#430

Committed 19 May 2025 01:23PM UTC coverage: 90.495% (+0.3%) from 90.169%

Build # #430

Build Type

Pull #258

github

Committed by

web-flow

Commit Message

Merge b22e70bd8 into 810d1995c

Pull Request Pull Request #258: Support .ELN-style crates in all zip readers and writers

Run Details

276 of 305 new or added lines in 21 files covered. (90.49%)

1 existing line in 1 file now uncovered.

2009 of 2220 relevant lines covered (90.5%)

0.9 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

95.0

/src/main/java/edu/kit/datamanager/ro_crate/reader/ReadZipStrategy.java

package edu.kit.datamanager.ro_crate.reader;

import com.fasterxml.jackson.databind.ObjectMapper;
import com.fasterxml.jackson.databind.node.ObjectNode;
import edu.kit.datamanager.ro_crate.entities.contextual.JsonDescriptor;
import edu.kit.datamanager.ro_crate.objectmapper.MyObjectMapper;
import net.lingala.zip4j.ZipFile;
import org.apache.commons.io.FileUtils;
import org.apache.commons.io.filefilter.FileFilterUtils;

import java.io.File;
import java.io.IOException;
import java.nio.file.Path;
import java.util.UUID;

/**
 * Reads a crate from a ZIP archive (file).
 * <p>
 * This class handles reading and extraction of RO-Crate content from ZIP archives
 * into a temporary directory structure on the file system,
 * which allows accessing the contained files.
 * <p>
 * Supports <a href=https://github.com/TheELNConsortium/TheELNFileFormat>ELN-Style crates</a>,
 * meaning the crate may be either in the zip archive directly or in a single,
 * direct subfolder beneath the root folder (/folder).
 * <p>
 * Note: This implementation checks for up to 50 subdirectories if multiple are present.
 * This is to avoid zip bombs, which may contain a lot of subdirectories,
 * and at the same time gracefully handle valid crated with hidden subdirectories
 * (for example, thumbnails).
 * <p>
 * NOTE: The resulting crate may refer to these temporary files. Therefore,
 * these files are only being deleted before the JVM exits. If you need to free
 * space because your application is long-running or creates a lot of
 * crates, you may use the getters to retrieve information which will help
 * you to clean up manually. Keep in mind that crates may refer to this
 * folder after extraction. Use RoCrateWriter to export it so some
 * persistent location and possibly read it from there, if required. Or use
 * the ZipWriter to write it back to its source.
 */
public class ReadZipStrategy implements GenericReaderStrategy<String> {

  protected final String ID = UUID.randomUUID().toString();
  protected Path temporaryFolder = Path.of(String.format("./.tmp/ro-crate-java/zipReader/%s/", ID));
  protected boolean isExtracted = false;

  /**
   * Crates an instance with the default configuration.
   * <p>
   * The default configuration is to extract the ZipFile to
   * `./.tmp/ro-crate-java/zipReader/$UUID/`.
   */
  public ReadZipStrategy() {}

  /**
   * Creates a ZipReader which will extract the contents temporary
   * to the given location instead of the default location.
   *
   * @param folderPath            the custom directory to extract
   *                              content to for temporary access.
   * @param shallAddUuidSubfolder if true, the reader will extract
   *                              into subdirectories of the given
   *                              directory. These subdirectories
   *                              will have UUIDs as their names.
   */
  public ReadZipStrategy(Path folderPath, boolean shallAddUuidSubfolder) {
    if (shallAddUuidSubfolder) {
      this.temporaryFolder = folderPath.resolve(ID);
    } else {
      this.temporaryFolder = folderPath;
    }
  }

  /**
   * @return the identifier which may be used as the name for a subfolder in the temporary directory.
   */
  public String getID() {
    return ID;
  }

  /**
   * @return the folder (considered temporary) where the zipped crate will be or has been extracted to.
   */
  public Path getTemporaryFolder() {
    return temporaryFolder;
  }

