# =============================================================================
# Examples of Using the Central Bank of Chile REST API (R)
# =============================================================================
# This script shows how to directly use the REST API to query
# series from the Central Bank's Statistical Database, building
# and using URLs explicitly.
#
# Main functions:
#   - SearchSeries: Search series by frequency
#   - GetSeries: Get data for a specific series
# =============================================================================

library("rjson")
library(httr)

# --- API Key Token (replace with your own) -------------------------------
token <- "your_token"

# =============================================================================
# EXAMPLE 1: QUERY OBSERVED DOLLAR
# =============================================================================
cat("\n=== EXAMPLE 1: QUERY OBSERVED DOLLAR ===\n")

# Step 1: Search for the series
cat("\nStep 1: Search for series related to 'observed dollar'\n")
cat("Building full URL for SearchSeries with frequency=DAILY...\n")

# Full URL to search daily series
url_search_dollar <- paste0("https://si3.bcentral.cl/SieteRestWS/SieteRestWS.ashx?token=", token, "&function=SearchSeries&frequency=DAILY")

cat("URL used:\n", url_search_dollar, "\n\n")

# Run the query
json_data_dolar <- rjson::fromJSON(file = url_search_dollar)

# Process SeriesInfos from JSON
daily_series <- as.data.frame(do.call(rbind, lapply(json_data_dolar$SeriesInfos, as.vector)))
daily_series$spanishTitle <- as.character(daily_series$spanishTitle)

# Filter series containing "dólar observado"
idx_dolar <- grep("dólar observado", daily_series$spanishTitle, ignore.case = TRUE)
dollar_series <- daily_series[idx_dolar, ]
cat("Found", nrow(dollar_series), "series related to 'dólar observado'\n")
cat("\nFirst matches:\n")
print(head(dollar_series, 3))

# Step 2: Query the data
cat("\nStep 2: Query data for observed dollar (last year)\n")
cat("We will use the code: F073.TCO.PRE.Z.D (daily observed dollar)\n")
cat("Building full URL for GetSeries...\n")

# Full URL to retrieve series data
url_dollar_data <- paste0("https://si3.bcentral.cl/SieteRestWS/SieteRestWS.ashx?token=", token, "&function=GetSeries&timeseries=F073.TCO.PRE.Z.D&firstdate=2024-09-01&lastdate=2025-09-30")

cat("URL utilizada:\n", url_dollar_data, "\n\n")

# Run the query
json_data_serie <- rjson::fromJSON(file = url_dollar_data)

# Process observations
# We use Obs from the JSON to access the value of the observations
observations <- as.data.frame(do.call(rbind, lapply(json_data_serie$Series$Obs, as.vector)))
observations$value <- as.numeric(observations$value)
observations$indexDateString <- as.character(observations$indexDateString)

cat("Data obtained:", nrow(observations), "observations\n")
cat("Period:", min(observations$indexDateString), "to", max(observations$indexDateString), "\n")
cat("\nLast 10 observations:\n")
print(tail(observations, 10))

# Step 3: Calculate basic statistics  
cat("\nStep 3: Basic statistics for observed dollar\n")
valid_values <- observations$value[!is.na(observations$value)]

cat("Minimum value: $", round(min(valid_values), 2), "\n")
cat("Maximum value: $", round(max(valid_values), 2), "\n") 
cat("Average value: $", round(mean(valid_values), 2), "\n")
cat("Last observation: $", round(tail(valid_values, 1), 2), "\n")


# =============================================================================
# Example 2: GDP AND IMACEC WITH DIFFERENT FREQUENCIES
# =============================================================================
cat("\n=== EXAMPLE 2: GDP AND IMACEC WITH DIFFERENT FREQUENCIES ===\n")
cat("In this example, we will work with two economic indicators that have\n")
cat("different frequencies: IMACEC (monthly) and GDP (quarterly).\n")
# Step 1: Search for the series
cat("\nStep 1: Search for the series\n")
cat("First, we search for both series separately according to their frequency.\n")

# Search for IMACEC (monthly)
cat("\nSearching for IMACEC (frequency MONTHLY):\n")
url_search_imacec <- paste0("https://si3.bcentral.cl/SieteRestWS/SieteRestWS.ashx?token=", token, "&function=SearchSeries&frequency=MONTHLY")

json_data_imacec <- rjson::fromJSON(file = url_search_imacec)

monthly_series <- as.data.frame(do.call(rbind, lapply(json_data_imacec$SeriesInfos, as.vector)))
monthly_series$spanishTitle <- as.character(monthly_series$spanishTitle)

idx_imacec <- grep("imacec empalmado", monthly_series$spanishTitle, ignore.case = TRUE)
imacec_series <- monthly_series[idx_imacec, ]
cat("Series found for IMACEC:", nrow(imacec_series), "\n")
cat("\nFirst matches:\n")
print(head(imacec_series, 5))

