CeON / dataverse / 987

Build # 987

Build Type

push

jenkins

Committed by GitHub

Commit Message

Closes #2339: Change File Labels API (#2347)

Run Details

124 of 124 new or added lines in 5 files covered. (100.0%)

21199 of 69091 relevant lines covered (30.68%)

0.31 hits per line

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

94.2

/dataverse-webapp/src/main/java/edu/harvard/iq/dataverse/dataset/FileLabelsService.java

package edu.harvard.iq.dataverse.dataset;

import edu.harvard.iq.dataverse.DataverseRequestServiceBean;
import edu.harvard.iq.dataverse.EjbDataverseEngine;
import edu.harvard.iq.dataverse.annotations.PermissionNeeded;
import edu.harvard.iq.dataverse.api.dto.FileLabelsChangeOptionsDTO;
import edu.harvard.iq.dataverse.engine.command.impl.UpdateDatasetVersionCommand;
import edu.harvard.iq.dataverse.ingest.IngestUtil;
import edu.harvard.iq.dataverse.interceptors.Restricted;
import edu.harvard.iq.dataverse.persistence.datafile.FileMetadata;
import edu.harvard.iq.dataverse.persistence.dataset.Dataset;
import edu.harvard.iq.dataverse.persistence.dataset.DatasetVersion;
import edu.harvard.iq.dataverse.persistence.dataset.DatasetVersionRepository;
import edu.harvard.iq.dataverse.persistence.user.Permission;

import javax.ejb.Stateless;
import javax.inject.Inject;
import java.util.ArrayList;
import java.util.Collections;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.function.Predicate;
import java.util.stream.Collectors;

@Stateless
public class FileLabelsService {

    private DatasetVersionRepository datasetVersionRepository;
    private EjbDataverseEngine dataverseEngine;
    private DataverseRequestServiceBean requestServiceBean;

    // -------------------- CONSTRUCTORS --------------------

    public FileLabelsService() { }

    @Inject
    public FileLabelsService(DatasetVersionRepository datasetVersionRepository, EjbDataverseEngine dataverseEngine,
                             DataverseRequestServiceBean requestServiceBean) {
        this.datasetVersionRepository = datasetVersionRepository;
        this.dataverseEngine = dataverseEngine;
        this.requestServiceBean = requestServiceBean;
    }

    // -------------------- LOGIC --------------------

    /**
     * Return all file labels from the latest version of dataset, and mark as
     * affected those matching the given pattern from {@link FileLabelsChangeOptionsDTO}.
     */
    @Restricted(@PermissionNeeded(needs = {Permission.ViewUnpublishedDataset}, allRequired = true))
    public List<FileLabelInfo> prepareFileLabels(@PermissionNeeded Dataset dataset, FileLabelsChangeOptionsDTO options) {
        DatasetVersion latestVersion = dataset.getLatestVersion();
        List<FileMetadata> fileMetadatas = latestVersion.getFileMetadatas();
        Predicate<String> labelMatcher = prepareFileLabelMatcher(options.getPattern());
        List<FileLabelInfo> result = new ArrayList<>();
        for (FileMetadata fileMetadata : fileMetadatas) {
            Long fileId = fileMetadata.getDataFile().getId();
            boolean labelMatches = labelMatcher.test(fileMetadata.getLabel());
            boolean affected = (labelMatches || options.getFilesToIncludeIds().contains(fileId))
                    && !options.getFilesToExcludeIds().contains(fileId);
            result.add(new FileLabelInfo(fileId, fileMetadata.getLabel(), affected));
        }
        Collections.sort(result);
        return result;
    }

