在R中提取文本及解压API传入的GZIP压缩Base64数据方法问询

Hey there! Let's tackle your two R questions one by one with actionable code snippets:

Text extraction covers a few common scenarios—here are the most practical methods for each:

• Extract specific patterns from strings
  For regex-based extraction, you can use base R functions or the more intuitive stringr package. Here's how:

  # Base R approach: Extract an email from a sample string
  sample_text <- "Reach me at jose@example.com for follow-ups"
  email_pattern <- "\\b[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\\.[A-Z|a-z]{2,}\\b"
  extracted_email <- regmatches(sample_text, gregexpr(email_pattern, sample_text))[[1]]
  print(extracted_email) # Output: "jose@example.com"
  
  # stringr approach (install first with install.packages("stringr"))
  library(stringr)
  str_extract(sample_text, email_pattern)
  

• Extract content from text files
  First read the file, then filter or extract using regex:

  # Read a text file into a vector of lines
  file_lines <- readLines("your_target_file.txt")
  # Extract all lines containing "error" (case-insensitive)
  error_lines <- file_lines[grepl("error", file_lines, ignore.case = TRUE)]
  print(error_lines)
  

• Extract text from HTML/XML
  Use the rvest package for web scraping or structured document parsing:

  # Install package first: install.packages("rvest")
  library(rvest)
  # Extract all paragraph text from a sample webpage
  page <- read_html("https://example.com")
  paragraph_text <- page %>% html_elements("p") %>% html_text()
  print(paragraph_text)
  

Your Java code first GZIP-compresses the string then encodes it to Base64. To reverse this in R, we need to decode the Base64 string first, then decompress the raw bytes. Here's a working example:

# First, install the required package if you haven't: install.packages("base64enc")
library(base64enc)

# Example encoded data (matches "who are you" from your Java code)
encoded_data <- "H4sIAAAAAAAAAEtUKC4pysxLBwCMN9M1NQAAAA="

# Step 1: Decode Base64 to raw bytes
raw_compressed <- base64decode(encoded_data)

# Step 2: Decompress the GZIP bytes
# Create a connection to the raw bytes, wrap it in a GZIP handler
con <- rawConnection(raw_compressed, open = "rb")
decompressed_text <- readLines(gzcon(con), warn = FALSE)
close(con)

# Check the result
print(decompressed_text) # Output: "who are you"

If you're working with non-text binary data instead of a string, replace readLines() with readBin() to handle the raw bytes appropriately.

内容的提问来源于stack exchange，提问作者Jose