Merge pull request #14 from richardrodgers/archive_ext

Refactor archive timestamp suppression - API change. Closes #13

README.md

@@ -6,15 +6,17 @@ commons compression library for support of tarred Gzip archive format (".tgz"),
 
 [![Build Status](https://travis-ci.org/richardrodgers/bagit.svg?branch=master)]
 (https://travis-ci.org/richardrodgers/bagit)
+[![Dependency Status](https://dependencyci.com/github/richardrodgers/bagit/badge)]
+(https://dependencyci.com/github/richardrodgers/bagit)
 
 ## Use Cases ##
 
 The library attempts to simplify a few of the most common use cases/patterns involving bag packages.
 The first (the _producer_ pattern) is where content is assembled and placed into a bag, and the bag is then serialized
 for transport/hand-off to another component or system. The goal here is to ensure that the constructed bag is correct.
-A helper class - bag _Filler_ - is used to orchestrate this assembly. Sequence: new Filler -> add content -> add more content -> serialize.
+A helper class - _Filler_ - is used to orchestrate this assembly. Sequence: new Filler -> add content -> add more content -> serialize.
 The second (the _consumer_ pattern) is where a bag serialization (or a loose directory) is given and must
-be interpreted and validated for use. Here another helper class _Loader_ is used to deserialize.
+be interpreted and validated for use. Here another helper class - _Loader_ - is used to deserialize.
 Sequence: new Loader -> load serialization -> convert to Bag -> process contents. If you have more complex needs
 in java, (e.g. support for multiple spec versions), you may wish to consider the [Library of Congress Java Library](https://github.com/LibraryOfCongress/bagit-java).
 
@@ -35,7 +37,7 @@ archive. To convert the same bag to use a compressed tar format:
 
     InputStream bagStream = filler.toStream("tgz");
 
-We don't always need bag I/O streams - suppose we wish obtain a reference to an archive file object instead:
+We don't always want bag I/O streams - suppose we wish to obtain a bag archive file package instead:
 
     Path bagPackage = new Filler().payload(file1).metadata("External-Identifier", "mit.edu.0001").toPackage();
 
@@ -93,10 +95,10 @@ See the [Javadoc](http://richardrodgers.github.io/bagit/javadoc/index.html) for
 ## Archive formats ##
 
 Bags are commonly serialized to standard archive formats such as ZIP. The library supports two archive formats:
-'zip' and 'tgz' and the variants 'zip.nt' and 'tgz.nt' (no time). If these variants are used, the library
-suppresses the file creation/modification time attributes, in order that checksums of archives produced at different times
-may accurately reflect only bag contents. That is, the checksum of a zip.nt bag (of the same name) is time-of-archiving-
-and filesystem-time-invariant, but content-sensitive.
+'zip' and 'tgz' and a variant in each of these. If the variant is used, the library suppresses the file
+creation/modification time attributes, in order that checksums of archives produced at different times
+may accurately reflect only bag contents. That is, the checksum of a zipped bag (with no timestamp variant) is
+time-of-archiving and filesystem-time-invariant, but content-sensitive. The variant is requested with an API call.
 
 ## Extras ##
 
@@ -131,14 +133,14 @@ Fat jars include all dependencies in a single executable jar (no classpath decla
 The distribution jars are kept at [Bintray](https://bintray.com), so make sure that repository is declared.
 Then (NB: using the most current version), for Gradle:
 
-    compile 'edu.mit.lib:bagit:0.7'
+    compile 'edu.mit.lib:bagit:0.8'
 
 or Maven:
 
     <dependency>
       <groupId>edu.mit.lib</groupId>
       <artifactId>bagit</artifactId>
-      <version>0.7</version>
+      <version>0.8</version>
     </dependency>
 
 in a standard pom.xml dependencies block.

