feat: support annotations in SQL file (#171)

* feat: support annotations in SQL file

Column Annotations are NOT a feature of Cloud Spanner.
 
This is an additional feature of the Cloud Spanner Schema parser exclusively in this tool so that users of this tool can add metadata to columns, and have that metadata represented in the parsed schema.

See description in src/main/jjtree-sources/ddl_annotation.jjt

---------

Co-authored-by: Niel Markwick <[email protected]>

pom.xml

@@ -220,6 +220,7 @@
                   src/main/jjtree-sources/ddl_keywords.jjt \
                   src/main/jjtree-sources/ddl_string_bytes_tokens.jjt \
                   src/main/jjtree-sources/ddl_expression.jjt \
+                  src/main/jjtree-sources/ddl_annotation.jjt \
                   src/main/jjtree-sources/ddl_parser.jjt \
                   &gt; ${project.build.directory}/generated-sources/jjtree-src/DdlParser.jjt</argument>
               </arguments>

src/main/java/com/google/cloud/solutions/spannerddl/diff/DdlDiff.java

@@ -55,6 +55,7 @@
 import java.util.Set;
 import java.util.TreeMap;
 import java.util.function.Function;
+import java.util.regex.Pattern;
 import java.util.stream.Collectors;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
@@ -707,6 +708,27 @@ private static String getDatabaseNameFromAlterDatabase(List<ASTddl_statement> st
    * @throws DdlDiffException if there is an error in parsing the DDL
    */
   public static List<ASTddl_statement> parseDdl(String original) throws DdlDiffException {
+    return parseDdl(original, false);
+  }
+
+  /**
+   * Parses the Cloud Spanner Schema (DDL) string to a list of AST DDL statements.
+   *
+   * @param original DDL to parse
+   * @param parseAnnotationInComments If true then the annotations that appear as comments
+   *     "-- @ANNOTATION annotation" will be parsed
+   * @return List of parsed DDL statements
+   */
+  public static List<ASTddl_statement> parseDdl(String original, boolean parseAnnotationInComments)
+      throws DdlDiffException {
+    // the annotations are prefixed with "--" so that SQL file remains valid.
+    // strip the comment prefix before so that annotations can be parsed.
+    // otherwise they will be ignored as comment lines
+    if (parseAnnotationInComments) {
+      original =
+          Pattern.compile("^\\s*--\\s+@", Pattern.MULTILINE).matcher(original).replaceAll("@");
+    }
+
     // Remove "--" comments and split by ";"
     List<String> statements = Splitter.on(';').splitToList(original.replaceAll("--.*(\n|$)", ""));
     ArrayList<ASTddl_statement> ddlStatements = new ArrayList<>(statements.size());

src/main/java/com/google/cloud/solutions/spannerddl/parser/ASTannotation.java

@@ -0,0 +1,74 @@
+/*
+ * Copyright 2024 Google LLC
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *     https://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package com.google.cloud.solutions.spannerddl.parser;
+
+import static com.google.cloud.solutions.spannerddl.diff.AstTreeUtils.getChildByType;
+
+import com.google.cloud.solutions.spannerddl.diff.AstTreeUtils;
+import java.util.ArrayList;
+import java.util.List;
+
+/**
+ * Abstract Syntax Tree parser object for "annotation" token
+ *
+ * <p>Column Annotations are NOT a feature of Cloud Spanner.
+ *
+ * <p>This is an additional feature of the Cloud Spanner Schema parser exclusively in this tool so
+ * that users of this tool can add metadata to colums, and have that metadata represented in the
+ * parsed schema.
+ *
+ * <p>To use Annotations, they should be added to a CREATE TABLE statement as follows:
+ *
+ * <pre>
+ *  CREATE TABLE Albums (
+ *   -- @ANNOTATION SOMETEXT,
+ *    id STRING(36),
+ *  ) PRIMARY KEY (id)
+ * </pre>
+ *
+ * Annotations need to be on their own line, and terminate with a comma. (This is because the '-- '
+ * prefix is removed before using the JJT parser).
+ *
+ * <p>As they are comments, they are ignored by the diff generator and by Spanner itself.
+ */
+public class ASTannotation extends SimpleNode {
+  public ASTannotation(int id) {
+    super(id);
+  }
+
+  public ASTannotation(DdlParser p, int id) {
+    super(p, id);
+  }
+
+  public String getName() {
+    return AstTreeUtils.tokensToString(getChildByType(children, ASTname.class));
+  }
+
+  public List<ASTannotation_param> getParams() {
+    List<ASTannotation_param> params = new ArrayList<>();
+    for (Node child : children) {
+      if (child instanceof ASTannotation_param) {
+        params.add((ASTannotation_param) child);
+      }
+    }
+    return params;
+  }
+
+  public String getAnnotation() {
+    return AstTreeUtils.tokensToString(this, false).replaceAll(" ", "");
+  }
+}

src/main/java/com/google/cloud/solutions/spannerddl/parser/ASTannotation_param.java

