R/omb.R

In abresler/govtrackR: Federal government tracking tools

#' DOD OMB Group
#'
#' @param data
#'
#' @return
#' @export
#'
#' @examples
add_dod_omb_group <-
  function(data) {

    match_cols <-
      data %>%
      select(one_of(
        c(
          "idCGAC",
          "nameAgencyCGAC",
          "nameBureauOMB",
          "nameSubFunctionOMB",
          "nameAccountOMB"
        )
      )) %>%
      ncol()

    if (match_cols < 3) {
      return(data)
    }
    data <-
      data %>%
      mutate_if(is.numeric, as.numeric) %>%
      select(-matches("groupDOD|isDefenseAccount"))

    df_accounts <-
      data %>%
      distinct(idCGAC,
               nameAgencyCGAC,
               nameBureauOMB,
               nameSubFunctionOMB,
               nameAccountOMB) %>%
      arrange(idCGAC) %>%
      mutate(isDefenseAccount = case_when(
        nameAgencyCGAC %>% str_detect("ARMY|NAVY|SPACE FORCE|AIR FORCE|DEFENSE") ~ TRUE,
        TRUE ~ FALSE
      ))

    df_accounts <-
      df_accounts %>%
      mutate(
        groupDOD = case_when(
          nameBureauOMB %>% str_detect("PROCUREMENT") ~ "PROCUREMENT",
          nameBureauOMB %>% str_detect("CONSTRUCTION|FAMILY HOUSING") ~ "CONSTRUCTION",
          nameBureauOMB %>% str_detect("RESEARCH") ~ "RDT&E",
          nameBureauOMB == "MILITARY PERSONNEL" ~ "MILITARY PERSONNEL",
          nameBureauOMB %>% str_detect("OPERATION AND MAINTENANCE") ~ "OPERATIONS AND MAINTENANCE",
          nameBureauOMB %>% str_detect("DEPARTMENT OF DEFENSE--MILITARY PROGRAMS") ~ "MILITARY PROGRAMS",
          nameSubFunctionOMB %>% str_detect(
            "RESEARCH, DEVELOPMENT,|SCIENCE AND TECHNOLOGY|GENERAL SCIENCE|AGRICULTURAL RESEARCH"
          ) ~ "RDT&E",
          nameAccountOMB %>% str_detect("PROCUREMENT") ~ "PROCUREMENT",
          nameAccountOMB %>% str_detect("CONSTRUCTION|FAMILY HOUSING") ~ "CONSTRUCTION",
          nameAccountOMB %>% str_detect("RETIREMENT") ~ "RETIREMENT",
          nameAccountOMB %>% str_detect("RESEARCH") ~ "RDT&E",
          nameAccountOMB == "MILITARY PERSONNEL" ~ "MILITARY PERSONNEL",
          nameAccountOMB %>% str_detect("WORKING CAPITAL") ~ "WORKING CAPITAL",
          nameAccountOMB %>% str_detect("OPERATION AND MAINTENANCE") ~ "OPERATIONS AND MAINTENANCE",
          nameAccountOMB %>% str_detect("DEPARTMENT OF DEFENSE--MILITARY PROGRAMS") ~ "MILITARY PROGRAMS",
          TRUE ~ "OTHER"
        )
      )

    data <-
      data %>%
      left_join(
        df_accounts,
        by =  c(
          "idCGAC",
          "nameAgencyCGAC",
          "nameBureauOMB",
          "nameAccountOMB",
          "nameSubFunctionOMB"
        )
      )


    data

  }



.join_tas_symbols <-
  function(data) {
    df_tas <- dictionary_treasury_account_symbols()

    df_tas <-
      df_tas %>%
      distinct(
        idFederalAccount,
        idCGAC,
        codeBureauTAS = codeBureauOMB,
        nameBureauTAS = nameBureauOMB,
        nameAgencyTAS = nameAgency,
        nameFinancialReportingEntity,
        codeFinancialReportingEntity,
        datetimeProgramEstablished,
        datetimeProgramEnded
      ) %>%
      group_by(idFederalAccount) %>%
      filter(datetimeProgramEstablished == max(datetimeProgramEstablished)) %>%
      slice(1) %>%
      ungroup()

    data <-
      data %>%
      mutate(idCGAC = as.numeric(idCGAC)) %>%
      left_join(df_tas, by = c("idFederalAccount", "idCGAC"))

    data <-
      data %>%
      munge_data()

    data
  }




# utils -------------------------------------------------------------------

.add_omb_account_group  <-
  function(data) {
    data <-
      data %>%
      mutate(
        groupOMBAccount = case_when(
          nameAccountOMB %>% str_detect("SALARIES|COMPENSATION OF|^COMPENSATION|SEPARATION PAY|PERSONNEL") ~ "SALARY",
          nameAccountOMB %>% str_detect(
            "RESEARCH PROJECTS|RESEARCH, DEVELOPMENT|^RESEARCH AND DEVELOPMENT|^SCIENCE|EXPLORATION TECHNOLOGY|INFORMATION TECHNOLOGY|RESEARCH AND DEVELOPMENT|^HEALTH RESEARCH|^RESEARCH AND|^MAJOR RESEARCH"
          ) ~ "RESEARCH, SCIENCE AND TECHNOLOGY",
          nameAccountOMB %>% str_detect("PROCUREMENT|PURCHASES|WEAPONS ACTIVITIES") ~ "PROCUREMENT",
          nameAccountOMB %>% str_detect("ASSISTANCE") ~ "ASSISTANCE",
          nameAccountOMB %>% str_detect("OPERATIONS AND SUPPORT|OPERATION AND MAINTENANCE") ~ "OPERATIONS AND MAINTENANCE",
          nameAccountOMB %>% str_detect(
            "PROGRAM ADMINISTRATION|POLICY AND ADMINISTRATION|DEPARTMENTAL ADMINISTRATION|GENERAL ADMINISTRATION"
          ) ~ "PROGRAM ADMINISTRATION",
          nameAccountOMB %>% str_detect("CAPITAL IMPROVEMENT") ~ "CAPITAL IMPROVMENT",
          nameAccountOMB %>% str_detect("WORKING CAPITAL|REVOLVING FUND") ~ "WORKING CAPITAL",
          nameAccountOMB %>% str_detect("OPERATING EXPENSES|^EXPENSES ") ~ "OPERATING EXPENSES",
          nameAccountOMB %>% str_detect("LEASES|SALES") ~ "SALES AND LEASES",
          nameAccountOMB %>% str_detect("MILITARY SALE") ~ "MILITARY SALES",
          nameAccountOMB %>% str_detect("FEES") ~ "FEES",
          nameAccountOMB %>% str_detect(
            "COURT SECURITY|BORDER SECURITY|CYBER SECURITY|CYBERSECURITY|MARITIME SECURITY|PROTECTION OF FOREIGN MISSIONS AND OFFICIALS"
          ) ~ "SECURITY",
          nameAccountOMB %>% str_detect("TRANSMISSION") ~ "POWER TRANSMISSION",
          nameAccountOMB %>% str_detect("HEALTH PROGRAM") ~ "HEALTH CARE",
          nameAccountOMB %>% str_detect("PENSION|RETIREMENT|READJUSTMENT BENEFITS|RETIREMENT BENEFITS|BENEFIT PLAN") ~ "RETIREMENT BENEFITS",
          nameAccountOMB %>% str_detect("ANNUITY") ~ "ANNUITIES",
          nameAccountOMB %>% str_detect(" FUND,| FUND |FUND ") ~ "FUND",
          nameAccountOMB %>% str_detect("EXPLORATION") ~ "EXPLORATION",
          nameAccountOMB %>% str_detect(
            "DRUG INTERDICTION|CRIME AND DRUG ENFORCEMENT AND ENFORCEMENT|DRUG TRAFFICKING|DRUG TRAFFICKING"
          ) ~ "DRUG INTERDICTION AND ENFORCEMENT",
          nameAccountOMB %>% str_detect("^RENT |^RENTS |^RENTAL PAYMENTS")  ~ "RENTS",
          nameAccountOMB %>% str_detect("SUBSIDIES|SUBSIDY") ~ "SUBSIDY",
          nameAccountOMB %>% str_detect("REFUNDS") ~ "REFUNDS",
          nameAccountOMB %>% str_detect("SCHOLARSHIP") ~ "SCHOLARSHIP",
          nameAccountOMB %>% str_detect(" GRANT |GRANT | GRANTS |GRANTS ") ~ "GRANTS",
          nameAccountOMB %>% str_detect("CONTRIBUTION TO|CONTRIBUTIONS TO|^CONTRIBUTIONS|ALLIED CONTRIB|AFGHANISTAN SECURITY|COOPERATIVE THREAT REDUCTION|COUNTER-ISLAMIC|BASE CLOSURE|WORKFORCE DEVELOPMENT") ~ "CONTRIBUTIONS",
          nameAccountOMB %>% str_detect(
            "COMMISSIONS|UNITED STATES-CHINA ECONOMIC AND SECURITY REVIEW COMMISSION|UNITED STATES COMMISSION ON INTERNATIONAL RELIGIOUS FREEDOM
DENALI COMMISSION"
          ) ~ "COMMISSION",
          nameAccountOMB %>% str_detect("MEDICARE") ~ "MEDICARE",

          nameAccountOMB %>% str_detect("NUTRITION") ~ "NUTRITION",
          nameAccountOMB %>% str_detect("FINANCING") ~ "FINANCING",
          nameAccountOMB %>% str_detect("LOAN PROGRAM") ~ "LOAN PROGRAM",
          nameAccountOMB %>% str_detect("FIRE") ~ "FIRE MANAGEMENT",
          nameAccountOMB %>% str_detect("SCHOOL IMPROVEMENT") ~ "SCHOOL IMPROVEMENT",
          nameAccountOMB %>% str_detect("PROFIT|EARNINGS") ~ "PROFITS AND EARNINGS",
          nameAccountOMB %>% str_detect("FRAUD") ~ "FRAUD PREVENTION",
          nameAccountOMB %>% str_detect("SUPPLEMENTAL") ~ "SUPPLEMENTAL ASSISTANCE",
          nameAccountOMB %>% str_detect("INSPECTOR GENERAL") ~ "INSPECTOR GENERAL",
          nameAccountOMB %>% str_detect("^PAYMENT TO") ~ "PAYMENTS",
          nameAccountOMB %>% str_detect(
            "ECONOMIC DEVELOPMENT|ECONOMIC SUPPORT AND DEVELOPMENT FUND|BUSINESS DEVELOPMENT|ENTREPRENEURIAL DEVELOPMENT|ECONOMIC SUPPORT |COMMUNITY DEVELOPMENT"
          ) ~ "ECONOMIC DEVELOPMENT",
          nameAccountOMB %>% str_detect("CAPITAL IMPROVEMENTS") ~ "CAPITAL IMPROVEMENTS",
          nameAccountOMB %>% str_detect("RECEIVABLES") ~ "RECEIVABLES",
          nameAccountOMB %>% str_detect("NOT OTHERWISE CLASSIFIED") ~ "NOT OTHERWISE CLASSIFIED",
          nameAccountOMB %>% str_detect("BUILDINGS AND FACILITIES|CARE OF THE BUILDING AND GROUNDS") ~ "BUILDINGS AND FACILITIES",
          nameAccountOMB %>% str_detect("INSURANCE") ~ "INSURANCE",
          nameAccountOMB %>% str_detect("LIQUIDATING ACCOUNT") ~ "LIQUIDATING ACCOUNT",
          nameAccountOMB %>% str_detect("CLEARNING ACCOUNT") ~ "CLEARNING ACCOUNT",
          nameAccountOMB %>% str_detect("INTEREST ON") ~ "INTEREST",
          nameAccountOMB %>% str_detect("^DONATIONS") ~ "DONATIONS",
          nameAccountOMB %>% str_detect("MILITARY CONSTRUCTION|FAMILY HOUSING") ~ "MILITARY CONSTRUCTION",
          nameAccountOMB %>% str_detect("REIMBURSEMENT") ~ "REIMBURSEMENTS",
          nameAccountOMB %>% str_detect("INNOVATION") ~ "INNOVATION",
          nameAccountOMB %>% str_detect("GUARANTEE") ~ "GUARANTEES",
          nameAccountOMB %>% str_detect("FORFEITURE") ~ "FORFEITURES",
          nameAccountOMB %>% str_detect("REVENUE") ~ "REVENUES",
          nameAccountOMB %>% str_detect("REFUGEE") ~ "REFUGEES",
          nameAccountOMB %>% str_detect("ALLOWANCE") ~ "ALLOWANCE",
          nameAccountOMB %>% str_detect("MEDICAID") ~ "MEDICAID",
          TRUE ~ "OTHER"
        )
      )

    data <-
      data %>%
      mutate(
        groupOMBAccount = case_when(
          nameSubFunctionOMB %>% str_detect("MEDICARE") ~ "MEDICARE",
          nameSubFunctionOMB %>% str_detect("CENTRAL FISCAL") ~ "SALARIES",
          nameSubFunctionOMB %>% str_detect("ATOMIC ENERGY|EMERGENCY ENERGY|ENERGY CONSERV|ENERGY SUPPLY") ~ "ENERGY",
          nameSubFunctionOMB %>% str_detect("SOCIAL SERVICES") ~ "SOCIAL SERVICES",
          nameSubFunctionOMB %>% str_detect("SOCIAL SECURITY") ~ "SOCIAL SECURITY",
          nameSubFunctionOMB %>% str_detect("INTEREST") ~ "INTEREST",
          nameSubFunctionOMB %>% str_detect("FARM INCOME") ~ "SUBSIDY",
          nameSubFunctionOMB %>% str_detect("TRANSPORT") ~ "TRANSPORTATION",
          nameSubFunctionOMB %>% str_detect("FOREIGN INFORMATION|FOREIGN AFFAIRS|INTERNATIONAL FINANC") ~ "FOREIGN AFFAIRS",
          nameSubFunctionOMB %>% str_detect("HEALTH CARE|HOSPITAL AND") ~ "HEALTH CARE",
          nameSubFunctionOMB %>% str_detect("EDUCATION") ~ "EDUCTATION",
          nameSubFunctionOMB %>% str_detect("ASSISTANCE$") ~ "ASSISTANCE",
          nameSubFunctionOMB %>% str_detect("OTHER ADVANCEMENT") ~ "CONTRIBUTIONS",
          nameSubFunctionOMB %>% str_detect("DISASTER RELIEF") ~ "DISASTER RELIEF",
          nameSubFunctionOMB %>% str_detect("INCOME SECURITY|BENEFITS AND SERVICES") ~ "RETIREMENT BENEFITS",
          nameSubFunctionOMB %>% str_detect("UNEMPLOYMENT|GENERAL RETIRE") ~ "UNEMPLOYMENT",
          nameSubFunctionOMB %>% str_detect("HEALTH RESEARCH|SPACE FLIGHT,|RESEARCH,") ~ "RESEARCH, SCIENCE AND TECHNOLOGY",
          nameSubFunctionOMB %>% str_detect("POLLUTION|ENVIRONMENTAL") ~ "ENVIRONMENT AND POLLUTION",
          nameBureauOMB %>% str_detect("PROCUREMENT") ~ "PROCUREMENT",
          nameSubFunctionOMB %>% str_detect("NATURAL RESOURCES") ~ "NATURAL RESOURCES",
          nameSubFunctionOMB %>% str_detect("RECREATION") ~ "RECREATION",
          nameSubFunctionOMB %>% str_detect("TRAINING AND EMPLOYMENT") ~ "JOB TRAINING",
          nameSubFunctionOMB %>% str_detect("AREA AND REGIONAL DEVELOPMENT|COMMUNITY DEVELOPMENT") ~ "ECONOMIC DEVELOPMENT",
          nameSubFunctionOMB %>% str_detect("LAND MANAGEMENT") ~ "LAND MANAGEMENT",
          nameSubFunctionOMB %>% str_detect("WATER") ~ "WATER",
          nameSubFunctionOMB %>% str_detect("LAW ENFORCEMENT") ~ "LAW ENFORCEMENT",
          nameSubFunctionOMB %>% str_detect("JUDICIAL ACTIV") ~ "JUDICIAL ACTIVITIES",
          nameSubFunctionOMB %>% str_detect("RESEARCH, DEVELOPMENT,|SCIENCE AND TECHNOLOGY|GENERAL SCIENCE|AGRICULTURAL RESEARCH") ~ "RESEARCH, SCIENCE AND TECHNOLOGY",
          TRUE ~ groupOMBAccount
        )
      )
    data
  }

