping invalid codepoint %d\n", code);
continue;
tprintf("Dropping invalid codepoint %d\n", code);
continue;
}
if (code < 0x10000) {
snprintf(utf16, sizeof(utf16), "<%04X>", code);
@ -567,7 +567,8 @@ bool TessPDFRenderer::BeginDocumentHandler() {
"<<\n"
" /Length %lu /Filter /FlateDecode\n"
">>\n"
"stream\n", (unsigned long)len);
"stream\n",
(unsigned long)len);
if (n >= sizeof(buf)) {
lept_free(comp);
return false;

29
api/renderer.cpp
View File

@ -155,11 +155,11 @@ TessHOcrRenderer::TessHOcrRenderer(const char *outputbase, bool font_info)
bool TessHOcrRenderer::BeginDocumentHandler() {
AppendString(
"<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n"
"<!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.0 Transitional//EN\"\n"
" \"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd\">\n"
"<html xmlns=\"http://www.w3.org/1999/xhtml\" xml:lang=\"en\" "
"lang=\"en\">\n <head>\n <title>");
"<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n"
"<!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.0 Transitional//EN\"\n"
" \"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd\">\n"
"<html xmlns=\"http://www.w3.org/1999/xhtml\" xml:lang=\"en\" "
"lang=\"en\">\n <head>\n <title>");
AppendString(title());
AppendString(
"</title>\n"
@ -198,25 +198,25 @@ bool TessHOcrRenderer::AddImageHandler(TessBaseAPI* api) {
/**********************************************************************
* TSV Text Renderer interface implementation
**********************************************************************/
TessTsvRenderer::TessTsvRenderer(const char *outputbase)
TessTsvRenderer::TessTsvRenderer(const char* outputbase)
: TessResultRenderer(outputbase, "tsv") {
font_info_ = false;
font_info_ = false;
}
TessTsvRenderer::TessTsvRenderer(const char *outputbase, bool font_info)
TessTsvRenderer::TessTsvRenderer(const char* outputbase, bool font_info)
: TessResultRenderer(outputbase, "tsv") {
font_info_ = font_info;
font_info_ = font_info;
}
bool TessTsvRenderer::BeginDocumentHandler() {
// Output TSV column headings
AppendString("level\tpage_num\tblock_num\tpar_num\tline_num\tword_num\tleft\ttop\twidth\theight\tconf\ttext\n");
AppendString(
"level\tpage_num\tblock_num\tpar_num\tline_num\tword_"
"num\tleft\ttop\twidth\theight\tconf\ttext\n");
return true;
}
bool TessTsvRenderer::EndDocumentHandler() {
return true;
}
bool TessTsvRenderer::EndDocumentHandler() { return true; }
bool TessTsvRenderer::AddImageHandler(TessBaseAPI* api) {
char* tsv = api->GetTSVText(imagenum());
@ -266,8 +266,7 @@ bool TessBoxTextRenderer::AddImageHandler(TessBaseAPI* api) {
* Osd Text Renderer interface implementation
**********************************************************************/
TessOsdRenderer::TessOsdRenderer(const char* outputbase)
: TessResultRenderer(outputbase, "osd") {
}
: TessResultRenderer(outputbase, "osd") {}
bool TessOsdRenderer::AddImageHandler(TessBaseAPI* api) {
char* osd = api->GetOsdText(imagenum());

18
api/renderer.h
View File

@ -153,13 +153,13 @@ class TESS_API TessHOcrRenderer : public TessResultRenderer {
explicit TessHOcrRenderer(const char *outputbase, bool font_info);
explicit TessHOcrRenderer(const char *outputbase);
protected:
protected:
virtual bool BeginDocumentHandler();
virtual bool AddImageHandler(TessBaseAPI* api);
virtual bool EndDocumentHandler();
private:
bool font_info_; // whether to print font information
private:
bool font_info_; // whether to print font information
};
/**
@ -167,15 +167,15 @@ private:
*/
class TESS_API TessTsvRenderer : public TessResultRenderer {
public:
explicit TessTsvRenderer(const char *outputbase, bool font_info);
explicit TessTsvRenderer(const char *outputbase);
explicit TessTsvRenderer(const char* outputbase, bool font_info);
explicit TessTsvRenderer(const char* outputbase);
protected:
protected:
virtual bool BeginDocumentHandler();
virtual bool AddImageHandler(TessBaseAPI* api);
virtual bool EndDocumentHandler();
private:
private:
bool font_info_; // whether to print font information
};
@ -188,12 +188,12 @@ class TESS_API TessPDFRenderer : public TessResultRenderer {
// we load a custom PDF font from this location.
TessPDFRenderer(const char *outputbase, const char *datadir);
protected:
protected:
virtual bool BeginDocumentHandler();
virtual bool AddImageHandler(TessBaseAPI* api);
virtual bool EndDocumentHandler();
private:
private:
// We don't want to have every image in memory at once,
// so we store some metadata as we go along producing
// PDFs one page at a time. At the end that metadata is

160
api/tesseractmain.cpp
View File

@ -40,28 +40,28 @@
static void Win32WarningHandler(const char* module, const char* fmt,
va_list ap) {
if (module != NULL) {
fprintf(stderr, "%s: ", module);
}
fprintf(stderr, "Warning, ");
vfprintf(stderr, fmt, ap);
fprintf(stderr, ".\n");
if (module != NULL) {
fprintf(stderr, "%s: ", module);
}
fprintf(stderr, "Warning, ");
vfprintf(stderr, fmt, ap);
fprintf(stderr, ".\n");
}
#endif /* HAVE_TIFFIO_H && _WIN32 */
void PrintVersionInfo() {
char *versionStrP;
char* versionStrP;
printf("tesseract %s\n", tesseract::TessBaseAPI::Version());
printf("tesseract %s\n", tesseract::TessBaseAPI::Version());
versionStrP = getLeptonicaVersion();
printf(" %s\n", versionStrP);
lept_free(versionStrP);
versionStrP = getLeptonicaVersion();
printf(" %s\n", versionStrP);
lept_free(versionStrP);
versionStrP = getImagelibVersions();
printf(" %s\n", versionStrP);
lept_free(versionStrP);
versionStrP = getImagelibVersions();
printf(" %s\n", versionStrP);
lept_free(versionStrP);
#ifdef USE_OPENCL
cl_platform_id platform;
@ -82,7 +82,7 @@ void PrintVersionInfo() {
printf(" Found %d devices.\n", num_devices);
for (i = 0; i < num_devices; ++i) {
clGetDeviceInfo(devices[i], CL_DEVICE_NAME, 256, info, 0);
printf(" Device %d name: %s.\n", i+1, info);
printf(" Device %d name: %s.\n", i + 1, info);
}
#endif
}
@ -100,27 +100,22 @@ void PrintUsage(const char* program) {
void PrintHelpForPSM() {
const char* msg =
"Page segmentation modes:\n"
" 0 Orientation and script detection (OSD) only.\n"
" 1 Automatic page segmentation with OSD.\n"
" 2 Automatic page segmentation, but no OSD, or OCR.\n"
" 3 Fully automatic page segmentation, but no OSD. (Default)\n"
" 4 Assume a single column of text of variable sizes.\n"
" 5 Assume a single uniform block of vertically aligned text.\n"
" 6 Assume a single uniform block of text.\n"
" 7 Treat the image as a single text line.\n"
" 8 Treat the image as a single word.\n"
" 9 Treat the image as a single word in a circle.\n"
" 10 Treat the image as a single character.\n"
//TODO: Consider publishing these modes.
#if 0
" 11 Sparse text. Find as much text as possible in no"
" particular order.\n"
" 12 Sparse text with OSD.\n"
" 13 Raw line. Treat the image as a single text line,\n"
"\t\t\tbypassing hacks that are Tesseract-specific.\n"
#endif
;
" 0 Orientation and script detection (OSD) only.\n"
" 1 Automatic page segmentation with OSD.\n"
" 2 Automatic page segmentation, but no OSD, or OCR.\n"
" 3 Fully automatic page segmentation, but no OSD. (Default)\n"
" 4 Assume a single column of text of variable sizes.\n"
" 5 Assume a single uniform block of vertically aligned text.\n"
" 6 Assume a single uniform block of text.\n"
" 7 Treat the image as a single text line.\n"
" 8 Treat the image as a single word.\n"
" 9 Treat the image as a single word in a circle.\n"
" 10 Treat the image as a single character.\n"
" 11 Sparse text. Find as much text as possible in no"
" particular order.\n"
" 12 Sparse text with OSD.\n"
" 13 Raw line. Treat the image as a single text line,\n"
"\t\t\tbypassing hacks that are Tesseract-specific.\n";
printf("%s", msg);
}
@ -137,31 +132,30 @@ void PrintHelpMessage(const char* program) {
" -c VAR=VALUE Set value for config variables.\n"
" Multiple -c arguments are allowed.\n"
" -psm NUM Specify page segmentation mode.\n"
"NOTE: These options must occur before any configfile.\n"
;
"NOTE: These options must occur before any configfile.\n";
printf("\n%s\n", ocr_options);
PrintHelpForPSM();
const char *single_options =
const char* single_options =
"Single options:\n"
" -h, --help Show this help message.\n"
" --help-psm Show page segmentation modes.\n"
" -v, --version Show version information.\n"
" --list-langs List available languages for tesseract engine.\n"
" --print-parameters Print tesseract parameters to stdout.\n"
;
" --print-parameters Print tesseract parameters to stdout.\n";
printf("\n%s", single_options);
}
void SetVariablesFromCLArgs(tesseract::TessBaseAPI* api, int argc, char** argv) {
void SetVariablesFromCLArgs(tesseract::TessBaseAPI* api, int argc,
char** argv) {
char opt1[256], opt2[255];
for (int i = 0; i < argc; i++) {
if (strcmp(argv[i], "-c") == 0 && i + 1 < argc) {
strncpy(opt1, argv[i + 1], 255);
opt1[255] = '\0';
char *p = strchr(opt1, '=');
char* p = strchr(opt1, '=');
if (!p) {
fprintf(stderr, "Missing = in configvar assignment\n");
exit(1);
@ -190,8 +184,8 @@ void PrintLangsList(tesseract::TessBaseAPI* api) {
}
void PrintBanner() {
tprintf("Tesseract Open Source OCR Engine v%s with Leptonica\n",
tesseract::TessBaseAPI::Version());
tprintf("Tesseract Open Source OCR Engine v%s with Leptonica\n",
tesseract::TessBaseAPI::Version());
}
/**
@ -209,31 +203,25 @@ void PrintBanner() {
* but that doesn't work.
*/
void FixPageSegMode(tesseract::TessBaseAPI* api,
tesseract::PageSegMode pagesegmode) {
tesseract::PageSegMode pagesegmode) {
if (api->GetPageSegMode() == tesseract::PSM_SINGLE_BLOCK)
api->SetPageSegMode(pagesegmode);
api->SetPageSegMode(pagesegmode);
}
// NOTE: arg_i is used here to avoid ugly *i so many times in this function
void ParseArgs(const int argc, char** argv,
const char** lang,
const char** image,
const char** outputbase,
const char** datapath,
bool* list_langs,
bool* print_parameters,
GenericVector<STRING>* vars_vec,
GenericVector<STRING>* vars_values,
int* arg_i,
tesseract::PageSegMode* pagesegmode) {
void ParseArgs(const int argc, char** argv, const char** lang,
const char** image, const char** outputbase,
const char** datapath, bool* list_langs, bool* print_parameters,
GenericVector<STRING>* vars_vec,
GenericVector<STRING>* vars_values, int* arg_i,
tesseract::PageSegMode* pagesegmode) {
if (argc == 1) {
PrintHelpMessage(argv[0]);
exit(0);
}
if (argc == 2) {
if ((strcmp(argv[1], "-h") == 0) ||
(strcmp(argv[1], "--help") == 0)) {
if ((strcmp(argv[1], "-h") == 0) || (strcmp(argv[1], "--help") == 0)) {
PrintHelpMessage(argv[0]);
exit(0);
}
@ -241,8 +229,7 @@ void ParseArgs(const int argc, char** argv,
PrintHelpForPSM();
exit(0);
}
if ((strcmp(argv[1], "-v") == 0) ||
(strcmp(argv[1], "--version") == 0)) {
if ((strcmp(argv[1], "-v") == 0) || (strcmp(argv[1], "--version") == 0)) {
PrintVersionInfo();
exit(0);
}
@ -298,10 +285,10 @@ void ParseArgs(const int argc, char** argv,
}
}
void PreloadRenderers(tesseract::TessBaseAPI* api,
tesseract::PointerVector<tesseract::TessResultRenderer>* renderers,
tesseract::PageSegMode pagesegmode,
const char* outputbase) {
void PreloadRenderers(
tesseract::TessBaseAPI* api,
tesseract::PointerVector<tesseract::TessResultRenderer>* renderers,
tesseract::PageSegMode pagesegmode, const char* outputbase) {
if (pagesegmode == tesseract::PSM_OSD_ONLY) {
renderers->push_back(new tesseract::TessOsdRenderer(outputbase));
} else {
@ -311,7 +298,7 @@ void PreloadRenderers(tesseract::TessBaseAPI* api,
bool font_info;
api->GetBoolVariable("hocr_font_info", &font_info);
renderers->push_back(
new tesseract::TessHOcrRenderer(outputbase, font_info));
new tesseract::TessHOcrRenderer(outputbase, font_info));
}
api->GetBoolVariable("tessedit_create_tsv", &b);
@ -324,8 +311,8 @@ void PreloadRenderers(tesseract::TessBaseAPI* api,
api->GetBoolVariable("tessedit_create_pdf", &b);
if (b) {
renderers->push_back(new tesseract::TessPDFRenderer(outputbase,
api->GetDatapath()));
renderers->push_back(
new tesseract::TessPDFRenderer(outputbase, api->GetDatapath()));
}
api->GetBoolVariable("tessedit_write_unlv", &b);
@ -359,8 +346,7 @@ void PreloadRenderers(tesseract::TessBaseAPI* api,
*
**********************************************************************/
int main(int argc, char **argv) {
int main(int argc, char** argv) {
const char* lang = "eng";
const char* image = NULL;
const char* outputbase = NULL;
@ -380,10 +366,8 @@ int main(int argc, char **argv) {
TIFFSetWarningHandler(Win32WarningHandler);
#endif /* HAVE_TIFFIO_H && _WIN32 */
ParseArgs(argc, argv,
&lang, &image, &outputbase, &datapath,
&list_langs, &print_parameters,
&vars_vec, &vars_values, &arg_i, &pagesegmode);
ParseArgs(argc, argv, &lang, &image, &outputbase, &datapath, &list_langs,
&print_parameters, &vars_vec, &vars_values, &arg_i, &pagesegmode);
bool banner = false;
if (outputbase != NULL && strcmp(outputbase, "-") &&
@ -396,8 +380,9 @@ int main(int argc, char **argv) {
api.SetOutputName(outputbase);
int init_failed = api.Init(datapath, lang, tesseract::OEM_DEFAULT,
&(argv[arg_i]), argc - arg_i, &vars_vec, &vars_values, false);
int init_failed =
api.Init(datapath, lang, tesseract::OEM_DEFAULT, &(argv[arg_i]),
argc - arg_i, &vars_vec, &vars_values, false);
if (init_failed) {
fprintf(stderr, "Could not initialize tesseract.\n");
exit(1);
@ -406,8 +391,8 @@ int main(int argc, char **argv) {
SetVariablesFromCLArgs(&api, argc, argv);
if (list_langs) {
PrintLangsList(&api);
exit(0);
PrintLangsList(&api);
exit(0);
}
if (print_parameters) {
@ -436,12 +421,13 @@ int main(int argc, char **argv) {
tesseract::TextlineOrder order;
float deskew_angle;
tesseract::PageIterator* it = api.AnalyseLayout();
tesseract::PageIterator* it = api.AnalyseLayout();
if (it) {
it->Orientation(&orientation, &direction, &order, &deskew_angle);
tprintf("Orientation: %d\nWritingDirection: %d\nTextlineOrder: %d\n" \
"Deskew angle: %.4f\n",
orientation, direction, order, deskew_angle);
tprintf(
"Orientation: %d\nWritingDirection: %d\nTextlineOrder: %d\n"
"Deskew angle: %.4f\n",
orientation, direction, order, deskew_angle);
} else {
ret_val = 1;
}
@ -456,14 +442,12 @@ int main(int argc, char **argv) {
// ambigs.train, box.train, box.train.stderr, linebox, rebox
bool b = false;
bool in_training_mode =
(api.GetBoolVariable("tessedit_ambigs_training", &b) && b) ||
(api.GetBoolVariable("tessedit_resegment_from_boxes", &b) && b) ||
(api.GetBoolVariable("tessedit_make_boxes_from_boxes", &b) && b);
(api.GetBoolVariable("tessedit_ambigs_training", &b) && b) ||
(api.GetBoolVariable("tessedit_resegment_from_boxes", &b) && b) ||
(api.GetBoolVariable("tessedit_make_boxes_from_boxes", &b) && b);
tesseract::PointerVector<tesseract::TessResultRenderer> renderers;
if (in_training_mode) {
renderers.push_back(NULL);
} else {

7
ccmain/control.cpp
View File

@ -1,8 +1,8 @@
/******************************************************************
* File: control.cpp (Formerly control.c)
* Description: Module-independent matcher controller.
* Author: Ray Smith
* Created: Thu Apr 23 11:09:58 BST 1992
* Author: Ray Smith
* Created: Thu Apr 23 11:09:58 BST 1992
* ReHacked: Tue Sep 22 08:42:49 BST 1992 Phil Cheatle
*
* (C) Copyright 1992, Hewlett-Packard Ltd.
@ -73,7 +73,6 @@ void Tesseract::recog_pseudo_word(PAGE_RES* page_res,
}
}
/**
* Recognize a single word in interactive mode.
*
@ -539,7 +538,7 @@ void Tesseract::bigram_correction_pass(PAGE_RES *page_res) {
}
}
}
if (overrides_word1.size() >= 1) {
if (!overrides_word1.empty()) {
// Excellent, we have some bigram matches.
if (EqualIgnoringCaseAndTerminalPunct(*w_prev->best_choice,
*overrides_word1[best_idx]) &&

16
ccmain/docqual.cpp
View File

@ -1,8 +1,8 @@
/******************************************************************
* File: docqual.cpp (Formerly docqual.c)
* Description: Document Quality Metrics
* Author: Phil Cheatle
* Created: Mon May 9 11:27:28 BST 1994
* Author: Phil Cheatle
* Created: Mon May 9 11:27:28 BST 1994
*
* (C) Copyright 1994, Hewlett-Packard Ltd.
** Licensed under the Apache License, Version 2.0 (the "License");
@ -98,8 +98,8 @@ void Tesseract::word_char_quality(WERD_RES *word,
ROW *row,
inT16 *match_count,
inT16 *accepted_match_count) {
if (word->bln_boxes == NULL ||
word->rebuild_word == NULL || word->rebuild_word->blobs.empty()) {
if (word->bln_boxes == NULL || word->rebuild_word == NULL ||
word->rebuild_word->blobs.empty()) {
*match_count = 0;
*accepted_match_count = 0;
return;
@ -132,7 +132,7 @@ inT16 Tesseract::count_outline_errs(char c, inT16 outline_count) {
int expected_outline_count;
if (STRING (outlines_odd).contains (c))
return 0; //Don't use this char
return 0; // Don't use this char
else if (STRING (outlines_2).contains (c))
expected_outline_count = 2;
else
@ -151,17 +151,16 @@ void Tesseract::quality_based_rejection(PAGE_RES_IT &page_res_it,
}
}
/*************************************************************************
* unrej_good_quality_words()
* Accept potential rejects in words which pass the following checks:
* - Contains a potential reject
* - Word looks like a sensible alpha word.
* - Word segmentation is the same as the original image
* - All characters have the expected number of outlines
* - All characters have the expected number of outlines
* NOTE - the rejection counts are recalculated after unrejection
* - CAN'T do it in a single pass without a bit of fiddling
* - keep it simple but inefficient
* - keep it simple but inefficient
*************************************************************************/
void Tesseract::unrej_good_quality_words( //unreject potential
PAGE_RES_IT &page_res_it) {
@ -403,7 +402,6 @@ void Tesseract::doc_and_block_rejection( //reject big chunks
} // namespace tesseract
/*************************************************************************
* reject_whole_page()
* Don't believe any of it - set the reject map to 00..00 in all words

5
ccmain/fixspace.cpp
View File

@ -3,8 +3,8 @@
* Description: Implements a pass over the page res, exploring the alternative
* spacing possibilities, trying to use context to improve the
* word spacing
* Author: Phil Cheatle
* Created: Thu Oct 21 11:38:43 BST 1993
* Author: Phil Cheatle
* Created: Thu Oct 21 11:38:43 BST 1993
*
* (C) Copyright 1993, Hewlett-Packard Ltd.
** Licensed under the Apache License, Version 2.0 (the "License");
@ -211,7 +211,6 @@ void Tesseract::match_current_words(WERD_RES_LIST &words, ROW *row,
}
}
/**
* @name eval_word_spacing()
* The basic measure is the number of characters in contextually confirmed

11
ccmain/ltrresultiterator.cpp
View File

@ -145,13 +145,12 @@ float LTRResultIterator::Confidence(PageIteratorLevel level) const {
return 0.0f;
}
void LTRResultIterator::RowAttributes(float* row_height,
float* descenders,
void LTRResultIterator::RowAttributes(float* row_height, float* descenders,
float* ascenders) const {
*row_height = it_->row()->row->x_height() + it_->row()-> row->ascenders()
- it_->row()->row->descenders();
*descenders = it_->row()->row->descenders();
*ascenders = it_->row()->row->ascenders();
*row_height = it_->row()->row->x_height() + it_->row()->row->ascenders() -
it_->row()->row->descenders();
*descenders = it_->row()->row->descenders();
*ascenders = it_->row()->row->ascenders();
}
// Returns the font attributes of the current word. If iterating at a higher

3
ccmain/ltrresultiterator.h
View File

@ -92,8 +92,7 @@ class TESS_API LTRResultIterator : public PageIterator {
float Confidence(PageIteratorLevel level) const;
// Returns the attributes of the current row.
void RowAttributes(float* row_height,
float* descenders,
void RowAttributes(float* row_height, float* descenders,
float* ascenders) const;
// ============= Functions that refer to words only ============.

