Merge pull request #499 from dehilsterlexis/NLP-ENGINE-422

NLP-ENGINE-422 Added ambig and meaning logic

.vscode/launch.json

@@ -5,11 +5,11 @@
     "version": "0.2.0",
     "configurations": [
         {
-            "name": "Corp Test (linux)",
+            "name": "Ambig (linux)",
             "type": "cppdbg",
             "request": "launch",
             "program": "/home/dehilster/nlp-engine/bin/nlp",
-            "args": ["-ANA","/home/dehilster/analyzers/Test","-WORK","/home/dehilster/nlp-engine/","/home/dehilster/analyzers/Test/input/após.txt","-DEV"],
+            "args": ["-ANA","/home/dehilster/analyzers/Ambig","-WORK","/home/dehilster/nlp-engine/","/home/dehilster/analyzers/Ambig/input/após.txt","-DEV"],
             "stopAtEntry": false,
             "cwd": "/home/dehilster/nlp-engine/",
             "environment": [],
@@ -23,6 +23,25 @@
                 }
             ]
         },
+        {
+            "name": "Test (win)",
+            "type": "cppvsdbg",
+            "request": "launch",
+            "program": "C:\\dev\\nlp-engine\\bin\\Debug\\nlp.exe",
+            "args": ["-ANA","C:\\dev\\analyzers\\Test","-WORK","C:\\dev\\nlp-engine","C:\\dev\\analyzers\\Test\\input\\test.txt","-DEV"],
+            "stopAtEntry": false,
+            "cwd": "C:\\dev\\nlp-engine",
+            "environment": [],
+            "externalConsole": false,
+            "MIMode": "gdb",
+            "setupCommands": [
+                {
+                    "description": "Enable pretty-printing for gdb",
+                    "text": "-enable-pretty-printing",
+                    "ignoreFailures": true
+                }
+            ]
+        },
         {
             "name": "Parse EN-US (win)",
             "type": "cppvsdbg",

cs/libconsh/cg.cpp

@@ -53,6 +53,7 @@ All rights reserved.
 #include "cmd.h"
 #include "dyn.h"					// 06/29/00 AM.
 #include "lite/dir.h"
+#include "lite/io.h"
 
 #include "prim/unicu.h"
 using namespace unicu;
@@ -628,37 +629,41 @@ _stprintf(path, _T("%s%c%s%c%s"), getAppdir(),DIR_CH, kbdir,DIR_CH, dir);
 _TCHAR infile[MAXPATH*2];
 _TCHAR *suff;
 suff = _T("kb");		// Kb file suffix.
-std::vector<std::filesystem::path> files;
+std::vector<std::filesystem::path> kbfiles;
+std::vector<std::filesystem::path> dictfiles;
 bool kbLoaded = false;
 bool bound = false;
 
-if (openDict(files)) {
-	_stprintf(infile, _T("%s%chier.%s"), path,DIR_CH, suff);
-	if (!readFile(infile))
-		return false;
-	con_add_root(this);
-	bind_sys(this);
-	bound = true;
-	readDicts(files);
-	outputTime(_T("[READ dict files time="),s_time);
-	s_time = clock();
-	kbLoaded = true;
-}
+openKBB(kbfiles);
+openDict(dictfiles);
 
-if (openKBB(files)) {
+if (kbfiles.size() > 0) {
 	if (!bound) {
 		_stprintf(infile, _T("%s%chier.%s"), path,DIR_CH, suff);
 		if (!readFile(infile))
 			return false;
 		bind_sys(this);
 		con_add_root(this);
 	}
-	readKBBs(files);
+	readKBBs(kbfiles);
 	outputTime(_T("[READ kbb files time="),s_time);
 	s_time = clock();
 	kbLoaded = true;
 }
 
+if (dictfiles.size() > 0) {
+	_stprintf(infile, _T("%s%chier.%s"), path,DIR_CH, suff);
+	if (!readFile(infile))
+		return false;
+	con_add_root(this);
+	bind_sys(this);
+	bound = true;
+	readDicts(dictfiles,kbfiles);
+	outputTime(_T("[READ dict files time="),s_time);
+	s_time = clock();
+	kbLoaded = true;
+}
+
 if (!kbLoaded) {
 
 	// Using a master take file for readin kb.			// 07/01/03 AM.
@@ -3481,6 +3486,31 @@ if (!(word = kbm_->dict_find_word(str)) )
 return word;
 }
 
+CONCEPT *CG::matchDictKB(std::string dictFilename, std::vector<std::filesystem::path> kbfiles) {
+	if (kbfiles.size() == 0) return NULL;
+	std::vector<std::filesystem::path>::iterator ptr;
+
+	_TCHAR buff[MAXSTR], buffkb[MAXSTR];
+	_TCHAR *head, *headkb;
+	CONCEPT *con, *dictcon;
+	_tcscpy(buff, dictFilename.c_str());
+	file_head(buff, head);
+
+    for (ptr = kbfiles.begin(); ptr < kbfiles.end(); ptr++) {
+		_tcscpy(buffkb, ptr->string().c_str());
+		file_head(buff, headkb);
+
+        if (!_tcscmp(head,headkb)) {
+			con = findRoot();
+			dictcon = findConcept(con,"dictionary");
+			if (dictcon) {
+				return dictcon;
+			}
+		}
+	}
+	return NULL;
+}
+
 bool CG::openDict(std::vector<std::filesystem::path>& files) {
 	bool found = false;
 	files.clear();
@@ -3501,16 +3531,16 @@ bool CG::openDict(std::vector<std::filesystem::path>& files) {
 	return found;
 }
 
