Sentence 21 works! ContentTypeSegmenter: if new segment type is align…

…ed, overwrite to return replaced instead (alignments would have been caught in the first round). Segment: added copy constructor, with overridable type.
bleekere · Mar 28, 2018 · 3be3509 · 3be3509
1 parent 3d74e98
commit 3be3509
Show file tree

Hide file tree

Showing 4 changed files with 21 additions and 11 deletions.
diff --git a/src/main/java/prioritised_xml_collation/ContentTypeSegmenter.java b/src/main/java/prioritised_xml_collation/ContentTypeSegmenter.java
@@ -23,7 +23,10 @@ public List<Segment> calculateSegmentation(EditGraphTable editTable) {
             Cell currentCell = iterateTable.next();
             // we change state based on type, in case of mixed that we look at the movement through the table
             if (editTable.determineUniqueCellType(currentCell) != editTable.determineUniqueCellType(lastCell)) {
-                Segment newSegment = editTable.createSegmentOfCellsWithType(currentCell, lastCell, Segment.Type.replacement);
+                Segment newSegment = editTable.createSegmentOfCells(currentCell, lastCell);
+                if (newSegment.type == Segment.Type.aligned) {
+                    newSegment = new Segment(newSegment, Segment.Type.replacement);
+                }
                 superwitness.add(0, newSegment);
                 lastCell = currentCell;
             }

diff --git a/src/main/java/prioritised_xml_collation/EditGraphTable.java b/src/main/java/prioritised_xml_collation/EditGraphTable.java
@@ -77,19 +77,13 @@ else if (segmentTokensA.isEmpty()) {
         }
     }
 
-    Segment createSegmentOfCellsWithType(Cell currentCell, Cell lastCell, Segment.Type type) {
-        List<XMLToken> segmentTokensA = tokensA.subList(currentCell.x, lastCell.x);
-        List<XMLToken> segmentTokensB = tokensB.subList(currentCell.y, lastCell.y);
-        return new Segment(segmentTokensA, segmentTokensB, type);
-    }
-
     private String cellToString(Cell cell) {
         XMLToken tokenA = tokensA.get(cell.x - 1);
         XMLToken tokenB = tokensB.get(cell.y - 1);
         return tokenA+" : "+tokenB;
     }
 
-    public CellType establishTypeOfCell(Cell cell){
+    CellType establishTypeOfCell(Cell cell){
         if (cell.x == 0 && cell.y == 0) {
             return CellType.root;
         }
@@ -110,7 +104,7 @@ else if (markupType) {
         else return CellType.mix;
     }
 
-    public CellType determineTypeOfToken(XMLToken tokenA) {
+    private CellType determineTypeOfToken(XMLToken tokenA) {
         boolean punctuationType = (tokenA.content.matches("\\W+"));
         boolean contentType = (tokenA.content.matches("\\w+") && tokenA instanceof TextToken);
         boolean markupType = (tokenA instanceof ElementToken);
@@ -126,7 +120,7 @@ else if (markupType) {
         else return CellType.mix;
     }
 
-    public CellType determineUniqueCellType(Cell cell) {
+    CellType determineUniqueCellType(Cell cell) {
         CellType type = establishTypeOfCell(cell);
         if (type == CellType.mix) {
             if (cell.movedVertical()) {

diff --git a/src/main/java/prioritised_xml_collation/Segment.java b/src/main/java/prioritised_xml_collation/Segment.java
@@ -26,6 +26,12 @@ public Segment(Type type) {
         this.type = type;
     }
 
+    public Segment(Segment original, Type type) {
+        this.tokensWa = original.tokensWa;
+        this.tokensWb = original.tokensWb;
+        this.type = type;
+    }
+
     // Factory method
     public static Segment s(Type type) {
         return new Segment(type);

diff --git a/src/test/java/prioritised_xml_collation/CoordinationUnitTest.java b/src/test/java/prioritised_xml_collation/CoordinationUnitTest.java
@@ -49,6 +49,13 @@ public void testSentence21() throws Exception {
         SegmentMatcher m4 = sM(addition).tokensWb("lb", "/lb");
         SegmentMatcher m5 = sM(aligned).tokensWa("vrouw");
         SegmentMatcher m6 = sM(replacement).tokensWa(",").tokensWb("!");
-        assertThat(segments, contains(m1, m2, m3, m4, m5, m6));
+        SegmentMatcher m7 = sM(addition).tokensWb("/s", "s");
+        SegmentMatcher m8 = sM(replacement).tokensWa("de", "ongewisheid").tokensWb("Die", "dagen", "van", "nerveuze","verwachting");
+        SegmentMatcher m9 = sM(aligned).tokensWa("voor", "de");
+        SegmentMatcher m10 = sM(addition).tokensWb("lb", "/lb");
+        SegmentMatcher m11 = sM(aligned).tokensWa("liefelijke", "toestemming");
+        SegmentMatcher m12 = sM(replacement).tokensWa("!", ".");
+        SegmentMatcher m13 = sM(aligned).tokensWa("/s", "/div", "/body", "/text");
+        assertThat(segments, contains(m1, m2, m3, m4, m5, m6, m7, m8, m9, m10, m11, m12, m13));
     }
 }