28
ccmain/output.cpp
View File

@ -1,8 +1,8 @@
/******************************************************************
* File: output.cpp (Formerly output.c)
* Description: Output pass
* Author: Phil Cheatle
* Created: Thu Aug 4 10:56:08 BST 1994
* Author: Phil Cheatle
* Created: Thu Aug 4 10:56:08 BST 1994
*
* (C) Copyright 1994, Hewlett-Packard Ltd.
** Licensed under the Apache License, Version 2.0 (the "License");
@ -78,18 +78,16 @@ void Tesseract::output_pass( //Tess output pass //send to api
while (page_res_it.word () != NULL) {
check_debug_pt (page_res_it.word (), 120);
if (target_word_box)
{
TBOX current_word_box=page_res_it.word ()->word->bounding_box();
FCOORD center_pt((current_word_box.right()+current_word_box.left())/2,(current_word_box.bottom()+current_word_box.top())/2);
if (!target_word_box->contains(center_pt))
{
page_res_it.forward ();
continue;
}
}
if (target_word_box) {
TBOX current_word_box = page_res_it.word()->word->bounding_box();
FCOORD center_pt(
(current_word_box.right() + current_word_box.left()) / 2,
(current_word_box.bottom() + current_word_box.top()) / 2);
if (!target_word_box->contains(center_pt)) {
page_res_it.forward();
continue;
}
}
if (tessedit_write_block_separators &&
block_of_last_word != page_res_it.block ()) {
block_of_last_word = page_res_it.block ();
@ -337,7 +335,7 @@ void Tesseract::set_unlv_suspects(WERD_RES *word_res) {
rating_per_ch = word.rating() / word_res->reject_map.length();
if (rating_per_ch >= suspect_rating_per_ch)
return; //Don't touch bad ratings
return; // Don't touch bad ratings
if ((word_res->tess_accepted) || (rating_per_ch < suspect_accept_rating)) {
/* Unreject any Tess Acceptable word - but NOT tess reject chs*/

7
ccmain/pagesegmain.cpp
View File

@ -412,9 +412,10 @@ ColumnFinder* Tesseract::SetupPageSegAndDetectOrientation(
"Don't rotate.\n", osd_margin);
osd_orientation = 0;
} else {
tprintf("OSD: Weak margin (%.2f) for %d blob text block, "
"but using orientation anyway: %d\n",
osd_margin, osd_blobs.length(), osd_orientation);
tprintf(
"OSD: Weak margin (%.2f) for %d blob text block, "
"but using orientation anyway: %d\n",
osd_margin, osd_blobs.length(), osd_orientation);
}
}
}

12
ccmain/paragraphs.cpp
View File

@ -2052,7 +2052,7 @@ void ConvertHypothesizedModelRunsToParagraphs(
bool single_line_paragraph = false;
SetOfModels models;
rows[start].NonNullHypotheses(&models);
if (models.size() > 0) {
if (!models.empty()) {
model = models[0];
if (rows[start].GetLineType(model) != LT_BODY)
single_line_paragraph = true;
@ -2190,17 +2190,17 @@ void LeftoverSegments(const GenericVector<RowScratchRegisters> &rows,
SetOfModels models_w_crowns;
rows[i].StrongHypotheses(&models);
rows[i].NonNullHypotheses(&models_w_crowns);
if (models.empty() && models_w_crowns.size() > 0) {
if (models.empty() && !models_w_crowns.empty()) {
// Crown paragraph. Is it followed by a modeled line?
for (int end = i + 1; end < rows.size(); end++) {
SetOfModels end_models;
SetOfModels strong_end_models;
rows[end].NonNullHypotheses(&end_models);
rows[end].StrongHypotheses(&strong_end_models);
if (end_models.size() == 0) {
if (end_models.empty()) {
needs_fixing = true;
break;
} else if (strong_end_models.size() > 0) {
} else if (!strong_end_models.empty()) {
needs_fixing = false;
break;
}
@ -2485,7 +2485,7 @@ void InitializeRowInfo(bool after_recognition,
info->ltr = ltr >= rtl;
info->has_leaders = num_leaders > 3;
info->num_words = werds.size();
if (werds.size() > 0) {
if (!werds.empty()) {
WERD_RES *lword = werds[0], *rword = werds[werds.size() - 1];
info->lword_text = lword->best_choice->unichar_string().string();
info->rword_text = rword->best_choice->unichar_string().string();
@ -2538,7 +2538,7 @@ void DetectParagraphs(int debug_level,
// If we're called before text recognition, we might not have
// tight block bounding boxes, so trim by the minimum on each side.
if (row_infos.size() > 0) {
if (!row_infos.empty()) {
int min_lmargin = row_infos[0].pix_ldistance;
int min_rmargin = row_infos[0].pix_rdistance;
for (int i = 1; i < row_infos.size(); i++) {

10
ccmain/paramsd.cpp
View File

@ -329,13 +329,19 @@ void ParamsEditor::WriteParams(char *filename,
fclose(fp);
sprintf (msg_str, "Overwrite file " "%s" "? (Y/N)", filename);
int a = sv_window_->ShowYesNoDialog(msg_str);
if (a == 'n') { return; } // don't write
if (a == 'n') {
return;
} // don't write
}
fp = fopen (filename, "wb"); // can we write to it?
if (fp == NULL) {
sv_window_->AddMessage("Can't write to file " "%s" "", filename);
sv_window_->AddMessage(
"Can't write to file "
"%s"
"",
filename);
return;
}

4
ccmain/reject.cpp
View File

@ -1,8 +1,8 @@
/**********************************************************************
* File: reject.cpp (Formerly reject.c)
* Description: Rejection functions used in tessedit
* Author: Phil Cheatle
* Created: Wed Sep 23 16:50:21 BST 1992
* Author: Phil Cheatle
* Created: Wed Sep 23 16:50:21 BST 1992
*
* (C) Copyright 1992, Hewlett-Packard Ltd.
** Licensed under the Apache License, Version 2.0 (the "License");

2
ccmain/tessedit.cpp
View File

@ -44,7 +44,7 @@
#include "params.h"
#define VARDIR "configs/" /*variables files */
//config under api
// config under api
#define API_CONFIG "configs/api_config"
ETEXT_DESC *global_monitor = NULL; // progress monitor

5
ccmain/tesseractclass.cpp
View File

@ -398,8 +398,8 @@ Tesseract::Tesseract()
"Don't suspect dict wds longer than this", this->params()),
BOOL_MEMBER(suspect_constrain_1Il, false, "UNLV keep 1Il chars rejected",
this->params()),
double_MEMBER(suspect_rating_per_ch, 999.9, "Don't touch bad rating limit",
this->params()),
double_MEMBER(suspect_rating_per_ch, 999.9,
"Don't touch bad rating limit", this->params()),
double_MEMBER(suspect_accept_rating, -999.9, "Accept good rating limit",
this->params()),
BOOL_MEMBER(tessedit_minimal_rejection, false,
@ -512,7 +512,6 @@ Tesseract::Tesseract()
"Page separator (default is form feed control character)",
this->params()),
// The following parameters were deprecated and removed from their
// original
// locations. The parameters are temporarily kept here to give Tesseract

5
ccmain/tesseractclass.h
View File

@ -38,7 +38,6 @@
class BLOB_CHOICE_LIST_CLIST;
class BLOCK_LIST;
class CharSamp;
struct OSResults;
class PAGE_RES;
class PAGE_RES_IT;
@ -98,6 +97,7 @@ namespace tesseract {
class ColumnFinder;
#ifndef NO_CUBE_BUILD
class CharSamp;
class CubeLineObject;
class CubeObject;
class CubeRecoContext;
@ -1010,8 +1010,7 @@ class Tesseract : public Wordrec {
INT_VAR_H(suspect_level, 99, "Suspect marker level");
INT_VAR_H(suspect_space_level, 100,
"Min suspect level for rejecting spaces");
INT_VAR_H(suspect_short_words, 2,
"Don't Suspect dict wds longer than this");
INT_VAR_H(suspect_short_words, 2, "Don't Suspect dict wds longer than this");
BOOL_VAR_H(suspect_constrain_1Il, false, "UNLV keep 1Il chars rejected");
double_VAR_H(suspect_rating_per_ch, 999.9, "Don't touch bad rating limit");
double_VAR_H(suspect_accept_rating, -999.9, "Accept good rating limit");

10
ccstruct/blobbox.cpp
View File

@ -1,8 +1,8 @@
/**********************************************************************
* File: blobbox.cpp (Formerly blobnbox.c)
* Description: Code for the textord blob class.
* Author: Ray Smith
* Created: Thu Jul 30 09:08:51 BST 1992
* Author: Ray Smith
* Created: Thu Jul 30 09:08:51 BST 1992
*
* (C) Copyright 1992, Hewlett-Packard Ltd.
** Licensed under the Apache License, Version 2.0 (the "License");
@ -31,7 +31,9 @@
#define PROJECTION_MARGIN 10 //arbitrary
#define EXTERN
ELISTIZE (BLOBNBOX) ELIST2IZE (TO_ROW) ELISTIZE (TO_BLOCK)
ELISTIZE(BLOBNBOX)
ELIST2IZE(TO_ROW)
ELISTIZE(TO_BLOCK)
// Up to 30 degrees is allowed for rotations of diacritic blobs.
const double kCosSmallAngle = 0.866;
@ -176,7 +178,7 @@ void BLOBNBOX::NeighbourGaps(int gaps[BND_COUNT]) const {
gaps[dir] = MAX_INT16;
BLOBNBOX* neighbour = neighbours_[dir];
if (neighbour != NULL) {
TBOX n_box = neighbour->bounding_box();
const TBOX& n_box = neighbour->bounding_box();
if (dir == BND_LEFT || dir == BND_RIGHT) {
gaps[dir] = box.x_gap(n_box);
} else {

3
ccstruct/boxread.cpp
View File

@ -34,8 +34,7 @@ FILE* OpenBoxFile(const STRING& fname) {
STRING filename = BoxFileName(fname);
FILE* box_file = NULL;
if (!(box_file = fopen(filename.string(), "rb"))) {
CANTOPENFILE.error("read_next_box", TESSEXIT,
"Can't open box file %s",
CANTOPENFILE.error("read_next_box", TESSEXIT, "Can't open box file %s",
filename.string());
}
return box_file;

4
ccstruct/boxword.h
View File

@ -82,9 +82,7 @@ class BoxWord {
const TBOX& bounding_box() const {
return bbox_;
}
int length() const {
return length_;
}
int length() const { return length_; }
const TBOX& BlobBox(int index) const {
return boxes_[index];
}

66
ccstruct/coutln.cpp
View File

@ -48,9 +48,9 @@ ICOORD C_OUTLINE::step_coords[4] = {
* @param length length of loop
*/
C_OUTLINE::C_OUTLINE (CRACKEDGE * startpt, ICOORD bot_left,
ICOORD top_right, inT16 length)
: box (bot_left, top_right), start (startpt->pos), offsets(NULL) {
C_OUTLINE::C_OUTLINE(CRACKEDGE* startpt, ICOORD bot_left, ICOORD top_right,
inT16 length)
: box(bot_left, top_right), start(startpt->pos), offsets(NULL) {
inT16 stepindex; //index to step
CRACKEDGE *edgept; //current point
@ -71,7 +71,6 @@ C_OUTLINE::C_OUTLINE (CRACKEDGE * startpt, ICOORD bot_left,
}
}
/**
* @name C_OUTLINE::C_OUTLINE
*
@ -139,7 +138,7 @@ inT16 length //length of loop
* @param rotation rotate to coord
*/
C_OUTLINE::C_OUTLINE(C_OUTLINE *srcline, FCOORD rotation) : offsets(NULL) {
C_OUTLINE::C_OUTLINE(C_OUTLINE* srcline, FCOORD rotation) : offsets(NULL) {
TBOX new_box; //easy bounding
inT16 stepindex; //index to step
inT16 dirdiff; //direction change
@ -300,7 +299,6 @@ inT32 C_OUTLINE::perimeter() const {
return total_steps;
}
/**
* @name C_OUTLINE::outer_area
*
@ -332,7 +330,6 @@ inT32 C_OUTLINE::outer_area() const {
return total;
}
/**
* @name C_OUTLINE::count_transitions
*
@ -459,7 +456,6 @@ inT32 C_OUTLINE::count_transitions(inT32 threshold) {
return total;
}
/**
* @name C_OUTLINE::operator<
*
@ -468,8 +464,7 @@ inT32 C_OUTLINE::count_transitions(inT32 threshold) {
*/
BOOL8
C_OUTLINE::operator< (const C_OUTLINE & other) const
{
C_OUTLINE::operator<(const C_OUTLINE& other) const {
inT16 count = 0; //winding count
ICOORD pos; //position of point
inT32 stepindex; //index to cstep
@ -495,7 +490,6 @@ C_OUTLINE::operator< (const C_OUTLINE & other) const
return count != 0;
}
/**
* @name C_OUTLINE::winding_number
*
@ -534,7 +528,6 @@ inT16 C_OUTLINE::winding_number(ICOORD point) const {
return count; //winding number
}
/**
* C_OUTLINE::turn_direction
*
@ -563,7 +556,6 @@ inT16 C_OUTLINE::turn_direction() const { //winding number
return count; //winding number
}
/**
* @name C_OUTLINE::reverse
*
@ -586,7 +578,6 @@ void C_OUTLINE::reverse() { //reverse drection
}
}
/**
* @name C_OUTLINE::move
*
@ -661,14 +652,27 @@ static void ComputeGradient(const l_uint32* data, int wpl,
int x, int y, int width, int height,
ICOORD* gradient) {
const l_uint32* line = data + y * wpl;
int pix_x_y = x < width && y < height ?
GET_DATA_BYTE(const_cast<void*> (reinterpret_cast<const void *>(line)), x) : 255;
int pix_x_prevy = x < width && y > 0 ?
GET_DATA_BYTE(const_cast<void*> (reinterpret_cast<const void *>(line - wpl)), x) : 255;
int pix_prevx_prevy = x > 0 && y > 0 ?
GET_DATA_BYTE(const_cast<void*> (reinterpret_cast<void const*>(line - wpl)), x - 1) : 255;
int pix_prevx_y = x > 0 && y < height ?
GET_DATA_BYTE(const_cast<void*> (reinterpret_cast<const void *>(line)), x - 1) : 255;
int pix_x_y =
x < width && y < height
? GET_DATA_BYTE(
const_cast<void*>(reinterpret_cast<const void*>(line)), x)
: 255;
int pix_x_prevy =
x < width && y > 0
? GET_DATA_BYTE(
const_cast<void*>(reinterpret_cast<const void*>(line - wpl)), x)
: 255;
int pix_prevx_prevy =
x > 0 && y > 0
? GET_DATA_BYTE(
const_cast<void*>(reinterpret_cast<void const*>(line - wpl)),
x - 1)
: 255;
int pix_prevx_y =
x > 0 && y < height
? GET_DATA_BYTE(
const_cast<void*>(reinterpret_cast<const void*>(line)), x - 1)
: 255;
gradient->set_x(pix_x_y + pix_x_prevy - (pix_prevx_y + pix_prevx_prevy));
gradient->set_y(pix_x_prevy + pix_prevx_prevy - (pix_x_y + pix_prevx_y));
}
@ -684,8 +688,10 @@ static bool EvaluateVerticalDiff(const l_uint32* data, int wpl, int diff_sign,
if (y <= 0 || y >= height)
return false;
const l_uint32* line = data + y * wpl;
int pixel1 = GET_DATA_BYTE(const_cast<void*> (reinterpret_cast<const void *>(line - wpl)), x);
int pixel2 = GET_DATA_BYTE(const_cast<void*> (reinterpret_cast<const void *>(line)), x);
int pixel1 = GET_DATA_BYTE(
const_cast<void*>(reinterpret_cast<const void*>(line - wpl)), x);
int pixel2 =
GET_DATA_BYTE(const_cast<void*>(reinterpret_cast<const void*>(line)), x);
int diff = (pixel2 - pixel1) * diff_sign;
if (diff > *best_diff) {
*best_diff = diff;
@ -705,8 +711,10 @@ static bool EvaluateHorizontalDiff(const l_uint32* line, int diff_sign,
int* best_diff, int* best_sum, int* best_x) {
if (x <= 0 || x >= width)
return false;
int pixel1 = GET_DATA_BYTE(const_cast<void*> (reinterpret_cast<const void *>(line)), x - 1);
int pixel2 = GET_DATA_BYTE(const_cast<void*> (reinterpret_cast<const void *>(line)), x);
int pixel1 = GET_DATA_BYTE(
const_cast<void*>(reinterpret_cast<const void*>(line)), x - 1);
int pixel2 =
GET_DATA_BYTE(const_cast<void*>(reinterpret_cast<const void*>(line)), x);
int diff = (pixel2 - pixel1) * diff_sign;
if (diff > *best_diff) {
*best_diff = diff;
@ -954,8 +962,7 @@ void C_OUTLINE::render_outline(int left, int top, Pix* pix) const {
*/
#ifndef GRAPHICS_DISABLED
void C_OUTLINE::plot(ScrollView* window,
ScrollView::Color colour) const {
void C_OUTLINE::plot(ScrollView* window, ScrollView::Color colour) const {
inT16 stepindex; // index to cstep
ICOORD pos; // current position
DIR128 stepdir; // direction of step
@ -1016,7 +1023,6 @@ void C_OUTLINE::plot_normed(const DENORM& denorm, ScrollView::Color colour,
}
#endif
/**
* @name C_OUTLINE::operator=
*
@ -1024,7 +1030,7 @@ void C_OUTLINE::plot_normed(const DENORM& denorm, ScrollView::Color colour,
* @param source assign from this
*/
C_OUTLINE & C_OUTLINE::operator= (const C_OUTLINE & source) {
C_OUTLINE& C_OUTLINE::operator=(const C_OUTLINE& source) {
box = source.box;
start = source.start;
if (steps != NULL)

17
ccstruct/mod128.cpp
View File

@ -1,8 +1,8 @@
/**********************************************************************
* File: mod128.c (Formerly dir128.c)
* Description: Code to convert a DIR128 to an ICOORD.
* Author: Ray Smith
* Created: Tue Oct 22 11:56:09 BST 1991
* Author: Ray Smith
* Created: Tue Oct 22 11:56:09 BST 1991
*
* (C) Copyright 1991, Hewlett-Packard Ltd.
** Licensed under the Apache License, Version 2.0 (the "License");
@ -86,16 +86,3 @@ DIR128::DIR128( //from fcoord
while (high - low > 1);
dir = low;
}
/**********************************************************************
* dir_to_gradient
*
* Convert a direction to a vector.
**********************************************************************/
#if 0 // code is buggy for negative dir and unused
ICOORD DIR128::vector() const { //convert to vector
return dirtab[dir]; //easy really
}
#endif

5
ccstruct/mod128.h
View File

@ -1,8 +1,8 @@
/**********************************************************************
* File: mod128.h (Formerly dir128.h)
* Description: Header for class which implements modulo arithmetic.
* Author: Ray Smith
* Created: Tue Mar 26 17:48:13 GMT 1991
* Author: Ray Smith
* Created: Tue Mar 26 17:48:13 GMT 1991
*
* (C) Copyright 1991, Hewlett-Packard Ltd.
** Licensed under the Apache License, Version 2.0 (the "License");
@ -77,7 +77,6 @@ class DLLSYM DIR128
inT8 get_dir() const { //access function
return dir;
}
ICOORD vector() const; //turn to vector
private:
inT8 dir; //a direction

26
ccstruct/otsuthr.cpp
View File

@ -51,23 +51,16 @@ int OtsuThreshold(Pix* src_pix, int left, int top, int width, int height,
// only use opencl if compiled w/ OpenCL and selected device is opencl
#ifdef USE_OPENCL
// all of channel 0 then all of channel 1...
int *histogramAllChannels = new int[kHistogramSize * num_channels];
// all of channel 0 then all of channel 1...
int* histogramAllChannels = new int[kHistogramSize * num_channels];
// Calculate Histogram on GPU
OpenclDevice od;
if (od.selectedDeviceIsOpenCL() &&
(num_channels == 1 || num_channels == 4) && top == 0 && left == 0 ) {
od.HistogramRectOCL(
(const unsigned char*)pixGetData(src_pix),
num_channels,
pixGetWpl(src_pix) * 4,
left,
top,
width,
height,
kHistogramSize,
histogramAllChannels);
// Calculate Histogram on GPU
OpenclDevice od;
if (od.selectedDeviceIsOpenCL() && (num_channels == 1 || num_channels == 4) &&
top == 0 && left == 0) {
od.HistogramRectOCL((const unsigned char*)pixGetData(src_pix), num_channels,
pixGetWpl(src_pix) * 4, left, top, width, height,
kHistogramSize, histogramAllChannels);
// Calculate Threshold from Histogram on cpu
for (int ch = 0; ch < num_channels; ++ch) {
@ -143,7 +136,6 @@ int OtsuThreshold(Pix* src_pix, int left, int top, int width, int height,
delete[] histogramAllChannels;
#endif // USE_OPENCL
if (!any_good_hivalue) {
// Use the best of the ones that were not good enough.
(*hi_values)[best_hi_index] = best_hi_value;

