Is that possible to use WebDAV or symbolic link toghther with ZotFile?
Dear everyone,
I am a freshman to Zotero. I want to get some advice on the choice of syncing at the beginning, because I think with the number of the items increasing in my library, it will be much harder to change from one method to another.
I have read the documentation and I'd prefer to use WebDAV or symbolic linked 'storage' folder in cloud-sync to sync my attached files at this moment.
However, the plugin ZotFile has some useful fucntions including rename the attached file and extract the annotations from PDFs. So, I want to use this plugin together with WebDAV or symbolic link. ZotFile can help me to rename the file name based on a customized rules and extract the annotations.
I am sorry I did not find much informations about how to use ZotFile with WebDAV or symbolic links toghther. Are there any possible conficts between them? Could you please give me some advices?
Any help would be much appreciated!
I am a freshman to Zotero. I want to get some advice on the choice of syncing at the beginning, because I think with the number of the items increasing in my library, it will be much harder to change from one method to another.
I have read the documentation and I'd prefer to use WebDAV or symbolic linked 'storage' folder in cloud-sync to sync my attached files at this moment.
However, the plugin ZotFile has some useful fucntions including rename the attached file and extract the annotations from PDFs. So, I want to use this plugin together with WebDAV or symbolic link. ZotFile can help me to rename the file name based on a customized rules and extract the annotations.
I am sorry I did not find much informations about how to use ZotFile with WebDAV or symbolic links toghther. Are there any possible conficts between them? Could you please give me some advices?
Any help would be much appreciated!
Thanks very much for your answer!
If I choose "Attach stored copy of file(s)", then ZotFile will just rename the files without moving them, is that right? So the files are still located in a subfolder named after 8 random characters?
And I am a little confused about the difference between WebDAV and symbolic link. I thought both of creating a subfolder in the cloud sync folder and using the cloud sync to sync them. Could you please tell me more about why WebDAV would generate less error?
WebDAV is natively supported by Zotero. You set it up in the Sync pane of Zotero preferences by choose “Sync attachment files in My Library using [WebDAV]” and entering your WebDAV service information. Then, Zotero syncing works as normal, but syncs your files using the WebDAV server instead of the Zotero server.
Setting up a symbolic link is a more complex and technical process. It involves setting a symbolic link between Zotero’s storage folder and another folder on your computer that you have syncing with a cloud provider. In Zotero, you would disable file syncing. This is more prone to error because it involves more steps to set up (which can be difficult if you’re not familiar with symbolic links in general) and because syncing is handled externally rather than within Zotero. If there is ever a sync error for your files, Zotero won’t be able to alert you to the issue.
Both setups are only about syncing the files as a backend for Zotero. In both, they will be organized in the eight character folders by Zotero, so you would only want to ever view them through Zotero.
Thanks very much! I followed your advice and have tried to sync with WebDAV. It seems that the files stored in the cloud-sync folder are some zipped files which cannot be read directly, and the original attached files are still stored in the storage file of zotero on my local computer as your said. Right now, syncing with WebDAV works fine with ZotFile.
It seems that it is not complicated to implement if I want to change to sync with symbolic link from syncing with WebDAV. In that case, I need to uncheck 'Sync attachments in My Library using WebDAV' and generate a symbolic link folder which linked to my current storatge folder of zotero, in my cloud-sync folder. By the way, I am using Ubuntu, so set a symbolic link seems simple, something like 'ln -s $cloud-sync_folder/storage ~/Zotero/storage'. Could please tell is that the right method to do that?
And no matter which method of syncing is used, it does not affect the file structures and the other things at local computer, so they should all works fine with ZotFile. Is that right?
Instead, just keep doing what you are already doing. Use Zotfile to move files and organize them in your cloud folder on your computer.
If you always want to access the files through Zotero, then ignore the zip files in WebDAV and let Zotero manage them.
I want human readable file hierarchies for my PDFs. (Incidentally, that Papers3 did away with this is why I never made that leap from Papers2.) From what I understand, human readable file hierarchies is where ZotFile comes in. Specifically, I will want to specify a directory for the field "linked attachment base directory" in Zotero and the same directory for the custom location of files field in ZotFile. After that, I make whatever mods I want to the naming conventions. That all worked fine. For PDF syncing, I would like to use WebDAV with pCloud. It is not obvious to me that this is compatible with my ZotFile scenario.
My library is about 8.5K items. I did an initial import from an RIS file today with synching turned off. That went smoothly enough, although I believe I encountered a memory leak--memory usage ballooned to over 10 GB before I killed the import and began importing in small batches, with memory usage expanding predictably for each batch until I restarted the app between batches. After the import was complete, I turned syncing on and made my initial commit to the Zotero server. The data went up fine. The client then seemed to check itself against the server. I believe it was not until this was done that a WebDAV upload began. Once the WebDAV sync completed, I then fired up a new Zotero client on a different machine. The database sync went fine. The PDF sync on the other hand seems not to have worked. The WebDAV transfer pulled down the gemisch of alphanumeric directories under Zotero/storage (originally on the source machine, now in the cloud) but nothing, seemingly, was synced that had been placed into the (ZotFile) human readable directory.
For reference, pCloud reports 160 MB of data used by the sync. The source machine has a "storage" directory with 529 MB of content. On the target machine this directory is 500 MB. The discrepancy between pCloud and the local machines is presumably compression. The ZotFile directory is 8.9 GB on the source and zero on the destination. On the target machine, Zotero shows open circles for the attachment marker for most items, which I take to mean 'there should be a PDF but it is not found'.
Is it possible to sync the human readable "ZotFile" directory via WebDAV (driven by Zotero) or am I required to use an external cloud sync service for that (i.e., still 3rd party storage provider but now their app is driving the file transfers)?