I guess that adding a folder import feature would invoke a lot of the same code paths, but might benefit from something like a progress bar, and might need some additional unit/functional tests. That would benefit in detecting any unreliability in the mutli-file case as well, where I've noticed some intermittent problems in the past.
Where should those tests live? Is test/tests/importExportTest.js a good place for it to go, or should it happen in a more integrated context?
No, this would be an importFromFolder() in attachments.js, calling importFromFile(), as well as Zotero.CollectionTreeView.prototype.drop() in collectionTreeView.js (which also calls importFromFile() currently). (But the latter file is about to be completely replaced, so you shouldn't touch that for the moment.)
If you want to look into implementing this, you can post further technical questions to zotero-dev.
Where should those tests live? Is test/tests/importExportTest.js a good place for it to go, or should it happen in a more integrated context?
importFromFolder()
in attachments.js, callingimportFromFile()
, as well asZotero.CollectionTreeView.prototype.drop()
in collectionTreeView.js (which also callsimportFromFile()
currently). (But the latter file is about to be completely replaced, so you shouldn't touch that for the moment.)If you want to look into implementing this, you can post further technical questions to zotero-dev.