# Search for GDP (quarterly)
cat("\nSearching for GDP (frequency QUARTERLY):\n")
url_search_gdp <- paste0("https://si3.bcentral.cl/SieteRestWS/SieteRestWS.ashx?token=", token, "&function=SearchSeries&frequency=QUARTERLY")

json_data_gdp <- rjson::fromJSON(file = url_search_gdp)

quarterly_series <- as.data.frame(do.call(rbind, lapply(json_data_gdp$SeriesInfos, as.vector)))
quarterly_series$spanishTitle <- as.character(quarterly_series$spanishTitle)

idx_gdp <- grep("PIB, volumen a precios del año anterior encadenado", quarterly_series$spanishTitle, ignore.case = TRUE)
series_gdp <- quarterly_series[idx_gdp, ]
cat("Series found for GDP:", nrow(series_gdp), "\n")
cat("\nLast matches:\n")
print(tail(series_gdp, 10))

# Step 2: Query each series separately
cat("\nStep 2: Query each series individually\n")
cat("From the previous results, we will use:\n")
cat("- IMACEC: F032.IMC.IND.Z.Z.EP18.Z.Z.0.M (monthly frequency)\n")
cat("- GDP: F032.PIB.FLU.R.CLP.EP18.Z.Z.0.T (quarterly frequency)\n")

# Query IMACEC
cat("\nQuerying IMACEC (monthly):\n")
url_imacec_data <- paste0("https://si3.bcentral.cl/SieteRestWS/SieteRestWS.ashx?token=", token, "&function=GetSeries&timeseries=F032.IMC.IND.Z.Z.EP18.Z.Z.0.M&firstdate=2024-01-01&lastdate=2025-09-30")
json_imacec_data <- rjson::fromJSON(file = url_imacec_data)

obs_imacec <- as.data.frame(do.call(rbind, lapply(json_imacec_data$Series$Obs, as.vector)))
obs_imacec$value <- as.numeric(obs_imacec$value)
obs_imacec$indexDateString <- as.character(obs_imacec$indexDateString)
colnames(obs_imacec)[colnames(obs_imacec) == "value"] <- "IMACEC"

cat("IMACEC:", nrow(obs_imacec), "monthly observations\n")
print(tail(obs_imacec, 5))

# Query GDP
cat("\nQuerying GDP (quarterly):\n")
url_gdp_data <- paste0("https://si3.bcentral.cl/SieteRestWS/SieteRestWS.ashx?token=", token, "&function=GetSeries&timeseries=F032.PIB.FLU.R.CLP.EP18.Z.Z.0.T&firstdate=2024-01-01&lastdate=2025-09-30")

json_gdp_data <- rjson::fromJSON(file = url_gdp_data)

obs_gdp <- as.data.frame(do.call(rbind, lapply(json_gdp_data$Series$Obs, as.vector)))
obs_gdp$value <- as.numeric(obs_gdp$value)
obs_gdp$indexDateString <- as.character(obs_gdp$indexDateString)
colnames(obs_gdp)[colnames(obs_gdp) == "value"] <- "GDP"

cat("GDP:", nrow(obs_gdp), "quarterly observations\n")
print(obs_gdp)
# Step 3: Combine both series into a DataFrame
cat("\nStep 3: Combine both series with different frequencies\n")
cat("The REST API allows querying series with different frequencies.\n")
cat("We create a combined DataFrame to show both series:\n")

# Prepare dates for merging 
obs_imacec$fecha_date <- as.Date(obs_imacec$indexDateString, format = "%d-%m-%Y")
obs_gdp$fecha_date <- as.Date(obs_gdp$indexDateString, format = "%d-%m-%Y")

# Create a sequence of monthly dates for the entire period
all_dates <- seq(from = as.Date("2024-01-01"), 
                       to = as.Date("2025-08-01"), 
                       by = "month")

# Base data frame with all dates
combined_data <- data.frame(
  fecha = format(all_dates, "%d-%m-%Y"),
  fecha_date = all_dates,
  stringsAsFactors = FALSE
)

# Add IMACEC (monthly - will have values for all dates)
combined_data <- merge(combined_data, obs_imacec[, c("fecha_date", "IMACEC")], 
                         by = "fecha_date", all.x = TRUE)

# Add GDP (quarterly - only in January, April, July, October)
combined_data <- merge(combined_data, obs_gdp[, c("fecha_date", "GDP")], 
                         by = "fecha_date", all.x = TRUE)

# Select final columns
combined_data <- combined_data[, c("fecha", "IMACEC", "GDP")]
cat("Combined data (original frequency):\n")
cat("IMACEC has monthly values, while GDP only has values\n")
cat("quarterly (January, April, July, October). That's why NA appear in the\n")
cat("intermediate dates for GDP.\n\n")
print(head(combined_data, 8))