Bringing things in · Chapter 6

Moving from Paperless-ngx

If you already use Paperless-ngx, you do not have to start from scratch with nexpaper. nexpaper fetches the documents through the interface of Paperless-ngx and takes the recognised text along, so nothing is read again. Paperless-ngx stays untouched: nexpaper only reads there. You can run both side by side until you have made up your mind. nexpaper is a project of its own and is not part of Paperless-ngx.

Before you start

• Paperless-ngx 1.14 or newer. nexpaper turns older versions away.
• Create the accounts in nexpaper. Whoever owns documents in Paperless needs an account in nexpaper, so that the documents can go into their vault. How to invite is in the chapter Inviting. It is best to create shared vaults beforehand too.
• An API token from Paperless. In Paperless you find it under your profile as the API token. A token sees in Paperless only what its account may see there. If you want to take everything over, use the token of an account that sees everything.
• An address that can be reached. The server running nexpaper has to be able to reach Paperless. That is the server itself, not your browser. nexpaper does not trust a self-signed certificate. In your own network use http instead.

Taking it over

Settings, Import

1. Connect and look

   The import is the operator's job. Open “Settings”, tab “Import”, card “Take over from Paperless”. Enter the address under “Address of Paperless” and the token under “API token”. If Paperless runs in your home network or on the same machine, tick “Paperless runs in the own network”, otherwise nexpaper cannot reach it. Then tap “Connect and look”.

2. Look at the result

   nexpaper shows the version of Paperless-ngx and counts how many documents, tags, correspondents and document types it has. Under “How it is translated” four lines say what turns into what (see below).

3. Assign the owners to people

   Under “Owner in Paperless → person here” you choose a person in nexpaper for every owner in Paperless. The person with the same name is suggested, otherwise the operator (“no account of that name, goes to …”). If the token may not list the accounts of Paperless, the owners show only as numbers.

4. Where the documents go

   Under “Where to” there are two choices. “Each owner into their vault” files every document in the personal vault of its owner. “Everything into …” sends everything into one vault of your choice, “Home”, for example. Everything can still be moved afterwards. What was shared with others in Paperless is shared with the same people in nexpaper.

5. Sorting

   Under “Sorting” you choose “As in Paperless” or “Everything into the inbox”. With “As in Paperless”, whatever sat in Paperless's inbox stays in the inbox here too, and the rest is filed right away. With “Everything into the inbox” everything waits in the inbox and you file it yourself. That pays off for a small collection you want to look through anyway.

6. Go

   The button carries the number: “Take over 1,250 documents”. After that the import runs in the background. You can close the page and it goes on, even across a restart of the server. “Stop” halts after the current document, and “Continue” picks up from there.

What comes along and what does not

nexpaper takes over the original, Paperless's searchable version, the recognised text, the title and the date. The correspondent becomes the sender and the document type the kind, with unknown types turning into new kinds. Tags stay tags and notes stay notes. Custom fields and the ASN end up as one note with all the values on the document.

Storage paths and workflows do not exist in nexpaper and are not taken over. nexpaper does not take Word, Excel or other office files. If Paperless made a PDF of one, that PDF comes as the document and the history says so. Without a PDF the file is skipped. If the original is missing in Paperless, the searchable version comes. After the run the card has a list “Skipped documents”, each with its Paperless number and the reason, so that you know what to look up in Paperless.

Afterwards

• Check. Documents taken over and documents skipped together should add up to the number nexpaper named when it looked. Open a few documents and compare them with Paperless.
• Run it again. “Run again” fetches only what has come into Paperless since the last run. What is already there does not come twice.
• Clean up. If you do not want to keep Paperless connected, “Remove connection” deletes the stored address and token.
• Who is told. Whoever received documents gets a notification when the import is through, if they have notifications switched on.