    /**
     * The method receives the output of {@link FileLabelsService#prepareFileLabels(Dataset, FileLabelsChangeOptionsDTO)}
     * method, and for the entries marked as affected it tries to change the
     * label according to the given options. If the label after change differs
     * from the original label the entry is marked as affected. At the end
     * all new labels are checked against the rest in order to find and handle
     * duplicates.
     */
    public List<FileLabelInfo> changeLabels(List<FileLabelInfo> inputLabels, FileLabelsChangeOptionsDTO options) {
        List<FileLabelInfo> changedLabels = new ArrayList<>();
        for (FileLabelInfo labelInfo : inputLabels) {
            if (!labelInfo.isAffected()) {
                changedLabels.add(labelInfo);
                continue;
            }
            String fileLabel = labelInfo.getLabel();
            String changed = fileLabel.replaceAll(options.getFrom(), options.getTo());
            boolean affected = !fileLabel.equals(changed);
            changedLabels.add(new FileLabelInfo(labelInfo.getId(), fileLabel,
                    affected ? changed : labelInfo.getLabelAfterChange(), affected));
        }
        // Handle duplicated names: first extract labels that aren't going to change to a map, then for each
        // changed label check whether it's a duplicate – if so keep generating new labels until it's not.
        // Then add this new label into the map, and check the next label.
        Map<String, FileLabelInfo> finalLabels = new HashMap<>(changedLabels.stream()
                .filter(e -> !e.isAffected())
                .collect(Collectors.toMap(FileLabelInfo::getLabel, e -> e)));
        for (FileLabelInfo labelInfo : changedLabels) {
            if (!labelInfo.isAffected()) {
                continue;
            }
            String newLabel = labelInfo.getLabelAfterChange();
            while (finalLabels.containsKey(newLabel)) {
                newLabel = IngestUtil.generateNewFileName(newLabel);
            }
            finalLabels.put(newLabel, new FileLabelInfo(labelInfo.getId(), labelInfo.getLabel(), newLabel, true));
        }
        return finalLabels.values().stream()
                .sorted()
                .collect(Collectors.toList());
    }

    /**
     * The method receives the output from {@link FileLabelsService#changeLabels(List, FileLabelsChangeOptionsDTO)}
     * and saves the labels marked as affected into the draft version of the
     * given dataset.
     */
    @Restricted(@PermissionNeeded(needs = {Permission.EditDataset}, allRequired = true))
    public List<FileLabelInfo> updateDataset(@PermissionNeeded Dataset dataset, List<FileLabelInfo> labelsToChange,
                                             FileLabelsChangeOptionsDTO options) {
        boolean shouldProcess = labelsToChange.stream().anyMatch(FileLabelInfo::isAffected);
        if (options.isPreview() || !shouldProcess) {
            return labelsToChange;
        }
        if (!dataset.getLatestVersion().isWorkingCopy()) {
            dataset.getEditVersion();
            dataverseEngine.submit(new UpdateDatasetVersionCommand(dataset, requestServiceBean.getDataverseRequest()));
        }
        DatasetVersion editVersion = dataset.getEditVersion();
        Map<Long, FileLabelInfo> labelsToChangeById = labelsToChange.stream()
                .filter(FileLabelInfo::isAffected)
                .collect(Collectors.toMap(FileLabelInfo::getId, l -> l));
        List<FileMetadata> fileMetadatas = editVersion.getFileMetadatas();
        for (FileMetadata metadata : fileMetadatas) {
            Long fileId = metadata.getDataFile().getId();
            if (!labelsToChangeById.containsKey(fileId)) {
                continue;
            }
            metadata.setLabel(labelsToChangeById.get(fileId).getLabelAfterChange());
        }
        datasetVersionRepository.save(editVersion);
        return labelsToChange;
    }

    // -------------------- PRIVATE --------------------

    /**
     * Label matching works in two modes. When wildcard (*) is used, the
     * pattern is matched against the whole label and the wildcard stands
     * for any number (including zero) of any characters. When the wildcard
     * is not present, there is a check whether the label contains the given
     * pattern.
     */
    private Predicate<String> prepareFileLabelMatcher(String labelPattern) {
        if ("*".equals(labelPattern)) {
            // that is all-matching pattern, so
            return s -> true;
        } else if (labelPattern.contains("*")) {
            String pattern = labelPattern.replaceAll("\\*", ".*");
            return s -> s.matches(pattern);
        } else {
            return s -> s.contains(labelPattern);
        }
    }
}

