使用两个记录分隔符读取 ASCII、固定长度字段/记录答案

【问题标题】：Read ASCII, fixed-length fields/records with two record-delimiters使用两个记录分隔符读取 ASCII、固定长度字段/记录
【发布时间】：2020-06-05 20:23:31
【问题描述】：

很抱歉，如果这太宽泛了，但我正在努力加载存档的人口普查县商业模式数据。文件格式描述为，

“ASCII，固定长度的字段/记录，带有两个记录分隔符（回车和换行）；记录长度包括分隔符。”

我尝试将其加载到 excel、R 和 Stata 中。我只是想以可读格式获取文件，以便以后可以使用。我尝试使用 read.fwf 将它读入 R，但我真的不清楚我应该使用什么宽度。我不太了解文件类型。我对 ASCII 文件不太熟悉，文件后缀并没有给我太多信息。任何建议将不胜感激。我在下面提供了我正在尝试使用的一组文件的链接。

https://catalog.archives.gov/id/873805

【问题讨论】：

标签： r excel ascii stata census

【解决方案1】：

有些行的数据不好，所以我推荐 read_delim 包中的 readr

library(readr)
data <- read_delim("https://catalog.archives.gov/OpaAPI/media/873805/content/arcmedia/electronic-records/rg-029/cbp-files/RG029.CBP85.T2I1?download=true",
                   delim = " ", col_names = FALSE)
data
## A tibble: 32,970 x 6
#      X1 X2    X3                                                                                                                   #X4    X5 X6    
#   <dbl> <chr> <chr>                                                                                                             #<dbl> <dbl> <chr> 
# 1 11001 ----  " 00000003482800000011566200000049404800248400140400048300029200018300006600003400001600000500000100000100000000… 11000 23001 " 424…
# 2 11001 07--  "B00000000000000000000000000000000000000001700001200000300000100000100000000000000000000000000000000000000000000… 11000 23001 " 424…
# 3 11001 0700  "B00000000000000000000000000000000000000001500001000000300000100000100000000000000000000000000000000000000000000… 11000 23001 " 424…
# 4 11001 0720  "B00000000000000000000000000000000000000000200000000000000000100000100000000000000000000000000000000000000000000… 11000 23001 " 424…
# 5 11001 0740  " 00000000001900000000003500000000019100000500000300000200000000000000000000000000000000000000000000000000000000… 11000 23001 " 424…
# 6 11001 0750  "A00000000000000000000000000000000000000000100000100000000000000000000000000000000000000000000000000000000000000… 11000 23001 " 424…
# 7 11001 0780  " 00000000001200000000002700000000035100000600000500000100000000000000000000000000000000000000000000000000000000… 11000 23001 " 424…
# 8 11001 0800  "A00000000000000000000000000000000000000000200000200000000000000000000000000000000000000000000000000000000000000… 11000 23001 " 424…
# 9 11001 10--  "A00000000000000000000000000000000000000000200000100000000000100000000000000000000000000000000000000000000000000… 11000 23001 " 424…
#10 11001 1400  "A00000000000000000000000000000000000000000200000100000000000100000000000000000000000000000000000000000000000000… 11000 23001 " 424…
## … with 32,960 more rows

【讨论】：

【解决方案2】：

我们可以在base R 中使用read.table 和fill = TRUE

data <- read.table("https://catalog.archives.gov/OpaAPI/media/873805/content/arcmedia/electronic-records/rg-029/cbp-files/RG029.CBP85.T2I1?download=true", fill = TRUE)

-输出

str(data)
#data.frame':   32970 obs. of  6 variables:
# $ V1: int  11001 11001 11001 11001 11001 11001 11001 11001 11001 11001 ...
# $ V2: chr  "----" "07--" "0700" "0720" ...
# $ V3: chr  "000000034828000000115662000000494048002484001404000483000292000183000066000034000016000005000001000001000000000000000000" "B000000000000000000000000000000000000000017000012000003000001000001000000000000000000000000000000000000000000000000000000" "B000000000000000000000000000000000000000015000010000003000001000001000000000000000000000000000000000000000000000000000000" "B000000000000000000000000000000000000000002000000000000000001000001000000000000000000000000000000000000000000000000000000" ...
# $ V4: int  11000 11000 11000 11000 11000 11000 11000 11000 11000 11000 ...
# $ V5: int  23001 23001 23001 23001 23001 23001 23001 23001 23001 23001 ...
# $ V6: int  4243 4243 4243 4243 4243 4243 4243 4243 4243 4243 ...

【讨论】：