Merge pull request apache#257 from pedro-w/csv-304-header-comments

garydgregory · web-flow · commit 2e851bc96029 · 2022-09-12T09:43:21.000-07:00
[CSV-304] Accessors for header/trailer comments
diff --git a/src/main/java/org/apache/commons/csv/CSVParser.java b/src/main/java/org/apache/commons/csv/CSVParser.java
@@ -353,6 +353,10 @@ public static CSVParser parse(final URL url, final Charset charset, final CSVFor
         return new CSVParser(new InputStreamReader(url.openStream(), charset), format);
     }
 
+    private String headerComment;
+
+    private String trailerComment;
+
     private final CSVFormat format;
 
     private final Headers headers;
@@ -480,10 +484,14 @@ private Headers createHeaders() throws IOException {
                 final CSVRecord nextRecord = this.nextRecord();
                 if (nextRecord != null) {
                     headerRecord = nextRecord.values();
+                    headerComment = nextRecord.getComment();
                 }
             } else {
                 if (this.format.getSkipHeaderRecord()) {
-                    this.nextRecord();
+                    final CSVRecord nextRecord = this.nextRecord();
+                    if (nextRecord != null) {
+                        headerComment = nextRecord.getComment();
+                    }
                 }
                 headerRecord = formatHeader;
             }
@@ -597,6 +605,57 @@ public List<String> getHeaderNames() {
         return Collections.unmodifiableList(headers.headerNames);
     }
 
+    /**
+     * Checks whether there is a header comment.
+     * The header comment appears before the header record.
+     * Note that if the parser's format has been given an explicit header
+     * (with {@link CSVFormat.Builder#setHeader(String... )} or another overload)
+     * and the header record is not being skipped
+     * ({@link CSVFormat.Builder#setSkipHeaderRecord} is false) then any initial comments
+     * will be associated with the first record, not the header.
+     *
+     * @return true if this parser has seen a header comment, false otherwise
+     * @since 1.10.0
+     */
+    public boolean hasHeaderComment() {
+        return headerComment != null;
+    }
+
+    /**
+     * Returns the header comment, if any.
+     * The header comment appears before the header record.
+     *
+     * @return the header comment for this stream, or null if no comment is available.
+     * @since 1.10.0
+     */
+    public String getHeaderComment() {
+        return headerComment;
+    }
+
+    /**
+     * Checks whether there is a trailer comment.
+     * Trailer comments are located between the last record and EOF.
+     * The trailer comments will only be available after the parser has
+     * finished processing this stream.
+     *
+     * @return true if this parser has seen a trailer comment, false otherwise
+     * @since 1.10.0
+     */
+    public boolean hasTrailerComment() {
+        return trailerComment != null;
+    }
+
+    /**
+     * Returns the trailer comment, if any.
+     * Trailer comments are located between the last record and EOF
+     *
+     * @return the trailer comment for this stream, or null if no comment is available.
+     * @since 1.10.0
+     */
+    public String getTrailerComment() {
+        return trailerComment;
+    }
+
     /**
      * Returns the current record number in the input stream.
      *
@@ -713,6 +772,8 @@ CSVRecord nextRecord() throws IOException {
             case EOF:
                 if (this.reusableToken.isReady) {
                     this.addRecordValue(true);
+                } else if (sb != null) {
+                    trailerComment = sb.toString();
                 }
                 break;
             case INVALID:
diff --git a/src/test/java/org/apache/commons/csv/CSVParserTest.java b/src/test/java/org/apache/commons/csv/CSVParserTest.java
@@ -1349,4 +1349,160 @@ private void validateRecordPosition(final String lineSeparator) throws IOExcepti
 