-bool CG::readDicts(std::vector<std::filesystem::path> files) {
+bool CG::readDicts(std::vector<std::filesystem::path> files, std::vector<std::filesystem::path> kbfiles) {
 	std::vector<std::filesystem::path>::iterator ptr;
 	if (files.size() == 0) return false;
     for (ptr = files.begin(); ptr < files.end(); ptr++) {
-        readDict(ptr->string());
+        readDict(ptr->string(), kbfiles);
 	}
 	return true;
 }
 
-bool CG::readDict(std::string file) {
+bool CG::readDict(std::string file, std::vector<std::filesystem::path> kbfiles) {
 	bool dirty;
 	CONCEPT *wordCon, *parentCon;
 	_TCHAR buf[MAXMSG];
@@ -3520,6 +3550,8 @@ bool CG::readDict(std::string file) {
 	int lineCount = 0;
 	DICT_CALL caller;
 
+	CONCEPT *ambigKB = matchDictKB(file, kbfiles);
+
 	// For error printouts
 	std::size_t botDirPos = file.find_last_of(DIR_CH);
 	std::string filename = file.substr(botDirPos+1, file.length()-2);
@@ -3705,6 +3737,16 @@ bool CG::readDict(std::string file) {
 				} else {
 					parentCon = wordCon;
 				}
+				// add ambigous concept to word concept
+				if (ambigKB) {
+					CONCEPT *kbcon = findConcept(ambigKB,token);
+					if (kbcon) {
+						CONCEPT *ambigCon = NULL;
+						if (!findVal(wordCon,"meaning",ambigCon)) {
+							addVal(wordCon,_T("meaning"),kbcon);
+						}
+					}
+				}
 
 			} else if (cc == '=') {
 				int donothing = 1;

include/Api/consh/cg.h

@@ -28,6 +28,7 @@ All rights reserved.
 
 #define FNAMESIZ 256
 #define NAMESIZ  256
+#define MAXPATH  2048
 //#define PATHSIZ  512
 // SIZES	// 03/25/13 AM.
 // Messes up concept display in Attribute Editor.	// 09/14/13 AM.
@@ -212,9 +213,10 @@ class LIBCONSH_API CG
 	// Find dictionary concept for given string.						// 06/29/03 AM.
 	CONCEPT *findWordConcept(_TCHAR*);								// 06/29/03 AM.
 
+	CONCEPT *matchDictKB(std::string dictFilename, std::vector<std::filesystem::path> kbfiles);
 	bool openDict(std::vector<std::filesystem::path>& files);
-	bool readDicts(std::vector<std::filesystem::path> files);
-	bool readDict(std::string file);
+	bool readDicts(std::vector<std::filesystem::path> files, std::vector<std::filesystem::path> kbfiles);
+	bool readDict(std::string file, std::vector<std::filesystem::path> kbfiles);
 
 	bool openKBB(std::vector<std::filesystem::path>& files);
 	bool readKBBs(std::vector<std::filesystem::path> files);
@@ -608,7 +610,7 @@ class LIBCONSH_API CG
 
 private:
 	_TCHAR appdir_[FNAMESIZ];		// Base directory path.
-	_TCHAR kbdir_[FNAMESIZ];
+	_TCHAR kbdir_[MAXPATH];
 
 	std::_t_ifstream allDictStream_;
 

include/Api/lite/Arun.h

@@ -1283,6 +1283,7 @@ class LITE_API Arun
 	static RFASem *pnprev(Nlppp*,NODE*);								// 01/08/01 AM.
 	static RFASem *pnprev(Nlppp*,RFASem*);								// 04/29/01 AM.
 
+	static bool pnremoveval(Nlppp*,NODE*,_TCHAR*);
 	static _TCHAR *pnrename(Nlppp*,NODE*,_TCHAR*);						// 01/08/01 AM.
 	static _TCHAR *pnrename(Nlppp*,RFASem*,_TCHAR*);						// 04/28/01 AM.
 	static _TCHAR *pnrename(Nlppp*,NODE*,RFASem*);						// 04/28/01 AM.

include/Api/lite/dir.h

@@ -17,11 +17,6 @@ All rights reserved.
 #include <filesystem>
 #include <regex>
 
-LITE_API bool plain_file(struct _tfinddata_t &fileinfo);
-LITE_API bool     is_dir(struct _tfinddata_t &fileinfo);
-LITE_API bool is_file(_TCHAR *name);
-LITE_API bool path_exists(_TCHAR *name);
-
 LITE_API void rm_path(
 	_TCHAR *infile,		// The path (file, directory, or directory tree).
 	bool tree			// True if removing subdirectories recursively.