6
ccstruct/pageres.h
View File

@ -1,7 +1,7 @@
/**********************************************************************
* File: pageres.h (Formerly page_res.h)
* Description: Results classes used by control.c
* Author: Phil Cheatle
* Author: Phil Cheatle
* Created: Tue Sep 22 08:42:49 BST 1992
*
* (C) Copyright 1992, Hewlett-Packard Ltd.
@ -327,7 +327,7 @@ class WERD_RES : public ELIST_LINK {
}
// Deep copies everything except the ratings MATRIX.
// To get that use deep_copy below.
WERD_RES(const WERD_RES &source) : ELIST_LINK(source) {
WERD_RES(const WERD_RES& source) : ELIST_LINK(source) {
InitPointers();
*this = source; // see operator=
}
@ -630,7 +630,7 @@ class WERD_RES : public ELIST_LINK {
static WERD_RES* deep_copy(const WERD_RES* src) {
WERD_RES* result = new WERD_RES(*src);
// That didn't copy the ratings, but we want a copy if there is one to
// begin width.
// begin with.
if (src->ratings != NULL)
result->ratings = src->ratings->DeepCopy();
return result;

130
ccstruct/pdblock.h
View File

@ -29,90 +29,76 @@ struct Pix;
CLISTIZEH (PDBLK)
///page block
class PDBLK
{
class PDBLK {
friend class BLOCK_RECT_IT; //< block iterator
public:
///empty constructor
PDBLK() {
hand_poly = NULL;
index_ = 0;
}
///simple constructor
PDBLK(inT16 xmin, //< bottom left
inT16 ymin,
inT16 xmax, //< top right
inT16 ymax);
public:
/// empty constructor
PDBLK() {
hand_poly = NULL;
index_ = 0;
}
/// simple constructor
PDBLK(inT16 xmin, //< bottom left
inT16 ymin,
inT16 xmax, //< top right
inT16 ymax);
///set vertex lists
///@param left list of left vertices
///@param right list of right vertices
void set_sides(ICOORDELT_LIST *left,
ICOORDELT_LIST *right);
/// set vertex lists
///@param left list of left vertices
///@param right list of right vertices
void set_sides(ICOORDELT_LIST *left, ICOORDELT_LIST *right);
///destructor
~PDBLK () {
if (hand_poly) delete hand_poly;
}
/// destructor
~PDBLK() {
if (hand_poly) delete hand_poly;
}
POLY_BLOCK *poly_block() const {
return hand_poly;
}
///set the poly block
void set_poly_block(POLY_BLOCK *blk) {
hand_poly = blk;
}
///get box
void bounding_box(ICOORD &bottom_left, //bottom left
ICOORD &top_right) const { //topright
bottom_left = box.botleft ();
top_right = box.topright ();
}
///get real box
const TBOX &bounding_box() const {
return box;
}
POLY_BLOCK *poly_block() const { return hand_poly; }
/// set the poly block
void set_poly_block(POLY_BLOCK *blk) { hand_poly = blk; }
/// get box
void bounding_box(ICOORD &bottom_left, // bottom left
ICOORD &top_right) const { // topright
bottom_left = box.botleft();
top_right = box.topright();
}
/// get real box
const TBOX &bounding_box() const { return box; }
int index() const {
return index_;
}
void set_index(int value) {
index_ = value;
}
int index() const { return index_; }
void set_index(int value) { index_ = value; }
///is pt inside block
BOOL8 contains(ICOORD pt);
/// is pt inside block
BOOL8 contains(ICOORD pt);
/// reposition block
void move(const ICOORD vec); // by vector
/// reposition block
void move(const ICOORD vec); // by vector
// Returns a binary Pix mask with a 1 pixel for every pixel within the
// block. Rotates the coordinate system by rerotation prior to rendering.
// If not NULL, mask_box is filled with the position box of the returned
// mask image.
Pix *render_mask(const FCOORD &rerotation, TBOX *mask_box);
// Returns a binary Pix mask with a 1 pixel for every pixel within the
// block. Rotates the coordinate system by rerotation prior to rendering.
// If not NULL, mask_box is filled with the position box of the returned
// mask image.
Pix *render_mask(const FCOORD &rerotation, TBOX *mask_box);
#ifndef GRAPHICS_DISABLED
///draw histogram
///@param window window to draw in
///@param serial serial number
///@param colour colour to draw in
void plot(ScrollView* window,
inT32 serial,
ScrollView::Color colour);
#endif // GRAPHICS_DISABLED
#ifndef GRAPHICS_DISABLED
/// draw histogram
///@param window window to draw in
///@param serial serial number
///@param colour colour to draw in
void plot(ScrollView *window, inT32 serial, ScrollView::Color colour);
#endif // GRAPHICS_DISABLED
///assignment
///@param source from this
PDBLK & operator= (const PDBLK & source);
/// assignment
///@param source from this
PDBLK &operator=(const PDBLK &source);
protected:
POLY_BLOCK *hand_poly; //< weird as well
ICOORDELT_LIST leftside; //< left side vertices
ICOORDELT_LIST rightside; //< right side vertices
TBOX box; //< bounding box
int index_; //< Serial number of this block.
protected:
POLY_BLOCK *hand_poly; //< weird as well
ICOORDELT_LIST leftside; //< left side vertices
ICOORDELT_LIST rightside; //< right side vertices
TBOX box; //< bounding box
int index_; //< Serial number of this block.
};
class DLLSYM BLOCK_RECT_IT //rectangle iterator

2
ccstruct/polyaprx.cpp
View File

@ -214,7 +214,7 @@ EDGEPT edgepts[] //output is array
void fix2( //polygonal approx
EDGEPT *start, /*loop to approimate */
int area) {
EDGEPT *edgept; /*current point */
EDGEPT *edgept; /*current point */
EDGEPT *edgept1;
EDGEPT *loopstart; /*modified start of loop */
EDGEPT *linestart; /*start of line segment */

2
ccstruct/polyblk.cpp
View File

@ -1,7 +1,7 @@
/**********************************************************************
* File: polyblk.c (Formerly poly_block.c)
* Description: Polygonal blocks
* Author: Sheelagh Lloyd?
* Author: Sheelagh Lloyd?
* Created:
*
* (C) Copyright 1993, Hewlett-Packard Ltd.

4
ccstruct/quspline.cpp
View File

@ -1,8 +1,8 @@
/**********************************************************************
* File: quspline.cpp (Formerly qspline.c)
* Description: Code for the QSPLINE class.
* Author: Ray Smith
* Created: Tue Oct 08 17:16:12 BST 1991
* Author: Ray Smith
* Created: Tue Oct 08 17:16:12 BST 1991
*
* (C) Copyright 1991, Hewlett-Packard Ltd.
** Licensed under the Apache License, Version 2.0 (the "License");

3
ccstruct/ratngs.h
View File

@ -288,7 +288,8 @@ class WERD_CHOICE : public ELIST_LINK {
src_certainty, src_permuter);
}
WERD_CHOICE(const char *src_string, const UNICHARSET &unicharset);
WERD_CHOICE(const WERD_CHOICE &word) : ELIST_LINK(word), unicharset_(word.unicharset_) {
WERD_CHOICE(const WERD_CHOICE &word)
: ELIST_LINK(word), unicharset_(word.unicharset_) {
this->init(word.length());
this->operator=(word);
}

12
ccstruct/rect.cpp
View File

@ -1,8 +1,8 @@
/**********************************************************************
* File: rect.c (Formerly box.c)
* Description: Bounding box class definition.
* Author: Phil Cheatle
* Created: Wed Oct 16 15:18:45 BST 1991
* Author: Phil Cheatle
* Created: Wed Oct 16 15:18:45 BST 1991
*
* (C) Copyright 1991, Hewlett-Packard Ltd.
** Licensed under the Apache License, Version 2.0 (the "License");
@ -29,10 +29,10 @@
*
**********************************************************************/
TBOX::TBOX( //constructor
const ICOORD pt1, //one corner
const ICOORD pt2 //the other corner
) {
TBOX::TBOX( // constructor
const ICOORD pt1, // one corner
const ICOORD pt2 // the other corner
) {
if (pt1.x () <= pt2.x ()) {
if (pt1.y () <= pt2.y ()) {
bot_left = pt1;

10
ccstruct/rect.h
View File

@ -1,8 +1,8 @@
/**********************************************************************
* File: rect.h (Formerly box.h)
* Description: Bounding box class definition.
* Author: Phil Cheatle
* Created: Wed Oct 16 15:18:45 BST 1991
* Author: Phil Cheatle
* Created: Wed Oct 16 15:18:45 BST 1991
*
* (C) Copyright 1991, Hewlett-Packard Ltd.
** Licensed under the Apache License, Version 2.0 (the "License");
@ -307,9 +307,9 @@ class DLLSYM TBOX { // bounding box
*
**********************************************************************/
inline TBOX::TBOX( // constructor
const FCOORD pt // floating centre
) {
inline TBOX::TBOX( // constructor
const FCOORD pt // floating centre
) {
bot_left = ICOORD ((inT16) floor (pt.x ()), (inT16) floor (pt.y ()));
top_right = ICOORD ((inT16) ceil (pt.x ()), (inT16) ceil (pt.y ()));
}

61
ccstruct/rejctmap.h
View File

@ -1,8 +1,8 @@
/**********************************************************************
* File: rejctmap.h (Formerly rejmap.h)
* Description: REJ and REJMAP class functions.
* Author: Phil Cheatle
* Created: Thu Jun 9 13:46:38 BST 1994
* Author: Phil Cheatle
* Created: Thu Jun 9 13:46:38 BST 1994
*
* (C) Copyright 1994, Hewlett-Packard Ltd.
** Licensed under the Apache License, Version 2.0 (the "License");
@ -48,46 +48,45 @@ OF THIS IMPLIED TEMPORAL ORDERING OF THE FLAGS!!!!
#include "bits16.h"
#include "params.h"
enum REJ_FLAGS
{
enum REJ_FLAGS {
/* Reject modes which are NEVER overridden */
R_TESS_FAILURE, // PERM Tess didn't classify
R_SMALL_XHT, // PERM Xht too small
R_EDGE_CHAR, // PERM Too close to edge of image
R_1IL_CONFLICT, // PERM 1Il confusion
R_POSTNN_1IL, // PERM 1Il unrejected by NN
R_REJ_CBLOB, // PERM Odd blob
R_MM_REJECT, // PERM Matrix match rejection (m's)
R_BAD_REPETITION, // TEMP Repeated char which doesn't match trend
R_TESS_FAILURE, // PERM Tess didn't classify
R_SMALL_XHT, // PERM Xht too small
R_EDGE_CHAR, // PERM Too close to edge of image
R_1IL_CONFLICT, // PERM 1Il confusion
R_POSTNN_1IL, // PERM 1Il unrejected by NN
R_REJ_CBLOB, // PERM Odd blob
R_MM_REJECT, // PERM Matrix match rejection (m's)
R_BAD_REPETITION, // TEMP Repeated char which doesn't match trend
/* Initial reject modes (pre NN_ACCEPT) */
R_POOR_MATCH, // TEMP Ray's original heuristic (Not used)
R_NOT_TESS_ACCEPTED, // TEMP Tess didn't accept WERD
R_CONTAINS_BLANKS, // TEMP Tess failed on other chs in WERD
R_BAD_PERMUTER, // POTENTIAL Bad permuter for WERD
R_POOR_MATCH, // TEMP Ray's original heuristic (Not used)
R_NOT_TESS_ACCEPTED, // TEMP Tess didn't accept WERD
R_CONTAINS_BLANKS, // TEMP Tess failed on other chs in WERD
R_BAD_PERMUTER, // POTENTIAL Bad permuter for WERD
/* Reject modes generated after NN_ACCEPT but before MM_ACCEPT */
R_HYPHEN, // TEMP Post NN dodgy hyphen or full stop
R_DUBIOUS, // TEMP Post NN dodgy chars
R_NO_ALPHANUMS, // TEMP No alphanumerics in word after NN
R_MOSTLY_REJ, // TEMP Most of word rejected so rej the rest
R_XHT_FIXUP, // TEMP Xht tests unsure
R_HYPHEN, // TEMP Post NN dodgy hyphen or full stop
R_DUBIOUS, // TEMP Post NN dodgy chars
R_NO_ALPHANUMS, // TEMP No alphanumerics in word after NN
R_MOSTLY_REJ, // TEMP Most of word rejected so rej the rest
R_XHT_FIXUP, // TEMP Xht tests unsure
/* Reject modes generated after MM_ACCEPT but before QUALITY_ACCEPT */
R_BAD_QUALITY, // TEMP Quality metrics bad for WERD
R_BAD_QUALITY, // TEMP Quality metrics bad for WERD
/* Reject modes generated after QUALITY_ACCEPT but before MINIMAL_REJ accep*/
R_DOC_REJ, // TEMP Document rejection
R_BLOCK_REJ, // TEMP Block rejection
R_ROW_REJ, // TEMP Row rejection
R_UNLV_REJ, // TEMP ~ turned to - or ^ turned to space
R_DOC_REJ, // TEMP Document rejection
R_BLOCK_REJ, // TEMP Block rejection
R_ROW_REJ, // TEMP Row rejection
R_UNLV_REJ, // TEMP ~ turned to - or ^ turned to space
/* Accept modes which occur between the above rejection groups */
R_NN_ACCEPT, //NN acceptance
R_HYPHEN_ACCEPT, //Hyphen acceptance
R_MM_ACCEPT, //Matrix match acceptance
R_QUALITY_ACCEPT, //Accept word in good quality doc
R_MINIMAL_REJ_ACCEPT //Accept EVERYTHING except tess failures
R_NN_ACCEPT, // NN acceptance
R_HYPHEN_ACCEPT, // Hyphen acceptance
R_MM_ACCEPT, // Matrix match acceptance
R_QUALITY_ACCEPT, // Accept word in good quality doc
R_MINIMAL_REJ_ACCEPT // Accept EVERYTHING except tess failures
};
/* REJECT MAP VALUES */

5
ccstruct/statistc.cpp
View File

@ -1,8 +1,8 @@
/**********************************************************************
* File: statistc.c (Formerly stats.c)
* Description: Simple statistical package for integer values.
* Author: Ray Smith
* Created: Mon Feb 04 16:56:05 GMT 1991
* Author: Ray Smith
* Created: Mon Feb 04 16:56:05 GMT 1991
*
* (C) Copyright 1991, Hewlett-Packard Ltd.
** Licensed under the Apache License, Version 2.0 (the "License");
@ -215,7 +215,6 @@ inT32 STATS::min_bucket() const { // Find min
return rangemin_ + min;
}
/**********************************************************************
* STATS::max_bucket
*

29
ccutil/clst.cpp
View File

@ -26,7 +26,7 @@
**********************************************************************/
/***********************************************************************
* CLIST::internal_deep_clear
* CLIST::internal_deep_clear
*
* Used by the "deep_clear" member function of derived list
* classes to destroy all the elements on the list.
@ -56,9 +56,8 @@ void (*zapper) (void *)) { //ptr to zapper functn
}
}
/***********************************************************************
* CLIST::shallow_clear
* CLIST::shallow_clear
*
* Used by the destructor and the "shallow_clear" member function of derived
* list classes to destroy the list.
@ -83,7 +82,7 @@ void CLIST::shallow_clear() { //destroy all links
}
/***********************************************************************
* CLIST::assign_to_sublist
* CLIST::assign_to_sublist
*
* The list is set to a sublist of another list. "This" list must be empty
* before this function is invoked. The two iterators passed must refer to
@ -107,9 +106,8 @@ void CLIST::assign_to_sublist( //to this list
last = start_it->extract_sublist (end_it);
}
/***********************************************************************
* CLIST::length
* CLIST::length
*
* Return count of elements on list
**********************************************************************/
@ -123,9 +121,8 @@ inT32 CLIST::length() const { //count elements
return count;
}
/***********************************************************************
* CLIST::sort
* CLIST::sort
*
* Sort elements on list
**********************************************************************/
@ -239,7 +236,7 @@ void CLIST::set_subtract(int comparator(const void*, const void*),
**********************************************************************/
/***********************************************************************
* CLIST_ITERATOR::forward
* CLIST_ITERATOR::forward
*
* Move the iterator to the next element of the list.
* REMEMBER: ALL LISTS ARE CIRCULAR.
@ -276,9 +273,8 @@ void *CLIST_ITERATOR::forward() {
return current->data;
}
/***********************************************************************
* CLIST_ITERATOR::data_relative
* CLIST_ITERATOR::data_relative
*
* Return the data pointer to the element "offset" elements from current.
* "offset" must not be less than -1.
@ -312,9 +308,8 @@ void *CLIST_ITERATOR::data_relative( //get data + or - ...
return ptr->data;
}
/***********************************************************************
* CLIST_ITERATOR::move_to_last()
* CLIST_ITERATOR::move_to_last()
*
* Move current so that it is set to the end of the list.
* Return data just in case anyone wants it.
@ -336,9 +331,8 @@ void *CLIST_ITERATOR::move_to_last() {
return current->data;
}
/***********************************************************************
* CLIST_ITERATOR::exchange()
* CLIST_ITERATOR::exchange()
*
* Given another iterator, whose current element is a different element on
* the same list list OR an element of another list, exchange the two current
@ -434,9 +428,8 @@ void CLIST_ITERATOR::exchange( //positions of 2 link
other_it->current = old_current;
}
/***********************************************************************
* CLIST_ITERATOR::extract_sublist()
* CLIST_ITERATOR::extract_sublist()
*
* This is a private member, used only by CLIST::assign_to_sublist.
* Given another iterator for the same list, extract the links from THIS to
@ -478,7 +471,7 @@ CLIST_LINK *CLIST_ITERATOR::extract_sublist( //from
temp_it.mark_cycle_pt ();
do { //walk sublist
if (temp_it.cycled_list ()) //can't find end pt
if (temp_it.cycled_list()) // can't find end pt
BAD_SUBLIST.error ("CLIST_ITERATOR.extract_sublist", ABORT, NULL);
if (temp_it.at_last ()) {

