将参数传递给R中的多个match_fun函数fuzzyjoin::fuzzy_join

Posted 2023-02-24

【中文标题】将参数传递给R中的多个match_fun函数fuzzyjoin::fuzzy_join

【英文标题】：Passing arguments into multiple match_fun functions in R fuzzyjoin::fuzzy_join

【发布时间】：2017-11-07 02:05:09

【问题描述】：

我正在回答这些twoquestions 并得到了适当的解决方案，但我无法将使用fuzzy_join 的参数传递到我从fuzzyjoin::stringdist_join 提取的match_fun 中。在这种情况下，我混合使用了多个 match_fun，包括自定义的 match_fun_stringdist 以及 == 和 <=，用于精确匹配和标准匹配。

我得到的错误信息是：

# Error in mf(rep(u_x, n_y), rep(u_y, each = n_x), ...): object 'ignore_case' not found

# Data:
library(data.table, quietly = TRUE)
Address1 <- c("786, GALI NO 5, XYZ","rambo, 45, strret 4, atlast, pqr","23/4, 23RD FLOOR, STREET 2, ABC-E, PQR","45-B, GALI NO5, XYZ","HECTIC, 99 STREET, PQR")
AREACODE <- c('10','10','14','20','30')
Year1 <- c(2001:2005)

Address2 <- c("abc, pqr, xyz","786, GALI NO 4 XYZ","45B, GALI NO 5, XYZ","del, 546, strret2, towards east, pqr","23/4, STREET 2, PQR","abc, pqr, xyz","786, GALI NO 4 XYZ","45B, GALI NO 5, XYZ","del, 546, strret2, towards east, pqr","23/4, STREET 2, PQR")
Year2 <- c(2001:2010)
AREA_CODE <- c('10','10','10','20','30','40','50','61','64', '99')

data1 <- data.table(Address1, Year1, AREACODE)
data2 <- data.table(Address2, Year2, AREA_CODE)
data2[, unique_id := sprintf("%06d", 1:nrow(data2))]

# Solution:
library(fuzzyjoin, quietly = TRUE); library(dplyr, quietly = TRUE)

# First, need to define match_fun_stringdist 
# Code from stringdist_join from https://github.com/dgrtwo/fuzzyjoin/blob/master/R/stringdist_join.R
match_fun_stringdist <- function(v1, v2, ...) 

  if (ignore_case) 
    v1 <- stringr::str_to_lower(v1)
    v2 <- stringr::str_to_lower(v2)
  

  dists <- stringdist::stringdist(v1, v2, method = method, ...)

  ret <- dplyr::data_frame(include = (dists <= max_dist))
  if (!is.null(distance_col)) 
    ret[[distance_col]] <- dists
  
  ret


# Call fuzzy_join
fuzzy_join(data1, data2, 
           by = list(x = c("Address1", "AREACODE", "Year1"), y = c("Address2", "AREA_CODE", "Year2")), 
           match_fun = list(match_fun_stringdist, `==`, `<=`),
           mode = "left",
           ignore_case = FALSE,
           method = "dl",
           max_dist = 99,
           distance_col = "dist"
) %>%
  group_by(Address1, Year1, AREACODE) %>%
  top_n(1, -Address1.dist) %>%
  top_n(1, Year2) %>%
  select(unique_id, Address1.dist, everything())
#> Error in mf(rep(u_x, n_y), rep(u_y, each = n_x), ...): object 'ignore_case' not found

【参考方案1】：

我认为错误是因为传递给每个多个 match_fun 的参数搞砸了，即不能将额外的参数（如 ignore_case，最初用于 string_dist match_fun）传递到 >= 的 match_fun 中

解决方案是使用固定参数定义我自己的 match_fun。请参阅下面我用固定参数定义自己的 match_fun_stringdist 的地方。我也在另一个问题/答案https://***.com/a/44383103/4663008 中实现了它。

# First, need to define match_fun_stringdist 
# Code from stringdist_join from https://github.com/dgrtwo/fuzzyjoin
match_fun_stringdist <- function(v1, v2) 

  # Can't pass these parameters in from fuzzy_join because of multiple incompatible match_funs, so I set them here.
  ignore_case = FALSE
  method = "dl"
  max_dist = 99
  distance_col = "dist"

  if (ignore_case) 
    v1 <- stringr::str_to_lower(v1)
    v2 <- stringr::str_to_lower(v2)
  

  # shortcut for Levenshtein-like methods: if the difference in
  # string length is greater than the maximum string distance, the
  # edit distance must be at least that large

  # length is much faster to compute than string distance
  if (method %in% c("osa", "lv", "dl")) 
    length_diff <- abs(stringr::str_length(v1) - stringr::str_length(v2))
    include <- length_diff <= max_dist

    dists <- rep(NA, length(v1))

    dists[include] <- stringdist::stringdist(v1[include], v2[include], method = method)
   else 
    # have to compute them all
    dists <- stringdist::stringdist(v1, v2, method = method)
  
  ret <- dplyr::data_frame(include = (dists <= max_dist))
  if (!is.null(distance_col)) 
    ret[[distance_col]] <- dists
  
  ret

并调用fuzzy_join

fuzzy_join(data1, data2, 
           by = list(x = c("Address1", "AREACODE", "Year1"), y = c("Address2", "AREA_CODE", "Year2")), 
           match_fun = list(match_fun_stringdist, `==`, `<=`),
           mode = "left")

【讨论】：

我得到同样的错误。此问题记录在这里：github.com/dgrtwo/fuzzyjoin/issues/50

事实证明，当我使用反引号而不是单引号时，我可以正常工作。我在 github 线程中添加了额外的 cmets，它显示了我的示例代码。

好的，我确实使用了反引号；但是，就我而言，我试图提供一个复杂的 match_fun 列表（match_fun = list(match_fun_stringdist, ==, <=)

如果没有在 match_fun_stringdist 中定义 max_dist ，你是如何让它工作的？我收到了错误Error in eval_tidy(xs[[i]], unique_output) : object 'max_dist' not found，即使我像您在模糊加入调用中所做的那样定义它。

我正在尝试做类似的事情，我认为“函数工厂”可能是一个不错的选择，因此您不必编写一堆类似的自定义函数。 adv-r.hadley.nz/function-factories.html