implement improved camel case

modules/speller/default/speller_impl.cpp

@@ -214,24 +214,64 @@ namespace aspeller {
     return NULL;
   }
 
+
+  /**
+    Returns false if camel case checking is not enabled.
+    Otherwise returns true when the provided string is acceptable camel case and false when not.
+  */
+  bool SpellerImpl::check_camel(const char * str, size_t len, CheckInfo & ci) {
+    if(!camel_case_) {
+      return false;
+    }
+
+    CompoundWord cw = lang_->split_word(str, len, true);
+    WordEntry we;
+    bool notSingle = false;
+    do {
+      size_t sz = cw.word_len();
+      char word[sz+1];
+      memcpy(word, cw.word, sz);
+      word[sz] = '\0';
+      if (!check_simple(word, we)) {
+        return false;
+      }
+      cw = lang_->split_word(cw.rest, cw.rest_len(), true);
+      if (cw.word_len()) {
+        notSingle = true;
+      }
+    } while (!cw.empty());
+    if (notSingle) {
+      ci.word = str;
+      ci.compound = true;
+      return true;
+    }
+    return false;
+  }
+
+
   PosibErr<bool> SpellerImpl::check(char * word, char * word_end, 
                                     /* it WILL modify word */
                                     bool try_uppercase,
                                     unsigned run_together_limit,
                                     CheckInfo * ci, CheckInfo * ci_end,
                                     GuessInfo * gi, CompoundInfo * cpi)
   {
+    clear_check_info(*ci);
+    if (check_camel(word, word_end - word, *ci)) {
+      return true;
+    }
+
     clear_check_info(*ci);
     bool res = check_runtogether(word, word_end, try_uppercase, run_together_limit, ci, ci_end, gi);
     if (res) return true;
-    
+
     CompoundWord cw = lang_->split_word(word, word_end - word, camel_case_);
     if (cw.single()) return false;
     bool ok = true;
     CheckInfo * ci_prev = NULL;
     do {
       unsigned len = cw.word_len();
-      
+
       char save = word[len];
       word[len] = '\0';
       CheckInfo * ci_last = check_runtogether(word, word + len, try_uppercase, run_together_limit, ci, ci_end, gi);
@@ -255,7 +295,7 @@ namespace aspeller {
 
       if (cpi)
         cpi->count++;
-      
+
       if (ci_prev) {
         ci_prev->compound = true;
         ci_prev->next = ci;
@@ -267,42 +307,13 @@ namespace aspeller {
         if (cpi) cpi->count = 0;
         return false;
       }
-      
+
       word = word + cw.rest_offset();
       cw = lang_->split_word(cw.rest, cw.rest_len(), camel_case_);
-      
+
     } while (!cw.empty());
-    
+
     return ok;
-
-    // for (;;) {
-    //   cw = lang_->split_word(cw.rest, cw.rest_len(), camel_case_);
-    //   if (cw.empty()) break;
-    //   if (ci + 1 >= ci_end) {
-    //     if (cpi) cpi->count = 0;
-    //     return false;
-    //   }
-    //   if (cpi) cpi->count++;
-    //   len = cw.word_len();
-    //   save = word[len];
-    //   word[len] = '\0';
-    //   ci_last = check_runtogether(word, word + len, try_uppercase, run_together_limit, ci + 1, ci_end, gi);
-    //   word[len] = save;
-    //   ci->compound = true;
-    //   ci->next = ci + 1;
-    //   if (ci_last) {
-    //     ci = ci_last;
-    //   } else if (cpi) {
-    //     ok = false;
-    //     ci->word.str = word;
-    //     ci->word.len = len;
-    //     ci->incorrect = true;
-    //     cpi->incorrect_count++;
-    //   } else {
-    //     return false;
-    //   }
-    //   word = word + cw.rest_offset();
-    // }
   }
 
   //////////////////////////////////////////////////////////////////////

modules/speller/default/speller_impl.hpp

@@ -125,6 +125,8 @@ namespace aspeller {
       return check(MutableString(&w.front(), sz));
     }
 
+    bool check_camel(const char * str, size_t len, CheckInfo & ci);
+
     CheckInfo * check_runtogether(char * word, char * word_end, /* it WILL modify word */
                                   bool try_uppercase,
                                   unsigned run_together_limit,

test/Makefile

@@ -8,8 +8,8 @@ ifdef SLOPPY
   EXTRA_CONFIG_FLAGS += --enable-sloppy-null-term-strings
 endif
 
-.PHONY: all prep sanity filter-test suggest wide cxx_warnings
-all: prep sanity filter-test suggest wide cxx_warnings
+.PHONY: all prep sanity camel-case filter-test suggest wide cxx_warnings
+all: prep sanity camel-case filter-test suggest wide cxx_warnings
 	cat test-res
 
 # warning-settings.mk defines EXTRA_CXXFLAGS
@@ -28,6 +28,10 @@ sanity: prep
 	./sanity
 	echo "all ok (sanity)" >> test-res
 
+camel-case: prep
+	sh ./camel-case
+	echo "all ok (camel-case)" >> test-res
+
 filter-test: prep
 	./filter-test "${ASPELL_WRAP} ${ASPELL}" < markdown.dat
 	echo "all ok (markdown filter-test)" >> test-res

test/camel-case

@@ -0,0 +1,22 @@
+#!/bin/sh
+
+set -e
+set -x
+
+echo 'itIsOKForACamelCaseWordToContainNumerousCOMPONENTS' | ${ASPELL} --camel-case -d en_US -a > tmp/camel-case-res
+if cat tmp/camel-case-res | fgrep -x '-'; then
+    echo "pass"
+else
+    echo "fail:"
+    cat tmp/camel-case-res
+    exit 1
+fi
+
+echo 'EvenIfTheFirstCOMPONENTBeginsWithACapitalLetter' | ${ASPELL} --camel-case -d en_US -a > tmp/camel-case-res
+if cat tmp/camel-case-res | fgrep -x '-'; then
+    echo "pass"
+else
+    echo "fail:"
+    cat tmp/camel-case-res
+    exit 1
+fi