  /**
   * @return whether the crate has already been extracted into the temporary folder.
   */
  public boolean isExtracted() {
    return isExtracted;
  }

  private void readCrate(String location) throws IOException {
    File folder = temporaryFolder.toFile();
    // ensure the directory is clean
    if (folder.isDirectory()) {
      FileUtils.cleanDirectory(folder);
    } else if (folder.isFile()) {
      FileUtils.delete(folder);
    }
    // extract
    try (ZipFile zf = new ZipFile(location)) {
      zf.extractAll(temporaryFolder.toAbsolutePath().toString());
      this.isExtracted = true;
    }
    // register deletion on exit
    FileUtils.forceDeleteOnExit(folder);
  }

  @Override
  public ObjectNode readMetadataJson(String location) throws IOException {
    if (!isExtracted) {
      this.readCrate(location);
    }

    ObjectMapper objectMapper = MyObjectMapper.getMapper();
    File jsonMetadata = this.temporaryFolder.resolve(JsonDescriptor.ID).toFile();
    if (!jsonMetadata.isFile()) {
      // Try to find the metadata file in subdirectories
      File firstSubdir = FileUtils.listFilesAndDirs(
            temporaryFolder.toFile(),
            FileFilterUtils.directoryFileFilter(),
            null // not recursive
        )
        .stream()
        .limit(50)
        .filter(file -> file.toPath().toAbsolutePath().resolve(JsonDescriptor.ID).toFile().isFile())
        .findFirst()
        .orElseThrow(() -> new IllegalStateException("No %s found in zip file".formatted(JsonDescriptor.ID)));
      jsonMetadata = firstSubdir.toPath().resolve(JsonDescriptor.ID).toFile();
    }

    return objectMapper.readTree(jsonMetadata).deepCopy();
  }

  @Override
  public File readContent(String location) throws IOException {
    if (!isExtracted) {
      this.readCrate(location);
    }
    return temporaryFolder.toFile();
  }
}