1	package edu.harvard.iq.dataverse.dataset;
2
3	import edu.harvard.iq.dataverse.DataverseRequestServiceBean;
4	import edu.harvard.iq.dataverse.EjbDataverseEngine;
5	import edu.harvard.iq.dataverse.annotations.PermissionNeeded;
6	import edu.harvard.iq.dataverse.api.dto.FileLabelsChangeOptionsDTO;
7	import edu.harvard.iq.dataverse.engine.command.impl.UpdateDatasetVersionCommand;
8	import edu.harvard.iq.dataverse.ingest.IngestUtil;
9	import edu.harvard.iq.dataverse.interceptors.Restricted;
10	import edu.harvard.iq.dataverse.persistence.datafile.FileMetadata;
11	import edu.harvard.iq.dataverse.persistence.dataset.Dataset;
12	import edu.harvard.iq.dataverse.persistence.dataset.DatasetVersion;
13	import edu.harvard.iq.dataverse.persistence.dataset.DatasetVersionRepository;
14	import edu.harvard.iq.dataverse.persistence.user.Permission;
15
16	import javax.ejb.Stateless;
17	import javax.inject.Inject;
18	import java.util.ArrayList;
19	import java.util.Collections;
20	import java.util.HashMap;
21	import java.util.List;
22	import java.util.Map;
23	import java.util.function.Predicate;
24	import java.util.stream.Collectors;
25
26	@Stateless
27	public class FileLabelsService {
28
29	private DatasetVersionRepository datasetVersionRepository;
30	private EjbDataverseEngine dataverseEngine;
31	private DataverseRequestServiceBean requestServiceBean;
32
33	// -------------------- CONSTRUCTORS --------------------
34
35	public FileLabelsService() { }	×
36
37	@Inject
38	public FileLabelsService(DatasetVersionRepository datasetVersionRepository, EjbDataverseEngine dataverseEngine,
39	DataverseRequestServiceBean requestServiceBean) {	1✔
40	this.datasetVersionRepository = datasetVersionRepository;	1✔
41	this.dataverseEngine = dataverseEngine;	1✔
42	this.requestServiceBean = requestServiceBean;	1✔
43	}	1✔
44
45	// -------------------- LOGIC --------------------
46
47	/**
48	* Return all file labels from the latest version of dataset, and mark as
49	* affected those matching the given pattern from {@link FileLabelsChangeOptionsDTO}.
50	*/
51	@Restricted(@PermissionNeeded(needs = {Permission.ViewUnpublishedDataset}, allRequired = true))
52	public List<FileLabelInfo> prepareFileLabels(@PermissionNeeded Dataset dataset, FileLabelsChangeOptionsDTO options) {
53	DatasetVersion latestVersion = dataset.getLatestVersion();	1✔
54	List<FileMetadata> fileMetadatas = latestVersion.getFileMetadatas();	1✔
55	Predicate<String> labelMatcher = prepareFileLabelMatcher(options.getPattern());	1✔
56	List<FileLabelInfo> result = new ArrayList<>();	1✔
57	for (FileMetadata fileMetadata : fileMetadatas) {	1✔
58	Long fileId = fileMetadata.getDataFile().getId();	1✔
59	boolean labelMatches = labelMatcher.test(fileMetadata.getLabel());	1✔
60	boolean affected = (labelMatches \|\| options.getFilesToIncludeIds().contains(fileId))	1✔
61	&& !options.getFilesToExcludeIds().contains(fileId);	1✔
62	result.add(new FileLabelInfo(fileId, fileMetadata.getLabel(), affected));	1✔
63	}	1✔
64	Collections.sort(result);	1✔
65	return result;	1✔
66	}
67
68	/**
69	* The method receives the output of {@link FileLabelsService#prepareFileLabels(Dataset, FileLabelsChangeOptionsDTO)}
70	* method, and for the entries marked as affected it tries to change the
71	* label according to the given options. If the label after change differs
72	* from the original label the entry is marked as affected. At the end
73	* all new labels are checked against the rest in order to find and handle
74	* duplicates.
75	*/
76	public List<FileLabelInfo> changeLabels(List<FileLabelInfo> inputLabels, FileLabelsChangeOptionsDTO options) {
77	List<FileLabelInfo> changedLabels = new ArrayList<>();	1✔
78	for (FileLabelInfo labelInfo : inputLabels) {	1✔
79	if (!labelInfo.isAffected()) {	1✔
