Create sca_e_projections by boro

[AmandaDoyle]

Aggregate HS data to boro level

1. Create a new folder named sca_e_projections_by_boro under the ceqr_app_date/ceqr/recipes directory
2. Create config.json, README.md, build.py, and requirements.txt
3. use sca_e_projections.2019 as the input data
4. target schema @bfreeds please confirm the schema for this CEQR table

   CREATE TABLE sca_e_projections_by_boro."2019" (
   school_year integer,
   borocode integer,
   hs integer
   );

5. build its ETL pipeline using sca_e_projections as an example
   • convert the input table from a wide table to a long table using a pandas syntax called melt
   • calculate the hs projections by summing up projected = [9,10,11,12], don't use projected = '9-12 Total' from sca_e_projections.2019 which has incorrect values
   • write a python function or dictionary to map district to its matchingborocode
   • aggregate the data to boro level
6. export the output to EDM_DATA
7. push your code to another branch

[bfreeds]

@AmandaDoyle following the schemas for 2017 and 2018 would be:

CREATE TABLE sca_enrollment_projections_by_boro."2019" (
    year varchar,
    borough text,
    hs integer
);

[baolingz]

@bfreeds @AmandaDoyle We've refactored the ETL for 2018 and 2019 tables based on the above schema using SCA's source data.

[bfreeds]

@baolingz copy, thank you! Apologies for my delayed response, Github's new notifications management (which is waaay more usable) helped me see that I was pinged on this issue.