dbagg3 TODO
===========

Overall
-------

  * Possible name: feedReactor

  * Profiles
    * Command-line, cron, & static HTML
    * Embedded in desktop app
    * Personal server
    * Multi-user web app

  * General
    * Work on much, much better error handling and reporting everywhere.
    * Think about restructuring to be more object-oriented?
    * Atom over Jabber
    * Update on received pings
    * Resolve permissions problem with logging, between web server user & CLI user

  * Data model
    * Notes for Subscriptions
    * Delicious-style tags for Subscriptions and Entries
    * Public user feed/subscriptions

  * Web UI
    * Use MUCH more unobtrusive JavaScript
      * http://www.onlinetools.org/articles/unobtrusivejavascript/chapter1.html
    * Sign up (by invitation?)
      * Email confirm?
      * Invitation code?
    * Login
      * Token in cookie?
    * Front / index page
    * Subscription management (add / edit / delete)
    * Need a plain reverse-chron entry listing
    * Convert favicons to GIF, include in UI
    
  * UI
    * Flash UI, OS X UI

  * REST API
    * Need by-URL ACLs.
    * Self-documenting API, pulled from docstrings
    * Make ALL urls in REST api relative, specify base in root tag xml:base attrib
    * Change REST api methods to have rest_* prefix.
    * Split rest API into separate topic-modules
    * Construct a query to list feeds along with new entry counts, ala bloglines
    * Think about Atom API for posting entries into feeds
    * Toss query parameters into XML produced by REST API, ala Amazon (parameters in XSL now)

  * CLI
    * Change cli-method docstrings to use OPTS: namecmd {one} {two} [{three}]
    * Distribute methods/functions collected in cli/api to respective modules.

  

Extracted
---------
./bin/update-todo.sh:2:# TODO: Make this suck less.
./htdocs/api:10:# TODO: Need XML-based traceback/error reporting
./htdocs/js/agg.js:127:  // TODO: Better busy-indicator for feed being marked as seen.
./htdocs/js/agg.js:135:  // TODO: Better busy-indicator for entry being marked as seen.
./htdocs/js/agg.js:143:  // TODO: Better busy-indicator for feed being marked as seen.
./htdocs/js/agg.js:151:  // TODO: Better busy-indicator for entry being marked as seen.
./htdocs/js/agg.js:163:  // TODO: Break up multiples into smaller chunks (of 10?)
./htdocs/js/agg.js:212:    // TODO: Some sort of loading animation for entries and XmlHTTPRequest
./htdocs/js/agg.js:256:            // TODO: Need an 'error' icon to switch handles to on loading error.
./lib/dbagg3/aggregate.py:300:        ### TODO: Collapse runs of empty whitespace & newlines in XML character copying.
./lib/dbagg3/aggregate.py:303:    # TODO: Do something sensible with errors encountered during XML copying.
./lib/dbagg3/handler/__init__.py:5:# TODO: Make more handlers (ie. pure Atom, xpath scraper), implement per-feed switching mechanism
./lib/dbagg3/handler/universal.py:213:            # TODO: Offer a preference to actually delete source 
./lib/dbagg3/model.py:22:    # TODO: This seems stupid & evil.  Learn about Unicode & fix.
./lib/dbagg3/model.py:27:        # TODO: Figure out how to handle Unicode encoding detection
./lib/dbagg3/rest.py:18:# TODO: Auto-generate API docs (in HTML?) from docstrings as an API call
./lib/dbagg3/rest.py:19:# TODO: Need better non-Pythonic error handling, in defined XML
./lib/dbagg3/rest.py:422:    # TODO: Find some way to abstract/automate these CRUD-style methods
./lib/dbagg3/rest.py:427:        # TODO: Figure out why XmlHTTPRequest is sending a null-terminated string
./lib/dbagg3/rest.py:487:                # TODO: Figure out proper way to present Location: header when multiple documents updated
./lib/dbagg3/rest.py:597:    # TODO: Output streams to stdout, should buffer (mid-stream exception?)
./lib/dbagg3/rest.py:666:# TODO: Find a better way to handle the EOL case in URL templates
./lib/dbagg3/scan.py:102:    # TODO: Per-source mix/max and period tweaking constants
./lib/dbagg3/scan.py:103:    # TODO: Per-source modular period tweaking algorithms (constant, AIMD, IM-driven)
./lib/dbagg3/web/api.py:18:# TODO: Auto-generate API docs (in HTML?) from docstrings as an API call
./lib/dbagg3/web/api.py:19:# TODO: Need better non-Pythonic error handling, in defined XML
./lib/dbagg3/web/api.py:542:    # TODO: Output streams to stdout, should buffer (mid-stream exception?)
./lib/dbagg3/web/api.py:601:# TODO: Find a better way to handle the EOL case in URL templates
./lib/dbagg3/web/httpd.py:21:        # TODO: Why doesn't os.path.join work here?
./lib/dbagg3/web/ui.py:115:    # TODO: Recognize attempts to create duplicate subscriptions
./lib/dbagg3/web/ui.py:116:    # TODO: Ask for confirmation before subscription.
./xsl/outliner/entry_include.xsl:11:      TODO: Make more tag-driven to handle cases of varying granularity. (ie. /feeds, /feed, /entry)
./xsl/outliner/entry_include_jsrs.xsl:11:      TODO: Make more tag-driven to handle cases of varying granularity. (ie. /feeds, /feed, /entry)
./xsl/outliner/index.xsl:11:      TODO: Make more tag-driven to handle cases of varying granularity. (ie. /feeds, /feed, /entry)