Changeset 2487 for trunk/gsdl/src/phind/generate/suffix.cpp

Timestamp:

2001-06-01T14:51:29+12:00 (23 years ago)

Author:

sjboddie

Message:

Changes to get phind working under windows

File:

• trunk/gsdl/src/phind/generate/suffix.cpp (modified) (16 diffs)

trunk/gsdl/src/phind/generate/suffix.cpp

@@ -28 +28 @@
 
 #include <assert.h>
-#include <fstream.h>
-#include <iostream.h>
 #include <math.h>
 #include <stdio.h>
@@ -35 +33 @@
 #include <string.h>
 
-#include <algo.h>
-#include <heap.h>
-#include <vector.h>
+#if defined(GSDL_USE_IOS_H)
+#  include <fstream.h>
+#  include <iostream.h>
+#else
+#  include <fstream>
+#  include <iostream>
+#endif
+
+#if defined(GSDL_USE_STL_H)
+#  if defined(GSDL_USE_ALGO_H)
+#    include <algo.h>
+#  else
+#    include <algorithm.h>
+#  endif
+#  include <vector.h>
+#else
+#  include <algorithm>
+#  include <vector>
+#endif
+#include <stl_heap.h>
+
 
 #include "suffix.h"
 #include "phrase.h"
-
-
 
 // Global variables declared in suffix.h
@@ -138 +152 @@
 
   // Create the suffix & prefix arrays
-  suffixArray = new (symbol *)[inputLength];
-  prefixArray = new (symbol *)[inputLength];
-  suffixCheck = new (check)[inputLength];
-  prefixCheck = new (check)[inputLength];
+  suffixArray = new symbol *[inputLength];
+  prefixArray = new symbol *[inputLength];
+  suffixCheck = new check[inputLength];
+  prefixCheck = new check[inputLength];
   if (prefixCheck == NULL) {
     cerr << "Suffix error: not enough memory to hold " << inputLength
@@ -169 +183 @@
   // each phrase occurs in each document.  The number of documents in
   // which a phrase occurs is stored in df.
-  frequency documentFrequency[numberOfDocuments];
+  frequency *documentFrequency = new frequency[numberOfDocuments];
   frequency df;
 
   // documentArray will be searched in order to discover which document
   // each phrase occurs in.
-  documentArray = new (symbol *)[numberOfDocuments];  
+  documentArray = new symbol *[numberOfDocuments];  
 
   // Discover all the DOCUMENTSTART symbols and store as a phrase
@@ -250 +264 @@
   // Iterate over the different symbols by working through the suffix array
   vector<Phrase> result;
-  cellindex i = 0;
+  cellindex ij = 0;
   char *tmpString;
 
-  while (i < inputLength) {
+  while (ij < inputLength) {
 
     // make a new phrase of length 1
-    p = Phrase(suffixArray[i], 1, SUFFIX);
-    p.findFirstAndLastSuffix(i, inputLength-1);
-
-    // cout << "cell " << i << " - " << p.toString() << endl;
+    p = Phrase(suffixArray[ij], 1, SUFFIX);
+    p.findFirstAndLastSuffix(ij, inputLength-1);
+
+    // cout << "cell " << ij << " - " << p.toString() << endl;
 
     // We ignore this symbol if it occurs only once, if it is a delimiter,
@@ -271 +285 @@
     // it explodes the size of the indexes.  So: would it be useful?  
     if (!((p.suffixFrequency <= 1) ||
-      // (*suffixArray[i] != 23054) ||
-      (*suffixArray[i] <= LASTDELIMITER) ||
-      ((phraseMode == STOPWORDS) && (*suffixArray[i] <= lastStopSymbol)))) {
+      // (*suffixArray[ij] != 23054) ||
+      (*suffixArray[ij] <= LASTDELIMITER) ||
+      ((phraseMode == STOPWORDS) && (*suffixArray[ij] <= lastStopSymbol)))) {
 
       // Get minimal expansions of the phrase
@@ -281 +295 @@
     
     // Remember that we have expanded this phrase
-    rememberThisPhrase(i, 1);
+    rememberThisPhrase(ij, 1);
 