#' Munge Treasury Account Symbol Data
#'
#' @param data
#' @param research_terms
#' @param research_columns
#'
#' @return
#' @export
#'
#' @examples
munge_account_symbols <-
  function(data,
           research_terms = c("research",
                              "science",
                              "technology",
                              "ARTIFICIAL INTELLIGENCE",
                              "technical"),
           research_columns = c("nameAgencyOMB", "nameBureauOMB", "nameAccountOMB")) {
    if (data %>% hasName("idCGAC")) {
      data <- data %>%
        mutate(hasNoCGAC = is.na(idCGAC))
    }

    if (data %>% hasName("nameAgencyCGAC") &
        data %>% hasName("idCGAC")) {
      data <- data %>%
        mutate(idCGAC = as.character(idCGAC)) %>%
        select(-one_of("slugCGAC")) %>%
        left_join(.cgac_codes() %>% mutate(idCGAC = as.character(idCGAC))
                  ,
                  by = "idCGAC") %>%
        select(names(data), everything())
    }

    has_research_cols <- length(research_columns) > 0

    has_research_search <-
      length(research_terms) > 0

    if (has_research_cols & has_research_search) {
      research_columns %>%
        walk(function(col) {
          data <<- .add_research_science_flags(data = data,
                                               terms = research_terms,
                                               column_name = col)
        })
    }

    data
  }

.add_research_science_flags <-
  function(data,
           terms = c("research",
                     "science",
                     "technology",
                     "ARTIFICIAL INTELLIGENCE",
                     "technical"),
           column_name = "nameAccountOMB") {
    if (!data %>% hasName(column_name)) {
      return(data)
    }
    terms <-
      terms %>% str_to_upper() %>% unique() %>% str_c(collapse = "|")

    new_col <-
      column_name %>% str_replace_all("name", "is") %>% str_c("ResearchFlag")

    data <-
      data %>%
      mutate(!!sym(new_col) := !!sym(column_name) %>% str_detect(terms))

    data
  }

.dictionary_omb_names <-
  function(data) {
    tibble(
      nameOMB = c(
        "Agency Code",
        "Agency Name",
        "Bureau Code",
        "Bureau Name",
        "Account Code",
        "Account Name",
        "Treasury Agency Code",
        "Subfunction Code",
        "Subfunction Title",
        "BEA Category",
        "On- or Off- Budget",
        "yearBudget",
        "amount",
        "Grant/non-grant split",
        "Source Category Code",
        "Source category name",
        "Source subcategory",
        "Source subcategory name",
        "Agency code",
        "Agency name",
        "Bureau code",
        "Bureau name",
        "Account code",
        "Account name",
        "Treasury Agency code",
        "On- or off- budget"
      ),
      nameActual = c(
        "codeAgencyOMB",
        "nameAgencyOMB",
        "codeBureauOMB",
        "nameBureauOMB",
        "codeAccountOMB",
        "nameAccountOMB",
        "idCGAC",
        "codeSubFunctionOMB",
        "nameSubFunctionOMB",
        "typeCategoryBEA",
        "isOnBudget",
        "yearBudget",
        "amount",
        "typeFunding",
        "codeSource",
        "nameSource",
        "codeSubFunctionBudget",
        "nameSubFunctionBudget",
        "codeAgencyOMB",
        "nameAgencyOMB",
        "codeBureauOMB",
        "nameBureauOMB",
        "codeAccountOMB",
        "nameAccountOMB",
        "idCGAC",
        "isOnBudget"
      )
    )

  }

.munge_omb_names <-
  function(data) {
    names_dict <- names(data)

    dict <- .dictionary_omb_names()
    actual_names <-
      names_dict %>%
      map_chr(function(name) {
        df_row <-
          dict %>% filter(nameOMB == name)
        if (nrow(df_row) == 0) {
          glue::glue("Missing {name}") %>% message()
          return(name)
        }

        df_row$nameActual
      })

    data %>%
      set_names(actual_names)
  }

.munge_general <-
  function(data,
           base_row = 2,
           table_name,
           year_plus = 2,
           year_plus_long = 1,
           is_wide = F) {
    df_types <-
      data %>% dplyr::slice(base_row) %>% select(2:ncol(data)) %>%
      gather(column, type) %>%
      fill(type) %>%
      mutate(idRow = 1:n()) %>%
      select(idRow, type) %>%
      .remove_na()

    if (df_types %>% ncol() == 1 | is_wide) {
      years <-
        data %>% dplyr::slice(base_row + year_plus_long) %>% select(2:ncol(data)) %>% as.character()

      df_features <-
        data %>%
        select(1) %>%
        slice((base_row + 2):nrow(data)) %>%
        mutate(idRow = 1:n()) %>%
        setNames(c("metric", "idRow"))

      df_subparents <-
        df_features %>%
        filter(metric %>% str_detect("\\:")) %>%
        rename(subparent = metric) %>%
        filter(!subparent %>% str_detect("Note:"))

      df_parents <-
        df_features %>%
        filter(metric %>% str_detect("\\:")) %>%
        select(idRow) %>%
        mutate(idRow = idRow + 1) %>%
        left_join(df_features, by = "idRow") %>%
        mutate(idRow = idRow - 1) %>%
        rename(parent = metric)

      data <-
        data %>%
        slice((base_row + 2):nrow(data)) %>%
        setNames(c("metric", years)) %>%
        mutate(idRow = 1:n()) %>%
        select(idRow, everything())

      df_parents <-
        data %>%
        select(idRow, metric) %>%
        filter(metric %>% str_detect(":")) %>%
        rename(parent = metric)

      d <-
        data %>%
        left_join(df_parents, by = "idRow") %>%
        select(idRow, parent, metric, everything()) %>%
        fill(parent)

      group <-
        d %>% filter(is.na(parent)) %>% select(metric) %>% pull()

      if (length(group) == 1) {
        d <-
          d %>%
          mutate(group = group)

        d <-
          d %>%
          filter(!is.na(parent)) %>%
          filter(!metric %>% str_detect(":")) %>%
          select(-idRow) %>%
          gather(yearBudget, amount, -c(group, parent, metric)) %>%
          mutate(amount = amount %>% readr::parse_number()) %>%
          filter(!is.na(amount)) %>%
          mutate(
            nameTable = table_name,
            isEstimate = yearBudget %>% str_detect("estimate"),
            yearBudget = yearBudget %>% str_remove_all("estimate")
          ) %>%
          select(nameTable,
                 isEstimate,
                 yearBudget,
                 group,
                 parent,
                 metric,
                 everything()) %>%
          filter(!metric %>% str_detect("Total")) %>%
          mutate(amount = as.character(amount)) %>%
          .convert_amount() %>%
          .summarise_sum(
            columns = c(
              "nameTable",
              "isEstimate",
              "yearBudget",
              "group",
              "parent",
              "metric"
            )
          )
      } else {
        d <- d %>%
          filter(!metric %>% str_detect(":")) %>%
          select(-idRow) %>%
          gather(yearBudget, amount, -c(parent, metric)) %>%
          mutate(amount = amount %>% readr::parse_number()) %>%
          filter(!is.na(amount)) %>%
          mutate(
            nameTable = table_name,
            isEstimate = yearBudget %>% str_detect("estimate"),
            yearBudget = yearBudget %>% str_remove_all("estimate")
          ) %>%
          select(nameTable,
                 isEstimate,
                 yearBudget,
                 parent,
                 metric,
                 everything()) %>%
          filter(!metric %>% str_detect("Total"))

        d <-
          d %>%
          mutate_at(c("parent",
                      "metric"),
                    list(function(x) {
                      x %>% str_remove_all("\\:|\\(|\\)") %>% str_replace_all("\\-", " ") %>% str_squish()
                    }))

        return(d)
      }


      d <- d %>%
        mutate_at(c("group",
                    "parent",
                    "metric"),
                  list(function(x) {
                    x %>% str_remove_all("\\:|\\(|\\)") %>% str_replace_all("\\-", " ") %>% str_squish()
                  }))

      return(d)
    }

    df_outlays <-
      data %>% dplyr::slice(base_row + 1) %>% select(2:ncol(data)) %>%
      gather(column, outlay) %>%
      mutate(idRow = 1:n(),
             outlay = coalesce(outlay, "")) %>%
      select(idRow, outlay) %>%
      mutate(outlay = outlay %>% str_replace_all("\n", " "))

    df_cols <-
      df_outlays %>%
      left_join(df_types, by = "idRow") %>%
      unite(type, type, outlay, sep = "_")
    d <-
      data %>%
      dplyr::slice((base_row + year_plus):nrow(data))


    data <-
      data %>%
      set_names(c("yearBudget", df_cols$type)) %>%
      gather(type, amount, -yearBudget) %>%
      separate(
        type,
        into = c("typeBudget", "typeCashFlow"),
        sep = "_",
        extra = "merge"
      ) %>%
      filter(typeBudget != "Total",
             !typeCashFlow %>% str_detect("Surplus")) %>%
      mutate(yearBudget = case_when(yearBudget == "TQ" ~ "1977",
                                    TRUE ~ yearBudget)) %>%
      mutate(amount = amount %>% readr::parse_number()) %>%
      filter(!is.na(amount)) %>%
      mutate(
        nameTable = table_name,
        isEstimate = yearBudget %>% str_detect("estimate"),
        yearBudget = yearBudget %>% str_remove_all("estimate")
      )

    data
  }



.unite_filter_metric <-
  function(data, budget_filter = NULL) {
    if (length(budget_filter) > 0) {
      data <- data %>%
        filter(typeBudget %in% budget_filter)
    }

    data %>%
      unite(metric, typeBudget, typeCashFlow, sep = "")
  }

.summarise_pct <-
  function(data,
           columns = c("yearBudget", "isEstimate", "metric")) {
    data %>%
      group_by(!!!syms(columns)) %>%
      summarise_all(mean) %>%
      ungroup() %>%
      arrange(yearBudget)
  }

.summarise_sum <-
  function(data,
           columns = c("yearBudget", "isEstimate", "metric"),
           amount_col = "amount") {
    data %>%
      group_by(!!!syms(columns)) %>%
      summarise_at(amount_col, sum) %>%
      ungroup() %>%
      arrange(yearBudget)
  }

.summarise_mean <-
  function(data,
           columns = c("yearBudget", "isEstimate", "metric"),
           amount_col = "amount") {
    data %>%
      group_by(!!!syms(columns)) %>%
      summarise_at(amount_col, mean) %>%
      ungroup() %>%
      arrange(yearBudget)
  }
.omb_munge <- function(data) {
  if (data %>% hasName("yearBudget") &
      !data %>% hasName("isEstimate")) {
    data <-
      data %>%
      mutate(
        isEstimate = yearBudget %>% str_detect("estimate"),
        yearBudget = yearBudget %>% str_remove_all("estimate") %>% parse_number()
      ) %>%
      fill(yearBudget)
  }

  amt_mill <- data %>% select(matches("amountMillions")) %>% names()
  amt_bil <-
    data %>% select(matches("amountBillions")) %>% names()
  if (length(amt_mill) > 0) {
    data <-
      data %>%
      mutate_at(amt_mill, function(x) {
        x * 1000000 %>% currency(digits = 0)
      })

    names(data) <-
      names(data) %>% str_replace_all("amountMillions", "amount")
  }

  pct_cols <- data %>% select(matches("pct")) %>% names()

  if (length(pct_cols) > 0) {
    data <- data %>% mutate_at(pct_cols, list(function(x) {
      x / 100
    }))
    data <-
      data %>%
      mutate_at(amt_bil, function(x) {
        x * 1000000000 %>% currency(digits = 0)
      })

    names(data) <-
      names(data) %>% str_replace_all("amountBillions", "amount")
  }
  data
}

.omb_chop <-
  function(data, base_col, start_row = 4) {
    df_base <-
      data %>%
      dplyr::slice(start_row:nrow(data))
    df_headers <-
      data %>% select(2:ncol(data)) %>% dplyr::slice(1:(start_row - 1))
    df_headers <- df_headers %>% t() %>%
      as_tibble() %>%
      .remove_na()
    names(df_headers) <- 1:ncol(df_headers) %>% str_c("V", .)
    other_cols <- df_headers %>%
      tidyr::fill(V1) %>%
      separate(V1, sep = "\\(", into = c("V1", "remove")) %>%
      select(-remove) %>%
      mutate_all(list(function(x) {
        x %>% str_remove_all("\\-|\\(|\\)|\\ ") %>% str_squish()
      })) %>%
      mutate_if(is.character,
                list(function(x) {
                  ifelse(is.na(x), "", x)
                })) %>%
      mutate(V1 = str_to_lower(V1)) %>%
      unite(type, V1, V2, sep = "") %>%
      pull()

    data <-
      df_base %>%
      setNames(c(base_col, other_cols)) %>%
      gather(item, amount, -c(base_col))

    data
  }

.add_table_no <- function(data, table_number = 1) {
  data %>%
    mutate(numberTable = table_number) %>%
    select(numberTable, everything())
}

