Welcome to OGeek Q&A Community for programmer and developer-Open, Learning and Share
Welcome To Ask or Share your Answers For Others

Categories

0 votes
4.0k views
in Technique[技术] by (71.8m points)

pdftools - Cleaning downloaded pdf dataset in R

I have downloaded the pdf file from this site (from the Table tab) and want to clean the dataset in R and convert it to a csv or excel file.

I am using pdftools package and have downloaded the other required packages. I want to focus on the data for the age groups. So far I have the dataset narrowed down by using these codes.

#Load the dataset 
PDF1 <- pdf_text("agegr_1-4-21.pdf") %>%
  readr::read_lines() #open the PDF inside your project folder
PDF1
PDF1.grass <-PDF1[-c(1:10,17:19)] # remove lines
PDF1.grass
write.table(PDF1.grass, file="docd_pdf.csv", sep=",", row.names=FALSE)

all_stat_lines <- PDF1.grass 

pdf_transpose = t(all_stat_lines)
write.table(pdf_transpose, file="docd_pdf.csv", sep=",", row.names=FALSE)

df <- plyr::ldply(pdf_transpose) #create a data frame
head(df)

However the data frame that I am getting includes everything on one variable. Is there is a way to efficiently break up the datasets and have different columns for the age groups? I downloaded the pdf file from the site and named it agegr_1-4-21.pdf.

The output I am getting is

enter image description here


与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…
Welcome To Ask or Share your Answers For Others

1 Reply

0 votes
by (71.8m points)

One approach to achieve this is via tidyr::extract. I first extract the header from the first row and second extract the data from the other rows.

library(dplyr)


regex_header <- paste0(
  "^(\w+)\s+",
  paste(rep("(\d+\-\d+ years)", 7), collapse = "\s+"), "\s+",
  "(\d+\+ years)\s+",
  "(\w+)"
)

header <- tidyr::extract(data = slice(df, 1), col = V1, into = paste0("var", 1:10), regex = regex_header) %>%
  t() %>%
  .[, 1]

regex_body <- paste0("^([\w\*]+)\s+", paste(rep("([\d,\.]+)", 9), collapse = "\s+"))

tidyr::extract(data = slice(df, 2:nrow(df)), col = V1, into = header, regex = regex_body)
#>        Outcome 0-17 years 18-29 years 30-39 years 40-49 years 50-59 years
#> 1         Case      2.090       3.435       2.706       2.190       1.887
#> 2 Hospitalized         20          81         133         188         264
#> 3         Died          0           4           4          11          36
#> 4    Missing**        612       1.740       1.369       1.076       1.013
#> 5  Gesamtsumme      2.722       5.260       4.212       3.465       3.200
#>   60-69 years 70-79 years 80+ years Gesamtsumme
#> 1       1.218         504       224      14.254
#> 2         299         219       151       1.355
#> 3          58          83       110         306
#> 4         674         295       208       6.987
#> 5       2.249       1.101       693      22.902

DATA For the data I downloaded one of the tables and cleaned it using your code.

df <- structure(list(V1 = c(
  "Outcome                 0-17 years       18-29 years     30-39 years      40-49 years   50-59 years     60-69 years 70-79 years     80+ years Gesamtsumme",
  "Case                      2.090            3.435            2.706            2.190        1.887           1.218        504            224        14.254",
  "Hospitalized                20               81              133              188          264             299         219            151         1.355",
  "Died                         0                4                4               11           36              58          83            110          306",
  "Missing**                  612             1.740            1.369            1.076        1.013            674         295            208         6.987",
  "Gesamtsumme               2.722            5.260            4.212            3.465        3.200           2.249       1.101           693        22.902"
)), class = "data.frame", row.names = c(NA, -6L))

与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…
OGeek|极客中国-欢迎来到极客的世界,一个免费开放的程序员编程交流平台!开放,进步,分享!让技术改变生活,让极客改变未来! Welcome to OGeek Q&A Community for programmer and developer-Open, Learning and Share
Click Here to Ask a Question

...