Restoring Scrivener files

Dear All,

I have used Scrivener for years for all my projects and business to run a biotechnology company. At times I’ve been careless about backing up and using the non-Dropbox cloud options. It caught up with me this week, all information in my two Scriveners is unreachable.

Can you help?

The size of the files is correct (all rtf’s in Snapshots) but all of the folders and files in the binder are empty. Is there any solution for this? I have solved this problem before but no joy this time, even bringing in >1000 entries and having to rename them.

I hope someone can help.

Desparate in London :slight_smile:

Sounds like the sync software is only downloading the index of your documents, and not the files that contain the text.

Before you open/close your projects any more, please find the backups folder (Scrivener->Preferences->Backup). Open that up and copy everything there to a safe place outside of that folder and outside of any folders that are being synced to the cloud. If you’re using iCloud Documents & Desktop, you’re going to want to put your projects in a new folder off of your /Users/[your login id here]/ at the same level with Music and other folders.

Try unzipping the most recent of each project, and opening it. If it’s not in the right state, just close it and delete it, then move to the second most-recent one.

Good luck.

Thanks very much for your quick help.

The Mac Scrivener files were downloaded as filename.scriv and are the expected size (27.5MB, 119Mb). When I look at package contents the files are in “Screenshots” (see below).

Is it possible to re-integrate these screenshots into the binder or do I start over (wiser, hopefully)?

Jeff

Look in the Files/Docs subfolder. That’s where your actual project contents should be. Snapshots are older versions from whenever a specific snapshot was taken.

You can import any .RTF file into any project, so one way to recover would be to create a new blank project and drag whatever files you find into it. Before going that route, though, I’d try creating a new blank project and using the File -> Import -> Scrivener Project command. Except in extreme circumstances, that’s usually as effective as any manual recovery you might try.

Katherine

The folder that has all of the information found in the binder is in Files/Docs, not Snapshots. Is that one empty? Snapshots is where anything you’ve created a snapshot for will be found. That may be up to date, but in most cases it won’t be, and will have a lot of clutter in older versions of each file. If all you have is the contents of the Snapshots folder, I would try other backups before using that.

Given the name of this project, there may be other copies? It looks like it is a copy of a backup already (but maybe that’s something that has happened in the process of trying to restore).

Did you have any luck digging up backup copies out of the automatic backup folder, as rdale suggested? That’s the #1 place to go if a sync goes wrong.

Thanks very much for your help, Katherine and Amber. On it this morning with renewed vigor! :smiley:

With your advice to look in data folders and for zip files, over 6 hours I went through every scriv file. Fortunately I found some data that I can rescue for early years. The problem came when transferring data from Sharefile to Sharepoint. I learned the hard way that drop and drag doesn’t take all the files.

I would be grateful for any final thoughts:

For “BUSINESS SCRIVENER” the file is the right size and I saw right amount of data in the data file but it’s all in the “content.rtf”, which only has 2 lines of information (not shown).

For “ADMINISTRATION SCRIVENER” I see no data but rtfs in Screenshots which is why I mentioned it earlier (see image).

I’m not familiar with Sharefile or Sharepoint, so I can’t help you out much there—but does that imply there is somewhere that has a good copy with their service, it’s just not copying correctly? Surely they have some method of restoring a folder entirely, rather than just downloading bits and pieces in an arbitrary fashion? Maybe getting on the phone with their technical support and seeing if they can recover your data from their end, or advise you on a better technique for transfer, would be another parallel course of investigation to take.

Sorry to hear that your automatic backups were out of date. It would certainly be a good thing to add reviewing those settings to your checklist, once things are cleared up. Could be the default settings are not conducive to how you work (they aren’t good for people who never close the software, for example, since they wait for you to close it before backing up).

Well, if Snapshots are all you have, then you’ll have to make do with that. You could probably even work from this incomplete copy as your starting point. There wouldn’t be too much of a difference between doing that and importing the project into a new blank one. Reason being: Scrivener’s a pretty distributed format—snapshots for example only exist because the file is there for them. If you delete the snapshots folder, it is as if the project never had snapshots and it won’t “hurt” anything to do that. Likewise if a cloud service wipes out (or never copies) the contents of the entire Files/Docs folder—well, from Scrivener’s point of view you’ve just done an awful lot of outlining and no writing. It doesn’t hurt it, or care that there is or isn’t content. There is no functional difference between a project you haven’t started writing in yet, and one that had all your writing nuked.

So with that in mind, here’s what I would do:

  1. First, working from the original downloaded .scriv folder, one that has never been opened if possible, duplicate it and rename the copy to “rebuild” suffix or whatever.
  2. In the new copy, go into the Files/ subfolder and look for a “search.indexes” file. This should be several megabytes of text. Drag that out to your desktop or somewhere convenient. This file may, depending on how things went down, have a lot of your most up to date data. As a search index, Scrivener does its best to keep that up to date as things change. The format itself is a fairly legible XML format. You shouldn’t have any problem navigating it by item title (you can open this in TextEdit if you don’t have an XML/coding editor). Just search for the binder name, and the text, synopsis, and notes should all be there. But if you don’t have a “never opened” copy it may already be gone—Scrivener will rebuild that file if it looks like the project has changed since the last time it opened it.
  3. When you first open the project without the index file, it will force a rebuild. You do want to do that, because that is the one place where the project can get out of sync with itself, if parts of it vanish “overnight”.
  4. At that point it’s a matter of going down the list of binder items with the Snapshots pane open in the inspector. First look for the item in the search index, and copy and paste from there if it is newer than the last snapshot—if not, restore from that snapshot.

For that, consider you can copy and paste from the search index into the editor, then use the Compare button in the inspector to run a quick word check. If they are both the same, the Snapshot may be better to roll back from, as it will also save formatting instead of pure text.

  1. And again, somewhere in here, go through the Backup preference pane! I like to have it keep 25 copies around, with date stamps (if you backup to cloud, that can be advantageous as each backup has a unique name which means you can even go back further than 25 if they track deleted files for a time). I also like the setting that backs up when pressing ⌘S. I can then do that a few times a day on a busy project and know I’ll only lose a few hours worst case.

Thanks, just seeing your post. I’ll work through your suggestions. Meanwhile, I’ve set up Scrivener properly!

Jeff