Merge pull request #16 from rcsb/builder

Builder Improvements: Add `addMasked(T[], ValueKind[])` and `addAll(Iterable<T>)`

Author: sbittrich

CHANGELOG.md

@@ -13,6 +13,11 @@ This project uses semantic versioning. Furthermore, this project provides code t
   * `diffrn_radiation_wavelength_id` removed
   * `geom_bond_distance_min` renamed to `geom_min_bond_distance_cutoff`
 
+### General
+* improved builder ergonomics
+  * added overloads that make it easier to build columns with missing values
+  * validate that all columns in a category are equal in size
+
 ciftools-java 7.0.2 - September 2025
 -------------
 ### Bug fixes

pom.xml

@@ -58,10 +58,15 @@
         <dependency>
             <groupId>org.junit.jupiter</groupId>
             <artifactId>junit-jupiter-api</artifactId>
-            <version>5.14.0</version>
+            <version>6.0.2</version>
+            <scope>test</scope>
+        </dependency>
+        <dependency>
+            <groupId>org.junit.jupiter</groupId>
+            <artifactId>junit-jupiter-engine</artifactId>
+            <version>6.0.2</version>
             <scope>test</scope>
         </dependency>
-
     </dependencies>
 
     <properties>
@@ -85,6 +90,9 @@
                 <groupId>org.apache.maven.plugins</groupId>
                 <artifactId>maven-surefire-plugin</artifactId>
                 <version>3.5.4</version>
+                <configuration>
+                    <failIfNoTests>true</failIfNoTests>
+                </configuration>
             </plugin>
         </plugins>
     </build>
@@ -98,7 +106,7 @@
                     <plugin>
                         <groupId>org.sonatype.central</groupId>
                         <artifactId>central-publishing-maven-plugin</artifactId>
-                        <version>0.9.0</version>
+                        <version>0.10.0</version>
                         <extensions>true</extensions>
                         <configuration>
                             <publishingServerId>central</publishingServerId>
@@ -137,7 +145,7 @@
                     <plugin>
                         <groupId>org.apache.maven.plugins</groupId>
                         <artifactId>maven-source-plugin</artifactId>
-                        <version>3.3.1</version>
+                        <version>3.4.0</version>
                         <executions>
                             <execution>
                                 <id>attach-sources</id>

src/main/java/org/rcsb/cif/model/CategoryBuilder.java

@@ -112,9 +112,9 @@ static <C extends Column<?>> C createColumnText(String columnName, List<?> value
             startToken[i] = builder.length();
             String value = String.valueOf(values.get(i));
             if (mask.get(i) == ValueKind.NOT_PRESENT) {
-                value = ".";
+                value = ValueKind.CIF_NOT_PRESENT;
             } else if (mask.get(i) == ValueKind.UNKNOWN) {
-                value = "?";
+                value = ValueKind.CIF_UNKNOWN;
             }
             builder.append(value);
             endToken[i] = builder.length();

src/main/java/org/rcsb/cif/model/FloatColumn.java

@@ -29,7 +29,7 @@ default DoubleStream values() {
      * @return a double
      */
     static double parseFloat(String text) {
-        if (text.isEmpty() || ".".equals(text) || "?".equals(text)) {
+        if (text.isEmpty() || ValueKind.isValueKindToken(text)) {
             return 0;
         }
         // some formats specify uncertain decimal places like: 0.00012(3) - ignore them (in agreement with Mol*)

src/main/java/org/rcsb/cif/model/FloatColumnBuilder.java

@@ -1,6 +1,7 @@
 package org.rcsb.cif.model;
 
 import java.util.List;
+import java.util.Objects;
 
