new orleans schools for loop

gistfile1.txt

library(jsonlite)
library(tidyr)

schools_list <- fromJSON("http://neworleansequityindex.org/jsoncache/schools.json")
schools_list <- schools_list$data

# for loop

# for i in 1 through number of rows of the dataframe schools_list

for (i in 1:nrow(schools_list)) {
  
  slug <- schools_list$slug[i]
  
  # paste0() is the base R equivalent to stringr's string_c
  # It concatenates strings together with nothing separating it
  
  # Rebuild the string to the school data json using the url we know
  school_url <- paste0("http://neworleansequityindex.org/api/schools/", slug, ".json")

  school_data <- fromJSON(school_url)
  
  # extracting the data only
  school_data <- school_data$data
  
  # transforming data from wide to tall
  # this is necesary because some schools don't have as many variables in the json as other schools
  # so this way it can still be joined with incomplete data
  
  school_data <- gather(school_data, "type", "data", 2:ncol(school_data))
  
  # building a big data frame of all the these pulled data frames
  
  if (i==1) { # if this is the first time this is looping...
    school_df <- school_data
    # then make the first pulled data frame the official data frame
  } else { # otherwise if it's not the first time this has looped
    
    # then rbind the new scraped dat frame to the older scraped data frame
    school_df <- rbind(school_df, school_data)
    
    # this is essentially appending the new data to the building data
  }
  
  # This prints out the status of the loop so I know how far it's progressed
  print(paste0(i, " of ", nrow(schools_list)))
  
}

# Spreads out the data wide like you'd see in a spreadsheet
# By turning the wide data to tall, combining, and then spreading it wide again
# We can get the wide data we had originally but won't be hindered by the missing data this way

school_df_wide <- spread(school_df, type, data)

# This is a cheat to deal with lists in the dataframe into normal arrays

df <- data.frame(apply(school_df_wide,2,as.character))

# This only happens when you're dealing with converting json into data frames


# saving the dataframe as a spreadsheet
write.csv(df, "school_df.csv", row.names=F)