.nest_omb <-
  function(data) {
    data %>%
      group_by(numberTable, nameTable, slugTable) %>%
      nest() %>%
      ungroup() %>%
      mutate(countRows = data %>% map_dbl(nrow),
             countColumns = data %>% map_dbl(ncol)) %>%
      select(-data, everything())
  }

.convert_amount <- function(data, amount_col = "amount") {
  data %>%
    mutate_at(amount_col,
              list(function(x) {
                x %>% parse_number() * 1000000 %>% as.integer() %>% currency(digits = 0)
              })) %>%
    filter(!is.na(!!!syms(amount_col)))
}

.unpivot_omb <-
  function(data,
           data_start_row = 5,
           column_names = c("metric", "parent", "item"),
           header_rows = 2:4,
           column_start = 2) {
    df_columns <-
      data %>% dplyr::slice(header_rows) %>% dplyr::select(column_start:ncol(data)) %>%
      t() %>%
      as_tibble() %>%
      mutate_all(as.character)

    names(df_columns) %>%
      walk(function(x) {
        df_columns <<- df_columns %>%
          fill(!!sym(x))
      })

    d <-
      data %>% dplyr::slice(data_start_row:nrow(data))

    years <-
      d %>% t() %>% as_tibble() %>% slice(1) %>% as.character()

    d <- d %>% t() %>% as_tibble()
    d <- d %>% slice(2:nrow(d)) %>%
      setNames(years)

    d <-
      df_columns %>%
      setNames(column_names) %>%
      bind_cols(d)

    d

  }

.fix_year <-
  function(data,
           year_column = "yearBudget",
           current_budget_year = 2020) {
    data <-
      data %>%
      mutate(
        UQ(year_column) := case_when(!!sym(year_column) == "TQ" ~ "1977",
                                     TRUE ~ !!sym(year_column)),
        isEstimate = !!sym(year_column) %>% str_detect("estimate"),
        UQ(year_column) := !!sym(year_column) %>% str_remove_all("estimate") %>% as.numeric()
      ) %>%
      select(isEstimate, one_of(year_column), everything())

    if (data %>% hasName("yearBudget")) {
      data <- data %>%
        mutate(
          dateFiscalYearEnd = case_when(
            yearBudget < 1977 ~ glue("{yearBudget}-12-31") %>% as.character() %>% ymd(),
            TRUE ~ glue("{yearBudget}-09-30") %>% as.character() %>% ymd()
          )
        ) %>%
        select(isEstimate,
               dateFiscalYearEnd,
               one_of(year_column),
               everything())
    }

    data <-
      data %>%
      mutate(isEstimate = yearBudget > current_budget_year)


    data
  }

.munge_metrics <-
  function(data) {
    data %>%
      mutate(
        metric = case_when(
          metric == "Addendum: Composite Deflator" ~ "ratioDeflator",
          metric == "In Constant (FY 2012) Dollars" ~ "amountBillions2012Dollars",
          metric %>% str_detect(
            "In Millions of Current Dollars|In Millions of Dollars|In Current Dollars"
          ) ~ "amountMillionsNominal",
          metric %in% c("In of Dollars") ~ "amountMillionsNominal",
          metric %>% str_detect("In Billions of Constant") ~ "amountBillions2012Dollars",
          metric %>% str_detect("Percentages of GDP") ~ "pctGDP",
          metric %>% str_detect(
            "Percentages of Total Outlays|As Percentages of Federal Outlays"
          ) ~ "pctBudget",
          metric %>% str_detect("Composite Outlay Deflators") ~ "ratio",
          metric %in% c("In Billions of Current Dollars", "In Billions of Dollars") ~ "amountBillionsNominal",
          metric == "GDP (Chained) Price Index" ~  "ratio",
          metric == "Addendum: Fiscal Year GDP" ~ "amountMillionsGDP",

          metric == "GDP (in billions of dollars)" ~ "amountBillions2012Dollars",
          TRUE ~ metric
        )
      )
  }

.munge_parents <-
  function(data) {
    if (!data %>% hasName("parent")) {
      return(data)
    }

    data <-
      data %>%
      mutate(
        parent = case_when(
          parent == "Addendum: Direct Capital" ~ "Direct Capital",
          parent %>% str_detect("Other nondefense") ~ "Non Defense",
          parent %>% str_detect("Natural resources and environment") ~ "Natural Resources and Environment",
          parent %>% str_detect("Community and regional development") ~ "Community Development",
          parent %>% str_detect("Direct Federal") ~ "Federal Direct",
          parent %>% str_detect("Atomic Energy and Other Defense") ~ "Atomic Energy",
          parent %>% str_detect("Department of Defense") ~ "Department of Defense",
          parent %>% str_detect("National defense") ~ "Defense",
          parent %>% str_detect("General science, space, and technology") ~ "Science, Space and Technology",
          TRUE ~ parent
        )
      )

    data
  }

.munge_omb_items <-
  function(data) {
    if (!data %>% hasName("item")) {
      return(data)
    }

    data <-
      data %>%
      mutate(item = item %>% str_squish()) %>%
      mutate(
        item = case_when(
          item %>% str_detect("Nondefense") ~ "Non Defense",
          item %>% str_detect("Urban renewal") ~ "Urban Renewal",
          item %>% str_detect("Pollution control facilities") ~ "Polluition Control Facilities",
          item %>% str_detect("Block grants") ~ "Block Grants",
          item %>% str_detect("Housing assistance") ~ "Housing
        Assistance",
          item %>% str_detect("National Defense|National defense") ~ "Defense",
          item %>% str_detect("Construction and Rehabilitation of Physical Assets") ~ "Asset Construction and Rehabilitation",
          item %>% str_detect(
            "Education, training, employment, and social services|Education, Training and Services"
          ) ~ "Education, Training and Services",
          item %>% str_detect("Public works acceleration/local public works") ~ "Public Works",
          item %>% str_detect("All other") ~ "Other",
          item %>% str_detect("DOT") ~ "DOT",
          item %>% str_detect("Other national defense") ~ "Other Defense",
          item %>% str_detect("Natural resources and environment") ~ "Natural Resources and Environment",
          item %>% str_detect("Atomic energy general science") ~ "Atomic Energy",
          TRUE ~ item
        )
      )

    data
  }

.parse_41 <-
  function(data, exclude_total = T) {
    table_name <-
      .add_table_name(data)

    df <-
      data %>% .unpivot_omb(data_start_row = 6, header_rows = 3:5)

    df <- df %>%
      mutate_if(is.character, str_squish)

    df <-
      df %>%
      mutate(
        item =
          case_when(
            metric == "GDP (in billions of dollars)" ~ "GDP",
            metric == "GDP (Chained) Price Index" ~ "ratioGDP",
            parent == "Total" ~ "Total",
            parent == "Total Defense" ~ "Defense",
            parent == "Total Nondefense" ~ "Non Defense",
            TRUE ~ item
          )
      ) %>%
      .munge_metrics() %>%
      mutate(
        parent = parent %>% str_remove_all("Total |\\Addendum: ") %>% str_replace_all(
          "Undis- tributed Offsetting Receipts",
          "Undistributed Recepits"
        )
      )

    df <-
      df %>%
      gather(yearBudget, amount, -c("metric", "parent", "item")) %>%
      .fix_year() %>%
      filter(!is.na(amount)) %>%
      .munge_omb_items() %>%
      .fix_omb_amount()


    df <-
      df %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    df <-
      df %>%
      filter(!is.na(yearBudget)) %>%
      .munge_parents()

    df
  }

.fix_omb_amount <-
  function(data) {
    has_cols <- data %>% hasName("metric") && data %>% hasName("amount")

    if (!has_cols) {
      return(data)
    }

    data <- data %>%
      mutate(
        amount = parse_number(amount),
        amount = case_when(
          metric %>% str_detect("^pct") ~ amount / 100,
          metric %>% str_detect("Millions") ~ amount * 1000000,
          metric %>% str_detect("Billions") ~ amount * 1000000000,
          TRUE ~ amount
        ),
        metric = metric %>% str_remove_all("Millions|Billions")
      )

    data
  }


.munge_for_column_parents <- function(df_items) {
  df_parents <- df_items %>%
    filter(item %>% str_detect(":")) %>%
    select(numberColumn, parent = item) %>%
    mutate(numberColumn = numberColumn + 1)

  df_items %>%
    left_join(df_parents, by = "numberColumn") %>%
    fill(parent) %>%
    mutate(
      parent = parent %>% str_remove_all("\\:"),
      item = item %>% str_remove_all("\\(|\\)|[0-9]")
    ) %>%
    mutate_if(is.character, str_squish) %>%
    filter(!item %>% str_detect(":"))

}

.munge_omb_wide <- function(data,
                            year_row = 3,
                            column_start = 2,
                            remove_column = NULL,
                            year_offset = 1,
                            gather_columns = c("parent", "item"),
                            metric = "amountMillionsNominal",
                            id_ratio_threshold = .015,
                            filter_total = T) {
  years <-
    data %>%
    dplyr::slice(year_row) %>%
    dplyr::select(column_start:ncol(data)) %>%
    as.character()

  if (years %>% is.na() %>% sum() / length(years) > .1) {
    years <-
      tibble(year = years) %>%
      fill(year) %>%
      pull(year)
  }

  d <-
    data %>% slice((year_row + year_offset):nrow(data))

  if (length(remove_column) > 0) {
    d <- d %>% select(-remove_column)
  }

  d <-
    d %>% t() %>% as_tibble() %>%
    mutate_all(as.character)

  items <-
    d %>% dplyr::slice(1) %>% t() %>% as.character()

  d <- d %>% dplyr::slice(2:nrow(d)) %>% .remove_na()

  df_items <- tibble(item = items) %>%
    mutate(numberColumn = 1:n()) %>%
    select(numberColumn, everything()) %>%
    mutate(idColumn = glue("V{numberColumn}") %>% as.character()) %>%
    select(idColumn, everything())

  id_ratio <-
    items %>% str_detect("[0-9][0-9][0-9]") %>% sum() / length(items)

  id_ratio <- case_when(is.na(id_ratio) ~ 0,
                        TRUE ~ id_ratio)

  if (id_ratio > id_ratio_threshold) {
    df_items <-
      df_items %>%
      mutate(
        isFunction =  item %>% str_detect("^[0-9][0-9][0-9]"),
        idFunction = case_when(isFunction ~ item %>% substr(1, 3),
                               TRUE ~ NA_character_),
        item = case_when(isFunction ~ item %>% substr(4, nchar(item)),
                         TRUE ~ item) %>% str_squish()
      ) %>%
      select(idColumn, numberColumn, idFunction, everything()) %>%
      select(-isFunction)
  }

  df_items <-
    df_items %>% .munge_for_column_parents()

  df_items <- df_items %>%
    mutate(
      parent = case_when(
        item == "Direct Payments for Individuals"  ~ "Direct Payments for Individuals",

        TRUE ~ parent
      )
    ) %>%
    fill(parent)

  d <-
    d %>% t() %>% data.frame() %>% tibble::rownames_to_column(var = "idColumn") %>% as_tibble() %>%
    mutate_all(as.character)

  bad_row <-
    d %>% slice(1) %>% select(2:ncol(d)) %>% as.character() %>%
    str_detect("[0-9]") %>% sum() == 0

  bad_row <- case_when(is.na(bad_row) ~ F,
                       TRUE ~ bad_row)

  if (bad_row) {
    d <- d %>% slice(2:nrow(d))
  }



  d <-
    d %>%
    left_join(df_items %>% select(-numberColumn), by = "idColumn") %>%
    select(-idColumn) %>%
    select(one_of(c("parent", "item", "idFunction")), everything())

  start_names <-
    names(d)[names(d) %in% c("parent", "item", "idFunction")]

  d <-
    d %>%
    setNames(c(start_names, years))

  d <-
    d %>%
    gather(yearBudget, amount , -c(start_names)) %>%
    filter(!is.na(amount)) %>%
    mutate(metric = metric) %>%
    select(metric, everything()) %>%
    .fix_year() %>%
    .munge_metrics() %>%
    .munge_parents() %>%
    .munge_omb_items() %>%
    filter(!is.na(amount))

  d <- d %>%
    mutate(
      metric = case_when(
        item %>% str_detect("As a percentage of total outlays") ~ "pctOutlays",
        item %>% str_detect("In billions of constant FY dollars") ~ "amountBillions2012Dollars",
        TRUE ~ metric
      )
    )

  d <-
    d %>%
    .fix_omb_amount() %>%
    filter(!is.na(amount))

  d <-
    d %>%
    mutate(item = item %>% str_remove_all("\\*"))

  if (filter_total) {
    d <-
      d %>%
      filter(!item %>% str_detect("Total|TOTAL|Subtotal"))
  }

  d


}

.calc_omb_summary <-
  function(data) {
    data <-
      data %>%
      filter(metric %>% str_detect("count|amount")) %>%
      .summarise_sum(amount_col = "amount",
                     columns =
                       data %>% select(-amount) %>% names()) %>%
      bind_rows(
        data %>%
          filter(metric %>% str_detect("pct|ratio")) %>%
          .summarise_mean(
            amount_col = "amount",
            columns =
              data %>% select(-amount) %>% names()
          )
      ) %>%
      mutate_if(is.character, str_squish) %>%
      .remove_na() %>%
      arrange(yearBudget)

    data
  }

#' OMB Budget Supplement URLs
#'
#' Includes links to the most recent supplemental
#' budget tables and links.
#'
#' @return \code{tibble}
#' @export
#'
#' @examples
#' dictionary_omb_supplements()
dictionary_omb_supplements <-
  memoise::memoise(function() {
    page <-
      read_html("https://www.whitehouse.gov/omb/supplemental-materials/")

    nodes <- page %>% html_nodes("p a")
    files <- nodes %>% html_text()

    urls <-
      nodes %>% html_attr("href")

    data <- tibble(nameFile = files, urlData = urls)

    data <-
      data %>%
      filter(urlData %>% str_detect(".pdf$|.xlsx$|.csv$")) %>%
      separate(
        nameFile,
        into = c("typeFile", "descriptionFile"),
        remove = F,
        extra = "merge",
        sep = "\\:",
        fill = "right"
      ) %>%
      mutate_if(is.character, str_squish) %>%
      mutate_at(c("nameFile", "typeFile", "descriptionFile"), str_to_upper)

    data
  })

