🐛 dinglehopper: Fix reading OCR-D workspace files when only URLs are provided

4 years ago · f699697eb3
parent ea1cc32b91
commit f699697eb3
1 changed files with 2 additions and 0 deletions
--- a/qurator/dinglehopper/ocrd_cli.py
+++ b/qurator/dinglehopper/ocrd_cli.py
@ -38,6 +38,8 @@ class OcrdDinglehopperEvaluate(Processor):
        for n, page_id in enumerate(self.workspace.mets.physical_pages):
            gt_file = self.workspace.mets.find_files(fileGrp=gt_grp, pageId=page_id)[0]
            ocr_file = self.workspace.mets.find_files(fileGrp=ocr_grp, pageId=page_id)[0]
+            gt_file = self.workspace.download_file(gt_file)
+            ocr_file = self.workspace.download_file(ocr_file)
            log.info("INPUT FILES %i / %s↔ %s", n, gt_file, ocr_file)

            file_id = self._make_file_id(ocr_file, ocr_grp, n)