@@ -0,0 +1,43 @@
+/*
+ * Copyright 2024 Google LLC
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *     https://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package com.google.cloud.solutions.spannerddl.parser;
+
+import com.google.cloud.solutions.spannerddl.diff.AstTreeUtils;
+
+public class ASTannotation_param extends SimpleNode {
+  public ASTannotation_param(int id) {
+    super(id);
+  }
+
+  public ASTannotation_param(DdlParser p, int id) {
+    super(p, id);
+  }
+
+  public String getKey() {
+    ASTparam_key key = AstTreeUtils.getChildByType(this, ASTparam_key.class);
+    return AstTreeUtils.tokensToString(key);
+  }
+
+  public String getValue() {
+    ASTparam_val val = AstTreeUtils.getOptionalChildByType(this, ASTparam_val.class);
+    if (val != null) {
+      return AstTreeUtils.tokensToString(val);
+    } else {
+      return null;
+    }
+  }
+}

src/main/java/com/google/cloud/solutions/spannerddl/parser/ASTcreate_table_statement.java

@@ -140,7 +140,8 @@ private void validateChildren() {
             ASTcheck_constraint.class,
             ASTprimary_key.class,
             ASTtable_interleave_clause.class,
-            ASTrow_deletion_policy_clause.class));
+            ASTrow_deletion_policy_clause.class,
+            ASTannotation.class));
   }
 
   @Override

src/main/jjtree-sources/DdlParser.head

@@ -25,6 +25,7 @@ cat src/main/jjtree-sources/DdlParser.head       \
       src/main/jjtree-sources/ddl_keywords.jjt   \
       src/main/jjtree-sources/ddl_string_bytes_tokens.jjt \
       src/main/jjtree-sources/ddl_expression.jjt \
+      src/main/jjtree-sources/ddl_annotation.jjt \
       src/main/jjtree-sources/ddl_parser.jjt     \
    > src/main/jjtree/DdlParser.jjt
 

src/main/jjtree-sources/ddl_annotation.jjt

@@ -0,0 +1,61 @@
+//
+// Copyright 2024 Google LLC
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//      http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+//
+
+// Column Annotations are NOT a feature of Cloud Spanner.
+//
+// This is an additional feature of the Cloud Spanner Schema parser exclusively
+// in this tool so that users of this tool can add metadata to colums, and have
+// that metadata represented in the parsed schema.
+//
+// To use Annotations, they should be added to a CREATE TABLE statement as
+// follows:
+//
+// CREATE TABLE Albums (
+//  -- @ANNOTATION SOMETEXT,
+//   id STRING(36),
+// ) PRIMARY KEY (id)
+//
+// Annotations need to be on their own line, and terminate with a comma.
+// (This is because the '-- ' prefix is removed before using the JJT parser).
+//
+// As they are comments, they are ignored by the diff generator and by
+// Spanner itself.
+//
+
+TOKEN:
+{
+  <ANNOTATION: "@ANNOTATION">
+}
+
+void column_annotation() #void: {}
+{
+  <ANNOTATION> annotation()
+}
+
+void annotation(): {}
+{
+  qualified_identifier() #name [ annotation_params() ]
+}
+
+void annotation_params() #void: {}
+{
+  "(" annotation_param() ( "," annotation_param() )* ")"
+}
+
+void annotation_param(): {}
+{
+  identifier() #param_key [ "=" identifier() #param_val ]
+}

src/main/jjtree-sources/ddl_parser.jjt

@@ -212,6 +212,11 @@ void table_element() #void :
     LOOKAHEAD(3) foreign_key()
   | LOOKAHEAD(3) check_constraint()
   | LOOKAHEAD(3) synonym_clause()
+
+  // Column annotations are not a Spanner feature.
+  // See file ddl_annotation.jjt for more details.
+  | column_annotation()
+
   | column_def()
 }
 

src/test/java/com/google/cloud/solutions/spannerddl/parser/DDLAnnotationTest.java

