Standardised bulk uploads are a good way of integrating substantial amounts of data into the repository. Irrespective of the procedure employed, the successful import into SSOAR requires both the metadata and the full texts, or the persistent identifiers (PIDs) – for example, URNs or DOIs – that refer to them. The metadata must be recorded in a structured and uniform way and unequivocally assigned to the respective full texts/PIDs.
Numerous services for the publication, management, and making available of literature on the Internet offer web interfaces for the exchange of data. To collect the data via an Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH) interface, either a so-called set must be created or it must be possible to identify the data sets of relevance to SSOAR on the basis of a metadata field and to filter and select them.
For an automatic download, each data set must contain an unequivocal link to the full text or a PID. It is sufficient if there is a possibility of automatically generating this link, for example from an ID.
.xml, .csv, .ods or .xls(x) Files
If the data records of a table (Excel, Libre/Open Office), a CSV file, or an XML file are available, they can be delivered to SSOAR in this format. Depending on the delivery variant, it is necessary to ensure that a full text can be unequivocally assigned to each metadata set. This can be ensured by means of a PID in the metadata or via the exact file name of the separately delivered PDF.
We offer our suppliers a corresponding template (78 kB) (including an explanatory legend (71 kB)) for the import of CSV formats. In this way, we ensure that the required data are imported into SSOAR in full and in the correct format.
Direct Delivery of the Full Texts
If the full texts are delivered directly to SSOAR (by DVD, CD-ROM, USB stick, or file upload), the metadata must have a field containing the relative file path with the file name of the PDF (e.g., …\Supplier\Documents\[PossibleSupplement]\document.pdf). In the case of a direct delivery, it is advisable to place the metadata and the documents in separate folders.
If none of the above-mentioned options apply to your system, other file formats could be used. In this case, you should determine what export possibilities exist in your system and contact us again, enclosing an example.