284
ccutil/clst.h
View File

@ -28,9 +28,9 @@
class CLIST_ITERATOR;
/**********************************************************************
* CLASS - CLIST_LINK
* CLASS - CLIST_LINK
*
* Generic link class for singly linked CONS cell lists
* Generic link class for singly linked CONS cell lists
*
* Note: No destructor - elements are assumed to be destroyed EITHER after
* they have been extracted from a list OR by the CLIST destructor which
@ -50,13 +50,13 @@ class DLLSYM CLIST_LINK
data = next = NULL;
}
CLIST_LINK( //copy constructor
const CLIST_LINK &) { //don't copy link
CLIST_LINK( // copy constructor
const CLIST_LINK &) { // don't copy link
data = next = NULL;
}
void operator= ( //don't copy links
const CLIST_LINK &) {
void operator=( // don't copy links
const CLIST_LINK &) {
data = next = NULL;
}
};
@ -89,8 +89,8 @@ class DLLSYM CLIST
void internal_deep_clear ( //destroy all links
void (*zapper) (void *)); //ptr to zapper functn
void shallow_clear(); //clear list but don't
//delete data elements
void shallow_clear(); // clear list but don't
// delete data elements
bool empty() const { //is list empty?
return !last;
@ -136,9 +136,10 @@ class DLLSYM CLIST
};
/***********************************************************************
* CLASS - CLIST_ITERATOR
* CLASS - CLIST_ITERATOR
*
* Generic iterator class for singly linked lists with embedded links
* Generic iterator class for singly linked lists with embedded
*links
**********************************************************************/
class DLLSYM CLIST_ITERATOR
@ -231,8 +232,8 @@ class DLLSYM CLIST_ITERATOR
BOOL8 cycled_list(); //Completed a cycle?
void add_to_end( //add at end &
void *new_data); //don't move
void add_to_end( // add at end &
void *new_data); // don't move
void exchange( //positions of 2 links
CLIST_ITERATOR *other_it); //other iterator
@ -246,7 +247,7 @@ class DLLSYM CLIST_ITERATOR
};
/***********************************************************************
* CLIST_ITERATOR::set_to_list
* CLIST_ITERATOR::set_to_list
*
* (Re-)initialise the iterator to point to the start of the list_to_iterate
* over.
@ -270,9 +271,8 @@ inline void CLIST_ITERATOR::set_to_list( //change list
ex_current_was_cycle_pt = FALSE;
}
/***********************************************************************
* CLIST_ITERATOR::CLIST_ITERATOR
* CLIST_ITERATOR::CLIST_ITERATOR
*
* CONSTRUCTOR - set iterator to specified list;
**********************************************************************/
@ -281,9 +281,8 @@ inline CLIST_ITERATOR::CLIST_ITERATOR(CLIST *list_to_iterate) {
set_to_list(list_to_iterate);
}
/***********************************************************************
* CLIST_ITERATOR::add_after_then_move
* CLIST_ITERATOR::add_after_then_move
*
* Add a new element to the list after the current element and move the
* iterator to the new element.
@ -329,9 +328,8 @@ inline void CLIST_ITERATOR::add_after_then_move( // element to add
current = new_element;
}
/***********************************************************************
* CLIST_ITERATOR::add_after_stay_put
* CLIST_ITERATOR::add_after_stay_put
*
* Add a new element to the list after the current element but do not move
* the iterator to the new element.
@ -380,9 +378,8 @@ inline void CLIST_ITERATOR::add_after_stay_put( // element to add
}
}
/***********************************************************************
* CLIST_ITERATOR::add_before_then_move
* CLIST_ITERATOR::add_before_then_move
*
* Add a new element to the list before the current element and move the
* iterator to the new element.
@ -425,9 +422,8 @@ inline void CLIST_ITERATOR::add_before_then_move( // element to add
current = new_element;
}
/***********************************************************************
* CLIST_ITERATOR::add_before_stay_put
* CLIST_ITERATOR::add_before_stay_put
*
* Add a new element to the list before the current element but don't move the
* iterator to the new element.
@ -471,11 +467,11 @@ inline void CLIST_ITERATOR::add_before_stay_put( // element to add
}
}
/***********************************************************************
* CLIST_ITERATOR::add_list_after
* CLIST_ITERATOR::add_list_after
*
* Insert another list to this list after the current element but don't move the
* Insert another list to this list after the current element but don't move
*the
* iterator.
**********************************************************************/
@ -518,9 +514,8 @@ inline void CLIST_ITERATOR::add_list_after(CLIST *list_to_add) {
}
}
/***********************************************************************
* CLIST_ITERATOR::add_list_before
* CLIST_ITERATOR::add_list_before
*
* Insert another list to this list before the current element. Move the
* iterator to the start of the inserted elements
@ -563,9 +558,8 @@ inline void CLIST_ITERATOR::add_list_before(CLIST *list_to_add) {
}
}
/***********************************************************************
* CLIST_ITERATOR::extract
* CLIST_ITERATOR::extract
*
* Do extraction by removing current from the list, deleting the cons cell
* and returning the data to the caller, but NOT updating the iterator. (So
@ -606,9 +600,8 @@ inline void *CLIST_ITERATOR::extract() {
return extracted_data;
}
/***********************************************************************
* CLIST_ITERATOR::move_to_first()
* CLIST_ITERATOR::move_to_first()
*
* Move current so that it is set to the start of the list.
* Return data just in case anyone wants it.
@ -626,9 +619,8 @@ inline void *CLIST_ITERATOR::move_to_first() {
return current != NULL ? current->data : NULL;
}
/***********************************************************************
* CLIST_ITERATOR::mark_cycle_pt()
* CLIST_ITERATOR::mark_cycle_pt()
*
* Remember the current location so that we can tell whether we've returned
* to this point later.
@ -651,9 +643,8 @@ inline void CLIST_ITERATOR::mark_cycle_pt() {
started_cycling = FALSE;
}
/***********************************************************************
* CLIST_ITERATOR::at_first()
* CLIST_ITERATOR::at_first()
*
* Are we at the start of the list?
*
@ -671,9 +662,8 @@ inline BOOL8 CLIST_ITERATOR::at_first() {
!ex_current_was_last)); //first and last
}
/***********************************************************************
* CLIST_ITERATOR::at_last()
* CLIST_ITERATOR::at_last()
*
* Are we at the end of the list?
*
@ -691,9 +681,8 @@ inline BOOL8 CLIST_ITERATOR::at_last() {
ex_current_was_last)); //first and last
}
/***********************************************************************
* CLIST_ITERATOR::cycled_list()
* CLIST_ITERATOR::cycled_list()
*
* Have we returned to the cycle_pt since it was set?
*
@ -709,9 +698,8 @@ inline BOOL8 CLIST_ITERATOR::cycled_list() {
}
/***********************************************************************
* CLIST_ITERATOR::length()
* CLIST_ITERATOR::length()
*
* Return the length of the list
*
@ -726,9 +714,8 @@ inline inT32 CLIST_ITERATOR::length() {
return list->length ();
}
/***********************************************************************
* CLIST_ITERATOR::sort()
* CLIST_ITERATOR::sort()
*
* Sort the elements of the list, then reposition at the start.
*
@ -747,9 +734,8 @@ const void *, const void *)) {
move_to_first();
}
/***********************************************************************
* CLIST_ITERATOR::add_to_end
* CLIST_ITERATOR::add_to_end
*
* Add a new element to the end of the list without moving the iterator.
* This is provided because a single linked list cannot move to the last as
@ -811,7 +797,7 @@ The macro generates:
- An element deletion function: CLASSNAME##_c1_zapper
- An element copier function:
CLASSNAME##_c1_copier
- A CLIST subclass: CLASSNAME##_CLIST
- A CLIST subclass: CLASSNAME##_CLIST
- A CLIST_ITERATOR subclass:
CLASSNAME##_C_IT
@ -830,114 +816,116 @@ CLISTIZEH is a concatenation of 3 fragments CLISTIZEH_A, CLISTIZEH_B and
CLISTIZEH_C.
***********************************************************************/
#define CLISTIZEH_A( CLASSNAME ) \
\
extern DLLSYM void CLASSNAME##_c1_zapper( /*delete a link*/ \
void* link); /*link to delete*/ \
\
extern DLLSYM void* CLASSNAME##_c1_copier( /*deep copy a link*/ \
void* old_element); /*source link */
#define CLISTIZEH_A(CLASSNAME) \
\
extern DLLSYM void CLASSNAME##_c1_zapper( /*delete a link*/ \
void *link); /*link to delete*/ \
\
extern DLLSYM void \
*CLASSNAME##_c1_copier( /*deep copy a link*/ \
void *old_element); /*source link */
#define CLISTIZEH_B( CLASSNAME ) \
\
/*********************************************************************** \
* CLASS - CLASSNAME##_CLIST \
* \
* List class for class CLASSNAME \
* \
**********************************************************************/ \
\
class DLLSYM CLASSNAME##_CLIST : public CLIST \
{ \
public: \
CLASSNAME##_CLIST():CLIST() {} \
/* constructor */ \
\
CLASSNAME##_CLIST( /* don't construct */ \
const CLASSNAME##_CLIST&) /*by initial assign*/ \
{ DONT_CONSTRUCT_LIST_BY_COPY.error( QUOTE_IT( CLASSNAME##_CLIST ), \
ABORT, NULL ); } \
\
void deep_clear() /* delete elements */ \
{ CLIST::internal_deep_clear( &CLASSNAME##_c1_zapper ); } \
\
void operator=( /* prevent assign */ \
const CLASSNAME##_CLIST&) \
{ DONT_ASSIGN_LISTS.error( QUOTE_IT( CLASSNAME##_CLIST ), \
ABORT, NULL ); }
#define CLISTIZEH_B(CLASSNAME) \
\
/*********************************************************************** \
* CLASS - \
*CLASSNAME##_CLIST \
* \
* List class for class \
*CLASSNAME \
* \
**********************************************************************/ \
\
class DLLSYM CLASSNAME##_CLIST : public CLIST { \
public: \
CLASSNAME##_CLIST() : CLIST() {} \
/* constructor */ \
\
CLASSNAME##_CLIST( /* don't construct */ \
const CLASSNAME##_CLIST &) /*by initial assign*/ \
{ \
DONT_CONSTRUCT_LIST_BY_COPY.error(QUOTE_IT(CLASSNAME##_CLIST), ABORT, \
NULL); \
} \
\
void deep_clear() /* delete elements */ \
{ \
CLIST::internal_deep_clear(&CLASSNAME##_c1_zapper); \
} \
\
void operator=(/* prevent assign */ \
const CLASSNAME##_CLIST &) { \
DONT_ASSIGN_LISTS.error(QUOTE_IT(CLASSNAME##_CLIST), ABORT, NULL); \
}
#define CLISTIZEH_C( CLASSNAME ) \
\
}; \
\
\
\
/*********************************************************************** \
* CLASS - CLASSNAME##_C_IT \
* \
* Iterator class for class CLASSNAME##_CLIST \
* \
* Note: We don't need to coerce pointers to member functions input \
* parameters as these are automatically converted to the type of the base \
* type. ("A ptr to a class may be converted to a pointer to a public base \
* class of that class") \
**********************************************************************/ \
\
class DLLSYM CLASSNAME##_C_IT : public CLIST_ITERATOR \
{ \
public: \
CLASSNAME##_C_IT():CLIST_ITERATOR(){} \
\
CLASSNAME##_C_IT( \
CLASSNAME##_CLIST* list):CLIST_ITERATOR(list){} \
\
CLASSNAME* data() \
{ return (CLASSNAME*) CLIST_ITERATOR::data(); } \
\
CLASSNAME* data_relative( \
inT8 offset) \
{ return (CLASSNAME*) CLIST_ITERATOR::data_relative( offset ); } \
\
CLASSNAME* forward() \
{ return (CLASSNAME*) CLIST_ITERATOR::forward(); } \
\
CLASSNAME* extract() \
{ return (CLASSNAME*) CLIST_ITERATOR::extract(); } \
\
CLASSNAME* move_to_first() \
{ return (CLASSNAME*) CLIST_ITERATOR::move_to_first(); } \
\
CLASSNAME* move_to_last() \
{ return (CLASSNAME*) CLIST_ITERATOR::move_to_last(); } \
};
#define CLISTIZEH_C(CLASSNAME) \
} \
; \
\
/*********************************************************************** \
* CLASS - CLASSNAME##_C_IT \
* \
* Iterator class for class CLASSNAME##_CLIST \
* \
* Note: We don't need to coerce pointers to member functions input \
* parameters as these are automatically converted to the type of the base \
* type. ("A ptr to a class may be converted to a pointer to a public base \
* class of that class") \
**********************************************************************/ \
\
class DLLSYM CLASSNAME##_C_IT : public CLIST_ITERATOR { \
public: \
CLASSNAME##_C_IT() : CLIST_ITERATOR() {} \
\
CLASSNAME##_C_IT(CLASSNAME##_CLIST *list) : CLIST_ITERATOR(list) {} \
\
CLASSNAME *data() { return (CLASSNAME *)CLIST_ITERATOR::data(); } \
\
CLASSNAME *data_relative(inT8 offset) { \
return (CLASSNAME *)CLIST_ITERATOR::data_relative(offset); \
} \
\
CLASSNAME *forward() { return (CLASSNAME *)CLIST_ITERATOR::forward(); } \
\
CLASSNAME *extract() { return (CLASSNAME *)CLIST_ITERATOR::extract(); } \
\
CLASSNAME *move_to_first() { \
return (CLASSNAME *)CLIST_ITERATOR::move_to_first(); \
} \
\
CLASSNAME *move_to_last() { \
return (CLASSNAME *)CLIST_ITERATOR::move_to_last(); \
} \
};
#define CLISTIZEH( CLASSNAME ) \
\
CLISTIZEH_A( CLASSNAME ) \
\
CLISTIZEH_B( CLASSNAME ) \
\
CLISTIZEH_C( CLASSNAME )
#define CLISTIZEH(CLASSNAME) \
\
CLISTIZEH_A(CLASSNAME) \
\
CLISTIZEH_B(CLASSNAME) \
\
CLISTIZEH_C(CLASSNAME)
/***********************************************************************
CLISTIZE( CLASSNAME ) MACRO
***********************************************************************/
#define CLISTIZE( CLASSNAME ) \
\
/*********************************************************************** \
* CLASSNAME##_c1_zapper \
* \
* A function which can delete a CLASSNAME element. This is passed to the \
* generic deep_clear list member function so that when a list is cleared the \
* elements on the list are properly destroyed from the base class, even \
* though we don't use a virtual destructor function. \
**********************************************************************/ \
\
DLLSYM void CLASSNAME##_c1_zapper( /*delete a link*/ \
void* link) /*link to delete*/ \
{ \
delete (CLASSNAME *) link; \
} \
#define CLISTIZE(CLASSNAME) \
\
/*********************************************************************** \
* CLASSNAME##_c1_zapper \
* \
* A function which can delete a CLASSNAME element. This is passed to the \
* generic deep_clear list member function so that when a list is cleared \
*the \
* elements on the list are properly destroyed from the base class, even \
* though we don't use a virtual destructor function. \
**********************************************************************/ \
\
DLLSYM void CLASSNAME##_c1_zapper( /*delete a link*/ \
void *link) /*link to delete*/ \
{ \
delete (CLASSNAME *)link; \
}
#endif

26
ccutil/elst.cpp
View File

@ -26,7 +26,7 @@
**********************************************************************/
/***********************************************************************
* ELIST::internal_clear
* ELIST::internal_clear
*
* Used by the destructor and the "clear" member function of derived list
* classes to destroy all the elements on the list.
@ -57,7 +57,7 @@ void (*zapper) (ELIST_LINK *)) {
}
/***********************************************************************
* ELIST::assign_to_sublist
* ELIST::assign_to_sublist
*
* The list is set to a sublist of another list. "This" list must be empty
* before this function is invoked. The two iterators passed must refer to
@ -81,9 +81,8 @@ void ELIST::assign_to_sublist( //to this list
last = start_it->extract_sublist (end_it);
}
/***********************************************************************
* ELIST::length
* ELIST::length
*
* Return count of elements on list
**********************************************************************/
@ -97,9 +96,8 @@ inT32 ELIST::length() const { // count elements
return count;
}
/***********************************************************************
* ELIST::sort
* ELIST::sort
*
* Sort elements on list
* NB If you don't like the const declarations in the comparator, coerce yours:
@ -187,7 +185,7 @@ ELIST_LINK *ELIST::add_sorted_and_find(
**********************************************************************/
/***********************************************************************
* ELIST_ITERATOR::forward
* ELIST_ITERATOR::forward
*
* Move the iterator to the next element of the list.
* REMEMBER: ALL LISTS ARE CIRCULAR.
@ -224,9 +222,8 @@ ELIST_LINK *ELIST_ITERATOR::forward() {
return current;
}
/***********************************************************************
* ELIST_ITERATOR::data_relative
* ELIST_ITERATOR::data_relative
*
* Return the data pointer to the element "offset" elements from current.
* "offset" must not be less than -1.
@ -260,9 +257,8 @@ ELIST_LINK *ELIST_ITERATOR::data_relative( //get data + or - ...
return ptr;
}
/***********************************************************************
* ELIST_ITERATOR::move_to_last()
* ELIST_ITERATOR::move_to_last()
*
* Move current so that it is set to the end of the list.
* Return data just in case anyone wants it.
@ -281,9 +277,8 @@ ELIST_LINK *ELIST_ITERATOR::move_to_last() {
return current;
}
/***********************************************************************
* ELIST_ITERATOR::exchange()
* ELIST_ITERATOR::exchange()
*
* Given another iterator, whose current element is a different element on
* the same list list OR an element of another list, exchange the two current
@ -379,9 +374,8 @@ void ELIST_ITERATOR::exchange( //positions of 2 link
other_it->current = old_current;
}
/***********************************************************************
* ELIST_ITERATOR::extract_sublist()
* ELIST_ITERATOR::extract_sublist()
*
* This is a private member, used only by ELIST::assign_to_sublist.
* Given another iterator for the same list, extract the links from THIS to
@ -425,7 +419,7 @@ ELIST_LINK *ELIST_ITERATOR::extract_sublist( //from
temp_it.mark_cycle_pt ();
do { //walk sublist
if (temp_it.cycled_list ()) //can't find end pt
if (temp_it.cycled_list()) // can't find end pt
BAD_SUBLIST.error ("ELIST_ITERATOR.extract_sublist", ABORT, NULL);
if (temp_it.at_last ()) {

62
ccutil/elst.h
View File

@ -98,8 +98,8 @@ class DLLSYM ELIST_LINK
next = NULL;
}
void operator= ( //don't copy links
const ELIST_LINK &) {
void operator=( // don't copy links
const ELIST_LINK &) {
next = NULL;
}
};
@ -273,8 +273,8 @@ class DLLSYM ELIST_ITERATOR
bool cycled_list(); //Completed a cycle?
void add_to_end( //add at end &
ELIST_LINK *new_link); //don't move
void add_to_end( // add at end &
ELIST_LINK *new_link); // don't move
void exchange( //positions of 2 links
ELIST_ITERATOR *other_it); //other iterator
@ -458,7 +458,6 @@ inline void ELIST_ITERATOR::add_before_then_move( // element to add
current = new_element;
}
/***********************************************************************
* ELIST_ITERATOR::add_before_stay_put
*
@ -501,11 +500,11 @@ inline void ELIST_ITERATOR::add_before_stay_put( // element to add
}
}
/***********************************************************************
* ELIST_ITERATOR::add_list_after
*
* Insert another list to this list after the current element but don't move the
* Insert another list to this list after the current element but don't move
*the
* iterator.
**********************************************************************/
@ -959,30 +958,29 @@ ELISTIZEH_C( CLASSNAME )
ELISTIZE( CLASSNAME ) MACRO
***********************************************************************/
#define ELISTIZE(CLASSNAME) \
\
/*********************************************************************** \
* CLASSNAME##_zapper \
* \
* A function which can delete a CLASSNAME element. This is passed to the \
* generic clear list member function so that when a list is cleared the \
* elements on the list are properly destroyed from the base class, even \
* though we don't use a virtual destructor function. \
**********************************************************************/ \
\
DLLSYM void CLASSNAME##_zapper(ELIST_LINK* link) { \
delete reinterpret_cast<CLASSNAME*>(link); \
} \
\
/* Become a deep copy of src_list*/ \
void CLASSNAME##_LIST::deep_copy(const CLASSNAME##_LIST* src_list, \
CLASSNAME* (*copier)(const CLASSNAME*)) { \
\
CLASSNAME##_IT from_it(const_cast<CLASSNAME##_LIST*>(src_list)); \
CLASSNAME##_IT to_it(this); \
\
for (from_it.mark_cycle_pt(); !from_it.cycled_list(); from_it.forward()) \
to_it.add_after_then_move((*copier)(from_it.data())); \
}
#define ELISTIZE(CLASSNAME) \
\
/*********************************************************************** \
* CLASSNAME##_zapper \
* \
* A function which can delete a CLASSNAME element. This is passed to the \
* generic clear list member function so that when a list is cleared the \
* elements on the list are properly destroyed from the base class, even \
* though we don't use a virtual destructor function. \
**********************************************************************/ \
\
DLLSYM void CLASSNAME##_zapper(ELIST_LINK *link) { \
delete reinterpret_cast<CLASSNAME *>(link); \
} \
\
/* Become a deep copy of src_list*/ \
void CLASSNAME##_LIST::deep_copy(const CLASSNAME##_LIST *src_list, \
CLASSNAME *(*copier)(const CLASSNAME *)) { \
CLASSNAME##_IT from_it(const_cast<CLASSNAME##_LIST *>(src_list)); \
CLASSNAME##_IT to_it(this); \
\
for (from_it.mark_cycle_pt(); !from_it.cycled_list(); from_it.forward()) \
to_it.add_after_then_move((*copier)(from_it.data())); \
}
#endif

26
ccutil/elst2.cpp
View File

@ -27,7 +27,7 @@
**********************************************************************/
/***********************************************************************
* ELIST2::internal_clear
* ELIST2::internal_clear
*
* Used by the destructor and the "clear" member function of derived list
* classes to destroy all the elements on the list.
@ -58,7 +58,7 @@ void (*zapper) (ELIST2_LINK *)) {
}
/***********************************************************************
* ELIST2::assign_to_sublist
* ELIST2::assign_to_sublist
*
* The list is set to a sublist of another list. "This" list must be empty
* before this function is invoked. The two iterators passed must refer to
@ -82,9 +82,8 @@ void ELIST2::assign_to_sublist( //to this list
last = start_it->extract_sublist (end_it);
}
/***********************************************************************
* ELIST2::length
* ELIST2::length
*
* Return count of elements on list
**********************************************************************/
@ -98,9 +97,8 @@ inT32 ELIST2::length() const { // count elements
return count;
}
/***********************************************************************
* ELIST2::sort
* ELIST2::sort
*
* Sort elements on list
* NB If you don't like the const declarations in the comparator, coerce yours:
@ -180,7 +178,7 @@ void ELIST2::add_sorted(int comparator(const void*, const void*),
**********************************************************************/
/***********************************************************************
* ELIST2_ITERATOR::forward
* ELIST2_ITERATOR::forward
*
* Move the iterator to the next element of the list.
* REMEMBER: ALL LISTS ARE CIRCULAR.
@ -218,9 +216,8 @@ ELIST2_LINK *ELIST2_ITERATOR::forward() {
return current;
}
/***********************************************************************
* ELIST2_ITERATOR::backward
* ELIST2_ITERATOR::backward
*
* Move the iterator to the previous element of the list.
* REMEMBER: ALL LISTS ARE CIRCULAR.
@ -257,9 +254,8 @@ ELIST2_LINK *ELIST2_ITERATOR::backward() {
return current;
}
/***********************************************************************
* ELIST2_ITERATOR::data_relative
* ELIST2_ITERATOR::data_relative
*
* Return the data pointer to the element "offset" elements from current.
* (This function can't be INLINEd because it contains a loop)
@ -289,9 +285,8 @@ ELIST2_LINK *ELIST2_ITERATOR::data_relative( //get data + or - ..
return ptr;
}
/***********************************************************************
* ELIST2_ITERATOR::exchange()
* ELIST2_ITERATOR::exchange()
*
* Given another iterator, whose current element is a different element on
* the same list list OR an element of another list, exchange the two current
@ -399,9 +394,8 @@ void ELIST2_ITERATOR::exchange( //positions of 2 li
other_it->current = old_current;
}
/***********************************************************************
* ELIST2_ITERATOR::extract_sublist()
* ELIST2_ITERATOR::extract_sublist()
*
* This is a private member, used only by ELIST2::assign_to_sublist.
* Given another iterator for the same list, extract the links from THIS to
@ -445,7 +439,7 @@ ELIST2_LINK *ELIST2_ITERATOR::extract_sublist( //fr
temp_it.mark_cycle_pt ();
do { //walk sublist
if (temp_it.cycled_list ()) //can't find end pt
if (temp_it.cycled_list()) // can't find end pt
BAD_SUBLIST.error ("ELIST2_ITERATOR.extract_sublist", ABORT, NULL);
if (temp_it.at_last ()) {