 /**
  * A builder instance for {@link FloatColumn} instances.
@@ -27,4 +28,60 @@ public interface FloatColumnBuilder<P extends CategoryBuilder<PP, PPP>, PP exten
      * @return this builder instance
      */
     FloatColumnBuilder<P, PP, PPP> add(double... values);
+
+    /**
+     * Add new values with fine-grained control.
+     * <p>
+     * For {@link ValueKind#PRESENT}, the corresponding entry from {@code values} is appended.
+     * For {@link ValueKind#NOT_PRESENT} and {@link ValueKind#UNKNOWN}, this method delegates to
+     * {@link #markNextNotPresent()} and {@link #markNextUnknown()} respectively.
+     * </p>
+     * @param values array of double values
+     * @param mask array of {@link ValueKind}, must be the same length as {@code values}
+     * @return this builder instance
+     * @throws IllegalArgumentException if arrays differ in size
+     * @throws NullPointerException if {@code values}, {@code mask}, or any {@code mask[i]} is null
+     */
+    default FloatColumnBuilder<P,PP,PPP> addMasked(double[] values, ValueKind[] mask) {
+        Objects.requireNonNull(values, "values");
+        Objects.requireNonNull(mask, "mask");
+        if (values.length != mask.length) {
+            throw new IllegalArgumentException("values.length (" + values.length + ") must equal mask.length (" + mask.length + ")");
+        }
+
+        for (int i = 0; i < values.length; i++) {
+            ValueKind k = Objects.requireNonNull(mask[i], "mask[" + i + "]");
+            switch (k) {
+                case PRESENT:
+                    add(values[i]);
+                    break;
+                case NOT_PRESENT:
+                    markNextNotPresent();
+                    break;
+                case UNKNOWN:
+                    markNextUnknown();
+                    break;
+                default:
+                    throw new IllegalStateException("Unhandled ValueKind: " + k);
+            }
+        }
+        return this;
+    }
+
+    /**
+     * Add values from an Iterable.
+     * @param values Double values, null is mapped to ValueKind.NOT_PRESENT (".")
+     * @return this builder instance
+     */
+    default FloatColumnBuilder<P,PP,PPP> addNullable(Iterable<Double> values) {
+        Objects.requireNonNull(values, "values");
+        for (Double v : values) {
+            if (v == null) {
+                markNextNotPresent();
+            } else {
+                add(v);
+            }
+        }
+        return this;
+    }
 }

src/main/java/org/rcsb/cif/model/IntColumn.java

@@ -29,7 +29,7 @@ default IntStream values() {
      * @return an int
      */
     static int parseInt(String text) {
-        if (text.isEmpty() || ".".equals(text) || "?".equals(text)) {
+        if (text.isEmpty() || ValueKind.isValueKindToken(text)) {
             return 0;
         }
         // some floats may omit decimal places and can be parsed as int: 88. - ignore the dot (in agreement with Mol*)

src/main/java/org/rcsb/cif/model/IntColumnBuilder.java

@@ -1,6 +1,7 @@
 package org.rcsb.cif.model;
 
 import java.util.List;
+import java.util.Objects;
 
 /**
  * A builder instance for {@link IntColumn} instances.
@@ -27,4 +28,60 @@ public interface IntColumnBuilder<P extends CategoryBuilder<PP, PPP>, PP extends
      * @return this builder instance
      */
     IntColumnBuilder<P, PP, PPP> add(int... values);
+
+    /**
+     * Add new values with fine-grained control.
+     * <p>
+     * For {@link ValueKind#PRESENT}, the corresponding entry from {@code values} is appended.
+     * For {@link ValueKind#NOT_PRESENT} and {@link ValueKind#UNKNOWN}, this method delegates to
+     * {@link #markNextNotPresent()} and {@link #markNextUnknown()} respectively.
+     * </p>
+     * @param values array of int values
+     * @param mask array of {@link ValueKind}, must be the same length as {@code values}
+     * @return this builder instance
+     * @throws IllegalArgumentException if arrays differ in size
+     * @throws NullPointerException if {@code values}, {@code mask}, or any {@code mask[i]} is null
+     */
+    default IntColumnBuilder<P,PP,PPP> addMasked(int[] values, ValueKind[] mask) {
+        Objects.requireNonNull(values, "values");
+        Objects.requireNonNull(mask, "mask");
+        if (values.length != mask.length) {
+            throw new IllegalArgumentException("values.length (" + values.length + ") must equal mask.length (" + mask.length + ")");
+        }
+
+        for (int i = 0; i < values.length; i++) {
+            ValueKind k = Objects.requireNonNull(mask[i], "mask[" + i + "]");
+            switch (k) {
+                case PRESENT:
+                    add(values[i]);
+                    break;
+                case NOT_PRESENT:
+                    markNextNotPresent();
+                    break;
+                case UNKNOWN:
+                    markNextUnknown();
+                    break;
+                default:
+                    throw new IllegalStateException("Unhandled ValueKind: " + k);
+            }
+        }
+        return this;
+    }
+
+    /**
+     * Add values from an Iterable.
+     * @param values Integer values, null is mapped to ValueKind.NOT_PRESENT (".")
+     * @return this builder instance
+     */
+    default IntColumnBuilder<P,PP,PPP> addNullable(Iterable<Integer> values) {
+        Objects.requireNonNull(values, "values");
+        for (Integer v : values) {
+            if (v == null) {
+                markNextNotPresent();
+            } else {
+                add(v);
+            }
+        }
+        return this;
+    }
 }

src/main/java/org/rcsb/cif/model/StrColumnBuilder.java

@@ -1,6 +1,7 @@
 package org.rcsb.cif.model;
 
 import java.util.List;
+import java.util.Objects;
 
