/todo/steal.todo
Unknown | 71 lines | 71 code | 0 blank | 0 comment | 0 complexity | 14a20f40a79006f08236fbe635a77098 MD5 | raw file
Possible License(s): BSD-3-Clause, LGPL-2.1
- - gui
- - plugins to handle making links
- - link
- - just a "page" paramager links to another page in the zone (you can tack on params too)
- - add in a "zone" paramater switches zones
- - baseLink
- - starts from the base
- - requires a zone and/or paramater
- - auto include templates
- - layouts
- - maybe have a way to insert multiple files, if they exist into multiple parts of the template
- - zone
- - routing???
- - auto include zone files???
- - auto generate files for people
- - zones (also have a zone inludes file that automatically gets an include tacked onto the end)
- + create a new application
- - migrations
- - templates
- - DbObject sublcasses
- - db
- - migrations
- - make it heirarchical so that you can insert migrations into old versions and still have them get applied to both dbs
- - use a decimal system type versioning (1, 1.1, 1.1.1, 1.1.2 etc)
- - always scan the entire list of migrations and figure out any that haven't been applied
- - allow applying or removing specific migrations
- - allow you to force the application of certain migrations even if they have already been applied??? (maybe it's best to always just remove the specific migration and then reapply it)
- - allow you to tell the db that a specific migration hasn't really been applied or needs to be reapplied???
- - have a way to separate: ddl, dml, or text (schema, data, text)
- - have a way to consolidate migrations???
- - indicate which data migrations are part of a specific install (project) and which are key to the app functionging (product)
- - this should alow you to create all of the data necessary for a test install without developers ever having production data
- - this should also facilitate setting up test cases
- - allow you to take an existing database and reverse engineer the schema and put it into migration syntax
- - this will allow us to port existing schemas to other databases
- - we will still need to extract product data however if we are to get it really working
- - list which migrations haven't been applied
- - dbobject
- - find
- - bundle a web server to get it started quickly
- - have a download for windows with the browser/server combo
- - have an interface for managing projects and updates to zoop
- - yaml based config files???
- - figure out a way to be able to move stuff around without breaking all of our links and redirects but without using routes
- - the first thing you need to do hear is figure out some legit use cases to test against
- - implement the routes examples in the book
- - implement the store example in the book
- - how to handle nested zones
- - rails seems to be built around a singular linear project with one live install, thus dev, test, and live are all it can handle well
- - it doesn't seem to be geared around a single product with lots of slightly different projects or installs, and it isn't designed to handle it well
- - scaffolding
- - standard list, edit, view, delete
- - login
- - convention over configuration
- - every coc features can be put into one of two categories
- - scaffolding
- - conform your domain to a restricted abstrated pattern in order to get started. later we remove the scaffolding and fill in the real building elements later
- - we force to domain into our pattern so that we can "get something working"
- - genunine pattern matching
- - the domain actually matches the pattern and so generalized code can be used
- - maybe there is a flow moving from extrememly generalized to extremely customized, on demand, agile
- - validation
- - put validation info into the model
- - how then do we get it into the form???
- - testing
- - unit testing?
- - regression testing?
- - design by contract, preconditions, postconditions
- - all of this should be validated against the existing bug database to see what is worth the effort
- - the flash
- - not sure we want this but we should probably think it through