317
ccutil/elst2.h
View File

@ -46,9 +46,9 @@ i) The duplication in source does not affect the run time code size - the
**********************************************************************/
/**********************************************************************
* CLASS - ELIST2_LINK
* CLASS - ELIST2_LINK
*
* Generic link class for doubly linked lists with embedded links
* Generic link class for doubly linked lists with embedded links
*
* Note: No destructor - elements are assumed to be destroyed EITHER after
* they have been extracted from a list OR by the ELIST2 destructor which
@ -68,13 +68,13 @@ class DLLSYM ELIST2_LINK
prev = next = NULL;
}
ELIST2_LINK( //copy constructor
const ELIST2_LINK &) { //don't copy link
ELIST2_LINK( // copy constructor
const ELIST2_LINK &) { // don't copy link
prev = next = NULL;
}
void operator= ( //don't copy links
const ELIST2_LINK &) {
void operator=( // don't copy links
const ELIST2_LINK &) {
prev = next = NULL;
}
};
@ -142,9 +142,10 @@ class DLLSYM ELIST2
};
/***********************************************************************
* CLASS - ELIST2_ITERATOR
* CLASS - ELIST2_ITERATOR
*
* Generic iterator class for doubly linked lists with embedded links
* Generic iterator class for doubly linked lists with embedded
*links
**********************************************************************/
class DLLSYM ELIST2_ITERATOR
@ -240,8 +241,8 @@ class DLLSYM ELIST2_ITERATOR
BOOL8 cycled_list(); //Completed a cycle?
void add_to_end( //add at end &
ELIST2_LINK *new_link); //don't move
void add_to_end( // add at end &
ELIST2_LINK *new_link); // don't move
void exchange( //positions of 2 links
ELIST2_ITERATOR *other_it); //other iterator
@ -255,7 +256,7 @@ class DLLSYM ELIST2_ITERATOR
};
/***********************************************************************
* ELIST2_ITERATOR::set_to_list
* ELIST2_ITERATOR::set_to_list
*
* (Re-)initialise the iterator to point to the start of the list_to_iterate
* over.
@ -279,9 +280,8 @@ inline void ELIST2_ITERATOR::set_to_list( //change list
ex_current_was_cycle_pt = FALSE;
}
/***********************************************************************
* ELIST2_ITERATOR::ELIST2_ITERATOR
* ELIST2_ITERATOR::ELIST2_ITERATOR
*
* CONSTRUCTOR - set iterator to specified list;
**********************************************************************/
@ -290,9 +290,8 @@ inline ELIST2_ITERATOR::ELIST2_ITERATOR(ELIST2 *list_to_iterate) {
set_to_list(list_to_iterate);
}
/***********************************************************************
* ELIST2_ITERATOR::add_after_then_move
* ELIST2_ITERATOR::add_after_then_move
*
* Add a new element to the list after the current element and move the
* iterator to the new element.
@ -339,9 +338,8 @@ inline void ELIST2_ITERATOR::add_after_then_move( // element to add
current = new_element;
}
/***********************************************************************
* ELIST2_ITERATOR::add_after_stay_put
* ELIST2_ITERATOR::add_after_stay_put
*
* Add a new element to the list after the current element but do not move
* the iterator to the new element.
@ -391,9 +389,8 @@ inline void ELIST2_ITERATOR::add_after_stay_put( // element to add
}
}
/***********************************************************************
* ELIST2_ITERATOR::add_before_then_move
* ELIST2_ITERATOR::add_before_then_move
*
* Add a new element to the list before the current element and move the
* iterator to the new element.
@ -438,9 +435,8 @@ inline void ELIST2_ITERATOR::add_before_then_move( // element to add
current = new_element;
}
/***********************************************************************
* ELIST2_ITERATOR::add_before_stay_put
* ELIST2_ITERATOR::add_before_stay_put
*
* Add a new element to the list before the current element but don't move the
* iterator to the new element.
@ -486,11 +482,11 @@ inline void ELIST2_ITERATOR::add_before_stay_put( // element to add
}
}
/***********************************************************************
* ELIST2_ITERATOR::add_list_after
* ELIST2_ITERATOR::add_list_after
*
* Insert another list to this list after the current element but don't move the
* Insert another list to this list after the current element but don't move
*the
* iterator.
**********************************************************************/
@ -537,9 +533,8 @@ inline void ELIST2_ITERATOR::add_list_after(ELIST2 *list_to_add) {
}
}
/***********************************************************************
* ELIST2_ITERATOR::add_list_before
* ELIST2_ITERATOR::add_list_before
*
* Insert another list to this list before the current element. Move the
* iterator to the start of the inserted elements
@ -586,9 +581,8 @@ inline void ELIST2_ITERATOR::add_list_before(ELIST2 *list_to_add) {
}
}
/***********************************************************************
* ELIST2_ITERATOR::extract
* ELIST2_ITERATOR::extract
*
* Do extraction by removing current from the list, returning it to the
* caller, but NOT updating the iterator. (So that any calling loop can do
@ -631,9 +625,8 @@ inline ELIST2_LINK *ELIST2_ITERATOR::extract() {
return extracted_link;
}
/***********************************************************************
* ELIST2_ITERATOR::move_to_first()
* ELIST2_ITERATOR::move_to_first()
*
* Move current so that it is set to the start of the list.
* Return data just in case anyone wants it.
@ -651,9 +644,8 @@ inline ELIST2_LINK *ELIST2_ITERATOR::move_to_first() {
return current;
}
/***********************************************************************
* ELIST2_ITERATOR::move_to_last()
* ELIST2_ITERATOR::move_to_last()
*
* Move current so that it is set to the end of the list.
* Return data just in case anyone wants it.
@ -671,9 +663,8 @@ inline ELIST2_LINK *ELIST2_ITERATOR::move_to_last() {
return current;
}
/***********************************************************************
* ELIST2_ITERATOR::mark_cycle_pt()
* ELIST2_ITERATOR::mark_cycle_pt()
*
* Remember the current location so that we can tell whether we've returned
* to this point later.
@ -696,9 +687,8 @@ inline void ELIST2_ITERATOR::mark_cycle_pt() {
started_cycling = FALSE;
}
/***********************************************************************
* ELIST2_ITERATOR::at_first()
* ELIST2_ITERATOR::at_first()
*
* Are we at the start of the list?
*
@ -716,9 +706,8 @@ inline BOOL8 ELIST2_ITERATOR::at_first() {
!ex_current_was_last)); //first and last
}
/***********************************************************************
* ELIST2_ITERATOR::at_last()
* ELIST2_ITERATOR::at_last()
*
* Are we at the end of the list?
*
@ -736,9 +725,8 @@ inline BOOL8 ELIST2_ITERATOR::at_last() {
ex_current_was_last)); //first and last
}
/***********************************************************************
* ELIST2_ITERATOR::cycled_list()
* ELIST2_ITERATOR::cycled_list()
*
* Have we returned to the cycle_pt since it was set?
*
@ -754,9 +742,8 @@ inline BOOL8 ELIST2_ITERATOR::cycled_list() {
}
/***********************************************************************
* ELIST2_ITERATOR::length()
* ELIST2_ITERATOR::length()
*
* Return the length of the list
*
@ -771,9 +758,8 @@ inline inT32 ELIST2_ITERATOR::length() {
return list->length ();
}
/***********************************************************************
* ELIST2_ITERATOR::sort()
* ELIST2_ITERATOR::sort()
*
* Sort the elements of the list, then reposition at the start.
*
@ -792,9 +778,8 @@ const void *, const void *)) {
move_to_first();
}
/***********************************************************************
* ELIST2_ITERATOR::add_to_end
* ELIST2_ITERATOR::add_to_end
*
* Add a new element to the end of the list without moving the iterator.
* This is provided because a single linked list cannot move to the last as
@ -854,7 +839,7 @@ will NOT work correctly for classes derived from this.
The macro generates:
- An element deletion function: CLASSNAME##_zapper
- An E_LIST2 subclass: CLASSNAME##_LIST
- An E_LIST2 subclass: CLASSNAME##_LIST
- An E_LIST2_ITERATOR subclass:
CLASSNAME##_IT
@ -873,132 +858,132 @@ ELIST2IZEH is a concatenation of 3 fragments ELIST2IZEH_A, ELIST2IZEH_B and
ELIST2IZEH_C.
***********************************************************************/
#define ELIST2IZEH_A( CLASSNAME ) \
\
extern DLLSYM void CLASSNAME##_zapper( /*delete a link*/ \
ELIST2_LINK* link); /*link to delete*/
#define ELIST2IZEH_A(CLASSNAME) \
\
extern DLLSYM void CLASSNAME##_zapper( /*delete a link*/ \
ELIST2_LINK *link); /*link to delete*/
#define ELIST2IZEH_B( CLASSNAME ) \
\
/*********************************************************************** \
* CLASS - CLASSNAME##_LIST \
* \
* List class for class CLASSNAME \
* \
**********************************************************************/ \
\
class DLLSYM CLASSNAME##_LIST : public ELIST2 \
{ \
public: \
CLASSNAME##_LIST():ELIST2() {} \
/* constructor */ \
\
CLASSNAME##_LIST( /* don't construct */ \
const CLASSNAME##_LIST&) /*by initial assign*/\
{ DONT_CONSTRUCT_LIST_BY_COPY.error( QUOTE_IT( CLASSNAME##_LIST ), \
ABORT, NULL ); } \
\
void clear() /* delete elements */\
{ ELIST2::internal_clear( &CLASSNAME##_zapper ); } \
\
~CLASSNAME##_LIST() /* destructor */ \
{ clear(); } \
\
/* Become a deep copy of src_list*/ \
void deep_copy(const CLASSNAME##_LIST* src_list, \
CLASSNAME* (*copier)(const CLASSNAME*)); \
\
void operator=( /* prevent assign */ \
const CLASSNAME##_LIST&) \
{ DONT_ASSIGN_LISTS.error( QUOTE_IT( CLASSNAME##_LIST ), \
ABORT, NULL ); }
#define ELIST2IZEH_B(CLASSNAME) \
\
/*********************************************************************** \
* CLASS - \
*CLASSNAME##_LIST \
* \
* List class for class \
*CLASSNAME \
* \
**********************************************************************/ \
\
class DLLSYM CLASSNAME##_LIST : public ELIST2 { \
public: \
CLASSNAME##_LIST() : ELIST2() {} \
/* constructor */ \
\
CLASSNAME##_LIST( /* don't construct */ \
const CLASSNAME##_LIST &) /*by initial assign*/ \
{ \
DONT_CONSTRUCT_LIST_BY_COPY.error(QUOTE_IT(CLASSNAME##_LIST), ABORT, \
NULL); \
} \
\
void clear() /* delete elements */ \
{ \
ELIST2::internal_clear(&CLASSNAME##_zapper); \
} \
\
~CLASSNAME##_LIST() /* destructor */ \
{ \
clear(); \
} \
\
/* Become a deep copy of src_list*/ \
void deep_copy(const CLASSNAME##_LIST *src_list, \
CLASSNAME *(*copier)(const CLASSNAME *)); \
\
void operator=(/* prevent assign */ \
const CLASSNAME##_LIST &) { \
DONT_ASSIGN_LISTS.error(QUOTE_IT(CLASSNAME##_LIST), ABORT, NULL); \
}
#define ELIST2IZEH_C( CLASSNAME ) \
}; \
\
\
\
/*********************************************************************** \
* CLASS - CLASSNAME##_IT \
* \
* Iterator class for class CLASSNAME##_LIST \
* \
* Note: We don't need to coerce pointers to member functions input \
* parameters as these are automatically converted to the type of the base \
* type. ("A ptr to a class may be converted to a pointer to a public base \
* class of that class") \
**********************************************************************/ \
\
class DLLSYM CLASSNAME##_IT : public ELIST2_ITERATOR \
{ \
public: \
CLASSNAME##_IT():ELIST2_ITERATOR(){} \
\
CLASSNAME##_IT( \
CLASSNAME##_LIST* list):ELIST2_ITERATOR(list){} \
\
CLASSNAME* data() \
{ return (CLASSNAME*) ELIST2_ITERATOR::data(); } \
\
CLASSNAME* data_relative( \
inT8 offset) \
{ return (CLASSNAME*) ELIST2_ITERATOR::data_relative( offset ); } \
\
CLASSNAME* forward() \
{ return (CLASSNAME*) ELIST2_ITERATOR::forward(); } \
\
CLASSNAME* backward() \
{ return (CLASSNAME*) ELIST2_ITERATOR::backward(); } \
\
CLASSNAME* extract() \
{ return (CLASSNAME*) ELIST2_ITERATOR::extract(); } \
\
CLASSNAME* move_to_first() \
{ return (CLASSNAME*) ELIST2_ITERATOR::move_to_first(); } \
\
CLASSNAME* move_to_last() \
{ return (CLASSNAME*) ELIST2_ITERATOR::move_to_last(); } \
};
#define ELIST2IZEH( CLASSNAME ) \
\
ELIST2IZEH_A( CLASSNAME ) \
\
ELIST2IZEH_B( CLASSNAME ) \
\
ELIST2IZEH_C( CLASSNAME )
#define ELIST2IZEH_C(CLASSNAME) \
} \
; \
\
/*********************************************************************** \
* CLASS - CLASSNAME##_IT \
* \
* Iterator class for class CLASSNAME##_LIST \
* \
* Note: We don't need to coerce pointers to member functions input \
* parameters as these are automatically converted to the type of the base \
* type. ("A ptr to a class may be converted to a pointer to a public base \
* class of that class") \
**********************************************************************/ \
\
class DLLSYM CLASSNAME##_IT : public ELIST2_ITERATOR { \
public: \
CLASSNAME##_IT() : ELIST2_ITERATOR() {} \
\
CLASSNAME##_IT(CLASSNAME##_LIST *list) : ELIST2_ITERATOR(list) {} \
\
CLASSNAME *data() { return (CLASSNAME *)ELIST2_ITERATOR::data(); } \
\
CLASSNAME *data_relative(inT8 offset) { \
return (CLASSNAME *)ELIST2_ITERATOR::data_relative(offset); \
} \
\
CLASSNAME *forward() { return (CLASSNAME *)ELIST2_ITERATOR::forward(); } \
\
CLASSNAME *backward() { return (CLASSNAME *)ELIST2_ITERATOR::backward(); } \
\
CLASSNAME *extract() { return (CLASSNAME *)ELIST2_ITERATOR::extract(); } \
\
CLASSNAME *move_to_first() { \
return (CLASSNAME *)ELIST2_ITERATOR::move_to_first(); \
} \
\
CLASSNAME *move_to_last() { \
return (CLASSNAME *)ELIST2_ITERATOR::move_to_last(); \
} \
};
#define ELIST2IZEH(CLASSNAME) \
\
ELIST2IZEH_A(CLASSNAME) \
\
ELIST2IZEH_B(CLASSNAME) \
\
ELIST2IZEH_C(CLASSNAME)
/***********************************************************************
ELIST2IZE( CLASSNAME ) MACRO
***********************************************************************/
#define ELIST2IZE( CLASSNAME ) \
\
/*********************************************************************** \
* CLASSNAME##_zapper \
* \
* A function which can delete a CLASSNAME element. This is passed to the \
* generic clear list member function so that when a list is cleared the \
* elements on the list are properly destroyed from the base class, even \
* though we don't use a virtual destructor function. \
**********************************************************************/ \
\
DLLSYM void CLASSNAME##_zapper( /*delete a link*/ \
ELIST2_LINK* link) /*link to delete*/ \
{ \
delete (CLASSNAME *) link; \
} \
\
/* Become a deep copy of src_list*/ \
void CLASSNAME##_LIST::deep_copy(const CLASSNAME##_LIST* src_list, \
CLASSNAME* (*copier)(const CLASSNAME*)) { \
\
CLASSNAME##_IT from_it(const_cast<CLASSNAME##_LIST*>(src_list)); \
CLASSNAME##_IT to_it(this); \
\
for (from_it.mark_cycle_pt(); !from_it.cycled_list(); from_it.forward()) \
to_it.add_after_then_move((*copier)(from_it.data())); \
}
#define ELIST2IZE(CLASSNAME) \
\
/*********************************************************************** \
* CLASSNAME##_zapper \
* \
* A function which can delete a CLASSNAME element. This is passed to the \
* generic clear list member function so that when a list is cleared the \
* elements on the list are properly destroyed from the base class, even \
* though we don't use a virtual destructor function. \
**********************************************************************/ \
\
DLLSYM void CLASSNAME##_zapper( /*delete a link*/ \
ELIST2_LINK *link) /*link to delete*/ \
{ \
delete (CLASSNAME *)link; \
} \
\
/* Become a deep copy of src_list*/ \
void CLASSNAME##_LIST::deep_copy(const CLASSNAME##_LIST *src_list, \
CLASSNAME *(*copier)(const CLASSNAME *)) { \
CLASSNAME##_IT from_it(const_cast<CLASSNAME##_LIST *>(src_list)); \
CLASSNAME##_IT to_it(this); \
\
for (from_it.mark_cycle_pt(); !from_it.cycled_list(); from_it.forward()) \
to_it.add_after_then_move((*copier)(from_it.data())); \
}
#endif

9
ccutil/errcode.h
View File

@ -87,11 +87,10 @@ const ERRCODE ASSERT_FAILED = "Assert failed";
__FILE__, __LINE__); \
}
#define ASSERT_HOST_MSG(x, ...) if (!(x)) \
{ \
tprintf(__VA_ARGS__); \
ASSERT_FAILED.error(#x, ABORT, "in file %s, line %d", \
__FILE__, __LINE__); \
#define ASSERT_HOST_MSG(x, ...) \
if (!(x)) { \
tprintf(__VA_ARGS__); \
ASSERT_FAILED.error(#x, ABORT, "in file %s, line %d", __FILE__, __LINE__); \
}
void signal_exit(int signal_code);

2
ccutil/helpers.h
View File

@ -73,7 +73,7 @@ class TRand {
// Remove newline (if any) at the end of the string.
inline void chomp_string(char *str) {
int last_index = (int)strlen(str) - 1;
int last_index = static_cast<int>(strlen(str)) - 1;
while (last_index >= 0 &&
(str[last_index] == '\n' || str[last_index] == '\r')) {
str[last_index--] = '\0';

12
ccutil/host.h
View File

@ -59,6 +59,7 @@
** of the computer and/or operating system.
******************************************************************************/
#include <limits>
#include "platform.h"
/* _WIN32 */
#ifdef _WIN32
@ -121,15 +122,16 @@ typedef unsigned char BOOL8;
#define MAX_UINT8 0xff
#define MAX_UINT16 0xffff
#define MAX_UINT32 0xffffffff
#define MAX_FLOAT32 ((float)3.40282347e+38)
#define MAX_FLOAT32 std::numeric_limits<float>::max()
#define MIN_INT8 0x80
#define MIN_INT16 0x8000
#define MIN_INT32 static_cast<int>(0x80000000)
#define MIN_INT8 static_cast<inT8>(0x80)
#define MIN_INT16 static_cast<inT16>(0x8000)
#define MIN_INT32 static_cast<inT32>(0x80000000)
#define MIN_UINT8 0x00
#define MIN_UINT16 0x0000
#define MIN_UINT32 0x00000000
#define MIN_FLOAT32 ((float)1.17549435e-38)
// Minimum positive value ie 1e-37ish.
#define MIN_FLOAT32 std::numeric_limits<float>::min()
// Defines
#ifndef TRUE

2
ccutil/lsterr.h
View File

@ -38,6 +38,6 @@ const ERRCODE NULL_PREV = "Previous element on the list is NULL";
const ERRCODE EMPTY_LIST = "List is empty";
const ERRCODE BAD_PARAMETER = "List parameter error";
const ERRCODE STILL_LINKED =
"Attempting to add an element with non NULL links, to a list";
"Attempting to add an element with non NULL links, to a list";
#endif
#endif

2
ccutil/mainblk.cpp
View File

@ -74,7 +74,7 @@ void CCUtil::main_setup(const char *argv0, const char *basename) {
#endif /* _WIN32 */
#if defined(TESSDATA_PREFIX)
} else {
/* Use tessdata prefix which was compiled in. */
/* Use tessdata prefix which was compiled in. */
#define _STR(a) #a
#define _XSTR(a) _STR(a)
datadir = _XSTR(TESSDATA_PREFIX);

