我有一个实现作为电子表格项目的一部分。

此代码尚未经过全面测试，但欢迎任何人使用它。

正如一些答案所指出的那样，如果您实际上有DSV或TSV文件，您的实现可以简单得多，因为它们不允许在值中使用记录和字段分隔符。另一方面，CSV实际上可以在字段中使用逗号和换行符，这打破了大多数正则表达式和基于分割的方法。

var CSV = {
    parse: function(csv, reviver) {
        reviver = reviver || function(r, c, v) { return v; };
        var chars = csv.split(''), c = 0, cc = chars.length, start, end, table = [], row;
        while (c < cc) {
            table.push(row = []);
            while (c < cc && '\r' !== chars[c] && '\n' !== chars[c]) {
                start = end = c;
                if ('"' === chars[c]){
                    start = end = ++c;
                    while (c < cc) {
                        if ('"' === chars[c]) {
                            if ('"' !== chars[c+1]) {
                                break;
                            }
                            else {
                                chars[++c] = ''; // unescape ""
                            }
                        }
                        end = ++c;
                    }
                    if ('"' === chars[c]) {
                        ++c;
                    }
                    while (c < cc && '\r' !== chars[c] && '\n' !== chars[c] && ',' !== chars[c]) {
                        ++c;
                    }
                } else {
                    while (c < cc && '\r' !== chars[c] && '\n' !== chars[c] && ',' !== chars[c]) {
                        end = ++c;
                    }
                }
                row.push(reviver(table.length-1, row.length, chars.slice(start, end).join('')));
                if (',' === chars[c]) {
                    ++c;
                }
            }
            if ('\r' === chars[c]) {
                ++c;
            }
            if ('\n' === chars[c]) {
                ++c;
            }
        }
        return table;
    },

    stringify: function(table, replacer) {
        replacer = replacer || function(r, c, v) { return v; };
        var csv = '', c, cc, r, rr = table.length, cell;
        for (r = 0; r < rr; ++r) {
            if (r) {
                csv += '\r\n';
            }
            for (c = 0, cc = table[r].length; c < cc; ++c) {
                if (c) {
                    csv += ',';
                }
                cell = replacer(r, c, table[r][c]);
                if (/[,\r\n"]/.test(cell)) {
                    cell = '"' + cell.replace(/"/g, '""') + '"';
                }
                csv += (cell || 0 === cell) ? cell : '';
            }
        }
        return csv;
    }
};

只需使用.split('，'):

var str = "How are you doing today?";
var n = str.split(" ");

正则表达式拯救你!这几行代码根据RFC 4180标准处理带有嵌入逗号、引号和换行符的正确引用字段。

function parseCsv(data, fieldSep, newLine) {
    fieldSep = fieldSep || ',';
    newLine = newLine || '\n';
    var nSep = '\x1D';
    var qSep = '\x1E';
    var cSep = '\x1F';
    var nSepRe = new RegExp(nSep, 'g');
    var qSepRe = new RegExp(qSep, 'g');
    var cSepRe = new RegExp(cSep, 'g');
    var fieldRe = new RegExp('(?<=(^|[' + fieldSep + '\\n]))"(|[\\s\\S]+?(?<![^"]"))"(?=($|[' + fieldSep + '\\n]))', 'g');
    var grid = [];
    data.replace(/\r/g, '').replace(/\n+$/, '').replace(fieldRe, function(match, p1, p2) {
        return p2.replace(/\n/g, nSep).replace(/""/g, qSep).replace(/,/g, cSep);
    }).split(/\n/).forEach(function(line) {
        var row = line.split(fieldSep).map(function(cell) {
            return cell.replace(nSepRe, newLine).replace(qSepRe, '"').replace(cSepRe, ',');
        });
        grid.push(row);
    });
    return grid;
}

const csv = 'A1,B1,C1\n"A ""2""","B, 2","C\n2"';
const separator = ',';      // field separator, default: ','
const newline = ' <br /> '; // newline representation in case a field contains newlines, default: '\n' 
var grid = parseCsv(csv, separator, newline);
// expected: [ [ 'A1', 'B1', 'C1' ], [ 'A "2"', 'B, 2', 'C <br /> 2' ] ]