build.gradle

@@ -17,7 +17,7 @@ sourceCompatibility = 1.8
 
 group = 'edu.mit.lib'
 archivesBaseName = 'bagit'
-version = '0.7'
+version = '0.8'
 description = 'Compact Java BagIt library'
 
 ext {
@@ -28,7 +28,7 @@ ext {
 }
 
 dependencies {
-    compile group: 'org.apache.commons', name: 'commons-compress', version: '1.9'
+    compile group: 'org.apache.commons', name: 'commons-compress', version: '1.13'
     testCompile group: 'junit', name: 'junit', version: '4.11'
 }
 

src/main/java/edu/mit/lib/bagit/Bag.java

@@ -14,7 +14,6 @@
 import java.nio.file.Path;
 import java.nio.file.attribute.BasicFileAttributes;
 import java.security.DigestInputStream;
-import java.security.DigestOutputStream;
 import java.security.MessageDigest;
 import java.security.NoSuchAlgorithmException;
 import java.util.ArrayList;
@@ -44,7 +43,7 @@ public class Bag {
     static final String ENCODING = "UTF-8";
     static final String CS_ALGO = "MD5";
     static final String BAGIT_VSN = "0.97";
-    static final String LIB_VSN = "0.7";
+    static final String LIB_VSN = "0.8";
     static final String DFLT_FMT = "zip";
     static final String TGZIP_FMT = "tgz";
     static final String SPACER = "  ";
@@ -77,9 +76,9 @@ public enum MetadataName {
         INTERNAL_SENDER_DESC("Internal-Sender-Description"),
         BAG_SOFTWARE_AGENT("Bag-Software-Agent");  // not in IETF spec
 
-        private String mdName;
+        private final String mdName;
 
-        private MetadataName(String name) {
+        MetadataName(String name) {
             mdName = name;
         }
 
@@ -168,11 +167,9 @@ public boolean isComplete() throws IOException {
         // # tag files and # manifest entries must agree
         // tag files consist of any top-level files except:
         // tagmanifest itself, and the payload directory.
-        DirectoryStream.Filter<Path> filter = new DirectoryStream.Filter<Path>() {
-            public boolean accept(Path file) throws IOException {
-                String name = file.getFileName().toString();
-                return ! (name.startsWith(TAGMANIF_FILE) || name.startsWith(DATA_DIR));
-            }
+        DirectoryStream.Filter<Path> filter = file -> {
+            String name = file.getFileName().toString();
+            return ! (name.startsWith(TAGMANIF_FILE) || name.startsWith(DATA_DIR));
         };
         int tagCount = 0;
         try (DirectoryStream<Path> stream = Files.newDirectoryStream(baseDir, filter)) {
@@ -277,10 +274,9 @@ public Map<String, String> payloadRefs() throws IOException {
      *
      * @param relPath the relative path of the file from the bag root directory
      * @return tagfile the tag file path, or null if no file at the specified path
-     * @throws IOException if unable to access tag file
      * @throws IllegalAccessException if bag is sealed
      */
-    public Path tagFile(String relPath) throws IOException, IllegalAccessException {
+    public Path tagFile(String relPath) throws IllegalAccessException {
         if (sealed) {
             throw new IllegalAccessException("Sealed Bag: no file access allowed");
         }
@@ -354,7 +350,7 @@ public List<String> property(String relPath, String name) throws IOException {
                 try (BufferedReader reader = Files.newBufferedReader(bagFile(relPath), StandardCharsets.UTF_8)) {
                     String propName = null;
                     StringBuilder valSb = new StringBuilder();
-                    String line = null;
+                    String line;
                     while ((line = reader.readLine()) != null) {
                         // if line does not start with spacer, it is a new property
                         if (! line.startsWith(SPACER)) {
@@ -414,7 +410,7 @@ public Map<String, String> tagManifest() throws IOException {
     public Map<String, String> manifest(String relPath) throws IOException {
         Map<String, String> mfMap = new HashMap<>();
         try (BufferedReader reader = Files.newBufferedReader(bagFile(relPath), StandardCharsets.UTF_8)) {
-            String line = null;
+            String line;
             while((line = reader.readLine()) != null) {
                 String[] parts = line.split(" ");
                 mfMap.put(parts[1], parts[0]);
@@ -439,11 +435,7 @@ private int fileCount(Path dir) throws IOException {
     }
 
