This module implements a simple high performance
CSV (
comma separated value) parser.
import os, parsecsv, streams
var s = newFileStream(ParamStr(1), fmRead)
if s == nil: quit("cannot open the file" & ParamStr(1))
var x: TCsvParser
open(x, s, ParamStr(1))
while readRow(x):
Echo "new row: "
for val in items(x.row):
Echo "##", val, "##"
close(x)
TCsvRow = seq[string]
-
a row in a CSV file
TCsvParser = object of TBaseLexer
row*: TCsvRow
filename: string
sep, quote, esc: char
skipWhite: bool
currRow: int
-
the parser object.
EInvalidCsv = object of EIO
-
exception that is raised if a parsing error occurs
proc open(my: var TCsvParser; input: PStream; filename: string; separator = ',';
quote = '\"'; escape = '\0'; skipInitialSpace = false) {.
raises: [E_Base], tags: [FReadIO].}-
initializes the parser with an input stream. Filename is only used for nice error messages. The parser's behaviour can be controlled by the diverse optional parameters:
- separator: character used to separate fields
- quote: Used to quote fields containing special characters like separator, quote or new-line characters. '0' disables the parsing of quotes.
- escape: removes any special meaning from the following character; '0' disables escaping; if escaping is disabled and quote is not '0', two quote characters are parsed one literal quote character.
- skipInitialSpace: If true, whitespace immediately following the separator is ignored.
proc processedRows(my: var TCsvParser): int {.raises: [], tags: [].}-
returns number of the processed rows
proc readRow(my: var TCsvParser; columns = 0): bool {.
raises: [EInvalidCsv, E_Base], tags: [FReadIO].}-
reads the next row; if columns > 0, it expects the row to have exactly this many columns. Returns false if the end of the file has been encountered else true.
proc close(my: var TCsvParser) {.inline, raises: [E_Base], tags: [].}-
closes the parser my and its associated input stream.