 /**
  * A builder instance for {@link StrColumn} instances.
@@ -27,4 +28,72 @@ public interface StrColumnBuilder<P extends CategoryBuilder<PP, PPP>, PP extends
      * @return this builder instance
      */
     StrColumnBuilder<P, PP, PPP> add(String... values);
+
+    /**
+     * Add new values with fine-grained control.
+     * <p>
+     * For {@link ValueKind#PRESENT}, the corresponding entry from {@code values} is appended.
+     * For {@link ValueKind#NOT_PRESENT} and {@link ValueKind#UNKNOWN}, this method delegates to
+     * {@link #markNextNotPresent()} and {@link #markNextUnknown()} respectively.
+     * </p>
+     * @param values array of String values
+     * @param mask array of {@link ValueKind}, must be the same length as {@code values}
+     * @return this builder instance
+     * @throws IllegalArgumentException if arrays differ in size
+     * @throws NullPointerException if {@code values}, {@code mask}, or any {@code mask[i]} is null
+     */
+    default StrColumnBuilder<P,PP,PPP> addMasked(String[] values, ValueKind[] mask) {
+        Objects.requireNonNull(values, "values");
+        Objects.requireNonNull(mask, "mask");
+        if (values.length != mask.length) {
+            throw new IllegalArgumentException("values.length (" + values.length + ") must equal mask.length (" + mask.length + ")");
+        }
+
+        for (int i = 0; i < values.length; i++) {
+            ValueKind k = Objects.requireNonNull(mask[i], "mask[" + i + "]");
+            if (k == ValueKind.PRESENT && (values[i] == null || ValueKind.isValueKindToken(values[i]))) {
+                throw new IllegalArgumentException("PRESENT value must not be null, '.' or '?': values[" + i + "]");
+            }
+            switch (k) {
+                case PRESENT:
+                    add(values[i]);
+                    break;
+                case NOT_PRESENT:
+                    markNextNotPresent();
+                    break;
+                case UNKNOWN:
+                    markNextUnknown();
+                    break;
+                default:
+                    throw new IllegalStateException("Unhandled ValueKind: " + k);
+            }
+        }
+        return this;
+    }
+
+    /**
+     * Add values from an Iterable.
+     * @param values String values, null is mapped to NOT_PRESENT ("."); "." and "?" are interpreted as CIF tokens.
+     * @return this builder instance
+     */
+    default StrColumnBuilder<P,PP,PPP> addNullable(Iterable<String> values) {
+        Objects.requireNonNull(values, "values");
+        for (String v : values) {
+            ValueKind valueKind = ValueKind.fromCifToken(v);
+            switch (valueKind) {
+                case PRESENT:
+                    add(v);
+                    break;
+                case NOT_PRESENT:
+                    markNextNotPresent();
+                    break;
+                case UNKNOWN:
+                    markNextUnknown();
+                    break;
+                default:
+                    throw new IllegalStateException("Unhandled ValueKind: " + valueKind);
+            }
+        }
+        return this;
+    }
 }

src/main/java/org/rcsb/cif/model/ValueKind.java

@@ -16,5 +16,28 @@ public enum ValueKind {
     /**
      * The value is unknown - <code>?</code> in CIF. String values will be empty, number values will be 0.
      */
-    UNKNOWN
+    UNKNOWN;
+
+    public static final String CIF_NOT_PRESENT = ".";
+    public static final String CIF_UNKNOWN = "?";
+
+    /**
+     * Checks whether a String matches "?" or ".", sequences with special meaning in CIF.
+     * @param s payload to evaluate
+     * @return true if this String indicates missing or undefined values
+     */
+    public static boolean isValueKindToken(String s) {
+        return CIF_NOT_PRESENT.equals(s) || CIF_UNKNOWN.equals(s);
+    }
+
+    /**
+     * Transforms a String into a ValueKind.
+     * @param s payload to evaluate
+     * @return appropriate ValueKind for "?" and ".", otherwise marked as PRESENT
+     */
+    public static ValueKind fromCifToken(String s) {
+        if (s == null || CIF_NOT_PRESENT.equals(s)) return NOT_PRESENT;
+        if (CIF_UNKNOWN.equals(s)) return UNKNOWN;
+        return PRESENT;
+    }
 }

src/main/java/org/rcsb/cif/model/binary/BinaryStrColumn.java

@@ -1,6 +1,7 @@
 package org.rcsb.cif.model.binary;
 
 import org.rcsb.cif.model.StrColumn;
+import org.rcsb.cif.model.ValueKind;
 
 public class BinaryStrColumn extends BinaryColumn<String[]> implements StrColumn {
     private final String[] data;
@@ -21,7 +22,7 @@ public String getStringData(int row) {
     }
 
     private String honorValueKind(String value) {
-        return (".".equals(value) || "?".equals(value)) ? "" : value;
+        return ValueKind.isValueKindToken(value) ? "" : value;
     }
 
     @Override