đ FairScan > Syncthing > Paperlees-ngx
from Teppichbrand@feddit.org to selfhosted@lemmy.world on 06 Feb 06:53
https://feddit.org/post/25409295
from Teppichbrand@feddit.org to selfhosted@lemmy.world on 06 Feb 06:53
https://feddit.org/post/25409295
Itâs perfect! Do you guys already do this?
I open a letter, I take itâs picture with FairScan. The FairScan-folder on my android device gets syncthinged to the ingest folder for Paperless-ngx on my âserverâ. Paperless imports it, deletes the file and sets the new documents tag to inbox. I decide if the document goes to the binder for important stuff, or if I just toss it in a binder with all the paper I most likely will never touch again. Next time I look at Paperless, I edit all documents with the inbox tag and remove the tag.
threaded - newest
I have configured an email address in paperless_ngx and send everything there. The appendix is then tagged as unsorted.
This is the transhumanism I like.
Jokes aside, how did you deal with your papers from the past? I do this with all new incoming mail, but I canât for the life of me find the time to scan the pile of taxes, bills, medical reports, etc. that has accumulated over 3 thousand years of family life.
You could buy an automatic scanner that takes a stack of docs and dumps the files to a network share.
Do you have any particular model in mind? What do you mean by âautomatic scannerâ? Any scanner I know needs quite a lot of manual preparation to scan a batch of documents.
Epson WorkForce DSâ730N
put 100 sheets on the tray, it scans them all and either puts them all into a single pdf or multiple pdfs. Then you split / merge them in software.
Not very automatic, is it? There are definitely better ways to scan high volumes of documents.
please elaborate
You can use e.g. barcodes, patch codes or separator sheets (which usually carry the patch code). Sometimes you can also separate documents by recognising some feature on the first page, e.g. a logo or a barcode thatâs already there. And of course itâs a good idea to put single page documents in a separate batch so you just separate them by page count. This of course also works if all documents are two or three pages long.
afaik you just listed features that the printer I mentioned (or if I am wrong, other similar printers) supports
itâs my bad for not mentioning all possible workflows, I was just a bit lazy and thinking of my personal documents only, which do not work well with further smart automation, because my batches are highly irregular. So the more manual approach is the best for me currently. Maybe possible with some future AI integration.
Most of those features are implemented in the scan software on the PC, not on the scanner itself. Although there is a tendency to integrate more and more features in the firmware, which is not always a good idea. Also, if youâre scanning low volumes, Iâd say doing the separation before the scan is generally more efficient. At least thatâs how I do it. But thatâs just me, of course. I wasnât in any way trying to criticize your approach. If it works for you, itâs great.
I took my Very Important Documents!!-folder to a neighbour with a decent scanner. Itâs not that much, we scanned for an hour or two. Older, less important stuff stays in binders I most likely will never touch. If I do have to look at something from there, I snap a quick FairScan before I put it back. So itâs not about perfection, I just try to make my live easier from now on. :)
Wait, this is genius. When paperless has processed it and moves the original, itâs also removed from your phone, right?
Yes, paperless deletes it from the import folder which is synced with syncthing, so on android the folder is also empty
Welp I guess thereâs another piece of software I have to setup now⌠đ
Iâm using CamScanner, but I have to share every doc to paperless. Should have a look at an automatic ingest as wellâŚ
This is awesome! Thank you for sharing this flow! đ
I use an hardware scanner which then puts the pdf via ftp on the inbox folder of paperless. So i can also all old stuff
Why donât you use the mobile app? Mine works just fine
At least I think it does. Maybe all the docs that I upload are lost. Who knows đ
Haha, thanks. I didnât know this existed! I already use Syncthing, so thatâs no problem. And I like that FairScan rotates my picture and adjusts itâs angle. If the Paperless Uploader-app does this as well, I might use it. Taking a look right now!
I use f-droid.org/âŚ/de.astubenbord.paperless_mobile
Does it detect the four corners and adjusts the angle to make it flat 2D?
It does, sometimes⌠but it is not a huge PITA.
I need to see if FairScan would do a better job.
Last updated 2024/01/26⌠HmmmâŚ
edit: checked the wiki, here are the apps for Android
Exactly what I came here to say, paperless app already have a scanner no need to fiddle with file sync and stuff đ
I do something similar but with the ADF scanner on my Brother printer. The scans automatically go to my server which processes them: deskew, combine with previous scan if it has the same amount of pages (because it might be me scanning the front pages first then the back pages), compress. After that, itâs put on my NAS. I left the step of importing it to paperless manual, in that I have to copy or move the file from the Scan directory on my NAS to the Paperless ingest directory. This is so that I can first check if the scan came out okay.
Similar approach here:
It doesnât take more than 10 seconds to scan a doc this way.
I have yet to use a consumer ADF scanner on a printer that didnât feed the paper at an angle until they are crushed and folded, doesnât matter if the guides are perfectly set for A4 either. It has never worked for me.
No, I use a printer/scanner combo for scanning my files. I trigger a shell script via HomeAssistant which starts the scanner, cleans and fixes the order of the pages for two sided documents, and then puts them into the paperless folder.
But your idea is great!
would you mind to elaborate more on the script? Also, what output format do you use from the scanner?