@@ -0,0 +1,76 @@
+package com.google.cloud.solutions.spannerddl.parser;
+
+import static com.google.common.truth.Truth.assertThat;
+import static com.google.common.truth.Truth.assertWithMessage;
+import static org.junit.Assert.fail;
+
+import com.google.cloud.solutions.spannerddl.diff.DdlDiff;
+import com.google.cloud.solutions.spannerddl.diff.DdlDiffException;
+import com.google.cloud.solutions.spannerddl.testUtils.ReadTestDatafile;
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.Iterator;
+import java.util.List;
+import java.util.Map;
+import java.util.Map.Entry;
+import org.junit.Test;
+
+public class DDLAnnotationTest {
+
+  @Test
+  public void validateAnnotations() throws IOException {
+    Map<String, String> tests = ReadTestDatafile.readDdlSegmentsFromFile("annotations.txt");
+    Map<String, String> expects =
+        ReadTestDatafile.readDdlSegmentsFromFile("expectedAnnotations.txt");
+
+    Iterator<Entry<String, String>> testIt = tests.entrySet().iterator();
+    Iterator<Entry<String, String>> expectedIt = expects.entrySet().iterator();
+
+    while (testIt.hasNext() && expectedIt.hasNext()) {
+      Entry<String, String> test = testIt.next();
+      String expected = expectedIt.next().getValue();
+      String segmentName = test.getKey();
+
+      try {
+        // first get all the annotations without removing the comment prefix
+        List<String> annotations = getTableAnnotations(test.getValue(), false);
+
+        // annotations should be empty
+        assertThat(annotations).isEmpty();
+
+        // now get all the annotations after removing the comment prefix
+        annotations = getTableAnnotations(test.getValue(), true);
+
+        List<String> expectedList =
+            expected != null ? Arrays.asList(expected.split("\n")) : Collections.emptyList();
+
+        assertWithMessage("Mismatch for section " + segmentName)
+            .that(annotations)
+            .isEqualTo(expectedList);
+      } catch (DdlDiffException e) {
+        fail("Failed to parse section: '" + segmentName + "': " + e);
+      }
+    }
+  }
+
+  private List<String> getTableAnnotations(String ddl, boolean parseAnnotations)
+      throws DdlDiffException {
+    List<String> annotations = new ArrayList<>();
+
+    List<ASTddl_statement> statements = DdlDiff.parseDdl(ddl, parseAnnotations);
+    for (ASTddl_statement statement : statements) {
+      if (statement.jjtGetChild(0).getId() == DdlParserTreeConstants.JJTCREATE_TABLE_STATEMENT) {
+        Node tableStatement = statement.jjtGetChild(0);
+        for (int i = 0, count = tableStatement.jjtGetNumChildren(); i < count; i++) {
+          Node child = tableStatement.jjtGetChild(i);
+          if (child instanceof ASTannotation) {
+            annotations.add(((ASTannotation) child).getAnnotation());
+          }
+        }
+      }
+    }
+    return annotations;
+  }
+}

src/test/java/com/google/cloud/solutions/spannerddl/testUtils/ReadTestDatafile.java

@@ -32,16 +32,26 @@ public abstract class ReadTestDatafile {
    * @return LinkedHashMap of segment name => contents
    */
   public static Map<String, String> readDdlSegmentsFromFile(String filename) throws IOException {
+    return readDdlSegmentsFromFile(filename, false);
+  }
+
+  /**
+   * Reads the test data file, parsing out the test titles and data from the file.
+   *
+   * @return LinkedHashMap of segment name => contents
+   */
+  public static Map<String, String> readDdlSegmentsFromFile(String filename, boolean preserveSpace)
+      throws IOException {
     File file = new File("src/test/resources/" + filename).getAbsoluteFile();
     LinkedHashMap<String, String> output = new LinkedHashMap<>();
 
     try (BufferedReader in = Files.newBufferedReader(file.toPath(), UTF_8)) {
 
       String sectionName = null;
       StringBuilder section = new StringBuilder();
-      String line;
-      while (null != (line = in.readLine())) {
-        line = line.replaceAll("#.*", "").trim();
+      String rawLine;
+      while (null != (rawLine = in.readLine())) {
+        String line = preserveSpace ? rawLine : rawLine.replaceAll("#.*", "").trim();
         if (line.isEmpty()) {
           continue;
         }

src/test/resources/annotations.txt

@@ -0,0 +1,12 @@
+== Test 1 all column annotations
+
+CREATE TABLE test1 (
+  -- @ANNOTATION DEPRECATED,
+  -- @ANNOTATION PII,
+  -- @ANNOTATION TAG.business(internal),
+  -- @ANNOTATION TAG.business(key1,key2),
+  -- @ANNOTATION TAG.business(key1=val1,key2=value),
+  id STRING(36),
+) PRIMARY KEY (id)
+
+==

src/test/resources/expectedAnnotations.txt

@@ -0,0 +1,9 @@
+== Test 1 all column annotations
+
+DEPRECATED
+PII
+TAG.business(internal)
+TAG.business(key1,key2)
+TAG.business(key1=val1,key2=value)
+
+==