# dicts -------------------------------------------------------------------


# https://www.whitehouse.gov/omb/historical-tables/

#' OMB Budget Tables
#'
#' List of OMB budget tables
#'
#' @param url url of file location
#'
#' @return
#' @export
#'
#' @examples
#' dictionary_omb_tables()
dictionary_omb_tables <-
  function(url = "https://www.whitehouse.gov/omb/historical-tables/") {
    page <- read_html(url)
    file_links <-
      page %>% html_nodes(".editor__module-eop+ p a") %>% html_attr("href") %>% .[[1]]

    zip <-
      page %>% html_nodes("h2+ p a") %>% html_attr("href")

    nodes <-
      page %>%
      html_nodes("p+ p a")

    files <-
      nodes %>%
      html_text() %>%
      str_squish() %>%
      str_replace("\\—", "\\|")

    urls <- nodes %>% html_attr("href")

    data <-
      tibble(files) %>%
      mutate(numberTable  = 1:n()) %>%
      select(numberTable, everything()) %>%
      separate(files, into = c("idTable", "file"), sep = "\\|") %>%
      mutate(idTable = idTable %>% str_remove_all("Table ")) %>%
      separate(file,
               into = c("nameFile", "yearsData"),
               sep = "\\:") %>%
      separate(yearsData,
               into = c("yearsData", "formatData"),
               sep = "\\(") %>%
      mutate_if(is.character, str_squish) %>%
      separate(
        yearsData,
        sep = "\\–",
        into = c("yearStart", "yearEnd"),
        remove = F
      ) %>%
      mutate(formatData = formatData %>% str_remove_all("\\)")) %>%
      mutate_at(c("yearStart", "yearEnd"), as.integer) %>%
      mutate(urlFile = urls,
             urlBudgetZIP = zip,
             urlBudgetPDF = file_links)

    data
  }

.add_table_name <- function(data) {
  data[1, 1] %>% as.character() %>% str_squish()
}

# parsers -----------------------------------------------------------------


.parse_32 <-
  function(data) {
    table_name <-
      .add_table_name(data)
    df_columns <-
      data %>% dplyr::slice(2:3) %>% select(2:ncol(data)) %>%
      t() %>%
      as_tibble() %>%
      fill(V1) %>%
      mutate(
        V1 = case_when(
          V1 %>% str_detect("In Millions of Dollars") ~ "amountMillions",
          V1 %>% str_detect("In Billions") ~ "amountBillions2012Dollars",
          V1 %>% str_detect("Percentages of GDP") ~ "pctGDP",
          V1 %>% str_detect("Percentages of Total Outlays") ~ "pctOutlays"
        )
      ) %>%
      mutate(V2 = V2 %>% str_remove_all("\n")) %>%
      unite(item, V1, V2, sep = "") %>%
      mutate(item = item %>% str_remove_all("\\ "))

    data <-
      data %>%
      dplyr::slice(4:nrow(data)) %>%
      setNames(c("yearBudget", df_columns$item)) %>%
      mutate(isEstimate = yearBudget %>% str_detect("estimate")) %>%
      select(isEstimate, everything()) %>%
      select(-matches("Total")) %>%
      mutate(
        nameTable = table_name,
        isEstimate = yearBudget %>% str_detect("estimate"),
        yearBudget = yearBudget %>% str_remove_all("estimate")
      ) %>%
      select(nameTable, everything()) %>%
      mutate(yearBudget = as.numeric(yearBudget)) %>%
      fill(yearBudget)

    amt_m <- data %>% select(matches("amountMillions")) %>%
      names()

    amt_b <-
      data %>% select(matches("amountBill")) %>%
      names()

    pct_cols <- data %>% select(matches("pct")) %>% names()

    data <-
      data %>%
      mutate_at(pct_cols, list(function(x) {
        x %>% parse_number() / 100
      })) %>%
      mutate_at(amt_m, list(function(x) {
        x %>% parse_number() * 1000000 %>% currency(digits = 0)
      })) %>%
      mutate_at(amt_b, list(function(x) {
        x %>% parse_number() * 1000000000 %>% currency(digits = 0)
      }))

    names(data) <-
      names(data) %>% str_remove_all("Millions|Billions")

    names(data) <-
      names(data) %>% str_replace_all("Nondefense", "NonDefense")

    data

    data <-
      data %>%
      select(nameTable:yearBudget, matches("pct")) %>%
      group_by(nameTable, isEstimate, yearBudget) %>%
      summarise_all(mean) %>%
      left_join(
        data %>%
          select(nameTable:yearBudget, matches("amount")) %>%
          group_by(nameTable, isEstimate, yearBudget) %>%
          summarise_all(sum),
        by = c("nameTable", "isEstimate", "yearBudget")
      ) %>%
      ungroup()

    data
  }

.parse_45 <-
  function(data, exclude_total = T) {
    table_name <-
      .add_table_name(data)

    df <-
      data %>% .unpivot_omb(
        data_start_row = 4,
        header_rows = 2:3,
        column_names = c("metric", "item")
      ) %>%
      .munge_metrics()

    df <- df %>%
      mutate_if(is.character, str_squish)

    df <-
      df %>%
      gather(yearBudget, amount, -c("metric", "item")) %>%
      .fix_year() %>%
      filter(!is.na(amount)) %>%
      .munge_omb_items() %>%
      .fix_omb_amount()

    if (exclude_total) {
      df <- df %>%
        filter(!item %>% str_detect("Total"))
    }


    df <-
      df %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    df <-
      df %>%
      filter(!is.na(yearBudget))

    df
  }


.parse_44 <-
  function(data) {
    table_name <-
      .add_table_name(data)

    df_years <-
      data %>% dplyr::slice(3) %>% select(2:ncol(data)) %>%
      gather(column, year) %>%
      fill(year) %>%
      mutate(idRow = 1:n()) %>%
      select(idRow, year)

    df_types <-
      data %>% dplyr::slice(4) %>% select(2:ncol(data)) %>%
      gather(column, type) %>%
      mutate(idRow = 1:n()) %>%
      select(idRow, type)

    df_cols <- df_years %>%
      left_join(df_types, by = "idRow") %>%
      mutate(type = str_to_lower(type)) %>%
      unite(type, type, year, sep = "_")

    data <- data %>%
      dplyr::slice(5:nrow(data))
    names(data)[[1]] <- "nameProgram"
    names(data)[2:ncol(data)] <- df_cols$type
    data <-
      data %>%
      gather(period, amount, -nameProgram) %>%
      separate(period,
               into = c("typeFunding", "yearBudget"),
               extra = "merge") %>%
      filter(typeFunding != "total") %>%
      filter(!nameProgram %>% str_detect("Total,"))

    data <- data %>% filter(!nameProgram %>% str_detect("\\*"))

    data <- data %>%
      mutate(idRow = 1:n())

    df_parents <-
      data %>%
      filter(is.na(amount) | nameProgram %>% str_detect(":")) %>%
      filter(is.na(amount)) %>%
      rename(parent = nameProgram) %>%
      mutate(idRow = 1 + idRow) %>%
      mutate(parent = parent %>% str_remove_all("\\:")) %>%
      select(idRow, parent)

    data <- data %>%
      left_join(df_parents, by = "idRow") %>%
      fill(parent) %>%
      select(parent, everything()) %>%
      select(-idRow) %>%
      filter(!is.na(parent)) %>%
      mutate(metric = "amountNominalMillions") %>%
      select(metric, everything()) %>%
      .fix_omb_amount() %>%
      .fix_year() %>%
      filter(!is.na(amount)) %>%
      mutate(
        nameProgram = nameProgram %>% str_remove_all("\\:|\\)|\\(") %>% str_remove_all("formerly Food stampsincluding Puerto Rico|WIC and CSFP|\\.") %>% str_replace_all("\\--|\\-|\\/", " ")
      )

    data <-
      data %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(amount = currency(amount, digits = 0)) %>%
      rename(item = nameProgram) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    data
  }

.parse_1 <-
  function(data) {
    table_name <-
      .add_table_name(data)

    df_types <-
      data %>% dplyr::slice(3) %>% select(2:ncol(data)) %>%
      gather(column, type) %>%
      fill(type) %>%
      mutate(idRow = 1:n()) %>%
      select(idRow, type) %>%
      mutate(
        type = case_when(
          type %>% str_detect("On-Budget") ~ "OnBudget",
          type %>% str_detect("Off-Budget") ~ "OffBudget" ,
          TRUE ~ "Total"
        )
      )

    df_outlays <-
      data %>% dplyr::slice(4) %>% select(2:ncol(data)) %>%
      gather(column, outlay) %>%
      mutate(idRow = 1:n()) %>%
      select(idRow, outlay)

    df_cols <-
      df_outlays %>%
      left_join(df_types, by = "idRow") %>%
      unite(type, type, outlay, sep = "_")

    data <-
      data %>% dplyr::slice(5:nrow(data)) %>%
      set_names(c("yearBudget", df_cols$type)) %>%
      gather(type, amount, -yearBudget) %>%
      separate(type,
               into = c("typeBudget", "typeCashFlow"),
               extra = "merge") %>%
      filter(typeBudget != "Total",
             !typeCashFlow %>% str_detect("Surplus")) %>%
      mutate(yearBudget = case_when(yearBudget == "TQ" ~ "1977",
                                    TRUE ~ yearBudget)) %>%
      .convert_amount() %>%
      group_by(yearBudget, typeBudget, typeCashFlow) %>%
      summarise_all(sum) %>%
      ungroup() %>%
      mutate(
        nameTable = table_name,
        isEstimate = yearBudget %>% str_detect("estimate"),
        yearBudget = yearBudget %>% str_remove_all("estimate")
      ) %>%
      mutate_if(is.character, str_squish) %>%
      arrange(yearBudget) %>%
      select(nameTable, everything())

    data
  }

.parse_2 <-
  function(data) {
    table_name <-
      .add_table_name(data)

    df_types <-
      data %>% dplyr::slice(2) %>% select(2:ncol(data)) %>%
      gather(column, type) %>%
      fill(type) %>%
      mutate(idRow = 1:n()) %>%
      select(idRow, type) %>%
      mutate(
        type = case_when(
          type %>% str_detect("On-Budget") ~ "OnBudget",
          type %>% str_detect("Off-Budget") ~ "OffBudget" ,
          type %>% str_detect("GDP") ~ "GDP",
          TRUE ~ "Total"
        )
      )

    df_outlays <-
      data %>% dplyr::slice(3) %>% select(2:ncol(data)) %>%
      gather(column, outlay) %>%
      mutate(idRow = 1:n(),
             outlay = coalesce(outlay, "")) %>%
      select(idRow, outlay)

    df_cols <-
      df_outlays %>%
      left_join(df_types, by = "idRow") %>%
      unite(type, type, outlay, sep = "_")

    data <-
      data %>% dplyr::slice(4:nrow(data)) %>%
      set_names(c("yearBudget", df_cols$type)) %>%
      gather(type, amount, -yearBudget) %>%
      separate(type,
               into = c("typeBudget", "typeCashFlow"),
               extra = "merge") %>%
      filter(typeBudget != "Total",
             !typeCashFlow %>% str_detect("Surplus")) %>%
      mutate(yearBudget = case_when(yearBudget == "TQ" ~ "1977",
                                    TRUE ~ yearBudget)) %>%
      mutate(amount = amount %>% readr::parse_number()) %>%
      filter(!is.na(amount))

    data <-
      data %>%
      .fix_year()



    df_gdp <-
      data %>%
      filter(typeBudget == "GDP") %>%
      select(-typeCashFlow) %>%
      mutate(amount = as.character(amount)) %>%
      .convert_amount() %>%
      rename(item = typeBudget) %>%
      mutate(metric = "amountNominal",
             amount = as.numeric(amount)) %>%
      .calc_omb_summary() %>%
      mutate(amount = amount * 1000)



    df_non_gdp <-
      data %>%
      filter(typeBudget != "GDP") %>%
      rename(item = typeCashFlow) %>%
      rename(parent = typeBudget) %>%
      mutate(amount = amount / 100) %>%
      mutate(metric = "pctGDP") %>%
      .calc_omb_summary() %>%
      select(isEstimate:yearBudget, metric, everything())

    data <-
      df_gdp %>%
      bind_rows(df_non_gdp) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, yearBudget, everything())

    data <- data %>%
      select(
        nameTable,
        isEstimate,
        yearBudget,
        dateFiscalYearEnd,
        metric,
        parent,
        item,
        everything()
      )

    data
  }

.parse_3 <-
  function(data) {
    table_name <-
      .add_table_name(data)

    df_types <-
      data %>% dplyr::slice(3) %>% select(2:ncol(data)) %>%
      gather(column, type) %>%
      fill(type) %>%
      mutate(idRow = 1:n()) %>%
      select(idRow, type) %>%
      mutate(
        type = case_when(
          type %>% str_detect("Current Dollars") ~ "amountNominal",
          type %>% str_detect("FY 2012") ~ "amount2012" ,
          type %>% str_detect("Percentages of GDP") ~ "pctGDP",
          type %>% str_detect("Addendum") ~ "ratioDeflator"
        )
      )

    df_outlays <-
      data %>% dplyr::slice(4) %>% select(2:ncol(data)) %>%
      gather(column, outlay) %>%
      mutate(idRow = 1:n(),
             outlay = coalesce(outlay, "")) %>%
      select(idRow, outlay)

    df_cols <-
      df_outlays %>%
      left_join(df_types, by = "idRow") %>%
      unite(type, type, outlay, sep = "_")

    data <-
      data %>%
      dplyr::slice(5:nrow(data)) %>%
      set_names(c("yearBudget", df_cols$type)) %>%
      gather(type, amount, -yearBudget) %>%
      separate(
        type,
        into = c("typeBudget", "typeCashFlow"),
        sep = "_",
        extra = "merge"
      ) %>%
      filter(typeBudget != "Total",
             !typeCashFlow %>% str_detect("Surplus")) %>%
      mutate(yearBudget = case_when(yearBudget == "TQ" ~ "1977",
                                    TRUE ~ yearBudget)) %>%
      mutate(amount = amount %>% readr::parse_number()) %>%
      filter(!is.na(amount)) %>%
      mutate(
        isEstimate = yearBudget %>% str_detect("estimate"),
        yearBudget = yearBudget %>% str_remove_all("estimate")
      )


