> > I think you're right that there is no way of doing that from the web > UI at least. > You probably could do it with a camtool search query though. >
Just had a quick look: https://github.com/camlistore/camlistore/blob/26d161db21e70fd3752a475a204c3e9ec1f4f356/pkg/search/query.go#L596 The `Attr` string field and the subsequent fields for the corresponding value of that attribute all appear to be value-related, as opposed to testing the presence of attribute (e.g. `tag`) in the first place. > The use case I have in mind here is creating a workflow whereby photos, > > scans (from a Fujitsu ScanSnap) etc are uploaded, and remain > "unclassified" > > until they have been tagged at least once. > > This reminds me of the user interface of Brad's scanning cabinet > (https://github.com/bradfitz/scanningcabinet). So maybe it would fit > your needs once it lands for Camlistore? > (https://camlistore-review.googlesource.com/5416) > Will certainly take a look - thanks for the pointer. > > Perhaps someone has a suggestion on a better workflow or tooling that > would > > achieve the same goal, else would a contribution be welcomed that added > > something along the lines of: > > > > hasattr:tag > > > > to the search lexicon? > > Maybe? :-) > You should probably start with opening an issue about it though please. > I can work around this for now by using camput with a tag of "unprocessed" (or similar), a tag that will then be removed as part of the later workflow. (Happy to still raise an issue if you think this is of broader interest?) My plan was then, subject to what scanningcabinet has to offer, to do something along the following lines: - documents will be scanned with my Fujitsu ScanSnap; this doesn't have huge capacity so will effectively be one document (which could be multi-page) at a time - text recognition is done by the software itself (scanned as PDFs) - files stored to local disk by the software - some watcher 'process' will `camput` the time-named scanned files with an "unprocessed" tag, then delete the local file post successful camput - create some sort of GopherJS-based frontend, using some React bindings that I'm working on, to make for a good UI/UX in the processing phase. Would also look to suggest a list of tags based on some correlation between the scanned document's recognised text and existing tags Would certainly be interested in hearing from anyone who has done something similar before, because if, like scanningcabinet, there's a working solution out there I'm certainly happy to save myself some time on the initial development and contribute! Thanks -- You received this message because you are subscribed to the Google Groups "Camlistore" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
