File I/O

## Learning Objectives - Read and write CSV files - Work with Excel files - Handle text files - Work with R data formats ## Working Directory ### Getting/Setting Directory ```r # Get current directory getwd() # Set working directory setwd("/path/to/directory") # RStudio shortcut: Session -> Set Working Directory ``` ### Relative Paths ```r # Within project directory read.csv("data/file.csv") # relative read.csv("/home/user/data/file.csv") # absolute # Check if file exists file.exists("data.csv") ``` ## CSV Files ### Reading CSV ```r # Base R read.csv("file.csv") read.csv("file.csv", header = TRUE) read.csv("file.csv", sep = ",") # With strings as characters (not factors) read.csv("file.csv", stringsAsFactors = FALSE) # Skip rows read.csv("file.csv", skip = 2) # Read first N rows read.csv("file.csv", nrows = 100) # NA strings read.csv("file.csv", na.strings = c("NA", "", "N/A")) ``` ### Reading with readr (tidyverse) ```r library(readr) # Read CSV read_csv("file.csv") # Read TSV read_tsv("file.tsv") # Read delimited read_delim("file.txt", delim = "|") ``` ### readr Options ```r library(readr) # Specify column types read_csv("file.csv", col_types = cols( name = col_character(), age = col_integer(), score = col_double() )) # Skip lines read_csv("file.csv", skip = 2) # No header read_csv("file.csv", col_names = FALSE) # Custom column names read_csv("file.csv", col_names = c("Name", "Age", "Score")) # File from URL read_csv("https://example.com/data.csv") ``` ### Writing CSV ```r # Base R write.csv(df, "output.csv") write.csv(df, "output.csv", row.names = FALSE) # Overwrite without quotes write.csv(df, "output.csv", quote = FALSE, row.names = FALSE) # With append write.csv(df, "output.csv", append = TRUE) ``` ### Writing with readr ```r library(readr) write_csv(df, "output.csv") write_tsv(df, "output.tsv") ``` ## Excel Files ### Reading Excel ```r # Install readxl package install.packages("readxl") library(readxl) # Read Excel file read_excel("file.xlsx") read_excel("file.xlsx", sheet = 1) read_excel("file.xlsx", sheet = "SheetName") # Read all sheets excel_sheets("file.xlsx") list_frames <- lapply(excel_sheets("file.xlsx"), read_excel, path = "file.xlsx") ``` ### Writing Excel ```r # Install writexl package install.packages("writexl") library(writexl) write_xlsx(df, "output.xlsx") # Multiple dataframes list_df <- list("Sheet1" = df1, "Sheet2" = df2) write_xlsx(list_df, "output.xlsx") ``` ### openxlsx Package ```r install.packages("openxlsx") library(openxlsx) # Write with formatting wb <- createWorkbook() addWorksheet(wb, "Data") writeData(wb, "Data", df) saveWorkbook(wb, "output.xlsx", overwrite = TRUE) # Read with formatting preserved read.xlsx("file.xlsx") ``` ## Text Files ### Reading Text ```r # Read entire file as lines lines <- readLines("file.txt") head(lines, 10) # Read delimited text read.table("file.txt", sep = "\t") # Scan (fast reading of numeric data) numbers <- scan("numbers.txt", what = numeric()) ``` ### Writing Text ```r # Write lines writeLines(c("Line 1", "Line 2", "Line 3"), "output.txt") # Append lines writeLines(c("Line 4"), "output.txt", append = TRUE) # cat (print to file) cat("Hello\n", file = "output.txt") cat("World\n", file = "output.txt", append = TRUE) ``` ## R Data Formats ### .RData files ```r # Save workspace save.image("my_session.RData") # Save specific objects save(df, model, file = "my_data.RData") # Load load("my_session.RData") ``` ### .rds files ```r # Save single object (more efficient) saveRDS(df, "df.rds") # Load df <- readRDS("df.rds") ``` ### save vs saveRDS ```r # save preserves names, loads to same names save(df, file = "df.RData") load("df.RData") # df appears in environment # saveRDS returns object, you assign it df <- readRDS("df.Rds") ``` ## File Operations ### Check File Info ```r # Check if file exists file.exists("file.csv") # File info file.info("file.csv") # size isdir mode mtime ctime atime exe # 12345 FALSE "rw-r--" ... # Get file extension tools::file_ext("file.csv") # "csv" # Get file without extension tools::file_path_sans_ext("file.csv") # "file" ``` ### File Paths ```r # Build paths file.path("dir", "subdir", "file.csv") # Normalize path normalizePath("~/file.csv") # basename and dirname basename("/path/to/file.csv") # "file.csv" dirname("/path/to/file.csv") # "/path/to" ``` ### Create/Delete Files ```r # Create directory dir.create("new_directory") # Create nested directories dir.create("a/b/c", recursive = TRUE) # Copy file file.copy("source.csv", "destination.csv") # Delete file file.remove("file.csv") # Rename file file.rename("old.csv", "new.csv") ``` ## Connection Interfaces ### Reading from URL ```r # Read from URL read.csv("https://example.com/data.csv") # Read from gzipped file read.csv(gzfile("data.csv.gz")) # Read from clipboard # Windows read.csv("clipboard") # Mac read.csv(pipe("pbpaste")) ``` ### Reading Large Files ```r # readr for large files library(readr) # Progress bar read_csv("large_file.csv", progress = TRUE) # Specify n_max to preview read_csv("large_file.csv", n_max = 1000) ``` ## Data Import Best Practices ### Import Checklist ```r # 1. Check file structure first readLines("file.csv", n = 5) # 2. Read with appropriate function # 3. Check structure with str() # 4. Convert types as needed # 5. Handle missing values ``` ### Common Import Issues ```r # Extra header row read.csv("file.csv", header = FALSE) # then remove # Wrong delimiter read.delim("file.txt", sep = "\t") # Encoding issues read.csv("file.csv", encoding = "UTF-8") # Trailing whitespace read.csv("file.csv", strip.white = TRUE) ``` ## Saving Plots ### Save as Image ```r # PNG png("plot.png", width = 800, height = 600) ggplot(df, aes(x = x, y = y)) + geom_point() dev.off() # JPEG jpeg("plot.jpg", width = 800, height = 600, quality = 90) # PDF pdf("plot.pdf", width = 10, height = 8) ``` ### Save with ggsave ```r library(ggplot2) p <- ggplot(df, aes(x = x, y = y)) + geom_point() ggsave("plot.png", p, width = 10, height = 8, dpi = 300) ggsave("plot.pdf", p, width = 10, height = 8) ggsave("plot.svg", p) ``` ## Summary - `getwd()` and `setwd()` manage working directory - `read.csv()` / `read_csv()` for CSV files - readxl package for Excel files - `readLines()` for text files - `saveRDS()` / `readRDS()` for single R objects - `save()` / `load()` for multiple R objects - `write.csv()` / `write_csv()` for writing CSV - `file.exists()` checks if file exists - `file.path()` builds paths safely - `ggsave()` for saving ggplot2 plots