Merge pull request #472 from metafacture/filemapIgnorePattern

Add `FileMap` option to ignore lines matching pattern.

Author: blackwinter

metamorph/src/main/java/org/metafacture/metamorph/maps/FileMap.java

@@ -35,6 +35,7 @@
 import java.util.Map;
 import java.util.Optional;
 import java.util.Set;
+import java.util.regex.Matcher;
 import java.util.regex.Pattern;
 
 /**
@@ -65,6 +66,7 @@ public final class FileMap extends AbstractReadOnlyMap<String, String> {
     private final Map<String, String> map = new HashMap<>();
 
     private ArrayList<String> filenames = new ArrayList<>();
+    private Matcher ignoreMatcher;
     private Pattern split = Pattern.compile("\t", Pattern.LITERAL);
     private boolean allowEmptyValues;
     private boolean isUninitialized = true;
@@ -109,6 +111,15 @@ public void setExpectedColumns(final int expectedColumns) {
         this.expectedColumns = expectedColumns;
     }
 
+    /**
+     * Sets the pattern which determines whether a line should be ignored.
+     *
+     * @param ignorePattern a Java regular expression
+     */
+    public void setIgnorePattern(final String ignorePattern) {
+        this.ignoreMatcher = Pattern.compile(ignorePattern).matcher("");
+    }
+
     /**
      * Sets a comma separated list of files which provides the {@link Map}.
      *
@@ -168,25 +179,31 @@ private void loadFile(final String file) {
 
             String line;
             while ((line = br.readLine()) != null) {
-                if (line.isEmpty()) {
+                if (ignore(line)) {
                     continue;
                 }
 
                 final String[] parts = allowEmptyValues ? split.split(line, -1) : split.split(line);
-                if (parts.length < minColumns) {
+                if (ignore(parts.length, minColumns, expColumns)) {
                     continue;
                 }
 
-                if (expColumns < 0 || parts.length == expColumns) {
-                    map.put(parts[keyColumn], parts[valueColumn]);
-                }
+                map.put(parts[keyColumn], parts[valueColumn]);
             }
         }
         catch (final IOException | UncheckedIOException e) {
             throw new MorphExecutionException("filemap: cannot read map file", e);
         }
     }
 
+    private boolean ignore(final String line) {
+        return line.isEmpty() || (ignoreMatcher != null && ignoreMatcher.reset(line).matches());
+    }
+
+    private boolean ignore(final int partsLength, final int minColumns, final int expColumns) {
+        return partsLength < minColumns || (expColumns > 0 && partsLength != expColumns);
+    }
+
     private InputStream openStream(final String file) {
         return openAsFile(file)
             .orElseGet(() -> openAsResource(file)

metamorph/src/main/resources/schemata/metamorph.xsd

@@ -660,6 +660,12 @@
                     </restriction>
                 </simpleType>
             </attribute>
+            <attribute name="ignorePattern" type="string" use="optional">
+                <annotation>
+                    <documentation>Sets the pattern which determines whether a line should
+                        be ignored.</documentation>
+                </annotation>
+            </attribute>
             <attribute ref="xml:base" />
         </complexType>
     </element>

metamorph/src/test/java/org/metafacture/metamorph/maps/FileMapTest.java

@@ -90,6 +90,25 @@ public void shouldLookupValuesInFileBasedMapWithColumnOptions() {
         );
     }
 
+    @Test
+    public void shouldLookupValuesInFileBasedMapWithIgnorePattern() {
+        assertMorph(receiver, buildMorph("lookup in", "ignorePattern=\"g.*\""),
+                i -> {
+                    i.startRecord("1");
+                    i.literal("1", "gw");
+                    i.literal("1", "fj");
+                    i.literal("1", "hk");
+                    i.endRecord();
+                },
+                o -> {
+                    o.get().startRecord("1");
+                    o.get().literal("1", "Fiji");
+                    o.get().literal("1", "HongKong");
+                    o.get().endRecord();
+                }
+        );
+    }
+
     @Test
     public void shouldWhitelistValuesInFileBasedMap() {
         assertMorph(receiver, buildMorph("whitelist map", ""),
@@ -308,6 +327,16 @@ public void shouldLoadFileWithArbitraryExpectedColumns() {
         });
     }
 
+    @Test
+    public void shouldLoadFileWithIgnorePattern() {
+        assertMap(369, i -> {
+            i.setIgnorePattern(".*New.*");
+
+            Assert.assertNull(i.get("pp"));
+            Assert.assertEquals("Puerto Rico", i.get("pr"));
+        });
+    }
+
     @Test
     public void shouldNotLoadFileWithOutOfRangeKeyColumn() {
         assertMap(0, i -> {