Intelligent Paperwork sorter
Hi All
Do you receive and store a lot of paperwork everyday in your mailbox ? on slack ? on whatsapp ?
Because I do, and for every paper I receive, I rename it and sort it to be able to find it back later. But Depending on the day I sort them, I classify them in different folders and end up with 3 different folders on my computer containing what should be in only one. And this is very boring
I made a CLI tool that will sort files that I put in my TOSORT folder and I feel this could be a tool that could help a lot of people.
This tool
- scans a directory
- opens pdf file and based on their content will classify them by:
-> date
-> entity(Company A/Company B/Rented Apartment/ House)
-> supplier (brand name)
-> title (invoice / contract / ...)
-> custom data
-> price - From this data, it will rename them and let you choose a folder in which put it (with a suggestion of previous folders used for this type of paper)
Would you be interested in that ?
Do you even classify your files or don't you care? And if yes, how do you do?
Thanks
This is a very nice idea. Thank you for sharing it with us! I am a little old fashion person, so for me is hard to enter this digital world. I work only with real papers and stamps. I'm thinking right now what will be with documents in the future? I'm very fond of my work, I have a lot of stamps, and I save them because I want to make a big collection. Recently, I discovered a constructor that allows you to design and create a custom stamp https://mystampready.com/en/constructor/. I think it's awesome! I ordered a stamp from them. The materials they used are very qualitative.
Ok well
I know have a first version of this tool
I plan to call it
What do you think?
I also had in mind
or do you have a more clever idea?
Thank you
I think this has some merit.
A couple years ago I decided to go completely paperless at home. I scanned everything to a
scannerfolder on my NAS (Network Attached Storage) as PDFs, and shredded the originals (obviously there are a few things I kept, like car titles etc). I named each file with the year, month, day, and keywords (e.g.20200131_us_bank_home_mortgage_statement.pdf) and manually moved it to its final folder on my NAS.I'm a Linux guy, so I set up a cron job to use
mlocate'supdatedbcommand to re-build an index of all the files on my NAS every 15 minutes. Then I wrote a command line tool callednaslocatethat usesmlocate'slocatecommand to find files on my NAS.Here's an example of using the command to find all my mortgage statements from 2018:
The whole process of setting it up was extremely time consuming (especially since the sheet feeder on my scanner was broken!). I took all the papers out of all the boxes and dumped them in a giant pile in the middle of my office (it came up to my waist and took up most of the floor space!), and spent a half an hour or more a day working on it. It took months to get through it! Anything that would have sped up the process would have been very welcome.
Now I have a sheet-fed scanner on my desk next to a shredder, so anything that comes in immediately gets scanned, re-named, and shredded.
Hi @CodeOfTheProgrammer and thanks for your reply
Does your scanner do OCR ? do you analyze the text in it to sort your files or do you just look at the file and tell your program what it is?
Thanks a lot, I really feel it's a pain for a lot of people
My scanner can do OCR, but I don't use that feature. I just name the file with the keywords I know I'll search on later. The important part is that it doesn't actually matter where I put my files. My program creates an index of all the files on my NAS every 15 minutes, so it can find anything anywhere. It has worked pretty well for me (I've always been able to find what I'm looking for).
It would be nice to have something a little more automated so I wouldn't have to name my files after I scan them, and that could let me find based on categories, keywords, date ranges, etc.
Agreed and that's what I'm aiming at!
But I still think a lot of people are very attached to where their documents are stored as they very likely already have their own way of sorting things (But maybe I'm wrong)
The tool I have is more like a virtual assistant that help you store your files where YOU want. Making it agnostic of your own classification system.
Not like itunes :-)
This way, if at some point you don't want to use my software anymore, you just bail out and you're fine
But I agree with you if you have a good locator service, this approach is not that useful (except for the bail out part)
Thanks !
I wonder if you could do some kind of machine learning classification on documents to tag them (e.g. this is a
finance,bank,mortgagedocument), and the user could have rules that define where a document should be moved to based on the tags assigned to it? I'm talking out my ass here because I don't know anything about ML. ;)