    df_gdp <-
      data %>%
      filter(typeBudget == "pctGDP") %>%
      unite(metric, typeBudget, typeCashFlow, sep = "") %>%
      .summarise_pct() %>%
      mutate(amount = amount / 100)

    df_deflator <-
      data %>%
      .unite_filter_metric(budget_filter = "ratioDeflator") %>%
      .summarise_pct()

    df_amts <-
      data %>%
      .unite_filter_metric(budget_filter = c("amount2012", "amountNominal")) %>%
      mutate(amount = as.character(amount)) %>%
      .convert_amount() %>%
      .summarise_sum() %>%
      mutate_if(is.numeric, as.numeric)

    data <-
      list(df_amts, df_gdp, df_deflator) %>%
      map_dfr(bind_rows) %>%
      mutate(yearBudget = as.numeric(yearBudget),
             nameTable = table_name) %>%
      select(nameTable, yearBudget, everything())

    data
  }

.parse_4 <-
  function(data,
           column_names = c("parent", "item"),
           metric = "amountNominalMillions") {
    table_name <-
      .add_table_name(data)

    df <-
      .unpivot_omb(
        data = data,
        data_start_row = 5,
        header_rows = 3:4,
        column_names = column_names
      ) %>%
      mutate(metric = metric) %>%
      gather(yearBudget, amount, -c("metric", column_names)) %>%
      .munge_metrics() %>%
      mutate_all(str_squish) %>%
      .fix_year() %>%
      filter(!is.na(amount)) %>%
      .munge_metrics() %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount)) %>%
      .calc_omb_summary() %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    df <- df %>%
      mutate_at(c("parent", "item"),
                list(function(x) {
                  x %>% str_remove_all("\\(") %>% str_remove_all("\\)") %>%
                    str_replace_all("\\ - ", " ") %>% str_remove_all("[0-9]") %>%
                    str_squish()
                }))

    df <-
      df %>%
      filter(item !=  "Total" | is.na(item)) %>%
      mutate(amount = currency(amount, digits = 0))

    df
  }


.parse_5 <-
  function(data,
           column_names = c("parent", "item"),
           metric = "amountNominalMillions") {
    table_name <-
      .add_table_name(data)

    df <-
      .unpivot_omb(
        data = data,
        data_start_row = 5,
        header_rows = 3:4,
        column_names = column_names
      ) %>%
      mutate(metric = metric) %>%
      gather(yearBudget, amount, -c("metric", column_names)) %>%
      .munge_metrics() %>%
      mutate_all(str_squish) %>%
      .fix_year() %>%
      filter(!is.na(amount)) %>%
      .munge_metrics() %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount)) %>%
      .calc_omb_summary() %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    df <- df %>%
      mutate_at(c("parent", "item"),
                list(function(x) {
                  x %>% str_remove_all("\\(") %>% str_remove_all("\\)") %>%
                    str_replace_all("\\ - ", " ") %>% str_remove_all("[0-9]") %>%
                    str_squish()
                }))

    df <-
      df %>%
      filter(item !=  "Total" | is.na(item)) %>%
      filter(parent != "Total Receipts") %>%
      mutate(amount = currency(amount, digits = 0))

    df
  }


.parse_7 <-
  function(data,
           column_names = c("parent", "item"),
           metric = "pctGDP") {
    table_name <-
      .add_table_name(data)

    df <-
      .unpivot_omb(
        data = data,
        data_start_row = 4,
        header_rows = 2:3,
        column_names = column_names
      ) %>%
      mutate(metric = metric) %>%
      gather(yearBudget, amount, -c("metric", column_names)) %>%
      .munge_metrics() %>%
      mutate_all(str_squish) %>%
      .fix_year() %>%
      filter(!is.na(amount)) %>%
      .munge_metrics() %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount)) %>%
      .calc_omb_summary() %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    df <- df %>%
      mutate_at(c("parent", "item"),
                list(function(x) {
                  x %>% str_remove_all("\\(") %>% str_remove_all("\\)") %>%
                    str_replace_all("\\ - ", " ")
                }))

    df <- df %>%
      filter(item !=  "Total" | is.na(item)) %>%
      filter(parent != "Total Receipts") %>%
      rename(pct = amount)

    df
  }

.parse_8 <-
  function(data) {
    table_name <-
      .add_table_name(data)

    df <-
      data %>%
      .munge_omb_wide(year_row = 3, gather_columns = "item") %>%
      .remove_na() %>%
      filter(!is.na(amount)) %>%
      .munge_metrics() %>%
      .calc_omb_summary() %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      filter(!is.na(amount)) %>%
      mutate_at(c("parent", "item"),
                list(function(x) {
                  x %>% str_replace_all("\\-", " ") %>% str_squish()
                }))

    df <- df %>%
      mutate(amount = currency(amount, digits = 0))


    df
  }


.parse_9 <-
  function(data,
           metric = "amountNominalMillions",
           exclude_total = T,
           column_names = c("parent", "item")) {
    table_name <-
      .add_table_name(data)

    df <-
      .unpivot_omb(
        data = data,
        data_start_row = 5,
        column_names = column_names,
        header_rows = 3:4
      ) %>%
      gather(yearBudget, amount, -c(column_names)) %>%
      mutate(metric = metric) %>%
      .munge_metrics() %>%
      mutate_all(str_squish) %>%
      .fix_year() %>%
      filter(!is.na(amount)) %>%
      .munge_metrics() %>%
      .fix_omb_amount() %>%
      .calc_omb_summary() %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      filter(!is.na(amount))

    df <-
      df %>%
      mutate(parent = parent %>%  str_remove_all('"')) %>%
      filter(item != "Total" | is.na(item)) %>%
      filter(parent != "Total Other Receipts")

    df <-
      df %>%
      .calc_omb_summary() %>%
      mutate(amount = currency(amount, digits = 0)) %>%
      mutate_at(column_names,
                list(function(x) {
                  x %>% str_remove_all("[0-9]") %>%
                    str_remove_all("\\(") %>%
                    str_remove_all("\\)") %>%
                    str_remove_all("\\Memorandum: ") %>%
                    str_squish()
                }))


    df
  }


.parse_10 <-
  function(data,
           metric = "amountNominalMillions",
           exclude_total = T) {
    table_name <-
      .add_table_name(data)

    df <-
      .munge_omb_wide(
        data = data,
        year_row = 2,
        gather_columns = "item",
        year_offset = 0,
        metric = metric
      ) %>%
      filter(
        !item %in% c(
          "Department or other unit",
          "Function and Subfunction",
          "Superfunction and Function"
        )
      ) %>%
      .remove_na() %>%
      mutate_if(is.character, str_squish)

    df <-
      df %>%
      mutate(
        metric = case_when(
          parent == "As percentages of outlays" ~ "pctOutlays",
          parent == "As percentages of GDP" ~ "pctGDP",
          TRUE ~ metric
        )
      )

    df <-
      df %>%
      select(-parent)

    df <-
      df %>%
      mutate(amount = case_when(metric %>% str_detect("pct") ~ amount / 100000000,
                                TRUE ~ amount))

    df <-
      df %>%
      .calc_omb_summary() %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    if (exclude_total) {
      df <-
        df %>%
        filter(!item %in% c("On-budget", "Off-budget"))
    }

    df
  }


.parse_11 <-
  function(data,
           metric = "amountNominalMillions",
           exclude_total = T) {
    table_name <-
      .add_table_name(data)

    df <-
      .munge_omb_wide(
        data = data,
        year_row = 3,
        gather_columns = "item",
        year_offset = 0,
        metric = metric
      ) %>%
      filter(!item %in% c("Department or other unit", "Function and Subfunction")) %>%
      .remove_na() %>%
      mutate_if(is.character, str_squish)

    df <-
      df %>%
      .calc_omb_summary() %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      mutate(amount = currency(amount, digits = 0))

    if (exclude_total) {
      df <- df %>%
        filter(!item %in% c("On-budget", "Off-budget"))
    }

    df
  }


.parse_12 <-
  function(data) {
    table_name <-
      .add_table_name(data)


    df <-
      .munge_omb_wide(data = data,
                      year_row = 3,
                      gather_columns = "item") %>%
      .remove_na() %>%
      mutate_if(is.character, str_squish)

    df <-
      df %>%
      .calc_omb_summary() %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      mutate(amount = currency(amount, digits = 0))

    df
  }


.parse_13 <-
  function(data, metric = "pctOutlay") {
    table_name <-
      .add_table_name(data)

    df <-
      .munge_omb_wide(
        data = data,
        year_row = 2,
        gather_columns = "item",
        year_offset = 0,
        metric = metric
      ) %>%
      filter(item != "Department or other unit") %>%
      .remove_na() %>%
      mutate_if(is.character, str_squish)

    df <-
      df %>%
      .calc_omb_summary() %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      rename(pct = amount)

    df
  }


.parse_18 <-
  function(data) {
    table_name <-
      .add_table_name(data)
    col_names <- data %>% dplyr::slice(2) %>% as.character()
    col_names[[1]] <- "nameDepartment"
    data <-
      data %>% dplyr::slice(3:nrow(data)) %>%
      setNames(col_names) %>%
      gather(yearBudget, amount, -c(nameDepartment)) %>%
      filter(!is.na(amount)) %>%
      .convert_amount() %>%
      filter(!is.na(yearBudget)) %>%
      filter(!nameDepartment %>% str_detect("Total")) %>%
      select(yearBudget, everything())

    return(data)
  }

.parse_20 <-
  function(data, base_row = 2) {
    table_name <-
      .add_table_name(data)

    data <-
      data %>%
      .munge_general(
        base_row = base_row,
        table_name = table_name,
        is_wide = T,
        year_plus_long  = 0
      )

    data <- data %>%
      mutate(
        measurement = case_when(
          is.na(parent) ~ "amountNominal",
          parent == "As percentages of GDP" ~ "pctGDP",
          parent == "As percentages of outlays" ~ "pctOutlays",
          parent == "In billions of constant FY 2012 dollars" ~ "amount2012",
          TRUE ~ NA_character_
        )
      ) %>%
      fill(measurement) %>%
      select(nameTable:yearBudget, measurement, metric, amount) %>%
      mutate(
        amount = case_when(
          measurement  %in% c("pctGDP", "pctOutlays") ~ amount / 100,
          measurement == "amountNominal"  ~  amount * 1000000,
          measurement == "amount2012" ~ amount * 1000000000,
          TRUE ~ amount
        )
      )

    data <-
      data %>%
      .fix_year() %>%
      mutate(metric = metric %>% str_remove_all("[0-9]$") %>% str_squish()) %>%
      select(nameTable, everything())


    data <-
      data %>%
      rename(item = metric,
             metric = measurement) %>%
      .calc_omb_summary()

    data
  }

.parse_21 <-
  function(data,
           column_names = c("metric", "parent", "item"),
           exclude_total = T) {
    table_name <-
      .add_table_name(data)
    df <-
      data %>%
      .unpivot_omb(
        data_start_row = 5,
        header_rows = 2:4,
        column_names = column_names
      ) %>%
      mutate_if(is.character, str_squish) %>%
      select(metric, everything()) %>%
      .remove_na()

    df <-
      df %>%
      gather(yearBudget, amount, -c("metric", column_names)) %>%
      .fix_year() %>%
      filter(!is.na(amount)) %>%
      mutate_all(str_squish) %>%
      .munge_metrics() %>%
      .fix_omb_amount() %>%
      .calc_omb_summary() %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())



    if (exclude_total) {
      df <-
        df %>%
        filter(item != "Total" | is.na(item))
    }

    df
  }

.parse_22 <-
  function(data) {
    table_name <-
      .add_table_name(data)

    d <-
      data %>%
      slice(4:nrow(data))

    df <-
      d %>% select(1, 3, 5, 7) %>%
      gather(col, year) %>%
      select(year)

    df <- df %>%
      bind_cols(
        d %>% select(2, 4, 6, 8) %>% gather(column, amount) %>%
          select(amountDebtSubjectToLimit = amount) %>%
          mutate(amountDebtSubjectToLimit = amountDebtSubjectToLimit %>% parse_number() * 1000000)
      ) %>%
      rename(yearBudget = year) %>%
      mutate(yearBudget = case_when(yearBudget == "TQ" ~ "1977",
                                    TRUE ~ yearBudget)) %>%
      mutate(
        nameTable = table_name,
        isEstimate = yearBudget %>% str_detect("estimate"),
        yearBudget = yearBudget %>% str_remove_all("estimate")
      ) %>%
      mutate(yearBudget = as.numeric(yearBudget)) %>%
      select(nameTable, isEstimate, yearBudget, everything()) %>%
      filter(!is.na(amountDebtSubjectToLimit))

    df <-
      df %>%
      .summarise_mean(
        columns = c("nameTable", "isEstimate", "yearBudget"),
        amount_col = "amountDebtSubjectToLimit"
      ) %>%
      mutate(amountDebtSubjectToLimit = currency(amountDebtSubjectToLimit, digits = 0))

  }

.parse_23 <-
  function(data) {
    table_name <-
      .add_table_name(data)

    d <-
      data %>%
      slice(4:nrow(data)) %>%
      setNames(c(
        "detailsStatute",
        "dateStatute",
        "descriptionStatute",
        "amountDebtCeiling"
      )) %>%
      mutate(dateStatute = mdy(dateStatute)) %>%
      fill(detailsStatute, dateStatute) %>%
      mutate(amountDebtCeiling =
               amountDebtCeiling %>% parse_number() * 1000000000 %>% currency(digits = 0)) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    d
  }