80	changedLabels.add(labelInfo);	1✔
81	continue;	1✔
82	}
83	String fileLabel = labelInfo.getLabel();	1✔
84	String changed = fileLabel.replaceAll(options.getFrom(), options.getTo());	1✔
85	boolean affected = !fileLabel.equals(changed);	1✔
86	changedLabels.add(new FileLabelInfo(labelInfo.getId(), fileLabel,	1✔
87	affected ? changed : labelInfo.getLabelAfterChange(), affected));	1✔
88	}	1✔
89	// Handle duplicated names: first extract labels that aren't going to change to a map, then for each
90	// changed label check whether it's a duplicate – if so keep generating new labels until it's not.
91	// Then add this new label into the map, and check the next label.
92	Map<String, FileLabelInfo> finalLabels = new HashMap<>(changedLabels.stream()	1✔
93	.filter(e -> !e.isAffected())	1✔
94	.collect(Collectors.toMap(FileLabelInfo::getLabel, e -> e)));	1✔
95	for (FileLabelInfo labelInfo : changedLabels) {	1✔
96	if (!labelInfo.isAffected()) {	1✔
97	continue;	1✔
98	}
99	String newLabel = labelInfo.getLabelAfterChange();	1✔
100	while (finalLabels.containsKey(newLabel)) {	1✔
101	newLabel = IngestUtil.generateNewFileName(newLabel);	1✔
102	}
103	finalLabels.put(newLabel, new FileLabelInfo(labelInfo.getId(), labelInfo.getLabel(), newLabel, true));	1✔
104	}	1✔
105	return finalLabels.values().stream()	1✔
106	.sorted()	1✔
107	.collect(Collectors.toList());	1✔
108	}
109
110	/**
111	* The method receives the output from {@link FileLabelsService#changeLabels(List, FileLabelsChangeOptionsDTO)}
112	* and saves the labels marked as affected into the draft version of the
113	* given dataset.
114	*/
115	@Restricted(@PermissionNeeded(needs = {Permission.EditDataset}, allRequired = true))
116	public List<FileLabelInfo> updateDataset(@PermissionNeeded Dataset dataset, List<FileLabelInfo> labelsToChange,
117	FileLabelsChangeOptionsDTO options) {
118	boolean shouldProcess = labelsToChange.stream().anyMatch(FileLabelInfo::isAffected);	1✔
119	if (options.isPreview() \|\| !shouldProcess) {	1✔
120	return labelsToChange;	×
121	}
122	if (!dataset.getLatestVersion().isWorkingCopy()) {	1✔
123	dataset.getEditVersion();	×
124	dataverseEngine.submit(new UpdateDatasetVersionCommand(dataset, requestServiceBean.getDataverseRequest()));	×
125	}
126	DatasetVersion editVersion = dataset.getEditVersion();	1✔
127	Map<Long, FileLabelInfo> labelsToChangeById = labelsToChange.stream()	1✔
128	.filter(FileLabelInfo::isAffected)	1✔
129	.collect(Collectors.toMap(FileLabelInfo::getId, l -> l));	1✔
130	List<FileMetadata> fileMetadatas = editVersion.getFileMetadatas();	1✔
131	for (FileMetadata metadata : fileMetadatas) {	1✔
132	Long fileId = metadata.getDataFile().getId();	1✔
133	if (!labelsToChangeById.containsKey(fileId)) {	1✔
134	continue;	1✔
135	}
136	metadata.setLabel(labelsToChangeById.get(fileId).getLabelAfterChange());	1✔
137	}	1✔
138	datasetVersionRepository.save(editVersion);	1✔
139	return labelsToChange;	1✔
140	}
141
142	// -------------------- PRIVATE --------------------
143
144	/**
145	* Label matching works in two modes. When wildcard (*) is used, the
146	* pattern is matched against the whole label and the wildcard stands
147	* for any number (including zero) of any characters. When the wildcard
148	* is not present, there is a check whether the label contains the given
149	* pattern.
150	*/
151	private Predicate<String> prepareFileLabelMatcher(String labelPattern) {
152	if ("*".equals(labelPattern)) {	1✔
153	// that is all-matching pattern, so
154	return s -> true;	1✔
155	} else if (labelPattern.contains("*")) {	1✔
156	String pattern = labelPattern.replaceAll("\\", ".");	1✔
157	return s -> s.matches(pattern);	1✔
158	} else {
159	return s -> s.contains(labelPattern);	1✔
160	}
161	}
162	}

CeON / dataverse / 987

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