Look for and extract a table anywhere within a CSV file.

Searches a CSV file for a table using one of two strategies:

By locating a marker row (e.g., "TABLE: RESULTS"), after which the next non-empty row is treated as the header.
By locating a header row that contains a known column name (start_column), analogous to extract_table() for Excel files.

Exactly one of table_marker or start_column must be supplied.

Usage

extract_table_csv(
  path,
  table_marker = NULL,
  start_column = NULL,
  header_regex = NULL,
  ignore_case = TRUE,
  exact_match = TRUE,
  max_blank_rows = Inf,
  na_values = c("", NA),
  safely = TRUE,
  id_cols = FALSE,
  verbose = FALSE,
  ...
)

Arguments

path: Path to CSV file
table_marker: Optional string or regex identifying the table marker row
start_column: Optional name of a known header column
header_regex: Optional regex to validate header row contents
ignore_case: Logical; ignore case when matching text
exact_match: Logical; exact matching for table_marker. Default TRUE.
max_blank_rows: Number of blank rows tolerated between marker and header. Default Inf.
na_values: Values treated as NA
safely: Logical; return NULL instead of error on failure
id_cols: Logical; add file/table metadata columns
verbose: Logical; emit detailed progress messages
...: Unused (for interface compatibility)

Value

A tibble, or NULL if safely = TRUE and no table found