Its a hacked together shell script. I wanted to learn shell scripting, but I shouldâve probably programmed it in python.
But it basically does this:
I wonât share it because itâs really ugly and some of the functionality can be done with paperless native, like combining the pages of two scans. Which wouldâve been easier.
Thanks! And thatâs ok, Iâm more interested in the steps than the code itself. Do you apply any optimizations with image magik in step 2?
Honestly, phone scanning is a total trap for anything you actually care about. I tried this exact setup for months and the OCR quality was absolute trash because of bad lighting or a slight tilt. If a document is important enough to keep, it is worth getting a dedicated scanner with an auto-feeder. Taking photos of letters feels like a massive chore that you will eventually just stop doing after the novelty wears off.
And letâs be real about that inbox tag. That is just a digital pile of shame. I have hundreds of documents in my own Paperless instance tagged for review that I havent touched in two years. You think you are staying organized, but you are really just moving the clutter from your desk to a database. Syncthing is the only part of this workflow that I actually trust to work every single time without failing.
Isnât syncthing for mobile discontinued?
This is the currently maintained android version: github.com/researchxxl/syncthing-android
wasnât there some recent controversy over the change of ownership for the repo?
That was all resolved.
I held off switching until very recently, but reading the githubs, etc. itâs all settled down now.
From memory it was just a bit of a nieve handover (ie âhiddenâ in the background, not out in public)
Thatâs a very cool idea, seems great for receipts and quick stuff.
There are companion apps for paperless (like Paperparrot for iOS) that simplify that process even more.
No need for syncthing.
Or âSwift Paperlessâ on iOS too, at least thatâs what I use
Paperparrot is flawless. The dev is also very responsive and helpful.
They also offer a similarly flawless app called Plappa as frontend for your audiobookserver.
Running both, super happy.
The best thing about this is that it enables you to stop organizing all this paper. I have a little box next to my scanner and everything that is scanned is just getting put on top of the pile. In most cases you never need the original document anyways. But if you need it, you can check Paperless when it was scanned and you will find that document from Feb 2023 in a few seconds.
Heads up if your Paperless is installed via Docker: be VERY CAREFUL about your database version and do an export often!
Mine has been down for a while because I did a pull and it doesnât support my version of Postgresql anymore. So itâs kinda a huge mess trying to figure out how to safely migrate it in the container.
I havenât been able to fix it yet so Iâve just left it disabled and gone without for a while. Itâs not fun.
Allegedly if you export from within Paperless, you can just start with a fresh updated database container and import when this happens. Oof.
your version of postgres meaning that you use a db external to the docker container?
No sadly the upgrade path is a bit ducked. Their compose, their versions, still not working without intervention.
Well, crap
Thank you, yes, exactly what I meant. Itâs a bit baffling such a mature project doesnât have some kind of migration script. :(
I donât think Postgres shouldnât be just left as
:latestor anything. At least the way I handle it, DB upgrades require manual intervention.I just got burned by accidental latest tag on a pg container for Nextcloud. They moved some paths internally and it could no longer find the db.
This is a reminder for self-hosters to put their apps (and their data) on snapshotting filesystems with automatic, regular snapahots turned on; and fix the app versions to at least the major version, across all containers. This should bring similar disruption to bare minimum and makes recovery always possible, without relying on specific app backup features.
Thank you for introducing me to FairScan! Great app. I have a scanner but being able to snap stuff on the go is so much quicker.
Purely awesome app if you scan a straight paper that are not cut in any way. I miss the ability to manually crop the images. Will still use the app, though.
Thanks for recommending it đ Have always used a flatbed scanner for my papers (receipts, bills, etc.), but now I can scan on the go đ
I havenât used FairScan (yet!), how does it compare to OSS Document Scanner? Other than having a much much better name, obviously.
I tried OSS Document Scanner a year ago but never really used it. It has way more features but I didnât need any of them. With FairScan, you just take pictures, give them a name and save them either as a single PDF or a couple PNGs. Repeat if necessary.
Oh man I just set this up. I just changed to a high deductible health plan with an HSA and need to save all of my medical receipts. I already had paperless set up but i was dreading setting up some automated way of getting files over to paperless. Thank you so much!
I keep having trouble getting paperless to actually injest any scans from the injest folder.
Sounds like a great workflow!
Unfortunately, I just canât get syncthing to run in the background of my phone without chewing up the battery.
You can get it to run at time intervals. E.g. once an hour for 5 minutes. Thatâs not bad on battery for me. I actually have mine once every 24 hours for 30 minutes so it can successfully transfer a few gigs of Signal backups.
I will look into this, thank you!
I luckily have a professional document scanner for this, which has wifi and is able to store the PDF on a cloud drive. Started using ASN numbers recently, so I put a sticker on the letter before scanning. The letter gets archived and is found quickly if needed because of the ASN.
Yesterday I wanted to introduce syncthing as my main backup solution. However, I am struggling with setting up the discovery server and want to fix that before installing it on all my devices.
I use a different app (though Iâll be checking out FairScan), but thatâs pretty much exactly my workflow for things that arenât convenient to do basically the same thing on my desktop PC & scanner.