         parser.close();
     }
+
+    // CSV with no header comments
+    static private final String CSV_INPUT_NO_COMMENT = "A,B"+CRLF+"1,2"+CRLF;
+    // CSV with a header comment
+    static private final String CSV_INPUT_HEADER_COMMENT = "# header comment" + CRLF + "A,B" + CRLF + "1,2" + CRLF;
+    // CSV with a single line header and trailer comment
+    static private final String CSV_INPUT_HEADER_TRAILER_COMMENT = "# header comment" + CRLF + "A,B" + CRLF + "1,2" + CRLF + "# comment";
+    // CSV with a multi-line header and trailer comment
+    static private final String CSV_INPUT_MULTILINE_HEADER_TRAILER_COMMENT = "# multi-line" + CRLF + "# header comment" + CRLF + "A,B" + CRLF + "1,2" + CRLF + "# multi-line" + CRLF + "# comment";
+    // Format with auto-detected header
+    static private final CSVFormat FORMAT_AUTO_HEADER = CSVFormat.Builder.create(CSVFormat.DEFAULT).setCommentMarker('#').setHeader().build();
+    // Format with explicit header
+    static private final CSVFormat FORMAT_EXPLICIT_HEADER = CSVFormat.Builder.create(CSVFormat.DEFAULT)
+            .setSkipHeaderRecord(true)
+            .setCommentMarker('#')
+            .setHeader("A", "B")
+            .build();
+    // Format with explicit header that does not skip the header line
+    CSVFormat FORMAT_EXPLICIT_HEADER_NOSKIP = CSVFormat.Builder.create(CSVFormat.DEFAULT)
+            .setCommentMarker('#')
+            .setHeader("A", "B")
+            .build();
+
+    @Test
+    public void testGetHeaderComment_NoComment1() throws IOException {
+        try (CSVParser parser = CSVParser.parse(CSV_INPUT_NO_COMMENT, FORMAT_AUTO_HEADER)) {
+            parser.getRecords();
+            // Expect no header comment
+            assertFalse(parser.hasHeaderComment());
+            assertNull(parser.getHeaderComment());
+        }
+    }
+
+    @Test
+    public void testGetHeaderComment_HeaderComment1() throws IOException {
+        try (CSVParser parser = CSVParser.parse(CSV_INPUT_HEADER_COMMENT, FORMAT_AUTO_HEADER)) {
+            parser.getRecords();
+            // Expect a header comment
+            assertTrue(parser.hasHeaderComment());
+            assertEquals("header comment", parser.getHeaderComment());
+        }
+    }
+
+    @Test
+    public void testGetHeaderComment_HeaderTrailerComment() throws IOException {
+        try (CSVParser parser = CSVParser.parse(CSV_INPUT_MULTILINE_HEADER_TRAILER_COMMENT, FORMAT_AUTO_HEADER)) {
+            parser.getRecords();
+            // Expect a header comment
+            assertTrue(parser.hasHeaderComment());
+            assertEquals("multi-line"+LF+"header comment", parser.getHeaderComment());
+        }
+    }
+
+    @Test
+    public void testGetHeaderComment_NoComment2() throws IOException {
+        try (CSVParser parser = CSVParser.parse(CSV_INPUT_NO_COMMENT, FORMAT_EXPLICIT_HEADER)) {
+            parser.getRecords();
+            // Expect no header comment
+            assertFalse(parser.hasHeaderComment());
+            assertNull(parser.getHeaderComment());
+        }
+    }
+
+    @Test
+    public void testGetHeaderComment_HeaderComment2() throws IOException {
+        try (CSVParser parser = CSVParser.parse(CSV_INPUT_HEADER_COMMENT, FORMAT_EXPLICIT_HEADER)) {
+            parser.getRecords();
+            // Expect a header comment
+            assertTrue(parser.hasHeaderComment());
+            assertEquals("header comment", parser.getHeaderComment());
+        }
+    }
+
+    @Test
+    public void testGetHeaderComment_NoComment3() throws IOException {
+        try (CSVParser parser = CSVParser.parse(CSV_INPUT_NO_COMMENT, FORMAT_EXPLICIT_HEADER_NOSKIP)) {
+            parser.getRecords();
+            // Expect no header comment
+            assertFalse(parser.hasHeaderComment());
+            assertNull(parser.getHeaderComment());
+        }
+    }
+
+    @Test
+    public void testGetHeaderComment_HeaderComment3() throws IOException {
+        try (CSVParser parser = CSVParser.parse(CSV_INPUT_HEADER_COMMENT, FORMAT_EXPLICIT_HEADER_NOSKIP)) {
+            parser.getRecords();
+            // Expect no header comment - the text "comment" is attached to the first record
+            assertFalse(parser.hasHeaderComment());
+            assertNull(parser.getHeaderComment());
+        }
+    }
+
+    @Test
+    public void testGetTrailerComment_HeaderComment1() throws IOException {
+        try (CSVParser parser = CSVParser.parse(CSV_INPUT_HEADER_COMMENT, FORMAT_AUTO_HEADER)) {
+            parser.getRecords();
+            assertFalse(parser.hasTrailerComment());
+            assertNull(parser.getTrailerComment());
+        }
+    }
+
+    @Test
+    public void testGetTrailerComment_HeaderTrailerComment1() throws IOException {
+        try (CSVParser parser = CSVParser.parse(CSV_INPUT_HEADER_TRAILER_COMMENT, FORMAT_AUTO_HEADER)) {
+            parser.getRecords();
+            assertTrue(parser.hasTrailerComment());
+            assertEquals("comment", parser.getTrailerComment());
+        }
+    }
+
+    @Test
+    public void testGetTrailerComment_MultilineComment() throws IOException {
+        try (CSVParser parser = CSVParser.parse(CSV_INPUT_MULTILINE_HEADER_TRAILER_COMMENT, FORMAT_AUTO_HEADER)) {
+            parser.getRecords();
+            assertTrue(parser.hasTrailerComment());
+            assertEquals("multi-line"+LF+"comment", parser.getTrailerComment());
+        }
+    }
+
+    @Test
+    public void testGetTrailerComment_HeaderComment2() throws IOException {
+        try (CSVParser parser = CSVParser.parse(CSV_INPUT_HEADER_COMMENT, FORMAT_EXPLICIT_HEADER)) {
+            parser.getRecords();
+            assertFalse(parser.hasTrailerComment());
+            assertNull(parser.getTrailerComment());
+        }
+    }
+
+    @Test
+    public void testGetTrailerComment_HeaderTrailerComment2() throws IOException {
+        try (CSVParser parser = CSVParser.parse(CSV_INPUT_HEADER_TRAILER_COMMENT, FORMAT_EXPLICIT_HEADER)) {
+            parser.getRecords();
+            assertTrue(parser.hasTrailerComment());
+            assertEquals("comment", parser.getTrailerComment());
+        }
+    }
+
+    @Test
+    public void testGetTrailerComment_HeaderComment3() throws IOException {
+        try (CSVParser parser = CSVParser.parse(CSV_INPUT_HEADER_COMMENT, FORMAT_EXPLICIT_HEADER_NOSKIP)) {
+            parser.getRecords();
+            assertFalse(parser.hasTrailerComment());
+            assertNull(parser.getTrailerComment());
+        }
+    }
+
+    @Test
+    public void testGetTrailerComment_HeaderTrailerComment3() throws IOException {
+        try (CSVParser parser = CSVParser.parse(CSV_INPUT_HEADER_TRAILER_COMMENT, FORMAT_EXPLICIT_HEADER_NOSKIP)) {
+            parser.getRecords();
+            assertTrue(parser.hasTrailerComment());
+            assertEquals("comment", parser.getTrailerComment());
+        }
+    }
+
 }