.parse_24 <-
  function(data,
           column_names = c("type", "roleBudget", "parent", "item"),
           exclude_total = T,
           metric = "amountBillionsNominal") {
    table_name <-
      .add_table_name(data)
    df <-
      data %>%
      .unpivot_omb(
        data_start_row = 7,
        header_rows = 3:6,
        column_names = column_names
      ) %>%
      mutate_if(is.character, str_squish) %>%
      mutate(metric = metric) %>%
      select(metric, everything()) %>%
      .remove_na()

    df <-
      df %>%
      gather(yearBudget, amount, -c("metric", column_names)) %>%
      .fix_year() %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount)) %>%
      .calc_omb_summary() %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    df <-
      df %>%
      filter(!is.na(roleBudget)) %>%
      mutate(
        roleBudget = case_when(
          roleBudget == "Non- defense" ~ "Non Defense",
          roleBudget == "National Defense" ~ "Defense",
          TRUE ~ roleBudget
        )
      ) %>%
      mutate(amount = amount %>% currency(digits = 0))

    df <- df %>%
      mutate_at(column_names,
                list(function(x) {
                  x %>% str_remove_all("[0-9]") %>%
                    str_remove_all("\\(") %>%
                    str_remove_all("\\)") %>%
                    str_squish()
                }))

    if (exclude_total) {
      df <-
        df %>%
        filter(roleBudget != "Total")
    }

    df
  }

.parse_25 <-
  function(data,
           column_names = c("roleBudget", "parent", "item"),
           exclude_total = T,
           metric = "amountBillionsNominal") {
    table_name <-
      .add_table_name(data)
    df <-
      data %>%
      .unpivot_omb(
        data_start_row = 7,
        header_rows = 3:5,
        column_names = column_names
      ) %>%
      mutate_if(is.character, str_squish) %>%
      mutate(metric = metric) %>%
      select(metric, everything()) %>%
      .remove_na()

    df <-
      df %>%
      gather(yearBudget, amount, -c("metric", column_names)) %>%
      .fix_year() %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount)) %>%
      .calc_omb_summary() %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    df <-
      df %>%
      filter(!is.na(parent)) %>%
      mutate(
        parent = case_when(
          parent == "Non- defense" ~ "Non Defense",
          parent == "National Defense" ~ "Defense",
          TRUE ~ parent
        )
      )

    df <-
      df %>%
      mutate(item = case_when(
        item %>% str_detect("Undistributed Offsetting Receipts") ~ "Undistributed Offsetting Receipts",
        TRUE ~ item
      )) %>%
      mutate(amount = amount %>% currency(digits = 0))

    if (exclude_total) {
      df <- df %>%
        filter(parent != "Total")
    }

    df
  }

.parse_26 <-
  function(data,
           column_names = c("type", "roleBudget", "parent", "item"),
           exclude_total = F,
           metric = "pctOutlay") {
    table_name <-
      .add_table_name(data)
    df <-
      data %>%
      .unpivot_omb(
        data_start_row = 6,
        header_rows = 2:5,
        column_names = column_names
      ) %>%
      mutate_if(is.character, str_squish) %>%
      mutate(metric = metric) %>%
      select(metric, everything())

    df <-
      df %>%
      gather(yearBudget, amount, -c("metric", column_names)) %>%
      .fix_year() %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount)) %>%
      .calc_omb_summary() %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    df <- df %>%
      filter(!is.na(roleBudget)) %>%
      mutate(
        roleBudget = case_when(
          roleBudget == "Non- defense" ~ "Non Defense",
          roleBudget == "National Defense" ~ "Defense",
          TRUE ~ roleBudget
        )
      ) %>%
      rename(pct = amount)

    df <-
      df %>%
      mutate(item = item %>% gsub("\\(1)", "", .) %>% str_squish())

    if (exclude_total) {
      df <- df %>%
        filter(roleBudget != "Total")
    }

    df
  }
.parse_28 <-
  function(data, metric = "amount") {
    table_name <-
      .add_table_name(data)


    df <-
      data %>%
      .munge_omb_wide(year_row = 3,
                      column_start = 2,
                      year_offset = 1) %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      mutate(amount = amount %>% currency(digits = 0))



    df <- df %>%
      mutate(item = item %>% str_remove_all("\\)|\\(|\\,|\\[0-9]") %>% str_replace_all("\\-", " "))

    df
  }

.parse_33 <-
  function(data,
           data_start_row = 5,
           column_names = c("metric", "parent", "item"),
           header_rows = 2:4,
           column_start = 2) {
    table_name <- .add_table_name(data)
    df <- .unpivot_omb(
      data = data,
      data_start_row = data_start_row,
      column_names = column_names,
      header_rows = header_rows,
      column_start = column_start
    )

    data <-
      df %>%
      gather(yearBudget, amount, -column_names)

    data <-
      data %>%
      .fix_year()

    data <-
      data %>%
      .munge_metrics() %>%
      .munge_parents() %>%
      .munge_omb_items()

    data <-
      data %>%
      .fix_omb_amount() %>%
      mutate(amount = currency(amount, digits = 0))

    data <-
      data %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      mutate(amount = amount %>% currency(digits = 0))

    data
  }

.parse_34 <-
  function(data,
           data_start_row = 5,
           column_names = c("metric", "parent", "item"),
           header_rows = 2:4,
           column_start = 2) {
    table_name <-
      .add_table_name(data)
    df <-
      .unpivot_omb(
        data = data,
        data_start_row = data_start_row,
        column_names = column_names,
        header_rows = header_rows,
        column_start = column_start
      )

    data <-
      df %>%
      gather(yearBudget, amount, -column_names)

    data <-
      data %>%
      .fix_year()

    data <-
      data %>%
      .munge_metrics() %>%
      .munge_parents() %>%
      .munge_omb_items()

    data <-
      data %>%
      .fix_omb_amount()

    data <-
      data %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      rename(pct = amount)

    data
  }

.parse_35 <-
  function(data,
           data_start_row = 5,
           column_names = c("parent", "item"),
           header_rows = 3:4,
           column_start = 2) {
    table_name <- .add_table_name(data)
    df <-
      .unpivot_omb(
        data = data,
        data_start_row = data_start_row,
        column_names = column_names,
        header_rows = header_rows,
        column_start = column_start
      )

    data <-
      df %>%
      gather(yearBudget, amount, -column_names) %>%
      filter(!is.na(amount)) %>%
      mutate(metric = "amountMillionsNominal") %>%
      select(metric, everything())

    data <-
      data %>%
      .fix_year()

    data <-
      data %>%
      .munge_metrics() %>%
      .munge_parents() %>%
      .munge_omb_items()

    data <-
      data %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount))

    data <-
      data %>%
      .summarise_sum(amount_col = "amount",
                     columns =
                       data %>% select(-amount) %>% names()) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    data
  }

.parse_36 <-
  function(data,
           data_start_row = 7,
           column_names = c("parent", "item", "groupAgency"),
           header_rows = 4:6,
           column_start = 2) {
    table_name <- .add_table_name(data)
    df <-
      .unpivot_omb(
        data = data,
        data_start_row = data_start_row,
        column_names = column_names,
        header_rows = header_rows,
        column_start = column_start
      )

    data <-
      df %>%
      gather(yearBudget, amount, -column_names) %>%
      filter(!is.na(amount)) %>%
      mutate(metric = "amountMillionsNominal") %>%
      select(metric, everything())

    data <-
      data %>%
      .fix_year()

    data <-
      data %>%
      .munge_metrics() %>%
      .munge_parents() %>%
      .munge_omb_items() %>%
      filter(!is.na(groupAgency))

    data <-
      data %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount))

    data <-
      data %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      mutate(amount = currency(amount, digits = 0))

    data
  }

.parse_37 <-
  function(data,
           year_row = 3,
           column_start = 2,
           gather_columns = c("parent", "item"),
           metric = "amountMillionsNominal",
           filter_total = T) {
    table_name <- .add_table_name(data)

    data <-
      .munge_omb_wide(
        data = data,
        year_row = year_row,
        gather_columns = gather_columns,
        metric = metric,
        filter_total = filter_total
      )
    data <-
      data %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      mutate(amount = currency(amount, digits = 0))

    data
  }

.parse_39 <-
  function(data,
           year_row = 3,
           column_start = 2,
           gather_columns = c("parent", "item"),
           metric = "amountMillionsNominal",
           filter_total = T) {
    table_name <- .add_table_name(data)

    data <-
      .munge_omb_wide(
        data = data,
        year_row = year_row,
        gather_columns = gather_columns,
        metric = metric,
        filter_total = filter_total
      )

    data <-
      data %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      rename(groupAgency = item) %>%
      .remove_na() %>%
      mutate(amount = currency(amount, digits = 0))

    data
  }

.parse_38 <-
  function(data,
           exclude_total = T,
           column_names = c("metric", "item")) {
    table_name <-
      .add_table_name(data)
    df <-
      .unpivot_omb(
        data = data,
        column_start = 2,
        data_start_row = 4,
        column_names = column_names,
        header_rows = 2:3
      )

    data <-
      df %>%
      gather(yearBudget, amount, -column_names)

    data <-
      data %>%
      .fix_year() %>%
      filter(!is.na(metric))

    data <-
      data %>%
      .munge_metrics() %>%
      .munge_parents() %>%
      .munge_omb_items() %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount))

    if (exclude_total) {
      data <-
        data %>% filter(!item %>% str_detect("Total"))
    }


    data <-
      data %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    data
  }



.parse_40 <- function(data, exclude_total = T) {
  table_name <-
    .add_table_name(data)

  df <-
    .munge_omb_wide(data = data)

  if (exclude_total) {
    df <-
      df %>% filter(!item %>% str_detect("Total"))
  }


  df <-
    df %>%
    .calc_omb_summary() %>%
    arrange(yearBudget) %>%
    mutate(nameTable = table_name) %>%
    select(nameTable, everything()) %>%
    .remove_na()

  df
}

.parse_42 <-
  function(data,
           exclude_total = T,
           column_names =  c("metric", "item")) {
    table_name <-
      .add_table_name(data)

    df <-
      .unpivot_omb(
        data = data,
        column_start = 2,
        data_start_row = 4,
        column_names = column_names,
        header_rows = 2:3
      )

    data <-
      df %>%
      gather(yearBudget, amount, -column_names)

    data <-
      data %>%
      .fix_year() %>%
      filter(!is.na(metric))

    data <-
      data %>%
      .munge_metrics() %>%
      .munge_parents() %>%
      .munge_omb_items() %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount))

    if (exclude_total) {
      data <-
        data %>% filter(!item %>% str_detect("Total"))
    }


    data <-
      data %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      .remove_na()

    data
  }

.parse_43 <- function(data, exclude_total = T) {
  table_name <-
    .add_table_name(data)

  df <-
    .munge_omb_wide(data = data)

  df <-
    df %>%
    filter(!is.na(metric))

  df <-
    df %>%
    .munge_metrics() %>%
    .munge_parents() %>%
    .munge_omb_items() %>%
    filter(!is.na(amount))

  df <-
    df %>%
    .calc_omb_summary() %>%
    arrange(yearBudget) %>%
    mutate(nameTable = table_name) %>%
    select(nameTable, everything()) %>%
    arrange(yearBudget) %>%
    mutate(amount = currency(amount, digits = 0))

  df <-
    df %>%
    filter(!is.na(yearBudget)) %>%
    filter(parent != "Total Payments for Individuals")

  df
}

.parse_46 <-
  function(data,
           exclude_total = T,
           year_offset = 1) {
    table_name <-
      .add_table_name(data)

    df <-
      data %>%
      .munge_omb_wide(year_offset = year_offset)

    df <-
      df %>%
      mutate_if(is.character, str_squish) %>%
      filter(!item %in% c("Federal funds", "Trust funds")) %>%
      .remove_na()

    df <-
      df %>%
      filter(!is.na(amount)) %>%
      .munge_omb_items()

    if (exclude_total) {
      df <-
        df %>%
        filter(!item %>% str_detect("Total"))
    }


    df <-
      df %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      mutate(amount = currency(amount, digits = 0))

    df <-
      df %>%
      filter(!is.na(yearBudget))

    df
  }
.parse_47 <-
  function(data,
           exclude_total = T,
           year_offset = 1,
           column_start = 3,
           remove_column = 2) {
    table_name <-
      .add_table_name(data)

    df <-
      data %>%
      .munge_omb_wide(
        year_offset = year_offset,
        column_start = column_start,
        filter_total = exclude_total,
        remove_column = remove_column
      )

    df <-
      df %>%
      mutate_if(is.character, str_squish) %>%
      filter(!item %in% c("Federal funds", "Trust funds")) %>%
      .remove_na()

    df <-
      df %>%
      filter(!is.na(amount)) %>%
      .munge_omb_items()

    if (exclude_total) {
      df <-
        df %>%
        filter(!item %>% str_detect("Total"))
    }


    df <-
      df %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    df <-
      df %>%
      filter(!is.na(yearBudget)) %>%
      mutate(amount = currency(amount, digits = 0))

    df
  }

.parse_48 <-
  function(data,
           exclude_total = T,
           year_offset = 1,
           column_start = 3,
           remove_column = 2) {
    table_name <-
      .add_table_name(data)

    df <-
      data %>%
      .munge_omb_wide(
        year_offset = year_offset,
        column_start = column_start,
        filter_total = exclude_total,
        remove_column = remove_column
      )

    df <-
      df %>%
      mutate_if(is.character, str_squish) %>%
      .remove_na()

    df <-
      df %>%
      filter(!is.na(amount)) %>%
      .munge_omb_items()

    if (exclude_total) {
      df <-
        df %>%
        filter(!item %>% str_detect("Total"))
    }

    df <-
      df %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      mutate(amount = currency(amount, digits = 0))

    df <-
      df %>%
      filter(!is.na(yearBudget))

    df
  }

.parse_49 <-
  function(data,
           exclude_total = T,
           column_names = c("metric", "parent", "item")) {
    table_name <-
      .add_table_name(data)

    df <-
      .unpivot_omb(
        data = data,
        column_start = 2,
        data_start_row = 6,
        column_names = column_names,
        header_rows = 3:5
      )

    df <-
      df %>%
      gather(yearBudget, amount, -column_names) %>%
      mutate_if(is.character, str_squish)

    data <-
      df %>%
      .fix_year() %>%
      filter(!is.na(item))

    data <-
      data %>%
      .munge_metrics() %>%
      .munge_parents() %>%
      .munge_omb_items() %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount))

    if (exclude_total) {
      data <-
        data %>% filter(!item %>% str_detect("Total"))
    }

    data <- data %>%
      mutate(
        parent = case_when(
          parent == "State and Local Government Non-Interest Receipts (NIPA Basis) (1)" ~ "State and Local Government Non-Interest Receipts",
          TRUE ~ parent
        )
      )


    data <-
      data %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    data
  }