39
ccutil/ocrclass.h
View File

@ -1,7 +1,7 @@
/**********************************************************************
* File: ocrclass.h
* Description: Class definitions and constants for the OCR API.
* Author: Hewlett-Packard Co
* Author: Hewlett-Packard Co
*
* (C) Copyright 1996, Hewlett-Packard Co.
** Licensed under the Apache License, Version 2.0 (the "License");
@ -110,28 +110,35 @@ typedef struct { /*single character */
* user words found. If it returns true then operation is cancelled.
**********************************************************************/
typedef bool (*CANCEL_FUNC)(void* cancel_this, int words);
typedef bool (*PROGRESS_FUNC)(int progress,
int left, int right, int top, int bottom);
typedef bool (*PROGRESS_FUNC)(int progress, int left, int right, int top,
int bottom);
class ETEXT_DESC { // output header
public:
inT16 count; /// chars in this buffer(0)
inT16 progress; /// percent complete increasing (0-100)
inT16 count; /// chars in this buffer(0)
inT16 progress; /// percent complete increasing (0-100)
/** Progress monitor covers word recognition and it does not cover layout
* analysis.
* See Ray comment in https://github.com/tesseract-ocr/tesseract/pull/27 */
inT8 more_to_come; /// true if not last
volatile inT8 ocr_alive; /// ocr sets to 1, HP 0
inT8 err_code; /// for errcode use
CANCEL_FUNC cancel; /// returns true to cancel
PROGRESS_FUNC progress_callback; /// called whenever progress increases
void* cancel_this; /// this or other data for cancel
struct timeval end_time; /** time to stop. expected to be set only by call
* to set_deadline_msecs() */
EANYCODE_CHAR text[1]; /// character data
inT8 more_to_come; /// true if not last
volatile inT8 ocr_alive; /// ocr sets to 1, HP 0
inT8 err_code; /// for errcode use
CANCEL_FUNC cancel; /// returns true to cancel
PROGRESS_FUNC progress_callback; /// called whenever progress increases
void* cancel_this; /// this or other data for cancel
struct timeval end_time; /// Time to stop. Expected to be set only
/// by call to set_deadline_msecs().
EANYCODE_CHAR text[1]; /// character data
ETEXT_DESC() : count(0), progress(0), more_to_come(0), ocr_alive(0),
err_code(0), cancel(NULL), cancel_this(NULL) {
ETEXT_DESC()
: count(0),
progress(0),
more_to_come(0),
ocr_alive(0),
err_code(0),
cancel(NULL),
progress_callback(NULL),
cancel_this(NULL) {
end_time.tv_sec = 0;
end_time.tv_usec = 0;
}

3
ccutil/params.cpp
View File

@ -31,8 +31,7 @@
#define EQUAL '='
tesseract::ParamsVectors *GlobalParams() {
static tesseract::ParamsVectors global_params =
tesseract::ParamsVectors();
static tesseract::ParamsVectors global_params = tesseract::ParamsVectors();
return &global_params;
}

4
ccutil/strngs.cpp
View File

@ -1,8 +1,8 @@
/**********************************************************************
* File: strngs.c (Formerly strings.c)
* Description: STRING class functions.
* Author: Ray Smith
* Created: Fri Feb 15 09:13:30 GMT 1991
* Author: Ray Smith
* Created: Fri Feb 15 09:13:30 GMT 1991
*
* (C) Copyright 1991, Hewlett-Packard Ltd.
** Licensed under the Apache License, Version 2.0 (the "License");

3
ccutil/unicharset.h
View File

@ -181,8 +181,7 @@ class UNICHARSET {
// Return the UNICHAR_ID of a given unichar representation within the
// UNICHARSET. Only the first length characters from unichar_repr are used.
UNICHAR_ID unichar_to_id(const char* const unichar_repr,
int length) const;
UNICHAR_ID unichar_to_id(const char* const unichar_repr, int length) const;
// Return the minimum number of bytes that matches a legal UNICHAR_ID,
// while leaving the rest of the string encodable. Returns 0 if the

5
classify/adaptmatch.cpp
View File

@ -908,8 +908,7 @@ void Classify::AdaptToChar(TBLOB* Blob, CLASS_ID ClassId, int FontinfoId,
NumFeatures = GetAdaptiveFeatures(Blob, IntFeatures, &FloatFeatures);
if (NumFeatures <= 0) {
FreeFeatureSet(FloatFeatures);
return;
return; // Features already freed by GetAdaptiveFeatures.
}
// Only match configs with the matching font.
@ -1008,8 +1007,6 @@ void Classify::DisplayAdaptedChar(TBLOB* blob, INT_CLASS_STRUCT* int_class) {
#endif
}
/**
* This routine adds the result of a classification into
* Results. If the new rating is much worse than the current

4
classify/classify.cpp
View File

@ -151,8 +151,8 @@ Classify::Classify()
INT_MEMBER(classify_integer_matcher_multiplier, 10,
"Integer Matcher Multiplier 0-255: ", this->params()),
EnableLearning(true),
INT_MEMBER(il1_adaption_test, 0, "Don't adapt to i/I at beginning of word",
this->params()),
INT_MEMBER(il1_adaption_test, 0,
"Don't adapt to i/I at beginning of word", this->params()),
BOOL_MEMBER(classify_bln_numeric_mode, 0,
"Assume the input is numbers [0-9].", this->params()),
double_MEMBER(speckle_large_max_size, 0.30, "Max large speckle size",

477
classify/cluster.cpp
View File

File diff suppressed because it is too large Load Diff

43
classify/clusttool.cpp
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: clustertool.c
** Purpose: Misc. tools for use with the clustering routines
** Author: Dan Johnson
** History: 6/6/89, DSJ, Created.
** Filename: clustertool.c
** Purpose: Misc. tools for use with the clustering routines
** Author: Dan Johnson
** History: 6/6/89, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at
@ -26,9 +26,10 @@
#include <math.h>
//---------------Global Data Definitions and Declarations--------------------
#define TOKENSIZE 80 //< max size of tokens read from an input file
#define MAXSAMPLESIZE 65535 //< max num of dimensions in feature space
//#define MAXBLOCKSIZE 65535 //< max num of samples in a character (block size)
#define TOKENSIZE 80 //< max size of tokens read from an input file
#define MAXSAMPLESIZE 65535 //< max num of dimensions in feature space
//#define MAXBLOCKSIZE 65535 //< max num of samples in a character (block
// size)
/**
* This routine reads a single integer from the specified
@ -37,7 +38,7 @@
* @param File open text file to read sample size from
* @return Sample size
* @note Globals: None
* @note Exceptions: ILLEGALSAMPLESIZE illegal format or range
* @note Exceptions: ILLEGALSAMPLESIZE illegal format or range
* @note History: 6/6/89, DSJ, Created.
*/
uinT16 ReadSampleSize(FILE *File) {
@ -293,7 +294,7 @@ FLOAT32* ReadNFloats(FILE * File, uinT16 N, FLOAT32 Buffer[]) {
if (NumFloatsRead != 1) {
if ((NumFloatsRead == EOF) && (i == 0)) {
if (needs_free) {
Efree(Buffer);
Efree(Buffer);
}
return NULL;
} else {
@ -315,8 +316,7 @@ FLOAT32* ReadNFloats(FILE * File, uinT16 N, FLOAT32 Buffer[]) {
* @note Exceptions: None
* @note History: 6/6/89, DSJ, Created.
*/
void
WriteParamDesc (FILE * File, uinT16 N, PARAM_DESC ParamDesc[]) {
void WriteParamDesc(FILE *File, uinT16 N, const PARAM_DESC ParamDesc[]) {
int i;
for (i = 0; i < N; i++) {
@ -446,15 +446,10 @@ void WriteProtoStyle(FILE *File, PROTOSTYLE ProtoStyle) {
* @note History: 6/12/89, DSJ, Created.
*/
void WriteProtoList(
FILE *File,
uinT16 N,
PARAM_DESC ParamDesc[],
LIST ProtoList,
BOOL8 WriteSigProtos,
BOOL8 WriteInsigProtos)
{
PROTOTYPE *Proto;
void WriteProtoList(FILE *File, uinT16 N, PARAM_DESC ParamDesc[],
LIST ProtoList, BOOL8 WriteSigProtos,
BOOL8 WriteInsigProtos) {
PROTOTYPE *Proto;
/* write file header */
fprintf(File,"%0d\n",N);
@ -464,8 +459,8 @@ void WriteProtoList(
iterate(ProtoList)
{
Proto = (PROTOTYPE *) first_node ( ProtoList );
if (( Proto->Significant && WriteSigProtos ) ||
( ! Proto->Significant && WriteInsigProtos ) )
WritePrototype( File, N, Proto );
if ((Proto->Significant && WriteSigProtos) ||
(!Proto->Significant && WriteInsigProtos))
WritePrototype(File, N, Proto);
}
}

22
classify/clusttool.h
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: clusttool.h
** Purpose: Definition of clustering utility tools
** Author: Dan Johnson
** History: 6/6/89, DSJ, Created.
** Filename: clusttool.h
** Purpose: Definition of clustering utility tools
** Author: Dan Johnson
** History: 6/6/89, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at
@ -36,7 +36,7 @@ PROTOSTYLE ReadProtoStyle(FILE *File);
FLOAT32 *ReadNFloats (FILE * File, uinT16 N, FLOAT32 Buffer[]);
void WriteParamDesc (FILE * File, uinT16 N, PARAM_DESC ParamDesc[]);
void WriteParamDesc(FILE *File, uinT16 N, const PARAM_DESC ParamDesc[]);
void WritePrototype(FILE *File, uinT16 N, PROTOTYPE *Proto);
@ -44,13 +44,9 @@ void WriteNFloats (FILE * File, uinT16 N, FLOAT32 Array[]);
void WriteProtoStyle(FILE *File, PROTOSTYLE ProtoStyle);
void WriteProtoList(
FILE *File,
uinT16 N,
PARAM_DESC ParamDesc[],
LIST ProtoList,
BOOL8 WriteSigProtos,
BOOL8 WriteInsigProtos);
void WriteProtoList(FILE *File, uinT16 N, PARAM_DESC ParamDesc[],
LIST ProtoList, BOOL8 WriteSigProtos,
BOOL8 WriteInsigProtos);
//--------------Global Data Definitions and Declarations---------------------
// define errors that can be trapped

10
classify/cutoffs.cpp
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: cutoffs.c
** Purpose: Routines to manipulate an array of class cutoffs.
** Author: Dan Johnson
** History: Wed Feb 20 09:28:51 1991, DSJ, Created.
** Filename: cutoffs.c
** Purpose: Routines to manipulate an array of class cutoffs.
** Author: Dan Johnson
** History: Wed Feb 20 09:28:51 1991, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at

14
classify/featdefs.cpp
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: featdefs.c
** Purpose: Definitions of currently defined feature types.
** Author: Dan Johnson
** History: Mon May 21 10:26:21 1990, DSJ, Created.
** Filename: featdefs.c
** Purpose: Definitions of currently defined feature types.
** Author: Dan Johnson
** History: Mon May 21 10:26:21 1990, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at
@ -289,13 +289,13 @@ CHAR_DESC ReadCharDescription(const FEATURE_DEFS_STRUCT &FeatureDefs,
* the feature type for the feature with the specified short
* name. Trap an error if the specified name is not found.
*
* Globals:
* Globals:
* - none
*
* @param FeatureDefs definitions of feature types/extractors
* @param ShortName short name of a feature type
* @return Feature type which corresponds to ShortName.
* @note Exceptions:
* @note Exceptions:
* - ILLEGAL_SHORT_NAME
* @note History: Wed May 23 15:36:05 1990, DSJ, Created.
*/

11
classify/fpoint.cpp
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: fpoint.c
** Purpose: Abstract data type for a 2D point (floating point coords)
** Author: Dan Johnson
** History: Thu Apr 12 10:44:15 1990, DSJ, Created.
** Filename: fpoint.c
** Purpose: Abstract data type for a 2D point (floating point coords)
** Author: Dan Johnson
** History: Thu Apr 12 10:44:15 1990, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at
@ -58,5 +58,4 @@ FLOAT32 NormalizedAngleFrom(FPOINT *Point1,
if (Angle < 0.0 || Angle >= FullScale)
Angle = 0.0;
return (Angle);
}

2
classify/intfx.cpp
View File

@ -520,7 +520,7 @@ bool ExtractIntFeat(const TBLOB& blob,
tesseract::Classify::ExtractFeatures(blob, nonlinear_norm,
&bl_features, &cn_features, results,
NULL);
if (bl_features.size() == 0 || cn_features.size() == 0 ||
if (bl_features.empty() || cn_features.empty() ||
bl_features.size() > MAX_NUM_INT_FEATURES ||
cn_features.size() > MAX_NUM_INT_FEATURES) {
return false; // Feature extraction failed.

26
classify/intmatcher.cpp
View File

@ -295,7 +295,8 @@ class ClassPruner {
HeapSort(num_classes_, sort_key_, sort_index_);
}
/** Prints debug info on the class pruner matches for the pruned classes only. */
/** Prints debug info on the class pruner matches for the pruned classes only.
*/
void DebugMatch(const Classify& classify,
const INT_TEMPLATES_STRUCT* int_templates,
const INT_FEATURE_STRUCT* features) const {
@ -370,8 +371,9 @@ class ClassPruner {
private:
/** Array[rounded_classes_] of initial counts for each class. */
int *class_count_;
/// Array[rounded_classes_] of modified counts for each class after normalizing
/// for expected number of features, disabled classes, fragments, and xheights.
/// Array[rounded_classes_] of modified counts for each class after
/// normalizing for expected number of features, disabled classes, fragments,
/// and xheights.
int *norm_count_;
/** Array[rounded_classes_ +1] of pruned counts that gets sorted */
int *sort_key_;
@ -402,8 +404,9 @@ class ClassPruner {
* normalization process (by CLASS_INDEX)
* @param expected_num_features Array of expected number of features
* for each class (by CLASS_INDEX)
* @param results Sorted Array of pruned classes. Must be an array
* of size at least int_templates->NumClasses.
* @param results Sorted Array of pruned classes. Must be an
* array of size at least
* int_templates->NumClasses.
* @param keep_this
*/
int Classify::PruneClasses(const INT_TEMPLATES_STRUCT* int_templates,
@ -606,7 +609,6 @@ int IntegerMatcher::FindGoodProtos(
return NumGoodProtos;
}
/**
* FindBadFeatures finds all features with maximum feature-evidence <
* AdaptFeatureThresh. The list is ordered by increasing feature number.
@ -701,7 +703,6 @@ void IntegerMatcher::Init(tesseract::IntParam *classify_debug_level) {
evidence_mult_mask_ = ((1 << kIntEvidenceTruncBits) - 1);
}
/*----------------------------------------------------------------------------
Private Code
----------------------------------------------------------------------------*/
@ -717,8 +718,6 @@ void ScratchEvidence::ClearFeatureEvidence(const INT_CLASS class_template) {
class_template->NumConfigs * sizeof(feature_evidence_[0]));
}
/**
* Print debugging information for Configuations
* @return none
@ -742,7 +741,6 @@ void IMDebugConfiguration(int FeatureNum,
cprintf ("\n");
}
/**
* Print debugging information for Configuations
* @return none
@ -795,10 +793,10 @@ int IntegerMatcher::UpdateTablesForFeature(
uinT32 XFeatureAddress;
uinT32 YFeatureAddress;
uinT32 ThetaFeatureAddress;
uinT8 *UINT8Pointer;
uinT8* UINT8Pointer;
int ProtoIndex;
uinT8 Temp;
int *IntPointer;
int* IntPointer;
int ConfigNum;
inT32 M3;
inT32 A3;
@ -916,7 +914,6 @@ int IntegerMatcher::UpdateTablesForFeature(
return SumOverConfigs;
}
/**
* Print debugging information for Configuations
* @return none
@ -1165,8 +1162,6 @@ void ScratchEvidence::UpdateSumOfProtoEvidences(
}
}
/**
* Normalize Sum of Proto and Feature Evidence by dividing by the sum of
* the Feature Lengths and the Proto Lengths for each configuration.
@ -1180,7 +1175,6 @@ void ScratchEvidence::NormalizeSums(
}
}
/**
* Find the best match for the current class and update the Result
* with the configuration and match rating.

10
classify/intmatcher.h
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: intmatcher.h
** Purpose: Interface to high level generic classifier routines.
** Author: Robert Moss
** History: Wed Feb 13 15:24:15 MST 1991, RWM, Created.
** Filename: intmatcher.h
** Purpose: Interface to high level generic classifier routines.
** Author: Robert Moss
** History: Wed Feb 13 15:24:15 MST 1991, RWM, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at

26
classify/intproto.cpp
View File

@ -46,9 +46,7 @@
#include "config_auto.h"
#endif
using tesseract::FontInfo;
using tesseract::FontSet;
using tesseract::FontSpacingInfo;
/* match debug display constants*/
#define PROTO_PRUNER_SCALE (4.0)
@ -326,10 +324,8 @@ int AddIntProto(INT_CLASS Class) {
Word < Proto->Configs + WERDS_PER_CONFIG_VEC; *Word++ = 0);
return (Index);
}
/**
* This routine adds Proto to the class pruning tables
* for the specified class in Templates.
@ -372,7 +368,6 @@ void AddProtoToClassPruner (PROTO Proto, CLASS_ID ClassId,
}
} /* AddProtoToClassPruner */
/**
* This routine updates the proto pruner lookup tables
* for Class to include a new proto identified by ProtoId
@ -432,7 +427,6 @@ void AddProtoToProtoPruner(PROTO Proto, int ProtoId,
FillPPLinearBits(ProtoSet->ProtoPruner[PRUNER_Y], Index, Y, Pad, debug);
} /* AddProtoToProtoPruner */
/**
* Returns a quantized bucket for the given param shifted by offset,
* notionally (param + offset) * num_buckets, but clipped and casted to the
@ -550,7 +544,6 @@ void Classify::ConvertProto(PROTO Proto, int ProtoId, INT_CLASS Class) {
P->A, P->B, P->C, Class->ProtoLengths[ProtoId]);
} /* ConvertProto */
/**
* This routine converts from the old floating point format
* to the new integer format.
@ -627,7 +620,7 @@ INT_TEMPLATES Classify::CreateIntTemplates(CLASSES FloatProtos,
* @note Exceptions: none
* @note History: Thu Mar 21 14:45:04 1991, DSJ, Created.
*/
void DisplayIntFeature(const INT_FEATURE_STRUCT* Feature, FLOAT32 Evidence) {
void DisplayIntFeature(const INT_FEATURE_STRUCT *Feature, FLOAT32 Evidence) {
ScrollView::Color color = GetMatchColorFor(Evidence);
RenderIntFeature(IntMatchWindow, Feature, color);
if (FeatureDisplayWindow) {
@ -635,7 +628,6 @@ void DisplayIntFeature(const INT_FEATURE_STRUCT* Feature, FLOAT32 Evidence) {
}
} /* DisplayIntFeature */
/**
* This routine renders the specified proto into a
* global display list.
@ -720,7 +712,6 @@ void free_int_class(INT_CLASS int_class) {
Efree(int_class);
}
/**
* This routine allocates a new set of integer templates
* initialized to hold 0 classes.
@ -1218,7 +1209,6 @@ FLOAT32 BucketStart(int Bucket, FLOAT32 Offset, int NumBuckets) {
} /* BucketStart */
/**
* This routine returns the parameter value which
* corresponds to the end of the specified bucket.
@ -1236,7 +1226,6 @@ FLOAT32 BucketEnd(int Bucket, FLOAT32 Offset, int NumBuckets) {
return (((FLOAT32) (Bucket + 1) / NumBuckets) - Offset);
} /* BucketEnd */
/**
* This routine fills in the section of a class pruner
* corresponding to a single x value for a single proto of
@ -1284,7 +1273,6 @@ void DoFill(FILL_SPEC *FillSpec,
}
} /* DoFill */
/**
* Return TRUE if the specified table filler is done, i.e.
* if it has no more lines to fill.
@ -1306,7 +1294,6 @@ BOOL8 FillerDone(TABLE_FILLER *Filler) {
} /* FillerDone */
/**
* This routine sets Bit in each bit vector whose
* bucket lies within the range Center +- Spread. The fill
@ -1349,7 +1336,6 @@ void FillPPCircularBits(uinT32 ParamTable[NUM_PP_BUCKETS][WERDS_PER_PP_VECTOR],
} /* FillPPCircularBits */
/**
* This routine sets Bit in each bit vector whose
* bucket lies within the range Center +- Spread. The fill
@ -1516,7 +1502,6 @@ void GetCPPadsForLevel(int Level,
} /* GetCPPadsForLevel */
/**
* @param Evidence evidence value to return color for
* @return Color which corresponds to specified Evidence value.
@ -1538,7 +1523,6 @@ ScrollView::Color GetMatchColorFor(FLOAT32 Evidence) {
return ScrollView::BLUE;
} /* GetMatchColorFor */
/**
* This routine returns (in Fill) the specification of
* the next line to be filled from Filler. FillerDone() should
@ -1589,7 +1573,6 @@ void GetNextFill(TABLE_FILLER *Filler, FILL_SPEC *Fill) {
} /* GetNextFill */
/**
* This routine computes a data structure (Filler)
* which can be used to fill in a rectangle surrounding
@ -1723,8 +1706,10 @@ void InitTableFiller (FLOAT32 EndPad, FLOAT32 SidePad,
/* translate into bucket positions and deltas */
Filler->X = Bucket8For(Start.x, XS, NB);
Filler->StartDelta = -(inT16) ((Sin / Cos) * 256);
Filler->EndDelta = (inT16) ((Cos / Sin) * 256);
Filler->StartDelta = static_cast<inT16>(ClipToRange<int>(
-IntCastRounded((Sin / Cos) * 256), MIN_INT16, MAX_INT16));
Filler->EndDelta = static_cast<inT16>(ClipToRange<int>(
IntCastRounded((Cos / Sin) * 256), MIN_INT16, MAX_INT16));
XAdjust = BucketEnd(Filler->X, XS, NB) - Start.x;
YAdjust = XAdjust * Sin / Cos;
@ -1787,7 +1772,6 @@ void RenderIntFeature(ScrollView *window, const INT_FEATURE_STRUCT* Feature,
window->DrawTo(X + Dx, Y + Dy);
} /* RenderIntFeature */
/**
* This routine extracts the parameters of the specified
* proto from the class description and adds a rendering of