include/Api/lite/io.h

@@ -0,0 +1,94 @@
+/*******************************************************************************
+Copyright (c) 2001-2010 by Text Analysis International, Inc.
+All rights reserved.
+********************************************************************************
+*
+* NAME:	IO.H
+* FILE:	lite\io.h
+* CR:		10/06/98 AM.
+* SUBJ:	Declares for I/O functions.
+*
+*******************************************************************************/
+
+//#include "fstream.h"
+
+void copy_file(const _TCHAR *, const _TCHAR *);
+bool file_exists(const _TCHAR *iname);				// 12/14/98 AM.
+void file_to_buffer(const _TCHAR *, _TCHAR *,
+	/*UP*/ long &len															// 05/28/00 /AM.
+	);
+_TCHAR *pretty_char(_TCHAR);
+_TCHAR *pretty_str(
+	_TCHAR *str,				// String to be prettified.
+	_TCHAR *buf,				// Buffer for placing prettified string.
+	long size
+	);
+_TCHAR *c_char(_TCHAR,_TCHAR*);												// 05/10/00 AM.
+_TCHAR *c_str(																	// 05/10/00 AM.
+	_TCHAR *str,				// String to be prettified.
+	_TCHAR *buf,				// Buffer for placing prettified string.
+	long size			// Buffer size. (-1 means count not used.)
+	);
+
+
+void read_file(
+	_TCHAR *fname,		// The filename
+	/*UP*/
+	long &len,			// Length of file.
+	_TCHAR* &buf			// Buffer to create.
+	)
+	;
+
+_TCHAR *next_token(/*DU*/ _TCHAR* &buf, bool &eol, _TCHAR *comment);
+
+bool fix_file_name(
+	_TCHAR *file,				// Buffer big enough to hold extension.
+	_TCHAR *suff				// File name extension needed.
+	)
+	;
+
+bool eq_str_range(_TCHAR *str, _TCHAR *ptr, long start, long end);
+
+_TCHAR *make_str(_TCHAR *str);
+_TCHAR *make_str(_TCHAR *str, long len);
+
+bool strcat_e(
+	/*DU*/ _TCHAR* &ptr,
+	/*DN*/ _TCHAR *str,
+	/*DU*/ long &count
+	);
+
+
+bool strncat_e(
+	/*DU*/ _TCHAR* &ptr,
+	/*DN*/ _TCHAR *str,
+	/*DN*/ long len,
+	/*DU*/ long &count
+	);
+
+
+bool file_name(															// 12/24/99 AM.
+	_TCHAR *file,				// Buffer with full file string.
+	/*UP*/
+	_TCHAR* &fname			// Pointer to the name in buffer.
+	);
+bool file_path(															// 12/24/99 AM.
+	_TCHAR *file,				// Buffer with full file string.
+	/*UP*/
+	_TCHAR* &fpath			// Pointer to the path in buffer.
+	);
+bool file_parent(
+	_TCHAR *file,				// Buffer with full file string.
+	/*UP*/
+	_TCHAR* &fpath			// Pointer to the path in buffer.
+	);
+bool file_head(															// 12/24/99 AM.
+	_TCHAR *file,				// Buffer with full file string.
+	/*UP*/
+	_TCHAR* &fhead			// Pointer to the file head in buffer.
+	);
+bool file_tail(															// 12/24/99 AM.
+	_TCHAR *file,				// Buffer with full file string.
+	/*UP*/
+	_TCHAR* &ftail			// Pointer to tail in buffer.
+	);

lite/dicttok.cpp

@@ -919,7 +919,11 @@ inline bool DICTTok::findAttrs(Node<Pn> *node, CONCEPT *con, _TCHAR *str, bool i
 			} else if (cg_->isValNum(vals)) {
 				long long num = 0L;
 				cg_->popVal(vals,num);
-				replaceNum(node,strattr,num);	
+				replaceNum(node,strattr,num);
+			} else if (cg_->isValCon(vals)) {
+				CONCEPT *con = NULL;
+				cg_->popVal(vals,con);
+				replaceCon(node,strattr,con);
 			} else
 				cg_->nextVal(vals);
 		}
@@ -991,6 +995,20 @@ return Ivar::nodeReplaceval(pn, name, val);
 }
 
 
+inline bool DICTTok::replaceCon(
+	Node<Pn> *node,
+	_TCHAR *name,	// variable name.
+	CONCEPT *con
+	)
+{
+if (!node)
+	return false;
+
+Pn *pn = node->getData();
+
+return Ivar::nodeReplaceval(pn, name, con);
+}
+
 /********************************************
 * FN:		POPSVAL
 * CR:		08/01/11 AM.

lite/dicttok.h

@@ -122,6 +122,11 @@ class DICTTok : public Algo
 		_TCHAR *name,	// variable name.
 		_TCHAR *str
 		);
+	inline bool replaceCon(
+		Node<Pn> *node,
+		_TCHAR *name,
+		CONCEPT *con
+		);
 	inline _TCHAR *popsval(
 		VAL *val
 		);