diff --git a/doc/release-notes/9763-versions-api-improvements.md b/doc/release-notes/9763-versions-api-improvements.md new file mode 100644 index 00000000000..8d7f6c7a20a --- /dev/null +++ b/doc/release-notes/9763-versions-api-improvements.md @@ -0,0 +1,8 @@ +# Improvements in the /versions API + +- optional pagination has been added to `/api/datasets/{id}/versions` that may be useful in datasets with a large number of versions; +- a new flag `includeFiles` is added to both `/api/datasets/{id}/versions` and `/api/datasets/{id}/versions/{vid}` (true by default), providing an option to drop the file information from the output; +- when files are requested to be included, some database lookup optimizations have been added to improve the performance on datasets with large numbers of files. + +This is reflected in the [Dataset Versions API](https://guides.dataverse.org/en/9763-lookup-optimizations/api/native-api.html#dataset-versions-api) section of the Guide. + diff --git a/doc/sphinx-guides/source/api/native-api.rst b/doc/sphinx-guides/source/api/native-api.rst index 2e1a878dce8..ff9e12db094 100644 --- a/doc/sphinx-guides/source/api/native-api.rst +++ b/doc/sphinx-guides/source/api/native-api.rst @@ -889,6 +889,10 @@ It returns a list of versions with their metadata, and file list: ] } +The optional ``includeFiles`` parameter specifies whether the files should be listed in the output. It defaults to ``true``, preserving backward compatibility. (Note that for a dataset with a large number of versions and/or files having the files included can dramatically increase the volume of the output). A separate ``/files`` API can be used for listing the files, or a subset thereof in a given version. + +The optional ``offset`` and ``limit`` parameters can be used to specify the range of the versions list to be shown. This can be used to paginate through the list in a dataset with a large number of versions. + Get Version of a Dataset ~~~~~~~~~~~~~~~~~~~~~~~~ @@ -901,13 +905,16 @@ Get Version of a Dataset export ID=24 export VERSION=1.0 - curl "$SERVER_URL/api/datasets/$ID/versions/$VERSION" + curl "$SERVER_URL/api/datasets/$ID/versions/$VERSION?includeFiles=false" The fully expanded example above (without environment variables) looks like this: .. code-block:: bash - curl "https://demo.dataverse.org/api/datasets/24/versions/1.0" + curl "https://demo.dataverse.org/api/datasets/24/versions/1.0?includeFiles=false" + +The optional ``includeFiles`` parameter specifies whether the files should be listed in the output (defaults to ``true``). Note that a separate ``/files`` API can be used for listing the files, or a subset thereof in a given version. + .. _export-dataset-metadata-api: diff --git a/src/main/java/edu/harvard/iq/dataverse/Dataset.java b/src/main/java/edu/harvard/iq/dataverse/Dataset.java index 21014d7f99c..245bdf0efd2 100644 --- a/src/main/java/edu/harvard/iq/dataverse/Dataset.java +++ b/src/main/java/edu/harvard/iq/dataverse/Dataset.java @@ -158,6 +158,23 @@ public void setCitationDateDatasetFieldType(DatasetFieldType citationDateDataset this.citationDateDatasetFieldType = citationDateDatasetFieldType; } + // Per DataCite best practices, the citation date of a dataset may need + // to be adjusted to reflect the latest embargo availability date of any + // file within the first published version. + // If any files are embargoed in the first version, this date will be + // calculated and cached here upon its publication, in the + // FinalizeDatasetPublicationCommand. + private Timestamp embargoCitationDate; + + public Timestamp getEmbargoCitationDate() { + return embargoCitationDate; + } + + public void setEmbargoCitationDate(Timestamp embargoCitationDate) { + this.embargoCitationDate = embargoCitationDate; + } + + @ManyToOne @JoinColumn(name="template_id",nullable = true) @@ -676,20 +693,10 @@ public Timestamp getCitationDate() { Timestamp citationDate = null; //Only calculate if this dataset doesn't use an alternate date field for publication date if (citationDateDatasetFieldType == null) { - List versions = this.versions; - // TODo - is this ever not version 1.0 (or draft if not published yet) - DatasetVersion