     // write the phrase text
     tmpString = p.toString();
-    phraseData << i << "-1:" << tmpString << ":" << p.suffixFrequency << ":"
+    phraseData << ij << "-1:" << tmpString << ":" << p.suffixFrequency << ":"
            << result.size() << ":";
     delete [] tmpString;
 
     // write the results
-    for (cellcount i = 0; i < result.size(); i++) {
-      if (i) {
+    for (cellcount k = 0; k < result.size(); k++) {
+      if (k) {
         phraseData << ",";
       }
-      phraseData << result[i].firstSuffixIndex << "-" << result[i].length;
-      outPhrase << result[i].firstSuffixIndex << " " << result[i].length << endl;
+      phraseData << result[k].firstSuffixIndex << "-" << result[k].length;
+      outPhrase << result[k].firstSuffixIndex << " " << result[k].length << endl;
       outPhraseCounter++;
     }
@@ -305 +319 @@
 
     // write the documents
-    for (cellcount i = 0, first = 1; i < numberOfDocuments; i++) {
-      if (documentFrequency[i]) {
+    for (cellcount m = 0, first = 1; m < numberOfDocuments; m++) {
+      if (documentFrequency[m]) {
         if (first) {
           first = 0;
@@ -315 +329 @@
         // N documents from 0 to N-1, but later they'll be 1-N.  Thus we
         // add 1 to the document id when we output it.
-        phraseData << "d" << (i+1);
+        phraseData << "d" << (m+1);
         // Next, output the frequency with which the document occurs, but
         // only if it is > 1.
-        if (documentFrequency[i] > 1) {
-          phraseData << "," << documentFrequency[i];
+        if (documentFrequency[m] > 1) {
+          phraseData << "," << documentFrequency[m];
         }
       }
@@ -338 +352 @@
       }
     }
-   i = p.lastSuffixIndex + 1;
+   ij = p.lastSuffixIndex + 1;
   }
   outPhrase.close();
@@ -472 +486 @@
   deletePhraseMemory();
 
+  delete [] documentFrequency;
   delete [] symbols;
   delete [] suffixArray;
@@ -548 +563 @@
       suffixCheck[i] = c.length;
     }
-    for (cellcount i = c.firstPrefixIndex; i <= c.lastPrefixIndex; i++) {
-      prefixCheck[i] = c.length;
+    for (cellcount ik = c.firstPrefixIndex; ik <= c.lastPrefixIndex; ik++) {
+      prefixCheck[ik] = c.length;
     }
       }
@@ -578 +593 @@
       suffixCheck[i] = c.length;
     }
-    for (cellcount i = c.firstPrefixIndex; i <= c.lastPrefixIndex; i++) {
-      prefixCheck[i] = c.length;
+    for (cellcount ijk = c.firstPrefixIndex; ijk <= c.lastPrefixIndex; ijk++) {
+      prefixCheck[ijk] = c.length;
     }
 
@@ -701 +716 @@
     cout << "Allocating symbol arrays for " << inputLength << " symbols" << endl;
   }
-  symbols = new (symbol)[inputLength];
+  symbols = new symbol[inputLength];
   if (symbols == NULL) {
     cerr << "Suffix error: not enough memory to hold " << inputLength
@@ -757 +772 @@
 
   // search for the document in which each occurence of the phrase is found
-  for (cellcount i = p.firstSuffixIndex; i <= p.lastSuffixIndex; i++) {
+  for (cellcount j = p.firstSuffixIndex; j <= p.lastSuffixIndex; j++) {
     
-    // cout << "looking for phrase at suffixArray[" << i << "]\n";
+    // cout << "looking for phrase at suffixArray[" << j << "]\n";
     
-    target = suffixArray[i];
+    target = suffixArray[j];
     begin = 0;
     end = numberOfDocuments - 1;
@@ -864 +879 @@
 void initialisePhraseMemory() {
 
-  phraseMemory = new (unsigned char)[inputLength];
+  phraseMemory = new unsigned char[inputLength];
 
   // to begin with, everything is empty