     private void addProp(String name, String value, Map<String, List<String>> mdSet) {
-        List<String> vals = mdSet.get(name);
-        if (vals == null) {
-            vals = new ArrayList<>();
-            mdSet.put(name, vals);
-        }
+        List<String> vals = mdSet.computeIfAbsent(name, k -> new ArrayList<>());
         vals.add(value.trim());
     }
 
@@ -484,7 +476,7 @@ static Map<String, String> payloadRefs(Path refFile) throws IOException {
         Map<String, String> refMap = new HashMap<>();
         if (Files.exists(refFile)) {
             try (BufferedReader reader = Files.newBufferedReader(refFile, StandardCharsets.UTF_8)) {
-                String line = null;
+                String line;
                 while((line = reader.readLine()) != null) {
                     String[] parts = line.split(" ");
                     refMap.put(parts[2], parts[0]);

src/main/java/edu/mit/lib/bagit/Bagger.java

@@ -26,13 +26,14 @@
 public class Bagger {
     /* A bit clunky in the cmd-line arg handling, but deliberately so as to limit
        external dependencies for those who want to only use the library API directly. */
-    private List<String> payloads = new ArrayList<>();
-    private List<String> references = new ArrayList<>();
-    private List<String> tags = new ArrayList<>();
-    private List<String> statements = new ArrayList<>();
+    private final List<String> payloads = new ArrayList<>();
+    private final List<String> references = new ArrayList<>();
+    private final List<String> tags = new ArrayList<>();
+    private final List<String> statements = new ArrayList<>();
     private String archFmt = "directory";
+    private boolean noTime = false;
     private String csAlg = "MD5";
-    private List<String> optFlags = new ArrayList<>();
+    private final List<String> optFlags = new ArrayList<>();
     private int verbosityLevel;
 
     public static void main(String[] args) throws IOException, IllegalAccessException {
@@ -47,6 +48,7 @@ public static void main(String[] args) throws IOException, IllegalAccessExceptio
                 case "-r": bagger.references.add(args[i+1]); break;
                 case "-t": bagger.tags.add(args[i+1]); break;
                 case "-m": bagger.statements.add(args[i+1]); break;
+                case "-n": bagger.noTime = Boolean.valueOf(args[i+1]); break;
                 case "-a": bagger.archFmt = args[i+1]; break;
                 case "-c": bagger.csAlg = args[i+1]; break;
                 case "-o": bagger.optFlags.add(args[i+1]); break;
@@ -80,7 +82,8 @@ public static void usage() {
             "-r    <bag path>=<URL> - payload reference\n" +
             "-t    [<bag path>=]<tag file>\n" +
             "-m    <name>=<value> - metadata statement\n" +
-            "-a    <archive format> - e.g. 'zip', 'zip.nt', 'tgz', 'tgz.nt' (default: loose directory)\n" +
+            "-a    <archive format> - e.g. 'zip', 'tgz', (default: loose directory)\n" +
+            "-n    <noTime> - 'true' or 'false'\n" +
             "-c    <checksum algorithm> - default: 'MD5'\n" +
             "-o    <optimization flag>\n" +
             "-v    <level> - output level to console (default: 0 = no output)");
@@ -121,11 +124,11 @@ private void fill(Path baseDir) throws IOException {
             String[] parts = statement.split("=");
             filler.metadata(parts[0], parts[1]);
         }
-        Path bagPath = null;
+        Path bagPath;
         if (archFmt.equals("directory")) {
             bagPath = filler.toDirectory();
         } else {
-            bagPath = filler.toPackage(archFmt);
+            bagPath = filler.toPackage(archFmt, noTime);
         }
         if (verbosityLevel > 0) {
             message(bagPath.getFileName().toString(), true, "created");