oldest = versions.get(versions.size() - 1); citationDate = super.getPublicationDate(); - if (oldest.isPublished()) { - List fms = oldest.getFileMetadatas(); - for (FileMetadata fm : fms) { - Embargo embargo = fm.getDataFile().getEmbargo(); - if (embargo != null) { - Timestamp embDate = Timestamp.valueOf(embargo.getDateAvailable().atStartOfDay()); - if (citationDate.compareTo(embDate) < 0) { - citationDate = embDate; - } - } + if (embargoCitationDate != null) { + if (citationDate.compareTo(embargoCitationDate) < 0) { + return embargoCitationDate; } } } diff --git a/src/main/java/edu/harvard/iq/dataverse/DatasetServiceBean.java b/src/main/java/edu/harvard/iq/dataverse/DatasetServiceBean.java index 29d3c46d59a..c6df2a2e1ab 100644 --- a/src/main/java/edu/harvard/iq/dataverse/DatasetServiceBean.java +++ b/src/main/java/edu/harvard/iq/dataverse/DatasetServiceBean.java @@ -137,7 +137,7 @@ public Dataset findDeep(Object pk) { .setHint("eclipselink.left-join-fetch", "o.files.roleAssignments") .getSingleResult(); } - + public List findByOwnerId(Long ownerId) { return findByOwnerId(ownerId, false); } diff --git a/src/main/java/edu/harvard/iq/dataverse/DatasetVersion.java b/src/main/java/edu/harvard/iq/dataverse/DatasetVersion.java index 93f45bd288e..5fd963f3931 100644 --- a/src/main/java/edu/harvard/iq/dataverse/DatasetVersion.java +++ b/src/main/java/edu/harvard/iq/dataverse/DatasetVersion.java @@ -68,7 +68,13 @@ query = "SELECT OBJECT(o) FROM DatasetVersion AS o WHERE o.dataset.harvestedFrom IS NULL and o.releaseTime IS NOT NULL and o.archivalCopyLocation IS NULL" ), @NamedQuery(name = "DatasetVersion.findById", - query = "SELECT o FROM DatasetVersion o LEFT JOIN FETCH o.fileMetadatas WHERE o.id=:id")}) + query = "SELECT o FROM DatasetVersion o LEFT JOIN FETCH o.fileMetadatas WHERE o.id=:id"), + @NamedQuery(name = "DatasetVersion.findByDataset", + query = "SELECT o FROM DatasetVersion o WHERE o.dataset.id=:datasetId ORDER BY o.versionNumber DESC, o.minorVersionNumber DESC"), + @NamedQuery(name = "DatasetVersion.findReleasedByDataset", + query = "SELECT o FROM DatasetVersion o WHERE o.dataset.id=:datasetId AND o.versionState=edu.harvard.iq.dataverse.DatasetVersion.VersionState.RELEASED ORDER BY o.versionNumber DESC, o.minorVersionNumber DESC")/*, + @NamedQuery(name = "DatasetVersion.findVersionElements", + query = "SELECT o.id, o.versionState, o.versionNumber, o.minorVersionNumber FROM DatasetVersion o WHERE o.dataset.id=:datasetId ORDER BY o.versionNumber DESC, o.minorVersionNumber DESC")*/}) @Entity diff --git a/src/main/java/edu/harvard/iq/dataverse/DatasetVersionServiceBean.java b/src/main/java/edu/harvard/iq/dataverse/DatasetVersionServiceBean.java index 5c43001dcb5..c2f9027a38a 100644 --- a/src/main/java/edu/harvard/iq/dataverse/DatasetVersionServiceBean.java +++ b/src/main/java/edu/harvard/iq/dataverse/DatasetVersionServiceBean.java @@ -166,9 +166,44 @@ public DatasetVersion findDeep(Object pk) { .setHint("eclipselink.left-join-fetch", "o.fileMetadatas.datasetVersion") .setHint("eclipselink.left-join-fetch", "o.fileMetadatas.dataFile.releaseUser") .setHint("eclipselink.left-join-fetch", "o.fileMetadatas.dataFile.creator") + .setHint("eclipselink.left-join-fetch", "o.fileMetadatas.dataFile.dataFileTags") .getSingleResult(); } - + + /** + * Performs the same database lookup as the one behind Dataset.getVersions(). + * Additionally, provides the arguments for selecting a partial list of + * (length-offset) versions for pagination, plus the ability to pre-select + * only the publicly-viewable versions. + * It is recommended that individual software components utilize the + * ListVersionsCommand, instead of calling this service method directly. + * @param datasetId + * @param offset for pagination through long lists of versions + * @param length for pagination through long lists of versions + * @param includeUnpublished retrieves all the versions, including drafts and deaccessioned. + * @return (partial) list of versions + */ + public List findVersions(Long datasetId, Integer offset, Integer length, boolean includeUnpublished) { + TypedQuery query; + if (includeUnpublished) { + query = em.createNamedQuery("DatasetVersion.findByDataset", DatasetVersion.class); + } else { + query = em.createNamedQuery("DatasetVersion.findReleasedByDataset", DatasetVersion.class) + .setParameter("datasetId", datasetId); + } + + query.setParameter("datasetId", datasetId); + + if (offset != null) { + query.setFirstResult(offset); + } + if (length != null) { + query.setMaxResults(length); + } + + return query.getResultList(); + } + public DatasetVersion findByFriendlyVersionNumber(Long datasetId, String friendlyVersionNumber) { Long majorVersionNumber = null; Long minorVersionNumber = null; diff --git a/src/main/java/edu/harvard/iq/dataverse/api/Datasets.java b/src/main/java/edu/harvard/iq/dataverse/api/Datasets.java index f8929c5e8d8..b2173c3f837 100644 --- a/src/main/java/edu/harvard/iq/dataverse/api/Datasets.java +++ b/src/main/java/edu/harvard/iq/dataverse/api/Datasets.java @@ -265,7 +265,7 @@ public Response getDataset(@Context ContainerRequestContext crc, @PathParam("id" MakeDataCountLoggingServiceBean.MakeDataCountEntry entry = new MakeDataCountEntry(uriInfo, headers, dvRequestService, retrieved); mdcLogService.logEntry(entry); } - return ok(jsonbuilder.add("latestVersion", (latest != null) ? json(latest) : null)); + return ok(jsonbuilder.add("latestVersion", (latest != null) ? json(latest, true) : null)); }, getRequestUser(crc)); } @@ -471,22 +471,33 @@ public Response useDefaultCitationDate(@Context ContainerRequestContext crc, @Pa @GET @AuthRequired @Path("{id}/versions") - public Response listVersions(@Context ContainerRequestContext crc, @PathParam("id") String id ) { - return response( req -> - ok( execCommand( new ListVersionsCommand(req, findDatasetOrDie(id)) ) + public Response listVersions(@Context ContainerRequestContext crc, @PathParam("id") String id, @QueryParam("includeFiles") Boolean includeFiles, @QueryParam("limit") Integer limit, @QueryParam("offset") Integer offset) { + + return response( req -> { + Dataset dataset = findDatasetOrDie(id); + + return ok( execCommand( new ListVersionsCommand(req, dataset, offset, limit, (includeFiles == null ? true : includeFiles)) ) .stream() - .map( d -> json(d) ) - .collect(toJsonArray())), getRequestUser(crc)); + .map( d -> json(d, includeFiles == null ? true : includeFiles) ) + .collect(toJsonArray())); + }, getRequestUser(crc)); } @GET @AuthRequired @Path("{id}/versions/{versionId}") - public Response getVersion(@Context ContainerRequestContext crc, @PathParam("id") String datasetId, @PathParam("versionId") String versionId, @Context UriInfo uriInfo, @Context HttpHeaders headers) { + public Response getVersion(@Context ContainerRequestContext crc, @PathParam("id") String datasetId, @PathParam("versionId") String versionId, @QueryParam("includeFiles") Boolean includeFiles, @Context UriInfo uriInfo, @Context HttpHeaders headers) { return response( req -> { DatasetVersion dsv = getDatasetVersionOrDie(req, versionId, findDatasetOrDie(datasetId), uriInfo, headers); - return (dsv == null || dsv.getId() == null) ? notFound("Dataset version not found") - : ok(json(dsv)); + + if (dsv == null || dsv.getId() == null) { + return notFound("Dataset version not found"); + } + + if (includeFiles == null ? true : includeFiles) { + dsv = datasetversionService.findDeep(dsv.getId()); + } + return ok(json(dsv, includeFiles == null ? true : includeFiles)); }, getRequestUser(crc)); } @@ -783,7 +794,7 @@ public Response updateDraftVersion(@Context ContainerRequestContext crc, String } managedVersion = execCommand(new CreateDatasetVersionCommand(req, ds, incomingVersion)); } - return ok( json(managedVersion) ); + return ok( json(managedVersion, true) ); } catch (JsonParseException ex) { logger.log(Level.SEVERE, "Semantic error parsing dataset version Json: " + ex.getMessage(), ex); @@ -1018,7 +1029,7 @@ private Response processDatasetFieldDataDelete(String jsonBody, String id, Datav DatasetVersion managedVersion = execCommand(new UpdateDatasetVersionCommand(ds, req)).getLatestVersion(); - return ok(json(managedVersion)); + return ok(json(managedVersion, true)); } catch (JsonParseException ex) { logger.log(Level.SEVERE, "Semantic error parsing dataset update Json: " + ex.getMessage(), ex); @@ -1167,7 +1178,7 @@ private Response processDatasetUpdate(String jsonBody, String id, DataverseReque } DatasetVersion managedVersion = execCommand(new UpdateDatasetVersionCommand(ds, req)).getLatestVersion(); - return ok(json(managedVersion)); + return ok(json(managedVersion, true)); } catch (JsonParseException ex) { logger.log(Level.SEVERE, "Semantic error parsing dataset update Json: " + ex.getMessage(), ex); @@ -3942,9 +3953,9 @@ public Response getPrivateUrlDatasetVersion(@PathParam("privateUrlToken") String JsonObjectBuilder responseJson; if (isAnonymizedAccess) { List anonymizedFieldTypeNamesList = new ArrayList<>(Arrays.asList(anonymizedFieldTypeNames.split(",\\s"))); - responseJson = json(dsv, anonymizedFieldTypeNamesList); + responseJson = json(dsv, anonymizedFieldTypeNamesList, true); } else { - responseJson = json(dsv); + responseJson = json(dsv, true); } return ok(responseJson); } diff --git a/src/main/java/edu/harvard/iq/dataverse/dataset/DatasetUtil.java b/src/main/java/edu/harvard/iq/dataverse/dataset/DatasetUtil.java index ac1567b24e5..096f1f87acc 100644 --- a/src/main/java/edu/harvard/iq/dataverse/dataset/DatasetUtil.java +++ b/src/main/java/edu/harvard/iq/dataverse/dataset/DatasetUtil.java @@ -523,7 +523,7 @@ public static boolean validateDatasetMetadataExternally(Dataset ds, String execu // for the filter to whitelist by these attributes. try { - jsonMetadata = json(ds).add("datasetVersion", json(ds.getLatestVersion())) + jsonMetadata = json(ds).add("datasetVersion", json(ds.getLatestVersion(), true)) .add("sourceAddress", sourceAddressLabel) .add("userIdentifier", userIdentifier) .add("parentAlias", ds.getOwner().getAlias()) diff --git a/src/main/java/edu/harvard/iq/dataverse/engine/command/impl/FinalizeDatasetPublicationCommand.java b/src/main/java/edu/harvard/iq/dataverse/engine/command/impl/FinalizeDatasetPublicationCommand.java index f5e70209744..3da087addd9 100644 --- a/src/main/java/edu/harvard/iq/dataverse/engine/command/impl/FinalizeDatasetPublicationCommand.java +++ b/src/main/java/edu/harvard/iq/dataverse/engine/command/impl/FinalizeDatasetPublicationCommand.java @@ -10,6 +10,7 @@ import edu.harvard.iq.dataverse.DatasetVersionUser; import edu.harvard.iq.dataverse.Dataverse; import edu.harvard.iq.dataverse.DvObject; +import edu.harvard.iq.dataverse.Embargo; import edu.harvard.iq.dataverse.UserNotification; import edu.harvard.iq.dataverse.authorization.Permission; import edu.harvard.iq.dataverse.authorization.users.AuthenticatedUser; @@ -117,9 +118,37 @@ public Dataset execute(CommandContext ctxt) throws CommandException { // is this the first publication of the dataset? if (theDataset.getPublicationDate() == null) { theDataset.setReleaseUser((AuthenticatedUser) getUser()); - } - if ( theDataset.getPublicationDate() == null ) { + theDataset.setPublicationDate(new Timestamp(new Date().getTime())); + + // if there are any embargoed files in this version, we will save + // the latest availability date as the "embargoCitationDate" for future + // reference (if the files are not available yet, as of publishing of + // the dataset, this date will be used as the "ciatation date" of the dataset, + // instead of the publicatonDate, in compliance with the DataCite + // best practices). + // the code below replicates the logic that used to be in the method + // Dataset.getCitationDate() that calculated this adjusted date in real time. + + Timestamp latestEmbargoDate = null; + for (DataFile dataFile : theDataset.getFiles()) { + // this is the first version of the dataset that is being published. + // therefore we can iterate through .getFiles() instead of obtaining + // the DataFiles by going through the FileMetadatas in the current version. + Embargo embargo = dataFile.getEmbargo(); + if (embargo != null) { + // "dataAvailable" is not nullable in the Embargo class, no need for a null check + Timestamp embargoDate = Timestamp.valueOf(embargo.getDateAvailable().atStartOfDay()); + if (latestEmbargoDate == null || latestEmbargoDate.compareTo(embargoDate) < 0) { + latestEmbargoDate = embargoDate; + } + } + } + // the above loop could be easily replaced with a database query; + // but we iterate through .getFiles() elsewhere in the command, when + // updating and/or registering the files, so it should not result in + // an extra performance hit. + theDataset.setEmbargoCitationDate(latestEmbargoDate); } //Clear any external status diff --git a/src/main/java/edu/harvard/iq/dataverse/engine/command/impl/ListVersionsCommand.java b/src/main/java/edu/harvard/iq/dataverse/engine/command/impl/ListVersionsCommand.java index 51283f29156..b93833ffdf9 100644 --- a/src/main/java/edu/harvard/iq/dataverse/engine/command/impl/ListVersionsCommand.java +++ b/src/main/java/edu/harvard/iq/dataverse/engine/command/impl/ListVersionsCommand.java @@ -14,6 +14,7 @@ import edu.harvard.iq.dataverse.engine.command.DataverseRequest; import edu.harvard.iq.dataverse.engine.command.RequiredPermissions; import edu.harvard.iq.dataverse.engine.command.exception.CommandException; +import edu.harvard.iq.dataverse.engine.command.exception.CommandExecutionException; import java.util.LinkedList; import java.util.List; @@ -23,23 +24,57 @@ */ // No permission needed to view published dvObjects @RequiredPermissions({}) -public class ListVersionsCommand extends AbstractCommand>{ - +public class ListVersionsCommand extends AbstractCommand> { + private final Dataset ds; + private final Integer limit; + private final Integer offset; + private final Boolean deepLookup; - public ListVersionsCommand(DataverseRequest aRequest, Dataset aDataset) { - super(aRequest, aDataset); - ds = aDataset; - } + public ListVersionsCommand(DataverseRequest aRequest, Dataset aDataset) { + this(aRequest, aDataset, null, null); + } + + public ListVersionsCommand(DataverseRequest aRequest, Dataset aDataset, Integer offset, Integer limit) { + this(aRequest, aDataset, null, null, false); + } + + public ListVersionsCommand(DataverseRequest aRequest, Dataset aDataset, Integer offset, Integer limit, boolean deepLookup) { + super(aRequest, aDataset); + ds = aDataset; + this.offset = offset; + this.limit = limit; + this.deepLookup = deepLookup; + } - @Override - public List execute(CommandContext ctxt) throws CommandException { - List outputList = new LinkedList<>(); - for ( DatasetVersion dsv : ds.getVersions() ) { - if (dsv.isReleased() || ctxt.permissions().request( getRequest() ).on(ds).has(Permission.EditDataset)) { - outputList.add(dsv); + @Override + public List execute(CommandContext ctxt) throws CommandException { + + boolean includeUnpublished = ctxt.permissions().request(getRequest()).on(ds).has(Permission.EditDataset); + + if (offset == null && limit == null) { + + List outputList = new LinkedList<>(); + for (DatasetVersion dsv : ds.getVersions()) { + if (dsv.isReleased() || includeUnpublished) { + if (deepLookup) { + // @todo: when "deep"/extended lookup is requested, and + // we call .findDeep() to look up each version again, + // there is probably a more economical way to obtain the + // numeric ids of the versions, by a direct single query, + // rather than go through ds.getVersions() like we are now. + dsv = ctxt.datasetVersion().findDeep(dsv.getId()); + if (dsv == null) { + throw new CommandExecutionException("Failed to look up full list of dataset versions", this); + } + } + outputList.add(dsv); + } } - } - return outputList; - } + return outputList; + } else { + // Only a partial list (one "page"-worth) of versions is being requested + return ctxt.datasetVersion().findVersions(ds.getId(), offset, limit, includeUnpublished); + } + } } diff --git a/src/main/java/edu/harvard/iq/dataverse/util/json/JsonPrinter.java b/src/main/java/edu/harvard/iq/dataverse/util/json/JsonPrinter.java index 6f34fe4bdb1..65fe749e554 100644 --- a/src/main/java/edu/harvard/iq/dataverse/util/json/JsonPrinter.java +++ b/src/main/java/edu/harvard/iq/dataverse/util/json/JsonPrinter.java @@ -351,11 +351,14 @@ public static JsonObjectBuilder json(FileDetailsHolder ds) { .add("mime",ds.getMime())); } - public static JsonObjectBuilder json(DatasetVersion dsv) { - return json(dsv, null); + public static JsonObjectBuilder json(DatasetVersion dsv, boolean includeFiles) { + return json(dsv, null, includeFiles); } - public static JsonObjectBuilder json(DatasetVersion dsv, List anonymizedFieldTypeNamesList) { + public