2
classify/intproto.h
View File

@ -52,7 +52,7 @@ class FCOORD;
#define NUM_CP_BUCKETS 24
#define CLASSES_PER_CP 32
#define NUM_BITS_PER_CLASS 2
#define CLASS_PRUNER_CLASS_MASK (~(~0 << NUM_BITS_PER_CLASS))
#define CLASS_PRUNER_CLASS_MASK (~(~0u << NUM_BITS_PER_CLASS))
#define CLASSES_PER_CP_WERD (CLASSES_PER_CP / NUM_BITS_PER_CLASS)
#define PROTOS_PER_PP_WERD BITS_PER_WERD
#define BITS_PER_CP_VECTOR (CLASSES_PER_CP * NUM_BITS_PER_CLASS)

28
classify/kdtree.cpp
View File

@ -70,11 +70,11 @@ class MinK {
const Element* elements() { return elements_; }
private:
const Key max_key_; //< the maximum possible Key
Element* elements_; //< unsorted array of elements
const Key max_key_; //< the maximum possible Key
Element *elements_; //< unsorted array of elements
int elements_count_; //< the number of results collected so far
int k_; //< the number of results we want from the search
int max_index_; //< the index of the result with the largest key
int k_; //< the number of results we want from the search
int max_index_; //< the index of the result with the largest key
};
template<typename Key, typename Value>
@ -117,7 +117,8 @@ bool MinK<Key, Value>::insert(Key key, Value value) {
//-----------------------------------------------------------------------------
/** Helper class for searching for the k closest points to query_point in tree. */
/** Helper class for searching for the k closest points to query_point in tree.
*/
class KDTreeSearch {
public:
KDTreeSearch(KDTREE* tree, FLOAT32 *query_point, int k_closest);
@ -241,14 +242,13 @@ void KDStore(KDTREE *Tree, FLOAT32 *Key, void *Data) {
*PtrToNode = MakeKDNode(Tree, Key, (void *) Data, Level);
} /* KDStore */
/**
* This routine deletes a node from Tree. The node to be
* deleted is specified by the Key for the node and the Data
* contents of the node. These two pointers must be identical
* to the pointers that were used for the node when it was
* originally stored in the tree. A node will be deleted from
* the tree only if its key and data pointers are identical
* This routine deletes a node from Tree. The node to be
* deleted is specified by the Key for the node and the Data
* contents of the node. These two pointers must be identical
* to the pointers that were used for the node when it was
* originally stored in the tree. A node will be deleted from
* the tree only if its key and data pointers are identical
* to Key and Data respectively. The tree is re-formed by removing
* the affected subtree and inserting all elements but the root.
*
@ -298,7 +298,6 @@ KDDelete (KDTREE * Tree, FLOAT32 Key[], void *Data) {
}
} /* KDDelete */
/**
* This routine searches the K-D tree specified by Tree and
* finds the QuerySize nearest neighbors of Query. All neighbors
@ -442,7 +441,7 @@ void KDTreeSearch::SearchRec(int level, KDNODE *sub_tree) {
/*---------------------------------------------------------------------------*/
/**
/**
*Returns the Euclidean distance squared between p1 and p2 for all essential
* dimensions.
* @param k keys are in k-space
@ -541,7 +540,6 @@ void Walk(KDTREE *tree, void_proc action, void *context,
Walk(tree, action, context, sub_tree->Right, NextLevel(tree, level));
}
/** Given a subtree nodes, insert all of its elements into tree. */
void InsertNodes(KDTREE *tree, KDNODE *nodes) {
if (nodes == NULL)

8
classify/mastertrainer.cpp
View File

@ -362,9 +362,11 @@ bool MasterTrainer::LoadFontInfo(const char* filename) {
fontinfo.name = font_name;
fontinfo.properties = 0;
fontinfo.universal_id = 0;
if (tfscanf(fp, "%1024s %i %i %i %i %i\n", font_name,
&italic, &bold, &fixed, &serif, &fraktur) != 6)
if (tfscanf(fp, "%1024s %i %i %i %i %i\n", font_name, &italic, &bold,
&fixed, &serif, &fraktur) != 6) {
delete[] font_name;
continue;
}
fontinfo.properties =
(italic << 0) +
(bold << 1) +
@ -373,6 +375,8 @@ bool MasterTrainer::LoadFontInfo(const char* filename) {
(fraktur << 4);
if (!fontinfo_table_.contains(fontinfo)) {
fontinfo_table_.push_back(fontinfo);
} else {
delete[] font_name;
}
}
fclose(fp);

12
classify/mf.cpp
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: mf.c
** Purpose: Micro-feature interface to flexible feature extractor.
** Author: Dan Johnson
** History: Thu May 24 09:08:38 1990, DSJ, Created.
** Filename: mf.c
** Purpose: Micro-feature interface to flexible feature extractor.
** Author: Dan Johnson
** History: Thu May 24 09:08:38 1990, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at
@ -36,7 +36,7 @@
* Call the old micro-feature extractor and then copy
* the features into the new format. Then deallocate the
* old micro-features.
* @param Blob blob to extract micro-features from
* @param Blob blob to extract micro-features from
* @param cn_denorm control parameter to feature extractor.
* @return Micro-features for Blob.
* @note Exceptions: none

16
classify/mfdefs.cpp
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: mfdefs.c
** Purpose: Basic routines for manipulating micro-features
** Author: Dan Johnson
** History: Mon Jan 22 08:48:58 1990, DSJ, Created.
** Filename: mfdefs.c
** Purpose: Basic routines for manipulating micro-features
** Author: Dan Johnson
** History: Mon Jan 22 08:48:58 1990, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at
@ -32,7 +32,7 @@
* @return New MICROFEATURE
* @note History: 7/27/89, DSJ, Created.
*/
MICROFEATURE NewMicroFeature() {
MICROFEATURE NewMicroFeature() {
return ((MICROFEATURE) Emalloc (sizeof (MFBLOCK)));
} /* NewMicroFeature */
@ -41,10 +41,10 @@ MICROFEATURE NewMicroFeature() {
/**
* This routine deallocates all of the memory consumed by
* a list of micro-features.
* @param MicroFeatures list of micro-features to be freed
* @param MicroFeatures list of micro-features to be freed
* @return none
* @note History: 7/27/89, DSJ, Created.
*/
void FreeMicroFeatures(MICROFEATURES MicroFeatures) {
void FreeMicroFeatures(MICROFEATURES MicroFeatures) {
destroy_nodes(MicroFeatures, Efree);
} /* FreeMicroFeatures */

5
classify/mfoutline.cpp
View File

@ -35,7 +35,8 @@
----------------------------------------------------------------------------*/
/*---------------------------------------------------------------------------*/
/** Convert a blob into a list of MFOUTLINEs (float-based microfeature format). */
/** Convert a blob into a list of MFOUTLINEs (float-based microfeature format).
*/
LIST ConvertBlob(TBLOB *blob) {
LIST outlines = NIL_LIST;
return (blob == NULL)
@ -344,7 +345,6 @@ void ChangeDirection(MFOUTLINE Start, MFOUTLINE End, DIRECTION Direction) {
} /* ChangeDirection */
/**
* This routine normalizes each point in Outline by
* translating it to the specified center and scaling it
@ -378,7 +378,6 @@ void CharNormalizeOutline(MFOUTLINE Outline, const DENORM& cn_denorm) {
} /* CharNormalizeOutline */
/**
* This routine computes the slope from Start to Finish and
* and then computes the approximate direction of the line

4
classify/mfx.cpp
View File

@ -128,7 +128,6 @@ FLOAT32 ComputeOrientation(MFEDGEPT *Start, MFEDGEPT *End) {
return (Orientation);
} /* ComputeOrientation */
/**
* Convert Outline to MicroFeatures
* @param Outline outline to extract micro-features from
@ -164,7 +163,6 @@ MICROFEATURES ConvertToMicroFeatures(MFOUTLINE Outline,
return (MicroFeatures);
} /* ConvertToMicroFeatures */
/**
* This routine computes the feature parameters which describe
* the micro-feature that starts and Start and ends at End.
@ -178,7 +176,7 @@ MICROFEATURES ConvertToMicroFeatures(MFOUTLINE Outline,
* @return New micro-feature or NULL if the feature was rejected.
* @note Globals: none
* @note Exceptions: none
* @note History:
* @note History:
* - 7/26/89, DSJ, Created.
* - 11/17/89, DSJ, Added handling for Start and End same point.
*/

10
classify/mfx.h
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: mfx.h
** Purpose: Definition of micro-feature extraction routines
** Author: Dan Johnson
** History: 5/29/89, DSJ, Created.
** Filename: mfx.h
** Purpose: Definition of micro-feature extraction routines
** Author: Dan Johnson
** History: 5/29/89, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at

11
classify/normfeat.cpp
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: normfeat.c
** Purpose: Definition of char normalization features.
** Author: Dan Johnson
** History: 12/14/90, DSJ, Created.
** Filename: normfeat.c
** Purpose: Definition of char normalization features.
** Author: Dan Johnson
** History: 12/14/90, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at
@ -33,7 +33,6 @@ FLOAT32 ActualOutlineLength(FEATURE Feature) {
return (Feature->Params[CharNormLength] * LENGTH_COMPRESSION);
}
/**
* Return the character normalization feature for a blob.
*

18
classify/normmatch.cpp
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: normmatch.c
** Purpose: Simple matcher based on character normalization features.
** Author: Dan Johnson
** History: Wed Dec 19 16:18:06 1990, DSJ, Created.
** Filename: normmatch.c
** Purpose: Simple matcher based on character normalization features.
** Author: Dan Johnson
** History: Wed Dec 19 16:18:06 1990, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at
@ -197,10 +197,10 @@ double NormEvidenceOf(register double NormAdj) {
/*---------------------------------------------------------------------------*/
/**
* This routine dumps out detailed normalization match info.
* @param File open text file to dump match debug info to
* @param NumParams # of parameters in proto and feature
* @param Proto[] array of prototype parameters
* @param Feature[] array of feature parameters
* @param File open text file to dump match debug info to
* @param NumParams # of parameters in proto and feature
* @param Proto[] array of prototype parameters
* @param Feature[] array of feature parameters
* Globals: none
* @return none
* @note Exceptions: none

33
classify/ocrfeatures.cpp
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: features.c
** Purpose: Generic definition of a feature.
** Author: Dan Johnson
** History: Mon May 21 10:49:04 1990, DSJ, Created.
** Filename: features.c
** Purpose: Generic definition of a feature.
** Author: Dan Johnson
** History: Mon May 21 10:49:04 1990, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at
@ -66,12 +66,11 @@ void FreeFeature(FEATURE Feature) {
} /* FreeFeature */
/**
* Release the memory consumed by the specified feature
* set. This routine also frees the memory consumed by the
* features contained in the set.
* @param FeatureSet set of features to be freed
* @param FeatureSet set of features to be freed
* @return none
* @note History: Mon May 21 13:59:46 1990, DSJ, Created.
*/
@ -85,11 +84,10 @@ void FreeFeatureSet(FEATURE_SET FeatureSet) {
}
} /* FreeFeatureSet */
/**
* Allocate and return a new feature of the specified
* type.
* @param FeatureDesc description of feature to be created.
* @param FeatureDesc description of feature to be created.
* @return New #FEATURE.
* @note History: Mon May 21 14:06:42 1990, DSJ, Created.
*/
@ -105,11 +103,10 @@ FEATURE NewFeature(const FEATURE_DESC_STRUCT* FeatureDesc) {
} /* NewFeature */
/**
* Allocate and return a new feature set large enough to
* hold the specified number of features.
* @param NumFeatures maximum # of features to be put in feature set
* @param NumFeatures maximum # of features to be put in feature set
* @return New #FEATURE_SET.
* @note History: Mon May 21 14:22:40 1990, DSJ, Created.
*/
@ -124,7 +121,6 @@ FEATURE_SET NewFeatureSet(int NumFeatures) {
} /* NewFeatureSet */
/**
* Create a new feature of the specified type and read in
* the value of its parameters from File. The extra penalty
@ -135,10 +131,11 @@ FEATURE_SET NewFeatureSet(int NumFeatures) {
* @param File open text file to read feature from
* @param FeatureDesc specifies type of feature to read from File
* @return New #FEATURE read from File.
* @note Exceptions: #ILLEGAL_FEATURE_PARAM if text file doesn't match expected format
* @note Exceptions: #ILLEGAL_FEATURE_PARAM if text file doesn't match expected
* format
* @note History: Wed May 23 08:53:16 1990, DSJ, Created.
*/
FEATURE ReadFeature(FILE *File, const FEATURE_DESC_STRUCT* FeatureDesc) {
FEATURE ReadFeature(FILE* File, const FEATURE_DESC_STRUCT* FeatureDesc) {
FEATURE Feature;
int i;
@ -153,7 +150,6 @@ FEATURE ReadFeature(FILE *File, const FEATURE_DESC_STRUCT* FeatureDesc) {
return (Feature);
} /* ReadFeature */
/**
* Create a new feature set of the specified type and read in
* the features from File. The correct text representation
@ -165,7 +161,7 @@ FEATURE ReadFeature(FILE *File, const FEATURE_DESC_STRUCT* FeatureDesc) {
* @return New feature set read from File.
* @note History: Wed May 23 09:17:31 1990, DSJ, Created.
*/
FEATURE_SET ReadFeatureSet(FILE *File, const FEATURE_DESC_STRUCT* FeatureDesc) {
FEATURE_SET ReadFeatureSet(FILE* File, const FEATURE_DESC_STRUCT* FeatureDesc) {
FEATURE_SET FeatureSet;
int NumFeatures;
int i;
@ -180,7 +176,6 @@ FEATURE_SET ReadFeatureSet(FILE *File, const FEATURE_DESC_STRUCT* FeatureDesc) {
return (FeatureSet);
} /* ReadFeatureSet */
/**
* Appends a textual representation of Feature to str.
* This representation is simply a list of the N parameters
@ -203,7 +198,6 @@ void WriteFeature(FEATURE Feature, STRING* str) {
*str += "\n";
} /* WriteFeature */
/**
* Write a textual representation of FeatureSet to File.
* This representation is an integer specifying the number of
@ -224,7 +218,6 @@ void WriteFeatureSet(FEATURE_SET FeatureSet, STRING* str) {
}
} /* WriteFeatureSet */
/**
* Write a textual representation of FeatureDesc to File
* in the old format (i.e. the format used by the clusterer).
@ -240,7 +233,7 @@ void WriteFeatureSet(FEATURE_SET FeatureSet, STRING* str) {
* @return none
* @note History: Fri May 25 15:27:18 1990, DSJ, Created.
*/
void WriteOldParamDesc(FILE *File, const FEATURE_DESC_STRUCT* FeatureDesc) {
void WriteOldParamDesc(FILE* File, const FEATURE_DESC_STRUCT* FeatureDesc) {
int i;
fprintf (File, "%d\n", FeatureDesc->NumParams);

14
classify/outfeat.cpp
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: outfeat.c
** Purpose: Definition of outline-features.
** Author: Dan Johnson
** History: 11/13/90, DSJ, Created.
** Filename: outfeat.c
** Purpose: Definition of outline-features.
** Author: Dan Johnson
** History: 11/13/90, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at
@ -40,7 +40,7 @@ namespace tesseract {
* @return Outline-features for Blob.
* @note Globals: none
* @note Exceptions: none
* @note History:
* @note History:
* - 11/13/90, DSJ, Created.
* - 05/24/91, DSJ, Updated for either char or baseline normalize.
*/
@ -115,7 +115,7 @@ void AddOutlineFeatureToSet(FPOINT *Start,
* @return none (results are returned in FeatureSet)
* @note Globals: none
* @note Exceptions: none
* @note History:
* @note History:
* - 11/13/90, DSJ, Created.
* - 5/24/91, DSJ, Added hidden edge capability.
*/

12
classify/picofeat.cpp
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: picofeat.c
** Purpose: Definition of pico-features.
** Author: Dan Johnson
** History: 9/4/90, DSJ, Created.
** Filename: picofeat.c
** Purpose: Definition of pico-features.
** Author: Dan Johnson
** History: 9/4/90, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at
@ -98,7 +98,7 @@ FEATURE_SET Classify::ExtractPicoFeatures(TBLOB *Blob) {
* nearest whole number of pico-features. The pico-features
* are spaced evenly over the entire segment.
* Globals:
* - classify_pico_feature_length length of a single pico-feature
* - classify_pico_feature_length length of a single pico-feature
* @param Start starting point of pico-feature
* @param End ending point of pico-feature
* @param FeatureSet set to add pico-feature to

2
classify/shapeclassifier.cpp
View File

@ -176,7 +176,7 @@ void ShapeClassifier::UnicharPrintResults(
for (int i = 0; i < results.size(); ++i) {
tprintf("%g: c_id=%d=%s", results[i].rating, results[i].unichar_id,
GetUnicharset().id_to_unichar(results[i].unichar_id));
if (results[i].fonts.size() != 0) {
if (!results[i].fonts.empty()) {
tprintf(" Font Vector:");
for (int f = 0; f < results[i].fonts.size(); ++f) {
tprintf(" %d", results[i].fonts[f].fontinfo_id);

2
cube/cube_search_object.cpp
View File

@ -400,7 +400,7 @@ bool CubeSearchObject::ComputeSpaceCosts() {
float prob = 0.0;
// gap is too small => no space
if (gap < min_spc_gap_) {
if (gap < min_spc_gap_ || max_spc_gap_ == min_spc_gap_) {
prob = 0.0;
} else if (gap > max_spc_gap_) {
// gap is too big => definite space

2
cube/hybrid_neural_net_classifier.cpp
View File

@ -330,7 +330,7 @@ bool HybridNeuralNetCharClassifier::LoadNets(const string &data_file_path,
// split into lines
vector<string> str_vec;
CubeUtils::SplitStringUsing(str, "\r\n", &str_vec);
if (str_vec.size() <= 0) {
if (str_vec.empty()) {
return false;
}

2
cube/word_unigrams.cpp
View File

@ -163,7 +163,7 @@ int WordUnigrams::Cost(const char_32 *key_str32,
CubeUtils::SplitStringUsing(key_str, " \t", &words);
// no words => no cost
if (words.size() <= 0) {
if (words.empty()) {
return 0;
}

42
cutil/bitvec.h
View File

@ -30,29 +30,29 @@ typedef uinT32 *BIT_VECTOR;
/*-----------------------------------------------------------------------------
Public Function Prototypes
-----------------------------------------------------------------------------*/
#define zero_all_bits(array,length) \
{\
int index; /*temporary index*/\
\
for (index=0;index<length;index++)\
array[index]=0; /*zero all bits*/\
}
#define zero_all_bits(array, length) \
{ \
int index; /*temporary index*/ \
\
for (index = 0; index < length; index++) \
array[index] = 0; /*zero all bits*/ \
}
#define set_all_bits(array,length) \
{\
int index; /*temporary index*/\
\
for (index=0;index<length;index++)\
array[index]= ~0; /*set all bits*/\
}
#define set_all_bits(array, length) \
{ \
int index; /*temporary index*/ \
\
for (index = 0; index < length; index++) \
array[index] = ~0; /*set all bits*/ \
}
#define copy_all_bits(source,dest,length) \
{\
int index; /*temporary index*/\
\
for (index=0;index<length;index++)\
dest[index]=source[index]; /*copy all bits*/\
}
#define copy_all_bits(source, dest, length) \
{ \
int index; /*temporary index*/ \
\
for (index = 0; index < length; index++) \
dest[index] = source[index]; /*copy all bits*/ \
}
#define SET_BIT(array,bit) (array[bit/BITSINLONG]|=1<<(bit&(BITSINLONG-1)))

10
cutil/danerror.cpp
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: danerror.c
** Purpose: Routines for managing error trapping
** Author: Dan Johnson
** History: 3/17/89, DSJ, Created.
** Filename: danerror.c
** Purpose: Routines for managing error trapping
** Author: Dan Johnson
** History: 3/17/89, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at

12
cutil/efio.cpp
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: efio.c
** Purpose: Utility I/O routines
** Author: Dan Johnson
** History: 5/21/89, DSJ, Created.
** Filename: efio.c
** Purpose: Utility I/O routines
** Author: Dan Johnson
** History: 5/21/89, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at
@ -40,7 +40,7 @@
* @note Exceptions: #FOPENERROR unable to open specified file
* @note History: 5/21/89, DSJ, Created.
*/
FILE *Efopen(const char *Name, const char *Mode) {
FILE *Efopen(const char *Name, const char *Mode) {
FILE *File;
char ErrorMessage[MAXERRORMESSAGE];

2
cutil/emalloc.cpp
View File

@ -10,7 +10,7 @@
** History:
4/3/89, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at

4
cutil/oldlist.cpp
View File

@ -206,8 +206,8 @@ void destroy_nodes(LIST list, void_dest destructor) {
destructor = memfree;
while (list != NIL_LIST) {
(*destructor) (first_node (list));
list = pop (list);
if (first_node(list) != NULL) (*destructor)(first_node(list));
list = pop(list);
}
}