.parse_50 <-
  function(data,
           exclude_total = T,
           column_names = c("parent", "item")) {
    table_name <-
      .add_table_name(data)

    df <-
      .unpivot_omb(
        data = data,
        column_start = 2,
        data_start_row = 6,
        column_names = column_names,
        header_rows = 3:4
      ) %>%
      mutate(metric = "amountNominalMillions") %>%
      select(metric, everything())

    df <-
      df %>%
      gather(yearBudget, amount, -c("metric", column_names)) %>%
      mutate_if(is.character, str_squish)

    data <-
      df %>%
      .fix_year() %>%
      filter(!is.na(item))

    data <-
      data %>%
      .munge_metrics() %>%
      .munge_parents() %>%
      .munge_omb_items() %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount))

    if (exclude_total) {
      data <-
        data %>% filter(!item %>% str_detect("Total")) %>%
        filter(!is.na(item))
    }

    data <-
      data %>%
      mutate(
        parent = case_when(
          parent == "State and Local Government Expenditures From Own Sources (NIPA Basis) (1)" ~ "State and Local Government Expenditures From Own Sources",
          parent == "Addendum: Federal Grants (NIPA Basis)" ~ "Federal Grants",
          TRUE ~ parent
        )
      )


    data <-
      data %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      mutate(amount = currency(amount, digits = 0))

    data
  }

.parse_51 <-
  function(data,
           exclude_total = T,
           column_names = c("parent", "item")) {
    table_name <-
      .add_table_name(data)

    df <-
      .unpivot_omb(
        data = data,
        column_start = 2,
        data_start_row = 4,
        column_names = column_names,
        header_rows = 2:3
      ) %>%
      mutate(metric = "pctGDP") %>%
      select(metric, everything())

    df <-
      df %>%
      gather(yearBudget, amount, -c("metric", column_names)) %>%
      mutate_if(is.character, str_squish)

    data <-
      df %>%
      .fix_year() %>%
      filter(!is.na(item))

    data <-
      data %>%
      .munge_metrics() %>%
      .munge_parents() %>%
      .munge_omb_items() %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount))

    if (exclude_total) {
      data <-
        data %>% filter(!item %>% str_detect("Total")) %>%
        filter(!is.na(item))
    }

    data <-
      data %>%
      mutate(
        parent = case_when(
          parent == "State and Local Government Expenditures From Own Sources (NIPA Basis) (1)" ~ "State and Local Government Expenditures From Own Sources",
          parent == "Addendum: Federal Grants (NIPA Basis)" ~ "Federal Grants",
          TRUE ~ parent
        )
      )


    data <-
      data %>%
      .calc_omb_summary() %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    data <- data %>%
      rename(pct = amount)

    data
  }

.parse_52 <-
  function(data,
           exclude_total = T,
           column_names = c("parent", "item")) {
    table_name <-
      .add_table_name(data)

    df <-
      .unpivot_omb(
        data = data,
        column_start = 2,
        data_start_row = 5,
        column_names = column_names,
        header_rows = 3:4
      ) %>%
      mutate(metric = "amountBillionsNominal") %>%
      select(metric, everything())

    df <-
      df %>%
      gather(yearBudget, amount, -c("metric", column_names)) %>%
      mutate_if(is.character, str_squish)

    data <-
      df %>%
      .fix_year() %>%
      filter(!is.na(item))

    data <-
      data %>%
      .munge_metrics() %>%
      .munge_parents() %>%
      .munge_omb_items() %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount))

    if (exclude_total) {
      data <-
        data %>% filter(!item %>% str_detect("Total")) %>%
        filter(!is.na(item))
    }

    data <-
      data %>%
      mutate(
        parent = case_when(
          parent == "State and Local Government Expenditures From Own Sources (NIPA Basis) (1)" ~ "State and Local Government Expenditures From Own Sources",
          parent == "Addendum: Federal Grants (NIPA Basis)" ~ "Federal Grants",
          TRUE ~ parent
        )
      )


    data <-
      data %>%
      .calc_omb_summary()

    data <-
      data %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    data
  }

.parse_53 <-
  function(data,
           exclude_total = F,
           column_names = c("parent", "item")) {
    table_name <-
      .add_table_name(data)

    df <-
      .unpivot_omb(
        data = data,
        column_start = 2,
        data_start_row = 4,
        column_names = column_names,
        header_rows = 2:3
      ) %>%
      mutate(metric = "pctGDP") %>%
      select(metric, everything())

    df <-
      df %>%
      gather(yearBudget, amount, -c("metric", column_names)) %>%
      mutate_if(is.character, str_squish)

    data <-
      df %>%
      .fix_year()

    data <-
      data %>%
      .munge_metrics() %>%
      .munge_parents() %>%
      .munge_omb_items() %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount))

    data <-
      data %>%
      mutate(item = case_when(is.na(item) ~ parent,
                              TRUE ~ item))

    if (exclude_total) {
      data <-
        data %>% filter(!item %>% str_detect("Total"))
    }

    data <-
      data %>%
      mutate(
        parent = case_when(
          parent == "State and Local Government Expenditures From Own Sources (NIPA Basis) (1)" ~ "State and Local Government Expenditures From Own Sources",
          parent == "Addendum: Federal Grants (NIPA Basis)" ~ "Federal Grants",
          TRUE ~ parent
        )
      )


    data <-
      data %>%
      .calc_omb_summary()

    data <-
      data %>%
      arrange(yearBudget) %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      rename(pct = amount)

    data
  }

.parse_54 <-
  function(data,
           exclude_total = T,
           column_names = c("metric", "parent", "item")) {
    table_name <-
      .add_table_name(data)

    df <-
      .unpivot_omb(
        data = data,
        column_start = 2,
        data_start_row = 5,
        column_names = column_names,
        header_rows = 2:4
      ) %>%
      select(metric, everything())

    df <-
      df %>%
      gather(yearBudget, amount, -c("metric", column_names)) %>%
      mutate_if(is.character, str_squish)

    data <-
      df %>%
      .fix_year()

    data <-
      data %>%
      .munge_metrics() %>%
      .munge_parents() %>%
      .munge_omb_items() %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount))

    data <- data %>%
      mutate(item = case_when(is.na(item) ~ parent,
                              TRUE ~ item))

    if (exclude_total) {
      data <-
        data %>% filter(!item %>% str_detect("Total")) %>%
        filter(!is.na(item))
    }

    data <-
      data %>%
      mutate(
        parent = case_when(
          parent == "State and Local Government Expenditures From Own Sources (NIPA Basis) (1)" ~ "State and Local Government Expenditures From Own Sources",
          parent %in% c(
            "Addendum: Federal Grants (NIPA Basis)",
            "State and Local (NIPA Basis)"
          ) ~ "Federal Grants",
          TRUE ~ parent
        )
      )


    data <-
      data %>%
      .calc_omb_summary() %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    data
  }

.parse_55 <-
  function(data,
           exclude_total = T,
           column_names = c("metric", "parent", "item")) {
    table_name <-
      .add_table_name(data)

    df <-
      .unpivot_omb(
        data = data,
        column_start = 2,
        data_start_row = 5,
        column_names = column_names,
        header_rows = 2:4
      ) %>%
      select(metric, everything())

    df <- df %>%
      mutate(metric = case_when(
        metric == "As Percentages of" ~ str_c(metric, parent, sep = " "),
        TRUE ~ metric
      ))

    df <-
      df %>%
      gather(yearBudget, amount, -c("metric", column_names)) %>%
      mutate_if(is.character, str_squish)

    data <-
      df %>%
      .fix_year()

    data <-
      data %>%
      .munge_metrics() %>%
      .munge_parents() %>%
      .munge_omb_items() %>%
      .fix_omb_amount() %>%
      filter(!is.na(amount))

    data <- data %>%
      mutate(item = case_when(is.na(item) ~ parent,
                              TRUE ~ item))


    if (exclude_total) {
      data <-
        data %>% filter(!item %>% str_detect("Total"))
    }

    data <-
      data %>%
      mutate(
        parent = case_when(
          parent == "Medicare (Excluding Premiums)" ~ "Medicare Excluding Premiums",
          parent == "Other Health (1)" ~ "Other Health",
          parent == "Federal Employees Health Benefits (FEHB)" ~ "Federal Employees Health Benefits",
          parent %in% c(
            "Addendum: Federal Grants (NIPA Basis)",
            "State and Local (NIPA Basis)"
          ) ~ "Federal Grants",
          TRUE ~ parent
        )
      )



    data <-
      data %>%
      .calc_omb_summary() %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything())

    data
  }

.parse_56 <-
  function(data,
           exclude_total = T,
           column_names = c("parent", "item")) {
    table_name <-
      .add_table_name(data)

    df <-
      .unpivot_omb(
        data = data,
        column_start = 2,
        data_start_row = 5,
        column_names = column_names,
        header_rows = 3:4
      ) %>%
      mutate(metric = "countEmployees") %>%
      select(metric, everything()) %>%
      .remove_na()


    data <-
      df %>%
      gather(yearBudget, amount, -c("metric", column_names)) %>%
      mutate_if(is.character, str_squish) %>%
      .fix_year() %>%
      mutate(amount = parse_number(amount) * 1000)

    data <-
      data %>%
      .munge_metrics() %>%
      .munge_parents() %>%
      .munge_omb_items() %>%
      filter(!is.na(amount))

    data <- data %>%
      mutate(item = case_when(is.na(item) ~ parent,
                              TRUE ~ item))


    if (exclude_total) {
      data <-
        data %>% filter(!parent %>% str_detect("Total")) %>%
        filter(item != "Total")
    }

    data <-
      data %>%
      mutate(
        item = case_when(
          item == "Transpor- tation" ~ "Transportation",
          item == "Treasure"  ~ "Treasury",
          TRUE ~ item
        )
      )

    data <-
      data %>%
      .calc_omb_summary()

    data <-
      data %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      rename(count = amount)


    data
  }

.parse_57 <-
  function(data,
           exclude_total = T,
           column_names = c("parent", "item")) {
    table_name <-
      .add_table_name(data)

    df <-
      .unpivot_omb(
        data = data,
        column_start = 2,
        data_start_row = 6,
        column_names = column_names,
        header_rows = 3:5
      ) %>%
      mutate(metric = "pctEmployees") %>%
      select(metric, everything()) %>%
      .remove_na()


    data <-
      df %>%
      gather(yearBudget, amount, -c("metric", column_names)) %>%
      mutate_if(is.character, str_squish) %>%
      .fix_year() %>%
      mutate(amount = parse_number(amount) / 100)

    data <-
      data %>%
      .munge_metrics() %>%
      .munge_parents() %>%
      .munge_omb_items() %>%
      filter(!is.na(amount))

    data <- data %>%
      mutate(item = case_when(is.na(item) ~ parent,
                              TRUE ~ item))


    if (exclude_total) {
      data <-
        data %>% filter(!parent %>% str_detect("Total")) %>%
        filter(item != "Total")
    }

    data <-
      data %>%
      mutate(
        item = case_when(
          item == "Transpor- tation" ~ "Transportation",
          item == "Treasure"  ~ "Treasury",
          TRUE ~ item
        )
      )

    data <-
      data %>%
      .calc_omb_summary()


    data <-
      data %>%
      mutate(nameTable = table_name) %>%
      select(nameTable, everything()) %>%
      rename(pct = amount)


    data
  }

.dl_omb <-
  function(url = "https://www.whitehouse.gov/wp-content/uploads/2019/03/hist-fy2020.zip") {
    tmp <-
      tempfile()

    file <- curl_download(url, tmp)

    unz_files <- unzip(file, exdir = "files")
    file_nos <-
      seq_along(unz_files)

    all_data <-
      file_nos %>%
      map_dfr(function(x) {
        options(scipen = 999999)
        glue("Table: {x}") %>% message()

        data <-
          unz_files[[x]] %>%
          read_excel(col_names = F)

        file <-
          unz_files[[x]] %>% str_split("/") %>% flatten_chr() %>%
          .[[2]] %>% str_remove_all(".xlsx")

        if (x  %in% c(1)) {
          df <-
            .parse_1(data = data)

          df <-
            df %>% mutate(metric = "amount")

          df <-
            df %>%
            filter(yearBudget %>% str_detect("-")) %>%
            bind_rows(
              df %>%
                filter(!yearBudget %>% str_detect("-")) %>%
                .fix_year() %>%
                .calc_omb_summary() %>%
                mutate(amount = amount %>% currency(digits = 0)) %>%
                mutate(yearBudget = as.character(yearBudget))
            ) %>%
            rename(yearGroupBudget = yearBudget) %>%
            mutate(yearBudget = year(dateFiscalYearEnd) %>% as.numeric()) %>%
            select(nameTable,
                   isEstimate,
                   yearGroupBudget,
                   yearBudget,
                   everything()) %>%
            mutate(amount = currency(amount, digits = 0)) %>%
            rename(parent = typeBudget,
                   item = typeCashFlow) %>%
            select(nameTable, metric, everything())

          data <- df %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(2)) {
          data <-
            .parse_2(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(3)) {
          data <-
            .parse_3(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(4)) {
          data <-
            .parse_4(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(5)) {
          data <-
            .parse_5(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(6)) {
          data <-
            .parse_7(data = data, metric = "pctOutlays") %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(7)) {
          data <-
            .parse_7(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(8)) {
          data <-
            .parse_8(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(9)) {
          data <-
            .parse_9(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(10)) {
          data <-
            .parse_10(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(11, 14)) {
          data <-
            .parse_11(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(12, 15, 17, 19)) {
          data <-
            .parse_12(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(13, 16, 18)) {
          data <-
            .parse_13(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(20)) {
          data <-
            .parse_20(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(21)) {
          data <-
            .parse_21(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(22)) {
          data <-
            .parse_22(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(23)) {
          data <-
            .parse_23(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(24)) {
          data <-
            .parse_24(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(25)) {
          data <-
            .parse_25(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(26)) {
          data <-
            .parse_26(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(27)) {
          data <-
            .parse_26(data = data, metric = "pctGDP") %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(28, 30)) {
          data <-
            .parse_28(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(29)) {
          data <-
            .parse_28(data = data) %>%
            mutate(amount = amount * 10000) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(31)) {
          data <-
            .parse_28(data = data) %>%
            mutate(amount = amount * 10000) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(32)) {
          data <-
            .parse_32(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% 33) {
          data <-
            .parse_33(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()

          return(data)
        }

        if (x %in% 34) {
          data <-
            .parse_34(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()

          return(data)
        }

        if (x %in% 35) {
          data <-
            .parse_35(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()

          return(data)
        }

        if (x %in% 36) {
          data <-
            .parse_36(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()

          return(data)
        }

        if (x %in% c(37)) {
          data <-
            .parse_37(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()

          return(data)

        }

        if (x %in% c(38)) {
          data <-
            .parse_38(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()

          return(data)