static JsonObjectBuilder json(DatasetVersion dsv, List anonymizedFieldTypeNamesList, boolean includeFiles) { + /* return json(dsv, null, includeFiles, null); + } + public static JsonObjectBuilder json(DatasetVersion dsv, List anonymizedFieldTypeNamesList, boolean includeFiles, Long numberOfFiles) {*/ Dataset dataset = dsv.getDataset(); JsonObjectBuilder bld = jsonObjectBuilder() .add("id", dsv.getId()).add("datasetId", dataset.getId()) @@ -371,6 +374,8 @@ public static JsonObjectBuilder json(DatasetVersion dsv, List anonymized .add("alternativePersistentId", dataset.getAlternativePersistentIdentifier()) .add("publicationDate", dataset.getPublicationDateFormattedYYYYMMDD()) .add("citationDate", dataset.getCitationDateFormattedYYYYMMDD()); + //.add("numberOfFiles", numberOfFiles); + License license = DatasetUtil.getLicense(dsv); if (license != null) { bld.add("license", jsonLicense(dsv)); @@ -398,7 +403,9 @@ public static JsonObjectBuilder json(DatasetVersion dsv, List anonymized jsonByBlocks(dsv.getDatasetFields(), anonymizedFieldTypeNamesList) : jsonByBlocks(dsv.getDatasetFields()) ); - bld.add("files", jsonFileMetadatas(dsv.getFileMetadatas())); + if (includeFiles) { + bld.add("files", jsonFileMetadatas(dsv.getFileMetadatas())); + } return bld; } @@ -430,8 +437,8 @@ public static JsonObjectBuilder jsonDataFileList(List dataFiles){ * to the regular `json` method for DatasetVersion? Will anything break? * Unit tests for that method could not be found. */ - public static JsonObjectBuilder jsonWithCitation(DatasetVersion dsv) { - JsonObjectBuilder dsvWithCitation = JsonPrinter.json(dsv); + public static JsonObjectBuilder jsonWithCitation(DatasetVersion dsv, boolean includeFiles) { + JsonObjectBuilder dsvWithCitation = JsonPrinter.json(dsv, includeFiles); dsvWithCitation.add("citation", dsv.getCitation()); return dsvWithCitation; } @@ -450,7 +457,7 @@ public static JsonObjectBuilder jsonWithCitation(DatasetVersion dsv) { */ public static JsonObjectBuilder jsonAsDatasetDto(DatasetVersion dsv) { JsonObjectBuilder datasetDtoAsJson = JsonPrinter.json(dsv.getDataset()); - datasetDtoAsJson.add("datasetVersion", jsonWithCitation(dsv)); + datasetDtoAsJson.add("datasetVersion", jsonWithCitation(dsv, true)); return datasetDtoAsJson; } diff --git a/src/main/resources/db/migration/V6.0.0.2__9763-embargocitationdate.sql b/src/main/resources/db/migration/V6.0.0.2__9763-embargocitationdate.sql new file mode 100644 index 00000000000..536798015ba --- /dev/null +++ b/src/main/resources/db/migration/V6.0.0.2__9763-embargocitationdate.sql @@ -0,0 +1,14 @@ +-- An aggregated timestamp which is the latest of the availability dates of any embargoed files in the first published version, if present +ALTER TABLE dataset ADD COLUMN IF NOT EXISTS embargoCitationDate timestamp without time zone; +-- ... and an update query that will populate this column for all the published datasets with embargoed files in the first released version: +UPDATE dataset SET embargocitationdate=o.embargocitationdate +FROM (SELECT d.id, MAX(e.dateavailable) AS embargocitationdate +FROM embargo e, dataset d, datafile f, datasetversion v, filemetadata m +WHERE v.dataset_id = d.id +AND v.versionstate = 'RELEASED' +AND v.versionnumber = 1 +AND v.minorversionnumber = 0 +AND f.embargo_id = e.id +AND m.datasetversion_id = v.id +AND m.datafile_id = f.id GROUP BY d.id) o WHERE o.id = dataset.id; +-- (the query follows the logic that used to be in the method Dataset.getCitationDate() that calculated this adjusted date in real time). diff --git a/src/test/java/edu/harvard/iq/dataverse/api/DatasetsIT.java b/src/test/java/edu/harvard/iq/dataverse/api/DatasetsIT.java index 34eccd3172a..d246fc3e62a 100644 --- a/src/test/java/edu/harvard/iq/dataverse/api/DatasetsIT.java +++ b/src/test/java/edu/harvard/iq/dataverse/api/DatasetsIT.java @@ -563,6 +563,103 @@ public void testCreatePublishDestroyDataset() { } + /** + * The apis (/api/datasets/{id}/versions and /api/datasets/{id}/versions/{vid} + * are already called from other RestAssured tests, in this class and also in FilesIT. + * But this test is dedicated to this api specifically, and focuses on the + * functionality