Something that will selectively inventory a file system and build clickable web page or some othe UI of interesting folders. 'Interesting' means there's a certain file in it with some metadata, but it's going to be intermingled with a whole bunch of the directories we don't care about.
Some\where\deep\really\deep\SuperCol\.project.meta:
Project: Super Collider
Owner: Heracles
Date Initiated: 0500-12-02
etc...
Possible index output (with links):
Project: Super Collider, <Heracles>, Started: 0500-12-02
Project: Super Zapper, <Zeus>, Started: 0502-10-02
Project: Dumb Collider, <Heracles>, Started: 0503-01-05
Asking for help elsewhere:
- https://softwarerecs.stackexchange.com/questions/47412/
- https://www.reddit.com/r/SomebodyMakeThis/comments/7hs1vy/
envapplist.py - by Jon Hourd, made this quick and dirty script to compile this list to save the effort of constantly browsing the directory structure there. Baby steps towards a better future!
2017-12-22
Minimum info:
Title/Name
Summary/tag line/purpose
Authors
Intitiation date
Latest milestone/release/published date
Clients
Nice to have:
Thumbnail
Extended description
(Status?)
Related projects
Tags
Speed tiles/dial (grid or list) Sort & filter by any attribute Find-as-you-type:
- for all fields, or
- just that field if in/above/below column Clickable - opens the folder on disk
Marker file in the interesting folder, plain text of at least the minimum elements. Format? Markdown/YAML/INI/...
A scheduled crawler locates the marker files and adds them to the index
- is the index a static page, like index.html or an app? (perhaps like Fossil SCM)
- is marker a single file? or folder? or a packaged file?
zinfo.inizinfo/*.*Project.zinfo
Only some folders of the thousands are Interesting or Projects. The can be located anywhere. We often don't know ahead of time if a thing will be interesting. They start small.
ENV.499 folders and the Map-Index.xls are too big to understand now. The system is very useful and has saved our bacon countless times, but it's collapsing under it's own weight.
Amazingly the name zinfo is not widely used. I found
- a polish website, http://zinfo.pl,
- a Google Apps series by Zinfo Enterprises in India, which seems to be very lightly used. (It's a location based search and meta-info app for local info. Cool idea.)
- Fashion watches for men
- 11 repositories on Github, only 1 of which has any significant activity or code and it's completely unrelated (https://github.com/ZRiemann/ZInfoTech)
Z - to sort to bottom of file explorer window in most instances. I'd rather not rely on hidden files because updating meta-data outside a management tool needs to be quick and easy, and approachable for less technical people.
2018-01-11
Trying .ini format and ConfigParser to start with.
Excel fields to migrate (* important)
Job Name*
Old Map ID
New Map ID
Series ID
Link*
Description*
Client*
Resource
SDR
Application
NOTES
New fields:
MapID*
Title*
Subtitle
Description*
Client*
Contacts*
Authors*
Derived fields, but INI overrides derived value:
Link
...
We want updating to work in both directions. Editing .ini file should update the master table, and editing table should update .ini. ...or is that just a bad idea? Why do I feel like it should go both ways?
Better: table is only read only, but has [edit this project] action button.