1	package edu.kit.datamanager.ro_crate.reader;
2
3	import com.fasterxml.jackson.databind.ObjectMapper;
4	import com.fasterxml.jackson.databind.node.ObjectNode;
5	import edu.kit.datamanager.ro_crate.entities.contextual.JsonDescriptor;
6	import edu.kit.datamanager.ro_crate.objectmapper.MyObjectMapper;
7	import net.lingala.zip4j.ZipFile;
8	import org.apache.commons.io.FileUtils;
9	import org.apache.commons.io.filefilter.FileFilterUtils;
10
11	import java.io.File;
12	import java.io.IOException;
13	import java.nio.file.Path;
14	import java.util.UUID;
15
16	/**
17	* Reads a crate from a ZIP archive (file).
18	* <p>
19	* This class handles reading and extraction of RO-Crate content from ZIP archives
20	* into a temporary directory structure on the file system,
21	* which allows accessing the contained files.
22	* <p>
23	* Supports <a href=https://github.com/TheELNConsortium/TheELNFileFormat>ELN-Style crates</a>,
24	* meaning the crate may be either in the zip archive directly or in a single,
25	* direct subfolder beneath the root folder (/folder).
26	* <p>
27	* Note: This implementation checks for up to 50 subdirectories if multiple are present.
28	* This is to avoid zip bombs, which may contain a lot of subdirectories,
29	* and at the same time gracefully handle valid crated with hidden subdirectories
30	* (for example, thumbnails).
31	* <p>
32	* NOTE: The resulting crate may refer to these temporary files. Therefore,
33	* these files are only being deleted before the JVM exits. If you need to free
34	* space because your application is long-running or creates a lot of
35	* crates, you may use the getters to retrieve information which will help
36	* you to clean up manually. Keep in mind that crates may refer to this
37	* folder after extraction. Use RoCrateWriter to export it so some
38	* persistent location and possibly read it from there, if required. Or use
39	* the ZipWriter to write it back to its source.
40	*/
41	public class ReadZipStrategy implements GenericReaderStrategy<String> {
42
43	protected final String ID = UUID.randomUUID().toString();	1✔
44	protected Path temporaryFolder = Path.of(String.format("./.tmp/ro-crate-java/zipReader/%s/", ID));	1✔
45	protected boolean isExtracted = false;	1✔
46
47	/**
48	* Crates an instance with the default configuration.
49	* <p>
50	* The default configuration is to extract the ZipFile to
51	* `./.tmp/ro-crate-java/zipReader/$UUID/`.
52	*/
53	public ReadZipStrategy() {}	1✔
54
55	/**
56	* Creates a ZipReader which will extract the contents temporary
57	* to the given location instead of the default location.
58	*
59	* @param folderPath the custom directory to extract
60	* content to for temporary access.
61	* @param shallAddUuidSubfolder if true, the reader will extract
62	* into subdirectories of the given
63	* directory. These subdirectories
64	* will have UUIDs as their names.
65	*/
66	public ReadZipStrategy(Path folderPath, boolean shallAddUuidSubfolder) {	1✔
67	if (shallAddUuidSubfolder) {	1✔
68	this.temporaryFolder = folderPath.resolve(ID);	1✔
69	} else {
70	this.temporaryFolder = folderPath;	1✔
71	}
72	}	1✔
73
74	/**
75	* @return the identifier which may be used as the name for a subfolder in the temporary directory.
76	*/
77	public String getID() {
78	return ID;	1✔
79	}
80
81	/**
82	* @return the folder (considered temporary) where the zipped crate will be or has been extracted to.
83	*/
84	public Path getTemporaryFolder() {
85	return temporaryFolder;	1✔
86	}
87
88	/**
89	* @return whether the crate has already been extracted into the temporary folder.
90	*/
91	public boolean isExtracted() {
92	return isExtracted;	1✔
93	}
94
95	private void readCrate(String location) throws IOException {
96	File folder = temporaryFolder.toFile();	1✔
97	// ensure the directory is clean
98	if (folder.isDirectory()) {	1✔
99	FileUtils.cleanDirectory(folder);	1✔
100	} else if (folder.isFile()) {	1✔
NEW 101	FileUtils.delete(folder);	×
102	}
103	// extract
104	try (ZipFile zf = new ZipFile(location)) {	1✔
105	zf.extractAll(temporaryFolder.toAbsolutePath().toString());	1✔
106	this.isExtracted = true;	1✔
107	}
108	// register deletion on exit
109	FileUtils.forceDeleteOnExit(folder);	1✔
110	}	1✔
111
112	@Override
113	public ObjectNode readMetadataJson(String location) throws IOException {
114	if (!isExtracted) {	1✔
115	this.readCrate(location);	1✔
116	}
117
118	ObjectMapper objectMapper = MyObjectMapper.getMapper();	1✔
119	File jsonMetadata = this.temporaryFolder.resolve(JsonDescriptor.ID).toFile();	1✔
120	if (!jsonMetadata.isFile()) {	1✔
121	// Try to find the metadata file in subdirectories
122	File firstSubdir = FileUtils.listFilesAndDirs(	1✔
123	temporaryFolder.toFile(),	1✔
124	FileFilterUtils.directoryFileFilter(),	1✔
125	null // not recursive
126	)
127	.stream()	1✔
128	.limit(50)	1✔
129	.filter(file -> file.toPath().toAbsolutePath().resolve(JsonDescriptor.ID).toFile().isFile())	1✔
130	.findFirst()	1✔
131	.orElseThrow(() -> new IllegalStateException("No %s found in zip file".formatted(JsonDescriptor.ID)));	1✔
132	jsonMetadata = firstSubdir.toPath().resolve(JsonDescriptor.ID).toFile();	1✔
133	}
134
135	return objectMapper.readTree(jsonMetadata).deepCopy();	1✔
136	}
137
138	@Override
139	public File readContent(String location) throws IOException {
140	if (!isExtracted) {	1✔
141	this.readCrate(location);	×
142	}
143	return temporaryFolder.toFile();	1✔
144	}
145	}

kit-data-manager / ro-crate-java / #430

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous