[LO extension] extensive optimization to realize full text check; sol…

…ves #194
languagetool-org · Jul 2, 2019 · 92270c5 · 92270c5
1 parent ef48d02
commit 92270c5
Show file tree

Hide file tree

Showing 23 changed files with 428 additions and 56 deletions.
diff --git a/languagetool-core/src/main/java/org/languagetool/rules/AbstractFillerWordsRule.java b/languagetool-core/src/main/java/org/languagetool/rules/AbstractFillerWordsRule.java
@@ -189,4 +189,9 @@ else if ((!isDirectSpeech || minPercent == 0) && !token.isWhitespace() && !token
     return toRuleMatchArray(ruleMatches);
   }
 
+  @Override
+  public int minToCheckParagraph() {
+    return 0;
+  }
+
 }
diff --git a/languagetool-core/src/main/java/org/languagetool/rules/AbstractStyleRepeatedWordRule.java b/languagetool-core/src/main/java/org/languagetool/rules/AbstractStyleRepeatedWordRule.java
@@ -319,4 +319,9 @@ public RuleMatch[] match(List<AnalyzedSentence> sentences) throws IOException {
     return toRuleMatchArray(ruleMatches);
   }
 
+  @Override
+  public int minToCheckParagraph() {
+    return maxDistanceOfSentences;
+  }
+
 }
diff --git a/languagetool-core/src/main/java/org/languagetool/rules/AbstractWordCoherencyRule.java b/languagetool-core/src/main/java/org/languagetool/rules/AbstractWordCoherencyRule.java
@@ -93,4 +93,9 @@ public RuleMatch[] match(List<AnalyzedSentence> sentences) {
     return toRuleMatchArray(ruleMatches);
   }
 
+  @Override
+  public int minToCheckParagraph() {
+    return -1;
+  }
+
 }
diff --git a/languagetool-core/src/main/java/org/languagetool/rules/EmptyLineRule.java b/languagetool-core/src/main/java/org/languagetool/rules/EmptyLineRule.java
@@ -86,4 +86,9 @@ public org.languagetool.rules.RuleMatch[] match(List<AnalyzedSentence> sentences
     return toRuleMatchArray(ruleMatches);
   }
 
+  @Override
+  public int minToCheckParagraph() {
+    return 1;
+  }
+
 }
diff --git a/languagetool-core/src/main/java/org/languagetool/rules/GenericUnpairedBracketsRule.java b/languagetool-core/src/main/java/org/languagetool/rules/GenericUnpairedBracketsRule.java
@@ -293,4 +293,9 @@ private String findCorrespondingSymbol(String symbol) {
     }
   }
 
+  @Override
+  public int minToCheckParagraph() {
+    return -1;
+  }
+
 }
diff --git a/languagetool-core/src/main/java/org/languagetool/rules/LongParagraphRule.java b/languagetool-core/src/main/java/org/languagetool/rules/LongParagraphRule.java
@@ -146,4 +146,9 @@ public RuleMatch[] match(List<AnalyzedSentence> sentences) throws IOException {
     return toRuleMatchArray(ruleMatches);
   }
 
+  @Override
+  public int minToCheckParagraph() {
+    return 0;
+  }
+
 }
diff --git a/languagetool-core/src/main/java/org/languagetool/rules/MultipleWhitespaceRule.java b/languagetool-core/src/main/java/org/languagetool/rules/MultipleWhitespaceRule.java
@@ -93,4 +93,9 @@ public RuleMatch[] match(List<AnalyzedSentence> sentences) {
     return toRuleMatchArray(ruleMatches);
   }
 
+  @Override
+  public int minToCheckParagraph() {
+    return 0;
+  }
+
 }
diff --git a/languagetool-core/src/main/java/org/languagetool/rules/ParagraphRepeatBeginningRule.java b/languagetool-core/src/main/java/org/languagetool/rules/ParagraphRepeatBeginningRule.java
@@ -140,5 +140,10 @@ public RuleMatch[] match(List<AnalyzedSentence> sentences) throws IOException {
     return toRuleMatchArray(ruleMatches);
   }
 
+  @Override
+  public int minToCheckParagraph() {
+    return 1;
+  }
+
 }  
 
diff --git a/languagetool-core/src/main/java/org/languagetool/rules/PunctuationMarkAtParagraphEnd.java b/languagetool-core/src/main/java/org/languagetool/rules/PunctuationMarkAtParagraphEnd.java
@@ -138,4 +138,9 @@ public RuleMatch[] match(List<AnalyzedSentence> sentences) throws IOException {
     return toRuleMatchArray(ruleMatches);
   }
 
+  @Override
+  public int minToCheckParagraph() {
+    return 0;
+  }
+
 }
diff --git a/languagetool-core/src/main/java/org/languagetool/rules/ReadabilityRule.java b/languagetool-core/src/main/java/org/languagetool/rules/ReadabilityRule.java
@@ -331,5 +331,10 @@ public RuleMatch[] match(List<AnalyzedSentence> sentences) throws IOException {
       return toRuleMatchArray(new ArrayList<>());
     }
   }
+
+  @Override
+  public int minToCheckParagraph() {
+    return 0;
+  }
 
 }
diff --git a/languagetool-core/src/main/java/org/languagetool/rules/SentenceWhitespaceRule.java b/languagetool-core/src/main/java/org/languagetool/rules/SentenceWhitespaceRule.java
@@ -88,5 +88,10 @@ public RuleMatch[] match(List<AnalyzedSentence> sentences) throws IOException {
     }
     return toRuleMatchArray(ruleMatches);
   }
+
+  @Override
+  public int minToCheckParagraph() {
+    return 0;
+  }
 
 }
diff --git a/languagetool-core/src/main/java/org/languagetool/rules/TextLevelRule.java b/languagetool-core/src/main/java/org/languagetool/rules/TextLevelRule.java
@@ -67,5 +67,18 @@ public int estimateContextForSureMatch() {
   public final RuleMatch[] match(AnalyzedSentence sentence) throws IOException {
     throw new RuntimeException("Not implemented for a text-level rule");
   }
+
+  /**
+   * Gives back the minimum number of paragraphs to check to give back a correct result
+   * only used by LO office extension
+   * return n;
+   * n == -1  --> need to check full text (use only if really needed / bad performance)
+   *              examples: AbstractWordCoherencyRule, GenericUnpairedBracketsRule, ...
+   * n == 0   --> need only to check the current paragraph
+   *              examples: MultipleWhitespaceRule, LongParagraphRule, ...
+   * n >= 1   --> need only to check n paragraphs around the current paragraph
+   *              examples: ParagraphRepeatBeginningRule (n == 1), WordRepeatBeginningRule (n == 2), ...
+   */
+  public abstract int minToCheckParagraph();
 
 }
diff --git a/languagetool-core/src/main/java/org/languagetool/rules/UppercaseSentenceStartRule.java b/languagetool-core/src/main/java/org/languagetool/rules/UppercaseSentenceStartRule.java
@@ -188,4 +188,9 @@ private boolean isSentenceEnd(String word) {
   private boolean isQuoteStart(String word) {
     return StringUtils.equalsAny(word, "\"", "'", "„", "»", "«", "“", "‘");
   }
+
+  @Override
+  public int minToCheckParagraph() {
+    return 0;
+  }
 }
diff --git a/languagetool-core/src/main/java/org/languagetool/rules/WhiteSpaceBeforeParagraphEnd.java b/languagetool-core/src/main/java/org/languagetool/rules/WhiteSpaceBeforeParagraphEnd.java
@@ -89,4 +89,9 @@ public RuleMatch[] match(List<AnalyzedSentence> sentences) throws IOException {
     return toRuleMatchArray(ruleMatches);
   }
 
+  @Override
+  public int minToCheckParagraph() {
+    return 0;
+  }
+
 }
diff --git a/languagetool-core/src/main/java/org/languagetool/rules/WordRepeatBeginningRule.java b/languagetool-core/src/main/java/org/languagetool/rules/WordRepeatBeginningRule.java
@@ -107,4 +107,9 @@ public RuleMatch[] match(List<AnalyzedSentence> sentences) throws IOException {
     return toRuleMatchArray(ruleMatches);
   }
 
+  @Override
+  public int minToCheckParagraph() {
+    return 2;
+  }
+
 }
diff --git a/...ol-language-modules/de/src/main/java/org/languagetool/rules/de/CompoundCoherencyRule.java b/...ol-language-modules/de/src/main/java/org/languagetool/rules/de/CompoundCoherencyRule.java
@@ -135,4 +135,9 @@ private String getLemma(AnalyzedTokenReadings atr) {
     }
   }
 
+  @Override
+  public int minToCheckParagraph() {
+    return -1;
+  }
+
 }
diff --git a/...ool-language-modules/de/src/main/java/org/languagetool/rules/de/DuUpperLowerCaseRule.java b/...ool-language-modules/de/src/main/java/org/languagetool/rules/de/DuUpperLowerCaseRule.java
@@ -117,4 +117,9 @@ public RuleMatch[] match(List<AnalyzedSentence> sentences) throws IOException {
     return toRuleMatchArray(ruleMatches);
   }
 
+  @Override
+  public int minToCheckParagraph() {
+    return -1;
+  }
+
 }
diff --git a/...uagetool-language-modules/de/src/main/java/org/languagetool/rules/de/SimilarNameRule.java b/...uagetool-language-modules/de/src/main/java/org/languagetool/rules/de/SimilarNameRule.java
@@ -111,5 +111,10 @@ private String similarName(String nameHere, Set<String> namesSoFar) {
     }
     return null;
   }
+
+  @Override
+  public int minToCheckParagraph() {
+    return -1;
+  }
 
 }
diff --git a/...getool-language-modules/de/src/main/java/org/languagetool/rules/de/VerbAgreementRule.java b/...getool-language-modules/de/src/main/java/org/languagetool/rules/de/VerbAgreementRule.java
@@ -594,4 +594,9 @@ private BooleanAndFiniteVerb(boolean verbDoesMatchPersonAndNumber, AnalyzedToken
       this.finiteVerb = finiteVerb;
     }
   }
+
+  @Override
+  public int minToCheckParagraph() {
+    return 0;
+  }
 }
diff --git a/languagetool-language-modules/en/src/test/java/org/languagetool/JLanguageToolTest.java b/languagetool-language-modules/en/src/test/java/org/languagetool/JLanguageToolTest.java
@@ -270,5 +270,9 @@ public String getId() {
     public String getDescription() {
       return "Test rule";
     }
+    @Override
+    public int minToCheckParagraph() {
+      return -1;
+    }
   }
 }
diff --git a/...ool-office-extension/src/main/java/org/languagetool/openoffice/MultiDocumentsHandler.java b/...ool-office-extension/src/main/java/org/languagetool/openoffice/MultiDocumentsHandler.java
@@ -34,6 +34,7 @@
 import org.languagetool.gui.Configuration;
 import org.languagetool.rules.CategoryId;
 import org.languagetool.rules.Rule;
+import org.languagetool.rules.TextLevelRule;
 import org.languagetool.tools.Tools;
 
 import com.sun.star.awt.XMenuBar;
@@ -78,6 +79,7 @@ public class MultiDocumentsHandler {
   private final String configFile;
   private Configuration config = null;
   private LinguisticServices linguServices = null;
+  private SortedTextRules sortedTextRules;
 
   private XComponentContext xContext;       //  The context of the document
   private List<SingleDocument> documents;   //  The List of LO documents to be checked
@@ -367,7 +369,7 @@ private int getNumDoc(String docID) {
         }
       }
     }
-    documents.add(new SingleDocument(xContext, config, docID, xComponent));
+    documents.add(new SingleDocument(xContext, config, docID, xComponent, this));
     setMenuTextForSwitchOff(xContext);
     if (debugMode) {
       MessageHandler.printToLogFile("Document " + docNum + " created; docID = " + docID);
@@ -469,10 +471,12 @@ void initCheck() {
         langTool.enableRule(ruleName);
       }
     }
+    recheck = false;
+    sortedTextRules = new SortedTextRules();
+    setConfigValues(config, langTool);
     for (SingleDocument document : documents) {
       document.resetCache();
     }
-    recheck = false;
   }
 
 /**
@@ -573,5 +577,128 @@ public boolean toggleSwitchedOff() throws IOException {
     return ret;
   }
 
