# tabular Parsing and formatting of tabular text data such as comma- and delimiter-separated values (CSV and DSV). ## Documentation The goal of the tabular library is to provide means of reading and writing tabular text data, such as comma- and delimiter-separated values (CSV and DSV), as well as fixed-width columns. ## Procedures ### Readers `reader:: INPUT [DELIMITER: #\,] [EOL: "\n"] [COMMENT-CHAR: #f] [NA: #f] [COLUMN-NAMES: #f] [COLUMN-WIDTHS: #f] [RETURN-NAMES: #f] -> READER STREAM` Constructs a reader procedure for the given input format specification and returns two values: the reader procedure and the input stream. The reader procedure is of the form `LAMBDA STRM -> RECORD STRM`. It reads one record from the given input stream and by default returns the record as a list of values and the remainder of the input stream. If optional argument `RETURN-NAMES` is true, the record will be returned as an alist where the keys are column names (if `COLUMN-NAMES` is provided) or indices, and the values are the column values. - Argument `INPUT` is a string or a port. - Argument `DELIMITER` specifies the delimiter character (default is `#\,`). If its value is #f, then `COLUMNN-WIDTHS` must be specified. - Argument `EOL` specifies the end-of-line string (default is `"\n"`). - Argument `COMMENT-CHAR` specifies a comment character: the reader will skip all lines that start with this character (default: none). - Argument `NA` specifies a default value to be supplied when a given field is empty (default: none). - Argument `COLUMN-NAMES` specifies column names which will be used if `RETURN-NAMES` is true (default: none). If `COLUMN-NAMES` is the symbol `'header` then the first line of the input will be used to infer the column names. - Argument `COLUMN-WIDTHS` specifies column widths which will be used if `DELIMITER` is not specified (default: none). - Argument `RETURN-NAMES` specifies that each returned record will have the form of an alist with column name and value pairs. If `COLUMN-NAMES` is not provided, the column index will be used as key instead. `reader*:: INPUT [DELIMITER: #\,] [EOL: "\n"] [COMMENT-CHAR: #f] [NA: #f] [COLUMN-NAMES: #f] [COLUMN-WIDTHS: #f] [RETURN-NAMES: #f] -> READER` Constructs a reader generator procedure for the given input format specification and returns the reader procedure. The reader procedure is of the form `LAMBDA () -> RECORD`. It reads one record from the given input stream and by default returns the record as a list of values. If optional argument `RETURN-NAMES` is true, the record will be returned as an alist where the keys are column names (if `COLUMN-NAMES` is provided) or indices, and the values are the column values. - Argument `INPUT` is a string or a port. - Argument `DELIMITER` specifies the delimiter character (default is `#\,`). If its value is #f, then `COLUMNN-WIDTHS` must be specified. - Argument `EOL` specifies the end-of-line string (default is `"\n"`). - Argument `COMMENT-CHAR` specifies a comment character: the reader will skip all lines that start with this character (default: none). - Argument `NA` specifies a default value to be supplied when a given field is empty (default: none). - Argument `COLUMN-NAMES` specifies column names which will be used if `RETURN-NAMES` is true (default: none). If `COLUMN-NAMES` is the symbol `'header` then the first line of the input will be used to infer the column names. - Argument `COLUMN-WIDTHS` specifies column widths which will be used if `DELIMITER` is not specified (default: none). - Argument `RETURN-NAMES` specifies that each returned record will have the form of an alist with column name and value pairs. If `COLUMN-NAMES` is not provided, the column index will be used as key instead. ### Writers ` writer:: OUTPUT [DELIMITER: #\,] [COLUMN-WIDTHS: #f] [ENDLINE: "\n"] -> WRITER ` Constructs a writer procedure that writes out records according to the given output specification. The writer procedure is of the form `LAMBDA RECORDS -> UNIT`. Each record is represented as a list. - Argument `OUTPUT` is a string file name or a port. - Argument `DELIMITER` specifies the delimiter character (default is `#\,`). If its value is #f, then `COLUMNN-WIDTHS` must be specified. - Argument `ENDLINE` specifies the end-of-line string (default is `"\n"`). - Argument `COLUMN-WIDTHS` specifies column widths which will be used if `DELIMITER` is not specified (default: none). ## Examples ```scheme (import tabular) (define (stream->list proc strm) (let recur ((ax '()) (strm strm)) (let ((value.strm (proc strm))) (if (eof-object? value.strm) (reverse ax) (recur (cons (car value.strm) ax) (cadr value.strm)) )) )) ;; Uses a reader generator to print all records in the given input string (let-values (((proc strm) (call-with-input-string "\"Test \n1\"|Test 2|Test 3\nTest 4|Test 5\n" (lambda (port) (reader port delimiter: #\|))))) (print (stream->list proc strm))) ;; Prints all the records from the given file, where column names are inferred from the first line ;; and each field in a record is prefixed by its column name (let ((input (open-input-file "file.csv"))) (let-values (((proc strm) (reader input delimiter: #\, column-names: 'header return-names: #t))) (let ((lst (stream->list proc strm))) (close-input-port input) (pretty-print lst) )) ) ;; Prints the first record of the given file (let ((input (open-input-file "file.csv")) (gen (reader* input delimiter: #\,))) (print (gen)) ) ;; Writes the given records to a string port (call-with-output-string (lambda (port) (let ((out (writer port))) (out (list "Test 1" "Test 2" "Test 3") (list "Test 4" "Test 5" ))) )) ``` ## Version history - 1.3 : Fix for return-names [thanks to Vasilij Schneidermann for reporting] - 1.2 : Fix for escaped double quotes [thanks to Matt Welland for reporting] - 1.0 : Initial release ## License > > Copyright 2019-2021 Ivan Raikov > > This program is free software: you can redistribute it and/or modify > it under the terms of the GNU General Public License as published by > the Free Software Foundation, either version 3 of the License, or (at > your option) any later version. > > This program is distributed in the hope that it will be useful, but > WITHOUT ANY WARRANTY; without even the implied warranty of > MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU > General Public License for more details. > > A full copy of the GPL license can be found at > .