Skip to content

Latest commit

 

History

History
76 lines (50 loc) · 2.71 KB

importing-data.md

File metadata and controls

76 lines (50 loc) · 2.71 KB
Raw

Pegasus

Importing Data

Data may be imported from Google Sheets or CSV files into SQL tables. Columns in these spreedsheets must be named using type suffixes.

Importing Google Sheets

  1. Create a sheet in the "Pegasus" Google Docs folder, naming your columns using type suffixes.
  2. In the Dropbox folder for a site, create a file named data/table_name.gsheet where table_name is the name you'd like the database table to have. The file should contain the "Pegasus" Google Docs path to the sheet you created in step 1. For example, code.org/data/about_team.gsheet would contain Data/About/Team and the data will be imported into a table named about_team.
  3. The data should import on staging automatically within a few minutes.

Importing CSV files

  1. In the Dropbox folder for a site, create a file named data/table_name.csv where table_name is the name you'd like the database table to have naming your columns using type suffixes.

Type Suffixes

  • _dt DateTime
  • _f Float
  • _i Integer
  • _s String (limited to 255 characters)
  • _t Text (limited to 65,535 characters)

A column can also, optionally, have one of the following modifiers:

  • * Index this field
  • ! Index this field, values must be unique.

Example

Here's an example spreadsheet for implementing a zipcode-based geocoding service, code.org/data/zip_codes.csv:

code_s! city_s* state_s* latitude_f longtude_f
10451 Bronx NY 40.84 -73.87
10452 Bronx NY 40.84 -73.87
10453 Bronx NY 40.84 -73.87
... ... ... ... ...

This spreadsheet defines a table with the following schema:

  • code_s - String - Unique, Indexed
  • city_s - String - Indexed
  • state_s - String - Indexed
  • latitude_f and longitude_f - Floats

Note that the column names include the type suffix, but do not include the index modifier (i.e. * or !).

And here's an example document using it:

code.org/zip_codes/splat.md:

<%
# Requests are prefixed with a '/' so turn '/98109' into '98109'
zip_code = request.splat_path_info[1..-1]

# Look for a row with that zip code and pass if we don't find one
pass unless row = DB[:zip_codes].where(code_s:zip_code).first

%>

Zip code: <%= row[:code_s] %>
City: <%= row[:city_s] %>
State: <%= row[:state_s] %>
Location: (<%= row[:latitude_f] %>, <%= row[:longitude_f] %>)

Developer Details

  • rake seed:sync will detect modified gsheets, download updated versions as .csv files, and then perform a seed:migrate.
  • rake seed:migrate will import any modified .csv files into the database.
  • rake seed:reset will import all .csv files into the database regardless of their modification status.