import error: insure that file is valid

ftrxx
Mixing up a Zot on FF and a new ZSA installation (because Zot on FF gets slow with 5500 entries) I exported the bib entries from Zot on FF with a RDF files incl. files and notes to sync the two.

Having imported the rdf file into ZSA after a long import process I get a javascript application error: insure that file is valid.
I always thought that a file exported from Zotero is valid when re-imported in Zotero !

Minutes later ZSA is still turning empty (Prog doesn't answer).
Weird.
  • ftrxx
    Now, after having killed the process (ZSA took some 700 KB) it seems that the file was in fact imported - but how !
    The intention was: just export the latest additions from the FF zot file, some 20 entries with notes and files. I filtered them on the base of the field Date last modified.
    What I got were some 1300 additional entries.
    I don't understand such a mishap.

    Sorry to say that but this does not instill trust in the prog. Besides my statistics and the text treatment programs this is my major research tool.
  • adamsmith
    I'm not entirely sure what you did, but it sounds like you exported your entire library and then tried to import that.

    If you tell us how you exported, we can likely tell you what happened, but there's no way Zotero will "accidentally" export thousands instead of 20 items.
    The right way to do this would be to select the items in questions, then right-click --> Export selected items.

    But also, I'd recommend against moving items between different Zotero installations using export/import. You're very likely to make a mess of things. Why not just run Zotero Standalone on a shared data directory and rely on that exclusively (using ZFF on a shared data directory, so that it's only in connector mode.).
  • ftrxx
    What I did was to select the bib entries of the last two days from Zotero on FF and exported it with the notes and files as a RDF files as this is the only one to allow exporting attached fles & notes. So, I expected some 20 items.
    This was for me the only imaginable way to sync the FF and the standalone versions.

    After the importation I got two things: a new collection called My Library with some 20 entries but not those of the last days. And the injection of the aforementioned some 1,300 items.

    Now, to get rid of the superfluous entries I understand the best is to delete say by bunch of a page (i.e. 30 entries) ? Right ? Or delete lists of 100s ?
    Should I check the database integrity of the file before ?

    Your recommandation interests my very much. Could be please be more explicit. What do you mean with a shared data directory ? for instance, on an online drive ? And using ZFF on a shared data directory, so that it's only in connector mode : I don't understand.
  • adamsmith
    Deleting sets of a couple of 100s at a time are fine. No need for integrity checks.

    But run us through your steps for exporting -- as I said, Zotero doesn't just randomly export 1000s of items.

    Sharing data directory: https://www.zotero.org/support/kb/sharing_data_directory
  • ftrxx
    Is it possible to re-use a backup copy of the sqlite file ? Say, delete the most recent file, and rename a backup file? But what happens to folders that have been created in for the more recent sqlite versions ? They just become orphaned ?
  • adamsmith
    edited September 17, 2016
    Say, delete the most recent file, and rename a backup file? But what happens to folders that have been created in for the more recent sqlite versions ? They just become orphaned ?
    yes, that'd work (though I'd not delete the most recent file, just move it somewhere else to be safe; and you should do all of this with Zotero/Firefox not running).
    https://www.zotero.org/support/zotero_data#restoring_your_zotero_library_from_a_backup
    New files in storage would indeed orphan that way though, yes. What's wrong with just deleting?
  • ftrxx
    It seems I got it. Thanks for your help !

    I backuped the t-1 file, deleted the corrupted current file, renamed the t-1 into t file, and oriented Zot for FF as well as ZSA to read the same folder.

    The only thing that differs between using ZSA and Zot/FF is that the latter shows unfiled items whereas ZSA shows none.

    I looked whether I can easily identify the now superfluous storage folders that were created when the import went wrong but I don't know how.
Sign In or Register to comment.