sunlightlabs / datacommons Goto Github PK

https://github.com/sunlightlabs/datacommons/blob/master/dcapi/aggregates/contributions/contribution_aggregates.sql#L376

https://github.com/sunlightlabs/datacommons/blob/master/dcapi/aggregates/contributions/contribution_aggregates.sql#L460

possibly related to recent formatting changes, emailed out by paul clark

need to have an all-cycles aggregation of the booleans in this table. this will be useful for search nav, but will have to be ignored with when aggregating totals for summary stats

need to include information about an organization enity's status as org/pol_group/lobbying_firm in the agg tables so that summaries will be accurate.

remember to ignore all_cycle totals from the metadata table, because it won't be of use, here, as mentioned in #4

set pagination and per-page limits to avoid truncating results.

cf redmine issue 1227 (used for sql update naming)

dependent on resolving #4 and #5

add field for...

is_org
is_pol_group

need to add -1 entries in dcdata/independentexpenditures/postload.sql

reported by @sckott:

Would be nice if when the APIs fail they give back somewhat useful stuff. For example, here's what happens for the transparency API:

https://gist.github.com/sckott/ba808869da0b60bf7677

currently linking to an empty results page for, eg, DLA Piper, who is never a client

needs to be addressed in dcdata/contribution/sources/crp.py

It would be great if the DataCommons API supported JSON-P output, which allows the JSON data to be loaded across domains in a browser.

This might be tricky, but I'd love to see how different candidates' contributions stack up against the local or national average.

For example, compared to other state senators-- does my state senator take in an unusual amount of money from the education sector? What about PAC money?

I come from a world of baseball statistics, where it's common to compare players against league averages. Similar metrics in politics would be hugely informative-- Both in comparing politicians against each other, and in watching the landscape change between election cycles.

caught thanks to user feedback:

now that you've loaded in Federal Lobbying Data: updated to CRP's
4/29/2013 release, please add 2013 to the "Year" pulldown. there may
be more places 2013 is missing.

iknow you've got a whole lot on your plate during staffing transition...

Allow sorting by each column in the preview, and possibly the download, for search results.

Hitting http://transparencydata.com/api/1.0/entities/id_lookup.json with a bioguide_id of B001293 returns an empty array. Should return entity_id for Rep. Mike Bishop.

agg_cycle query populates with unique values for cycle, but in an odd year the max value will be the current year. agg_contribution_even_cycles will round that up to 2014, and later queries that use agg_cycle to filter will leave out 2014 entries because 2014 won't be in agg_cycle