        }

        if (x %in% c(39)) {
          data <-
            .parse_39(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()

          return(data)

        }

        if (x %in% c(40)) {
          data <-
            .parse_40(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()

          return(data)

        }

        if (x %in% c(41)) {
          data <-
            .parse_41(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()

          return(data)

        }

        if (x %in% c(42)) {
          data <-
            .parse_42(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()

          return(data)

        }

        if (x %in% c(43)) {
          data <-
            .parse_43(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()

          return(data)

        }

        if (x %in% c(44)) {
          data <-
            .parse_44(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(45)) {
          data <-
            .parse_45(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(46)) {
          data <-
            .parse_46(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(47)) {
          data <-
            .parse_47(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(48)) {
          data <-
            .parse_48(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(49)) {
          data <-
            .parse_49(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(50)) {
          data <-
            .parse_50(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(51)) {
          data <-
            .parse_51(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(52)) {
          data <-
            .parse_52(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(53)) {
          data <-
            .parse_53(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(54)) {
          data <-
            .parse_54(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(55)) {
          data <-
            .parse_55(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(56)) {
          data <-
            .parse_56(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }

        if (x %in% c(57)) {
          data <-
            .parse_57(data = data) %>%
            .add_table_no(table_number = x) %>%
            mutate(slugTable = file) %>%
            .nest_omb()
          return(data)
        }


        data
      })


    file %>% unlink()
    unlink("xml", recursive = T)
    unz_files %>% unlink()

    all_data <- all_data %>%
      mutate(nameTable = nameTable %>% sub("\\ - ", "|", .)) %>%
      separate(
        nameTable,
        into = c("idTable", "nameTable"),
        sep = "\\|",
        extra = "merge"
      ) %>%
      mutate(idTable = idTable %>% str_remove_all("\\Table ") %>% as.numeric()) %>%
      separate(
        nameTable,
        into = c("typeTable", "periodTable"),
        sep = "\\: ",
        remove = F
      ) %>%
      mutate(periodTable = periodTable %>% str_replace_all("\\ - ", "\\-"))

    all_data
  }

#' Office of Budget Management Tables
#'
#' @param assign_to_environment if \code{TRUE} assigns each table
#' to environment
#'
#' @return \code{nested_tibble}
#' @export
#'
#' @examples
omb_tables <-
  function(assign_to_environment = T) {
    df_tables <- dictionary_omb_tables()
    url <-
      df_tables %>%
      filter(numberTable == 1) %>%
      pull(urlBudgetZIP)

    all_data <- .dl_omb(url = url)

    if (assign_to_environment) {
      dict_tables <-
        all_data %>%
        select(-matches("data"))

      assign("tbl_000_omb_data", dict_tables, envir = .GlobalEnv)

      1:nrow(all_data) %>%
        walk(function(x) {
          zeros <- 3 - nchar(x)
          zeros <- rep(0, zeros) %>% str_c(collapse = "")
          zeros <- str_c(zeros, x, sep = "")
          df_row <- all_data %>% dplyr::slice(x)
          table_slug <- df_row$slugTable
          tn <- glue("tbl_{zeros}_{table_slug}")
          df <-
            df_row %>%
            select(slugTable, nameTable, data) %>%
            unnest_legacy()

          if (df %>% hasName("amount")) {
            df <- df %>%
              mutate(amount = currency(amount, digits = 0))
          }

          assign(x = tn, df, envir = .GlobalEnv)
        })

    }

    all_data
  }


# other -------------------------------------------------------------------
.parse_omb_budget_authority <-
  memoise::memoise(function(url = "https://www.whitehouse.gov/wp-content/uploads/2019/03/budauth-fy2020.xlsx",
                            filter_zero_balance = T) {
    data <-
      url %>%
      rio::import()

    data <-
      data %>%
      as_tibble()

    end_col <- data %>% names() %>% grep("On-", .)
    gather_cols <- data %>% select(1:(end_col)) %>% names()
    data <-
      data %>%
      gather(yearBudget, amount, -c(gather_cols))

    data <-
      data %>%
      .munge_omb_names()

    data <-
      data %>%
      mutate(isOnBudget = isOnBudget %>% str_detect("On-budget")) %>%
      mutate(amount = 1000 * amount,
             amount = currency(amount, digits = 0))

    data <-
      data %>%
      .fix_year() %>%
      mutate(metric = "amountNominal")

    if (filter_zero_balance) {
      data <-
        data %>%
        filter(amount != 0)
    }

    data <-
      data %>%
      .calc_omb_summary()

    data <-
      data %>%
      select(-metric)

    data <- data %>%
      mutate(amount = currency(amount, digits = 0))

    data <-
      data %>%
      mutate(slugCGAC =
               case_when(
                 !is.na(idCGAC) ~  glue("0{idCGAC}") %>% as.character(),
                 TRUE ~ NA_character_
               )) %>%
      .munge_data()

    data
  })

.parse_omb_outlays <-
  memoise::memoise(function(url = "https://www.whitehouse.gov/wp-content/uploads/2019/03/outlays-fy2020.xlsx",
                            filter_zero_balance = T) {
    data <- rio::import(url) %>% as_tibble()
    data <-
      data %>%
      as_tibble()

    end_col <- data %>% names() %>% grep("On-", .)
    gather_cols <- data %>% select(1:(end_col)) %>% names()
    data <-
      data %>%
      gather(yearBudget, amount, -c(gather_cols))

    data <-
      data %>%
      .munge_omb_names()

    data <-
      data %>%
      mutate(isOnBudget = isOnBudget %>% str_detect("On-budget")) %>%
      mutate(amount = 1000 * amount,
             amount = currency(amount, digits = 0))

    data <-
      data %>%
      .fix_year() %>%
      mutate(metric = "amountNominal")

    if (filter_zero_balance) {
      data <-
        data %>%
        filter(amount != 0)
    }

    data <-
      data %>%
      .calc_omb_summary()

    data <-
      data %>%
      mutate(slugCGAC =
               case_when(
                 !is.na(idCGAC) ~  glue("0{idCGAC}") %>% as.character(),
                 TRUE ~ NA_character_
               )) %>%
      .munge_data() %>%
      select(-metric)

    data

  })

.parse_omb_receipts <-
  memoise::memoise(function(url = "https://www.whitehouse.gov/wp-content/uploads/2019/03/receipts-fy2020.xlsx",
                            filter_zero_balance = T) {
    data <-
      rio::import(url) %>% as_tibble()
    data <-
      data %>%
      as_tibble()

    end_col <- data %>% names() %>% grep("On-", .)
    gather_cols <- data %>% select(1:(end_col)) %>% names()
    data <-
      data %>%
      gather(yearBudget, amount, -c(gather_cols))

    data <-
      data %>%
      .munge_omb_names()

    data <-
      data %>%
      mutate(isOnBudget = isOnBudget %>% str_detect("On-budget")) %>%
      mutate(amount = 1000 * amount,
             amount = currency(amount, digits = 0))

    data <-
      data %>%
      .fix_year() %>%
      mutate(metric = "amountNominal")

    if (filter_zero_balance) {
      data <-
        data %>%
        filter(amount != 0)
    }

    data <-
      data %>%
      .calc_omb_summary()

    data <-
      data %>%
      mutate(slugCGAC =
               case_when(
                 !is.na(idCGAC) ~  glue("0{idCGAC}") %>% as.character(),
                 TRUE ~ NA_character_
               )) %>%
      .munge_data() %>%
      select(-metric)

    data

  })



#' OMB Historic Budget Authority
#'
#' Returns data containing information
#' about the historic OMB budget authority
#'
#' @param filter_zero_balance if \code{TRUE} excludes
#' accounts with zero authority
#' @param research_terms
#' @param research_columns
#' @param join_tas  if \code{TRUE} join extra treasury symbol data
#'
#' @return \code{tibble()}
#' @export
#'
#' @examples
#' omb_budget_authority()
omb_budget_authority <-
  function(filter_zero_balance = T,
           join_tas = T,
           research_terms = c("research",
                              "science",
                              "technology",
                              "ARTIFICIAL INTELLIGENCE",
                              "technical"),
           research_columns = c("nameAgencyOMB",
                                "nameBureauOMB", "nameAccountOMB")) {
    url <-
      dictionary_omb_supplements() %>%
      filter(typeFile %>% str_detect("BUDGET AUTHORITY XLS")) %>%
      pull(urlData)

    data <-
      .parse_omb_budget_authority(url = url, filter_zero_balance = filter_zero_balance)

    data <-
      data %>% .generate_federal_account_ids(cgac_column = "idCGAC", account_column = "codeAccountOMB")



    data <-
      data %>%
      munge_account_symbols(research_columns = research_columns,
                            research_terms = research_terms) %>%
      .add_omb_account_group()

    data <-
      data %>%
      mutate(idCGAC = as.numeric(idCGAC)) %>%
      select(-one_of("slugCGAC")) %>%
      left_join(dictionary_omb_cgac_accounts(), by = "idCGAC") %>%
      distinct()

    if (join_tas) {
      data <-
        data %>%
        .join_tas_symbols()
    }
    data <-
      data %>% mutate_if(is.numeric, as.numeric) %>%
      add_dod_omb_group()

    data

  }

#' OMB Budget Outlays
#'
#' Returns information about all outflows
#' as since 1962 as recorded by the OMB
#'
#' @param filter_zero_balance if \code{TRUE} excludes
#' accounts with zero authority
#' @param research_terms
#' @param research_columns
#' @param join_tas
#'
#' @return
#' @export
#'
#' @examples
#' omb_outlays()
omb_outlays <- function(filter_zero_balance = T,
                        research_terms = c("research",
                                           "science",
                                           "technology",
                                           "ARTIFICIAL INTELLIGENCE",
                                           "technical"),
                        join_tas = T,
                        research_columns = c("nameAgencyOMB",
                                             "nameBureauOMB", "nameAccountOMB")) {
  url <-
    dictionary_omb_supplements() %>%
    filter(typeFile %>% str_detect("OUTLAYS XLS")) %>%
    pull(urlData)

  data <-
    .parse_omb_outlays(url = url, filter_zero_balance = filter_zero_balance)

  data <-
    data %>% .generate_federal_account_ids(cgac_column = "idCGAC", account_column = "codeAccountOMB")



  data <-
    data %>%
    munge_account_symbols(research_columns = research_columns,
                          research_terms = research_terms) %>%
    .add_omb_account_group()

  data <-
    data %>%
    mutate(idCGAC = as.numeric(idCGAC)) %>%
    select(-one_of("slugCGAC")) %>%
    left_join(dictionary_omb_cgac_accounts(), by = "idCGAC") %>%
    distinct()

  if (join_tas) {
    data <-
      data %>%
      .join_tas_symbols()
  }
  data <-
    data %>% mutate_if(is.numeric, as.numeric) %>%
    add_dod_omb_group()

  data
}

#' OMB Budget Receipts
#'
#' Returns data about all monies received by
#' the United States Government since 1962 as recorded by the OMB
#'
#' @param filter_zero_balance if \code{TRUE} excludes
#' accounts with zero authority
#'
#' @return \code{tibble}
#' @export
#'
#' @examples
#' omb_receipts()
omb_receipts <-
  function(filter_zero_balance = T,
           join_tas = T,
           research_terms = c("research",
                              "science",
                              "technology",
                              "ARTIFICIAL INTELLIGENCE",
                              "technical"),
           research_columns = c("nameAgencyOMB",
                                "nameBureauOMB", "nameAccountOMB")) {
    url <-
      dictionary_omb_supplements() %>%
      filter(typeFile %>% str_detect("RECEIPTS XLS")) %>%
      pull(urlData)

    data <-
      data %>% .generate_federal_account_ids(cgac_column = "idCGAC", account_column = "codeAccountOMB")



    data <-
      data %>%
      munge_account_symbols(research_columns = research_columns,
                            research_terms = research_terms) %>%
      .add_omb_account_group()

    data <-
      data %>%
      mutate(idCGAC = as.numeric(idCGAC)) %>%
      select(-one_of("slugCGAC")) %>%
      left_join(dictionary_omb_cgac_accounts(), by = "idCGAC") %>%
      distinct()

    if (join_tas) {
      data <-
        data %>%
        .join_tas_symbols()
    }
    data <- add_dod_omb_group(data = data)
    data
  }

#' All OMB Data
#'
#' @param types
#' @param filter_zero_balance
#' @param research_terms
#' @param research_columns
#' @param widen_data
#'
#' @return
#' @export
#'
#' @examples
omb_data <-
  function(types = c("budget", "receipts", "outlays"),
           filter_zero_balance = T,
           join_tas = F,
           research_terms = c("research",
                              "science",
                              "technology",
                              "ARTIFICIAL INTELLIGENCE",
                              "technical"),
           research_columns = c("nameAgencyOMB",
                                "nameBureauOMB", "nameAccountOMB"),
           widen_data = F
           ) {
    slug_types <- str_to_lower(types)
    all_data <-
      tibble()

    if (types %>% str_to_lower() %>% str_detect("budget") %>% sum(na.rm = T) > 0) {
      all_data <-
        all_data %>% bind_rows(
          omb_budget_authority(
            filter_zero_balance = filter_zero_balance,
            research_terms = research_terms,
            research_columns = research_columns
          ) %>%
            mutate(typeBudget = "Budget")
        )
    }

    if (types %>% str_to_lower() %>% str_detect("receipts") %>% sum(na.rm = T) > 0) {
      all_data <-
        all_data %>% bind_rows(
          omb_receipts(
            filter_zero_balance = filter_zero_balance,
            research_terms = research_terms,
            research_columns = research_columns,

          ) %>%
            mutate(typeBudget = "Receipts")
        )
    }

    if (types %>% str_to_lower() %>% str_detect("outlays") %>% sum(na.rm = T) > 0) {
      all_data <-
        all_data %>% bind_rows(
          omb_outlays(
            filter_zero_balance = filter_zero_balance,
            research_terms = research_terms,
            research_columns = research_columns
          ) %>%
            mutate(typeBudget = "Outlays")
        )
    }
    all_data <- all_data %>%
      select(-typeFunding)

    all_data <- all_data %>%
      group_by(!!!syms(all_data %>% select(-amount) %>% names())) %>%
      summarise(amount = sum(amount, na.rm = T)) %>%
      ungroup()

    if (widen_data) {
      all_data %>%
        mutate(typeBudget = str_c("amount", typeBudget)) %>%
        spread(typeBudget, amount)
    }

    if (join_tas) {
      data <-
        data %>%
        .join_tas_symbols()
    }
    all_data
  }