lsmyrnaios
/
UrlsWorker
1
0
Fork 1
Code Issues Pull Requests Releases 12 Wiki Activity

- Handle the case, where, from a group of related records, the initial record which led to a publication-url, failed to have its full-text downloaded. Now we make sure the file-related data for all those related records is kept "null" and a special error is written. 

- Code optimization.
This commit is contained in:
Lampros Smyrnaios 2022-04-05 17:51:45 +03:00
parent d682298850
commit a1f750a0aa
1 changed files with 30 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

48
src/main/java/eu/openaire/urls_worker/plugins/PublicationsRetrieverPlugin.java
View File

@ -156,6 +156,9 @@ public class PublicationsRetrieverPlugin {
}
private static final int lengthOfAlreadyDownloadedFromSourceUrlContinuedMessage = ConnSupportUtils.alreadyDownloadedFromSourceUrlContinuedMessage.length();
private static final int lengthOfAlreadyDownloadedFromIDMessage = ConnSupportUtils.alreadyDownloadedFromIDMessage.length();
public static void addUrlReportsToWorkerReport(Collection<Assignment> assignments)
{
// Index the UrlIds with the DatasourceIds for quick-search later.
@ -176,31 +179,40 @@ public class PublicationsRetrieverPlugin {
if ( "true".equals(data.getWasDocumentOrDatasetAccessible()) ) // The reversed order defends against a potential NPE.
{
status = UrlReport.StatusType.accessible;
if ( comment.startsWith(UrlUtils.alreadyDownloadedFromIDMessage, 0) ) {
if ( comment.startsWith(ConnSupportUtils.alreadyDownloadedFromIDMessage, 0) ) {
// The file of this docUrl was already downloaded by another docUrl.
int indexOfAlreadyDownloadedFromSourceUrlMessage = comment.indexOf(UrlUtils.alreadyDownloadedFromSourceUrlContinuedMessage);
int indexOfAlreadyDownloadedFromSourceUrl = indexOfAlreadyDownloadedFromSourceUrlMessage + UrlUtils.alreadyDownloadedFromSourceUrlContinuedMessage.length();
String initialId = comment.substring(UrlUtils.alreadyDownloadedFromIDMessage.length(), indexOfAlreadyDownloadedFromSourceUrlMessage); // The fileName starts right after the "message".
int indexOfAlreadyDownloadedFromSourceUrlMessage = comment.indexOf(ConnSupportUtils.alreadyDownloadedFromSourceUrlContinuedMessage);
int indexOfAlreadyDownloadedFromSourceUrl = indexOfAlreadyDownloadedFromSourceUrlMessage + lengthOfAlreadyDownloadedFromSourceUrlContinuedMessage;
String initialId = comment.substring(lengthOfAlreadyDownloadedFromIDMessage, indexOfAlreadyDownloadedFromSourceUrlMessage); // The fileName starts right after the "message".
String initialSourceUrl = comment.substring(indexOfAlreadyDownloadedFromSourceUrl);
//logger.debug("initialId: " + initialId + " | sourceUrl: " + initialSourceUrl); // DEBUG!
// Search that ID and sourceUrl inside the list, if that instance is the first-found one, then get the file-data (there might be duplicate ID-sourceUrl instances, but only one of them has the file-data).
boolean foundAlreadyDownloadedFullText = false;
for ( DataToBeLogged data_2 : FileUtils.dataToBeLoggedList ) {
if ( data_2.getUrlId().equals(initialId) && (data_2.getSourceUrl().equals(initialSourceUrl))
&& ! data_2.getComment().startsWith(UrlUtils.alreadyDownloadedFromIDMessage) ) {
fileLocation = data_2.getComment();
size = data_2.getSize();
hash = data_2.getHash();
mimeType = "application/pdf"; // TODO - If support is added for other doc-formats other than "pdf", then make sure the "mime_type" is assigned correctly.
foundAlreadyDownloadedFullText = true;
break;
}
boolean foundIDUrlInWorkerReport = false;
for ( DataToBeLogged data_2 : FileUtils.dataToBeLoggedList )
{
if ( ! (data_2.getUrlId().equals(initialId) && (data_2.getSourceUrl().equals(initialSourceUrl))) )
continue;
foundIDUrlInWorkerReport = true;
String tempFileLocation = data_2.getComment();
if ( tempFileLocation.startsWith(ConnSupportUtils.alreadyDownloadedFromIDMessage, 0) || tempFileLocation.startsWith(HttpConnUtils.docFileNotRetrievedMessage, 0) )
continue;
fileLocation = tempFileLocation;
size = data_2.getSize();
hash = data_2.getHash();
mimeType = "application/pdf"; // TODO - If support is added for other doc-formats other than "pdf", then make sure the "mime_type" is assigned correctly.
foundAlreadyDownloadedFullText = true;
break;
}
// In case the "alreadyDownloaded" full-text is not found, we have an error. All file-related data is "null".
if ( !foundAlreadyDownloadedFullText ) {
String addErrorMessage = ((!foundIDUrlInWorkerReport) ? " | That ID-sourceUrl was not found inside the WorkerReport!" : " | The file was not downloaded!");
error = new Error(Error.ErrorType.couldRetry, comment + addErrorMessage); // We can still try to download it from the found docUrl, in the future.
}
// In case the "alreadyDownloaded" full-text is not found, we have an error.
if ( !foundAlreadyDownloadedFullText )
error = new Error(Error.ErrorType.couldRetry, comment + " | That ID-sourceUrl was not found inside the WorkerReport!"); // We can still try to download it from the found docUrl, in the future.
}
else if ( ! comment.contains(HttpConnUtils.docFileNotRetrievedMessage) ) { // If it was downloaded without an error.
else if ( ! comment.startsWith(HttpConnUtils.docFileNotRetrievedMessage, 0) ) { // If it was downloaded without an error.
fileLocation = comment; // This is the full-file-path.
mimeType = "application/pdf";
} else // Else the file was not retrieved, so all file-related data are kept "null".