added to it in 6.1. + */ + @Test + public void testDatasetVersionsAPI() { + // Create user + String apiToken = UtilIT.createRandomUserGetToken(); + + // Create user with no permission + String apiTokenNoPerms = UtilIT.createRandomUserGetToken(); + + // Create Collection + String collectionAlias = UtilIT.createRandomCollectionGetAlias(apiToken); + + // Create Dataset + Response createDataset = UtilIT.createRandomDatasetViaNativeApi(collectionAlias, apiToken); + createDataset.then().assertThat() + .statusCode(CREATED.getStatusCode()); + + Integer datasetId = UtilIT.getDatasetIdFromResponse(createDataset); + String datasetPid = JsonPath.from(createDataset.asString()).getString("data.persistentId"); + + // Upload file + String pathToFile = "src/main/webapp/resources/images/dataverseproject.png"; + Response uploadResponse = UtilIT.uploadFileViaNative(datasetId.toString(), pathToFile, apiToken); + uploadResponse.then().assertThat().statusCode(OK.getStatusCode()); + + // Check that the file we just uploaded is shown by the versions api: + Response unpublishedDraft = UtilIT.getDatasetVersion(datasetPid, ":draft", apiToken); + unpublishedDraft.prettyPrint(); + unpublishedDraft.then().assertThat() + .body("data.files.size()", equalTo(1)) + .statusCode(OK.getStatusCode()); + + // Now check that the file is NOT shown, when we ask the versions api to + // skip files: + boolean skipFiles = true; + unpublishedDraft = UtilIT.getDatasetVersion(datasetPid, ":draft", apiToken, skipFiles); + unpublishedDraft.prettyPrint(); + unpublishedDraft.then().assertThat() + .body("data.files", equalTo(null)) + .statusCode(OK.getStatusCode()); + + // Publish collection and dataset + UtilIT.publishDataverseViaNativeApi(collectionAlias, apiToken).then().assertThat().statusCode(OK.getStatusCode()); + UtilIT.publishDatasetViaNativeApi(datasetId, "major", apiToken).then().assertThat().statusCode(OK.getStatusCode()); + + // Upload another file: + String pathToFile2 = "src/main/webapp/resources/images/cc0.png"; + Response uploadResponse2 = UtilIT.uploadFileViaNative(datasetId.toString(), pathToFile2, apiToken); + uploadResponse2.then().assertThat().statusCode(OK.getStatusCode()); + + // We should now have a published version, and a draft. + + // Call /versions api, *with the owner api token*, make sure both + // versions are listed; also check that the correct numbers of files + // are shown in each version (2 in the draft, 1 in the published). + Response versionsResponse = UtilIT.getDatasetVersions(datasetPid, apiToken); + versionsResponse.prettyPrint(); + versionsResponse.then().assertThat() + .statusCode(OK.getStatusCode()) + .body("data.size()", equalTo(2)) + .body("data[0].files.size()", equalTo(2)) + .body("data[1].files.size()", equalTo(1)); + + // Now call this api with the new (as of 6.1) pagination parameters + Integer offset = 0; + Integer howmany = 1; + versionsResponse = UtilIT.getDatasetVersions(datasetPid, apiToken, offset, howmany); + // (the above should return only one version, the draft) + versionsResponse.prettyPrint(); + versionsResponse.then().assertThat() + .statusCode(OK.getStatusCode()) + .body("data.size()", equalTo(1)) + .body("data[0].files.size()", equalTo(2)); + + // And now call it with an un-privileged token, to make sure only one + // (the published) version is shown: + + versionsResponse = UtilIT.getDatasetVersions(datasetPid, apiTokenNoPerms); + versionsResponse.prettyPrint(); + versionsResponse.then().assertThat() + .statusCode(OK.getStatusCode()) + .body("data.size()", equalTo(1)); + + // And now call the "short", no-files version of the same api + versionsResponse = UtilIT.getDatasetVersions(datasetPid, apiTokenNoPerms, skipFiles); + versionsResponse.prettyPrint(); + versionsResponse.then().assertThat() + .statusCode(OK.getStatusCode()) + .body("data[0].files", equalTo(null)); + } + + /** * This test requires the root dataverse to be published to pass. */ diff --git a/src/test/java/edu/harvard/iq/dataverse/api/UtilIT.java b/src/test/java/edu/harvard/iq/dataverse/api/UtilIT.java index 4421e9280b3..fa4e9861ff6 100644 --- a/src/test/java/edu/harvard/iq/dataverse/api/UtilIT.java +++ b/src/test/java/edu/harvard/iq/dataverse/api/UtilIT.java @@ -11,6 +11,7 @@ import jakarta.json.JsonObjectBuilder; import jakarta.json.JsonArrayBuilder; import jakarta.json.JsonObject; +import static jakarta.ws.rs.core.Response.Status.CREATED; import java.nio.charset.StandardCharsets; import java.nio.file.Files; @@ -116,6 +117,16 @@ public static Response createRandomUser() { return createRandomUser("user"); } + + /** + * A convenience method for creating a random test user, when all you need + * is the api token. + * @return apiToken + */ + public static String createRandomUserGetToken(){ + Response createUser = createRandomUser(); + return getApiTokenFromResponse(createUser); + } public static Response createUser(String username, String email) { logger.info("Creating user " + username); @@ -366,6 +377,20 @@ static Response createRandomDataverse(String apiToken) { String category = null; return createDataverse(alias, category, apiToken); } + + /** + * A convenience method for creating a random collection and getting its + * alias in one step. + * @param apiToken + * @return alias + */ + static String createRandomCollectionGetAlias(String apiToken){ + + Response createCollectionResponse = createRandomDataverse(apiToken); + //createDataverseResponse.prettyPrint(); + createCollectionResponse.then().assertThat().statusCode(CREATED.getStatusCode()); + return UtilIT.getAliasFromResponse(createCollectionResponse); + } static Response showDataverseContents(String alias, String apiToken) { return given() @@ -1400,9 +1425,17 @@ static Response nativeGetUsingPersistentId(String persistentId, String apiToken) } static Response getDatasetVersion(String persistentId, String versionNumber, String apiToken) { + return getDatasetVersion(persistentId, versionNumber, apiToken, false); + } + + static Response getDatasetVersion(String persistentId, String versionNumber, String apiToken, boolean skipFiles) { return given() .header(API_TOKEN_HTTP_HEADER, apiToken) - .get("/api/datasets/:persistentId/versions/" + versionNumber + "?persistentId=" + persistentId); + .get("/api/datasets/:persistentId/versions/" + + versionNumber + + "?persistentId=" + + persistentId + + (skipFiles ? "&includeFiles=false" : "")); } static Response getMetadataBlockFromDatasetVersion(String persistentId, String versionNumber, String metadataBlock, String apiToken) { @@ -1764,6 +1797,18 @@ static Response removeDatasetThumbnail(String datasetPersistentId, String apiTok } static Response getDatasetVersions(String idOrPersistentId, String apiToken) { + return getDatasetVersions(idOrPersistentId, apiToken, false); + } + + static Response getDatasetVersions(String idOrPersistentId, String apiToken, boolean skipFiles) { + return getDatasetVersions(idOrPersistentId, apiToken, null, null, skipFiles); + } + + static Response getDatasetVersions(String idOrPersistentId, String apiToken, Integer offset, Integer limit) { + return getDatasetVersions(idOrPersistentId, apiToken, offset, limit, false); + } + + static Response getDatasetVersions(String idOrPersistentId, String apiToken, Integer offset, Integer limit, boolean skipFiles) { logger.info("Getting Dataset Versions"); String idInPath = idOrPersistentId; // Assume it's a number. String optionalQueryParam = ""; // If idOrPersistentId is a number we'll just put it in the path. @@ -1771,6 +1816,27 @@ static Response getDatasetVersions(String idOrPersistentId, String apiToken) { idInPath = ":persistentId"; optionalQueryParam = "?persistentId=" + idOrPersistentId; } + if (skipFiles) { + if ("".equals(optionalQueryParam)) { + optionalQueryParam = "?includeFiles=false"; + } else { + optionalQueryParam = optionalQueryParam.concat("&includeFiles=false"); + } + } + if (offset != null) { + if ("".equals(optionalQueryParam)) { + optionalQueryParam = "?offset="+offset; + } else { + optionalQueryParam = optionalQueryParam.concat("&offset="+offset); + } + } + if (limit != null) { + if ("".equals(optionalQueryParam)) { + optionalQueryParam = "?limit="+limit; + } else { + optionalQueryParam = optionalQueryParam.concat("&limit="+limit); + } + } RequestSpecification requestSpecification = given(); if (apiToken != null) { requestSpecification = given()