zam-haus / inventory_management Goto Github PK

Unique Items would be one-of-a-kind things, usually no meant to be given away, but need tracking. For example, a Lasercutter, a Laptop and so forth.

An interface to fill out all the fields, that weren't filled out during the inventory was taken, has to be established.

first ideas about the layout and functionality:

https://erlangen.ccc.de/markdown/1zO_x7CfRJe9FJ8aezzaYg

Drop down without search is not a solution

Provide a dedicated data entry interface to fill item info from photos.

@danieloeh some thing you would like to have look at?
Also to present the ocr results in the front end.

We need a interface to register changes in the inventory when someone takes some parts out of the storage.

first ideas about forms and workflows:

https://erlangen.ccc.de/markdown/Uoix1jP4QaSaa4Z2dPvy0A

1. Search for something with lots of search results: https://inv.zam.haus/item/?q=Rolle
2. click on "next" page link
3. Search text has disappeared, now all items are shown: https://inv.zam.haus/item/?page=2

Expected result: "next" link should point to https://inv.zam.haus/item/?q=Rolle&page=2

i cannot get the dropdown list when changing the description for an image to work.

im using firefox mobile 97.2.0

how to reproduce:
point the cursor in an empty image-description input-field and try to bring the suggestions up. what does not work:

• start typing, e.g. "pack" should yield "packaged" and "Single item (unpacked)"
• tap onto the input field multiple times when it is empty

what should happen:
drop-down suggestions should show up

i tried the lineageOs included browser "org.lineageos.jelly" v11. The dropdown-list works but the getUserMedia does not work. when using the input type=file button, a file-selection screen appears instead of the os-camera-app. so this browser is useless, too.

Currently ~75 MB of image data are transferred on a single list views of items, e.g., https://inv.zam.haus/item/?page=50

To improve this, thumbs should be generated and used.

Goal: make items easier findable, without having to manually describe them in detail.

Idea:

1. Run OCR on all photos and extract text
2. Store extracts with photo
3. Extend item search to include photo OCR-extracts

Things to consider:

• Allow this process to be done image-by-image, so that it can be improved in future
• Primary goal would be to run this process in batch, possibly outside of django, but if the implementation is capable of doing the same live on newly uploaded photos this would be a nice feature
• Test cases to determine quality of extracts would be good to have, e.g. compare automatic extract to manual extractions
• Consider comparing multiple OCR systems
• Also extract and store EANs or other barcodes present in photos

A tracked item (vs. a stock item) is an item that is individually tracked, for example a computer or an extension cord, which may have serial number or a regular test protocol and log. These items may be rented, have a responsible person, a location (also outside ZAM) and other meta data.

Tracked items (specific item) may be derived from a specific instance of a stock item (model description).

Either run at/right after upload or alternatively allow the manage.py ocr to only run on new images.
@danieloeh :)

• seperate edit from view (maybe responsive?)
• show ocr_text with image to easily grep for it

Run on new photos during upload and on old photos using management command