+  /**
+   * Returns a list of different numbers of paragraphs to check for text level rules
+   * (currently only -1 for full text check and n for max number for other text level rules)
+   */
+  public List<Integer> getNumMinToCheckParas() {
+    return sortedTextRules.minToCheckParagraph;
+  }
+
+  /**
+   * activate all rules stored under a given index related to the list of getNumMinToCheckParas
+   * deactivate all other text level rules
+   */
+  public void activateTextRulesByIndex(int index) {
+    sortedTextRules.activateTextRulesByIndex(index);
+  }
+
+  /**
+   * reactivate all text level rules
+   */
+  public void reactivateTextRules() {
+    sortedTextRules.reactivateTextRules();;
+  }
 
+  /**
+   * class to store all text level rules sorted by the minimum to check paragraphs
+   * (currently only full text check and all other text level rules)
+   *
+   */
+  class SortedTextRules { 
+    List<Integer> minToCheckParagraph;
+    List<List<String>> textLevelRules;
+    SortedTextRules () {
+      minToCheckParagraph = new ArrayList<Integer>();
+      textLevelRules = new ArrayList<List<String>>();
+      List<Rule> rules = langTool.getAllActiveOfficeRules();
+      for(Rule rule : rules) {
+        if(rule instanceof TextLevelRule) {
+          insertRule(((TextLevelRule) rule).minToCheckParagraph(), rule.getId());
+        }
+      }
+      if(debugMode) {
+        MessageHandler.printToLogFile("Number different minToCheckParagraph: " + minToCheckParagraph.size());
+        for( int i = 0; i < minToCheckParagraph.size(); i++) {
+          MessageHandler.printToLogFile("minToCheckParagraph: " + minToCheckParagraph.get(i));
+          for (int j = 0; j < textLevelRules.get(i).size(); j++) {
+            MessageHandler.printToLogFile("RuleId: " + textLevelRules.get(i).get(j));
+          }
+        }
+      }
+    }
+
+    private void insertRule (int minPara, String RuleId) {
+      if(minPara < 0) {
+        int n = minToCheckParagraph.indexOf(minPara);
+        if( n >= 0) {
+        } else {
+          minToCheckParagraph.add(minPara);
+          textLevelRules.add(new ArrayList<String>());
+        }
+        textLevelRules.get(textLevelRules.size() - 1).add(new String(RuleId));
+      } else {
+        int n = minToCheckParagraph.indexOf(-1);
+        if( n == 0 || minToCheckParagraph.size() == 0) {
+          minToCheckParagraph.add(0, minPara);
+          textLevelRules.add(0, new ArrayList<String>());
+        } else if(minPara > minToCheckParagraph.get(0)) {
+          minToCheckParagraph.set(0, minPara);
+        }
+        textLevelRules.get(0).add(new String(RuleId));
+      }
+    }
+/*
+ * This rule was commented out for performance reasons and replaced by the same named rule before 
+ * 
+    private void insertRule (int minPara, String RuleId) {
+      int n = minToCheckParagraph.indexOf(minPara); 
+      if( n >= 0) {
+        textLevelRules.get(n).add(new String(RuleId));
+        return;
+      } else {
+        if(minPara < 0) {
+          n = minToCheckParagraph.size();
+        } else {
+          for (n = 0; n < minToCheckParagraph.size(); n++) {
+            if(minPara < minToCheckParagraph.get(n) || minToCheckParagraph.get(n) < 0) {
+              break;
+            }
+          }
+        }
+        minToCheckParagraph.add(n, minPara);
+        textLevelRules.add(n, new ArrayList<String>());
+        textLevelRules.get(n).add(new String(RuleId));
+      }
+    }
+*/
+    public List<Integer> getMinToCheckParas() {
+      return minToCheckParagraph;
+    }
+
+    public void activateTextRulesByIndex(int index) {
+      for(int i = 0; i < textLevelRules.size(); i++) {
+        if(i == index) {
+          for (String ruleId : textLevelRules.get(i)) {
+            langTool.enableRule(ruleId);
+          }
+        } else {
+          for (String ruleId : textLevelRules.get(i)) {
+            langTool.disableRule(ruleId);
+          }
+        }
+      }
+    }
+
+    public void reactivateTextRules() {
+      for(List<String> textRules : textLevelRules) {
+        for (String ruleId : textRules) {
+          langTool.enableRule(ruleId);
+        }
+      }
+    }
+
+  }
+
+
 }