38
dict/context.cpp
View File

@ -32,28 +32,22 @@ namespace tesseract {
static const int kMinAbsoluteGarbageWordLength = 10;
static const float kMinAbsoluteGarbageAlphanumFrac = 0.5f;
const int case_state_table[6][4] = { {
/* 0. Beginning of word */
/* P U L D */
/* -1. Error on case */
0, 1, 5, 4
},
{ /* 1. After initial capital */
0, 3, 2, 4
},
{ /* 2. After lower case */
0, -1, 2, -1
},
{ /* 3. After upper case */
0, 3, -1, 4
},
{ /* 4. After a digit */
0, -1, -1, 4
},
{ /* 5. After initial lower case */
5, -1, 2, -1
},
};
const int case_state_table[6][4] = {
{/* 0. Beginning of word */
/* P U L D */
/* -1. Error on case */
0, 1, 5, 4},
{/* 1. After initial capital */
0, 3, 2, 4},
{/* 2. After lower case */
0, -1, 2, -1},
{/* 3. After upper case */
0, 3, -1, 4},
{/* 4. After a digit */
0, -1, -1, 4},
{/* 5. After initial lower case */
5, -1, 2, -1},
};
int Dict::case_ok(const WERD_CHOICE &word, const UNICHARSET &unicharset) {
int state = 0;

66
dict/dict.cpp
View File

@ -30,13 +30,12 @@ namespace tesseract {
class Image;
Dict::Dict(CCUtil* ccutil)
Dict::Dict(CCUtil *ccutil)
: letter_is_okay_(&tesseract::Dict::def_letter_is_okay),
probability_in_context_(&tesseract::Dict::def_probability_in_context),
params_model_classify_(NULL),
ccutil_(ccutil),
STRING_MEMBER(user_words_file, "",
"A filename of user-provided words.",
STRING_MEMBER(user_words_file, "", "A filename of user-provided words.",
getCCUtil()->params()),
STRING_INIT_MEMBER(user_words_suffix, "",
"A suffix of user-provided words located in tessdata.",
@ -54,33 +53,41 @@ Dict::Dict(CCUtil* ccutil)
getCCUtil()->params()),
BOOL_INIT_MEMBER(load_unambig_dawg, true, "Load unambiguous word dawg.",
getCCUtil()->params()),
BOOL_INIT_MEMBER(load_punc_dawg, true, "Load dawg with punctuation"
" patterns.", getCCUtil()->params()),
BOOL_INIT_MEMBER(load_number_dawg, true, "Load dawg with number"
" patterns.", getCCUtil()->params()),
BOOL_INIT_MEMBER(load_bigram_dawg, true, "Load dawg with special word "
"bigrams.", getCCUtil()->params()),
BOOL_INIT_MEMBER(load_punc_dawg, true,
"Load dawg with punctuation"
" patterns.",
getCCUtil()->params()),
BOOL_INIT_MEMBER(load_number_dawg, true,
"Load dawg with number"
" patterns.",
getCCUtil()->params()),
BOOL_INIT_MEMBER(load_bigram_dawg, true,
"Load dawg with special word "
"bigrams.",
getCCUtil()->params()),
double_MEMBER(xheight_penalty_subscripts, 0.125,
"Score penalty (0.1 = 10%) added if there are subscripts "
"or superscripts in a word, but it is otherwise OK.",
getCCUtil()->params()),
double_MEMBER(xheight_penalty_inconsistent, 0.25,
"Score penalty (0.1 = 10%) added if an xheight is "
"inconsistent.", getCCUtil()->params()),
"inconsistent.",
getCCUtil()->params()),
double_MEMBER(segment_penalty_dict_frequent_word, 1.0,
"Score multiplier for word matches which have good case and"
"are frequent in the given language (lower is better).",
getCCUtil()->params()),
double_MEMBER(segment_penalty_dict_case_ok, 1.1,
"Score multiplier for word matches that have good case "
"(lower is better).", getCCUtil()->params()),
"(lower is better).",
getCCUtil()->params()),
double_MEMBER(segment_penalty_dict_case_bad, 1.3125,
"Default score multiplier for word matches, which may have "
"case issues (lower is better).",
getCCUtil()->params()),
double_MEMBER(segment_penalty_ngram_best_choice, 1.24,
"Multipler to for the best choice from the ngram model.",
getCCUtil()->params()),
"Multipler to for the best choice from the ngram model.",
getCCUtil()->params()),
double_MEMBER(segment_penalty_dict_nonword, 1.25,
"Score multiplier for glyph fragment segmentations which "
"do not match a dictionary word (lower is better).",
@ -88,11 +95,13 @@ Dict::Dict(CCUtil* ccutil)
double_MEMBER(segment_penalty_garbage, 1.50,
"Score multiplier for poorly cased strings that are not in"
" the dictionary and generally look like garbage (lower is"
" better).", getCCUtil()->params()),
" better).",
getCCUtil()->params()),
STRING_MEMBER(output_ambig_words_file, "",
"Output file for ambiguities found in the dictionary",
getCCUtil()->params()),
INT_MEMBER(dawg_debug_level, 0, "Set to 1 for general debug info"
INT_MEMBER(dawg_debug_level, 0,
"Set to 1 for general debug info"
", to 2 for more details, to 3 to see all the debug messages",
getCCUtil()->params()),
INT_MEMBER(hyphen_debug_level, 0, "Debug level for hyphenated words.",
@ -109,12 +118,12 @@ Dict::Dict(CCUtil* ccutil)
"Certainty threshold for non-dict words",
getCCUtil()->params()),
double_MEMBER(stopper_phase2_certainty_rejection_offset, 1.0,
"Reject certainty offset",
getCCUtil()->params()),
"Reject certainty offset", getCCUtil()->params()),
INT_MEMBER(stopper_smallword_size, 2,
"Size of dict word to be treated as non-dict word",
getCCUtil()->params()),
double_MEMBER(stopper_certainty_per_char, -0.50, "Certainty to add"
double_MEMBER(stopper_certainty_per_char, -0.50,
"Certainty to add"
" for each dict char above small word size.",
getCCUtil()->params()),
double_MEMBER(stopper_allowable_character_badness, 3.0,
@ -130,9 +139,9 @@ Dict::Dict(CCUtil* ccutil)
"Deprecated- backward compatibility only",
getCCUtil()->params()),
INT_MEMBER(tessedit_truncate_wordchoice_log, 10,
"Max words to keep in list",
getCCUtil()->params()),
STRING_MEMBER(word_to_debug, "", "Word for which stopper debug"
"Max words to keep in list", getCCUtil()->params()),
STRING_MEMBER(word_to_debug, "",
"Word for which stopper debug"
" information should be printed to stdout",
getCCUtil()->params()),
STRING_MEMBER(word_to_debug_lengths, "",
@ -141,10 +150,10 @@ Dict::Dict(CCUtil* ccutil)
INT_MEMBER(fragments_debug, 0, "Debug character fragments",
getCCUtil()->params()),
BOOL_MEMBER(segment_nonalphabetic_script, false,
"Don't use any alphabetic-specific tricks."
"Set to true in the traineddata config file for"
" scripts that are cursive or inherently fixed-pitch",
getCCUtil()->params()),
"Don't use any alphabetic-specific tricks."
"Set to true in the traineddata config file for"
" scripts that are cursive or inherently fixed-pitch",
getCCUtil()->params()),
BOOL_MEMBER(save_doc_words, 0, "Save Document Words",
getCCUtil()->params()),
double_MEMBER(doc_dict_pending_threshold, 0.0,
@ -152,8 +161,10 @@ Dict::Dict(CCUtil* ccutil)
getCCUtil()->params()),
double_MEMBER(doc_dict_certainty_threshold, -2.25,
"Worst certainty for words that can be inserted into the"
"document dictionary", getCCUtil()->params()),
INT_MEMBER(max_permuter_attempts, 10000, "Maximum number of different"
"document dictionary",
getCCUtil()->params()),
INT_MEMBER(max_permuter_attempts, 10000,
"Maximum number of different"
" character choices to consider during permutation."
" This limit is especially useful when user patterns"
" are specified, since overly generic patterns can result in"
@ -179,6 +190,7 @@ Dict::Dict(CCUtil* ccutil)
}
Dict::~Dict() {
End();
if (hyphen_word_ != NULL) delete hyphen_word_;
if (output_ambig_words_file_ != NULL) fclose(output_ambig_words_file_);
}

11
dict/stopper.cpp
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: stopper.c
** Purpose: Stopping criteria for word classifier.
** Author: Dan Johnson
** History: Mon Apr 29 14:56:49 1991, DSJ, Created.
** Filename: stopper.c
** Purpose: Stopping criteria for word classifier.
** Author: Dan Johnson
** History: Mon Apr 29 14:56:49 1991, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at
@ -41,7 +41,6 @@
#pragma warning(disable:4800) // int/bool warnings
#endif
using tesseract::ScriptPos;
/*----------------------------------------------------------------------------
Private Code
----------------------------------------------------------------------------*/

10
dict/stopper.h
View File

@ -1,10 +1,10 @@
/******************************************************************************
** Filename: stopper.h
** Purpose: Stopping criteria for word classifier.
** Author: Dan Johnson
** History: Wed May 1 09:42:57 1991, DSJ, Created.
** Filename: stopper.h
** Purpose: Stopping criteria for word classifier.
** Author: Dan Johnson
** History: Wed May 1 09:42:57 1991, DSJ, Created.
**
** (c) Copyright Hewlett-Packard Company, 1988.
** (c) Copyright Hewlett-Packard Company, 1988.
** Licensed under the Apache License, Version 2.0 (the "License");
** you may not use this file except in compliance with the License.
** You may obtain a copy of the License at

6
opencl/oclkernels.h
View File

@ -1092,7 +1092,8 @@ void kernel_ThresholdRectToPix(
for ( int c = 0; c < NUM_CHANNELS; c++) {
unsigned char pixChan = pixels.s[p*NUM_CHANNELS + c];
if (pHi_Values[c] >= 0 && (pixChan > pThresholds[c]) == (pHi_Values[c] == 0)) {
word |= (((uint)0x80000000) >> ((b*PIXELS_PER_BURST+p)&31));
const uint kTopBit = 0x80000000;
word |= (kTopBit >> ((b*PIXELS_PER_BURST+p)&31));
}
}
}
@ -1157,7 +1158,8 @@ void kernel_ThresholdRectToPix_OneChan(
\n#endif\n
unsigned char pixChan = pixels.s[idx];
if (pHi_Values[0] >= 0 && (pixChan > pThresholds[0]) == (pHi_Values[0] == 0)) {
word |= (0x80000000 >> ((b*PIXELS_PER_BURST+p)&31));
const uint kTopBit = 0x80000000;
word |= (kTopBit >> ((b*PIXELS_PER_BURST+p)&31));
}
}
}

6
opencl/opencl_device_selection.h
View File

@ -63,8 +63,10 @@ static ds_status releaseDSProfile(ds_profile* profile, ds_score_release sr) {
if (profile->devices!=NULL && sr!=NULL) {
unsigned int i;
for (i = 0; i < profile->numDevices; i++) {
if (profile->devices[i].oclDeviceName) free(profile->devices[i].oclDeviceName);
if (profile->devices[i].oclDriverVersion) free(profile->devices[i].oclDriverVersion);
if (profile->devices[i].oclDeviceName)
free(profile->devices[i].oclDeviceName);
if (profile->devices[i].oclDriverVersion)
free(profile->devices[i].oclDriverVersion);
status = sr(profile->devices[i].score);
if (status != DS_SUCCESS)
break;

1848
opencl/openclwrapper.cpp
View File

File diff suppressed because it is too large Load Diff

93
opencl/openclwrapper.h
View File

@ -10,7 +10,8 @@
// including CL/cl.h doesn't occur until USE_OPENCL defined below
// platform preprocessor commands
#if defined( WIN32 ) || defined( __WIN32__ ) || defined( _WIN32 ) || defined( __CYGWIN__ ) || defined( __MINGW32__ )
#if defined(WIN32) || defined(__WIN32__) || defined(_WIN32) || \
defined(__CYGWIN__) || defined(__MINGW32__)
#define ON_WINDOWS 1
#define ON_LINUX 0
#define ON_APPLE 0
@ -80,21 +81,23 @@
time_sub_start = time_funct_start; \
time_sub_end = time_funct_start;
#define PERF_COUNT_END \
QueryPerformanceCounter(&time_funct_end); \
elapsed_time_sec = (time_funct_end.QuadPart-time_funct_start.QuadPart)/(double)(freq.QuadPart); \
printf(PERF_COUNT_REPORT_STR, funct_name, "total", elapsed_time_sec);
#define PERF_COUNT_END \
QueryPerformanceCounter(&time_funct_end); \
elapsed_time_sec = (time_funct_end.QuadPart - time_funct_start.QuadPart) / \
(double)(freq.QuadPart); \
printf(PERF_COUNT_REPORT_STR, funct_name, "total", elapsed_time_sec);
#else
#define PERF_COUNT_START(FUNCT_NAME)
#define PERF_COUNT_END
#endif
#if PERF_COUNT_VERBOSE >= 3
#define PERF_COUNT_SUB(SUB) \
QueryPerformanceCounter(&time_sub_end); \
elapsed_time_sec = (time_sub_end.QuadPart-time_sub_start.QuadPart)/(double)(freq.QuadPart); \
printf(PERF_COUNT_REPORT_STR, funct_name, SUB, elapsed_time_sec); \
time_sub_start = time_sub_end;
#define PERF_COUNT_SUB(SUB) \
QueryPerformanceCounter(&time_sub_end); \
elapsed_time_sec = (time_sub_end.QuadPart - time_sub_start.QuadPart) / \
(double)(freq.QuadPart); \
printf(PERF_COUNT_REPORT_STR, funct_name, SUB, elapsed_time_sec); \
time_sub_start = time_sub_end;
#else
#define PERF_COUNT_SUB(SUB)
#endif
@ -112,21 +115,25 @@
time_sub_start = time_funct_start; \
time_sub_end = time_funct_start;
#define PERF_COUNT_END \
clock_gettime( CLOCK_MONOTONIC, &time_funct_end ); \
elapsed_time_sec = (time_funct_end.tv_sec - time_funct_start.tv_sec)*1.0 + (time_funct_end.tv_nsec - time_funct_start.tv_nsec)/1000000000.0; \
printf(PERF_COUNT_REPORT_STR, funct_name, "total", elapsed_time_sec);
#define PERF_COUNT_END \
clock_gettime(CLOCK_MONOTONIC, &time_funct_end); \
elapsed_time_sec = \
(time_funct_end.tv_sec - time_funct_start.tv_sec) * 1.0 + \
(time_funct_end.tv_nsec - time_funct_start.tv_nsec) / 1000000000.0; \
printf(PERF_COUNT_REPORT_STR, funct_name, "total", elapsed_time_sec);
#else
#define PERF_COUNT_START(FUNCT_NAME)
#define PERF_COUNT_END
#endif
#if PERF_COUNT_VERBOSE >= 3
#define PERF_COUNT_SUB(SUB) \
clock_gettime( CLOCK_MONOTONIC, &time_sub_end ); \
elapsed_time_sec = (time_sub_end.tv_sec - time_sub_start.tv_sec)*1.0 + (time_sub_end.tv_nsec - time_sub_start.tv_nsec)/1000000000.0; \
printf(PERF_COUNT_REPORT_STR, funct_name, SUB, elapsed_time_sec); \
time_sub_start = time_sub_end;
#define PERF_COUNT_SUB(SUB) \
clock_gettime(CLOCK_MONOTONIC, &time_sub_end); \
elapsed_time_sec = \
(time_sub_end.tv_sec - time_sub_start.tv_sec) * 1.0 + \
(time_sub_end.tv_nsec - time_sub_start.tv_nsec) / 1000000000.0; \
printf(PERF_COUNT_REPORT_STR, funct_name, SUB, elapsed_time_sec); \
time_sub_start = time_sub_end;
#else
#define PERF_COUNT_SUB(SUB)
#endif
@ -253,12 +260,12 @@ public:
// OpenCL implementation of Morphology (Hollow = Closed - Open)
static PIX* pixHollowCL(PIX *pixd, PIX *pixs, l_int32 close_hsize, l_int32 close_vsize, l_int32 open_hsize, l_int32 open_vsize, bool reqDataCopy);
static void pixGetLinesCL(PIX *pixd, PIX *pixs,
PIX** pix_vline, PIX** pix_hline,
PIX** pixClosed, bool getpixClosed,
l_int32 close_hsize, l_int32 close_vsize,
l_int32 open_hsize, l_int32 open_vsize,
l_int32 line_hsize, l_int32 line_vsize);
static void pixGetLinesCL(PIX *pixd, PIX *pixs, PIX **pix_vline,
PIX **pix_hline, PIX **pixClosed,
bool getpixClosed, l_int32 close_hsize,
l_int32 close_vsize, l_int32 open_hsize,
l_int32 open_vsize, l_int32 line_hsize,
l_int32 line_vsize);
//int InitOpenclAttr( OpenCLEnv * env );
//int ReleaseKernel( KernelEnv * env );
@ -279,34 +286,24 @@ public:
static void FreeOpenclDll();
#endif
inline static int AddKernelConfig( int kCount, const char *kName );
/* for binarization */
static int HistogramRectOCL(
const unsigned char *imagedata,
int bytes_per_pixel,
int bytes_per_line,
int left,
int top,
int width,
int height,
int kHistogramSize,
int *histogramAllChannels);
static int HistogramRectOCL(const unsigned char *imagedata,
int bytes_per_pixel, int bytes_per_line,
int left, int top, int width, int height,
int kHistogramSize, int *histogramAllChannels);
static int ThresholdRectToPixOCL(
const unsigned char* imagedata,
int bytes_per_pixel,
int bytes_per_line,
const int* thresholds,
const int* hi_values,
Pix** pix,
int rect_height,
int rect_width,
int rect_top,
int rect_left);
static int ThresholdRectToPixOCL(const unsigned char *imagedata,
int bytes_per_pixel, int bytes_per_line,
const int *thresholds,
const int *hi_values, Pix **pix,
int rect_height, int rect_width,
int rect_top, int rect_left);
static Pix * pixConvertRGBToGrayOCL( Pix *pix, float weightRed = 0.3, float weightGreen = 0.5, float weightBlue = 0.2 );
static Pix *pixConvertRGBToGrayOCL(Pix *pix, float weightRed = 0.3,
float weightGreen = 0.5,
float weightBlue = 0.2);
static ds_device getDeviceSelection();
static ds_device selectedDevice;

8
tessdata/configs/box.train.stderr
View File

@ -1,7 +1,7 @@
file_type .bl
#tessedit_use_nn F
textord_fast_pitch_test T
tessedit_single_match 0
file_type .bl
#tessedit_use_nn F
textord_fast_pitch_test T
tessedit_single_match 0
tessedit_zero_rejection T
tessedit_minimal_rejection F
tessedit_write_rep_codes F

2
textord/alignedblob.cpp
View File

@ -188,7 +188,7 @@ ScrollView* AlignedBlob::DisplayTabs(const char* window_name,
gsearch.StartFullSearch();
BLOBNBOX* bbox;
while ((bbox = gsearch.NextFullSearch()) != NULL) {
TBOX box = bbox->bounding_box();
const TBOX& box = bbox->bounding_box();
int left_x = box.left();
int right_x = box.right();
int top_y = box.top();

4
textord/bbgrid.cpp
View File

@ -231,7 +231,7 @@ Pix* GridReducedPix(const TBOX& box, int gridsize,
// Note that the Pix is used upside-down, with (0, 0) being the bottom-left.
Pix* TraceOutlineOnReducedPix(C_OUTLINE* outline, int gridsize,
ICOORD bleft, int* left, int* bottom) {
TBOX box = outline->bounding_box();
const TBOX& box = outline->bounding_box();
Pix* pix = GridReducedPix(box, gridsize, bleft, left, bottom);
int wpl = pixGetWpl(pix);
l_uint32* data = pixGetData(pix);
@ -257,7 +257,7 @@ Pix* TraceOutlineOnReducedPix(C_OUTLINE* outline, int gridsize,
// As TraceOutlineOnReducedPix above, but on a BLOCK instead of a C_OUTLINE.
Pix* TraceBlockOnReducedPix(BLOCK* block, int gridsize,
ICOORD bleft, int* left, int* bottom) {
TBOX box = block->bounding_box();
const TBOX& box = block->bounding_box();
Pix* pix = GridReducedPix(box, gridsize, bleft, left, bottom);
int wpl = pixGetWpl(pix);
l_uint32* data = pixGetData(pix);

2
textord/bbgrid.h
View File

@ -623,7 +623,7 @@ void BBGrid<BBC, BBC_CLIST, BBC_C_IT>::DisplayBoxes(ScrollView* tab_win) {
gsearch.StartFullSearch();
BBC* bbox;
while ((bbox = gsearch.NextFullSearch()) != NULL) {
TBOX box = bbox->bounding_box();
const TBOX& box = bbox->bounding_box();
int left_x = box.left();
int right_x = box.right();
int top_y = box.top();

20
textord/blkocc.h
View File

@ -72,12 +72,12 @@ ELISTIZEH (REGION_OCC)
Adapted from the following procedure so that it can be used in the bands
class in an include file...
BOOL8 range_in_band[
BOOL8 range_in_band[
range within band?
inT16 band_max,
inT16 band_min,
inT16 range_max,
inT16 range_min]
inT16 band_max,
inT16 band_min,
inT16 range_max,
inT16 range_min]
{
if ( (range_min >= band_min) && (range_max < band_max) )
return TRUE;
@ -91,12 +91,12 @@ inT16 range_min]
Adapted from the following procedure so that it can be used in the bands
class in an include file...
BOOL8 range_overlaps_band[
BOOL8 range_overlaps_band[
range crosses band?
inT16 band_max,
inT16 band_min,
inT16 range_max,
inT16 range_min]
inT16 band_max,
inT16 band_min,
inT16 range_max,
inT16 range_min]
{
if ( (range_max >= band_min) && (range_min < band_max) )
return TRUE;

Some files were not shown because too many files have changed in this diff Show More