您不需要像lex/yacc这样的解析器-生成器。正则表达式可以正确地处理RFC 4180，这要归功于正向向后查找、反向向后查找和正向向前查找。

克隆/下载代码https://github.com/peterthoeny/parse-csv-js

我不知道为什么我不能让Kirtan的例子对我有用。它似乎在空字段或带尾随逗号的字段上失败了……

这个似乎可以同时处理这两个问题。

我没有编写解析器代码，只是对解析器函数进行了包装，以使其适用于文件。看到归因。

    var Strings = {
        /**
         * Wrapped CSV line parser
         * @param s      String delimited CSV string
         * @param sep    Separator override
         * @attribution: http://www.greywyvern.com/?post=258 (comments closed on blog :( )
         */
        parseCSV : function(s,sep) {
            // http://stackoverflow.com/questions/1155678/javascript-string-newline-character
            var universalNewline = /\r\n|\r|\n/g;
            var a = s.split(universalNewline);
            for(var i in a){
                for (var f = a[i].split(sep = sep || ","), x = f.length - 1, tl; x >= 0; x--) {
                    if (f[x].replace(/"\s+$/, '"').charAt(f[x].length - 1) == '"') {
                        if ((tl = f[x].replace(/^\s+"/, '"')).length > 1 && tl.charAt(0) == '"') {
                            f[x] = f[x].replace(/^\s*"|"\s*$/g, '').replace(/""/g, '"');
                          } else if (x) {
                        f.splice(x - 1, 2, [f[x - 1], f[x]].join(sep));
                      } else f = f.shift().split(sep).concat(f);
                    } else f[x].replace(/""/g, '"');
                  } a[i] = f;
        }
        return a;
        }
    }

下面是我的PEG(.js)语法，它在RFC 4180中似乎做得不错(即它处理http://en.wikipedia.org/wiki/Comma-separated_values):上的示例)

start
  = [\n\r]* first:line rest:([\n\r]+ data:line { return data; })* [\n\r]* { rest.unshift(first); return rest; }

line
  = first:field rest:("," text:field { return text; })*
    & { return !!first || rest.length; } // ignore blank lines
    { rest.unshift(first); return rest; }

field
  = '"' text:char* '"' { return text.join(''); }
  / text:[^\n\r,]* { return text.join(''); }

char
  = '"' '"' { return '"'; }
  / [^"]

在http://jsfiddle.net/knvzk/10或http://pegjs.majda.cz/online上试试吧。从https://gist.github.com/3362830下载生成的解析器。

我已经构造了这个JavaScript脚本来解析字符串到数组对象中的CSV。我发现最好将整个CSV分解成行、字段并相应地处理它们。我认为这将使您更容易更改代码以满足您的需要。

    //
    //
    // CSV to object
    //
    //

    const new_line_char = '\n';
    const field_separator_char = ',';

    function parse_csv(csv_str) {

        var result = [];

        let line_end_index_moved = false;
        let line_start_index = 0;
        let line_end_index = 0;
        let csr_index = 0;
        let cursor_val = csv_str[csr_index];
        let found_new_line_char = get_new_line_char(csv_str);
        let in_quote = false;

        // Handle \r\n
        if (found_new_line_char == '\r\n') {
            csv_str = csv_str.split(found_new_line_char).join(new_line_char);
        }
        // Handle the last character is not \n
        if (csv_str[csv_str.length - 1] !== new_line_char) {
            csv_str += new_line_char;
        }

        while (csr_index < csv_str.length) {
            if (cursor_val === '"') {
                in_quote = !in_quote;
            } else if (cursor_val === new_line_char) {
                if (in_quote === false) {
                    if (line_end_index_moved && (line_start_index <= line_end_index)) {
                        result.push(parse_csv_line(csv_str.substring(line_start_index, line_end_index)));
                        line_start_index = csr_index + 1;
                    } // Else: just ignore line_end_index has not moved or line has not been sliced for parsing the line
                } // Else: just ignore because we are in a quote
            }
            csr_index++;
            cursor_val = csv_str[csr_index];
            line_end_index = csr_index;
            line_end_index_moved = true;
        }

        // Handle \r\n
        if (found_new_line_char == '\r\n') {
            let new_result = [];
            let curr_row;
            for (var i = 0; i < result.length; i++) {
                curr_row = [];
                for (var j = 0; j < result[i].length; j++) {
                    curr_row.push(result[i][j].split(new_line_char).join('\r\n'));
                }
                new_result.push(curr_row);
            }
            result = new_result;
        }
        return result;
    }

    function parse_csv_line(csv_line_str) {

        var result = [];

        //let field_end_index_moved = false;
        let field_start_index = 0;
        let field_end_index = 0;
        let csr_index = 0;
        let cursor_val = csv_line_str[csr_index];
        let in_quote = false;

        // Pretend that the last char is the separator_char to complete the loop
        csv_line_str += field_separator_char;

        while (csr_index < csv_line_str.length) {
            if (cursor_val === '"') {
                in_quote = !in_quote;
            } else if (cursor_val === field_separator_char) {
                if (in_quote === false) {
                    if (field_start_index <= field_end_index) {
                        result.push(parse_csv_field(csv_line_str.substring(field_start_index, field_end_index)));
                        field_start_index = csr_index + 1;
                    } // Else: just ignore field_end_index has not moved or field has not been sliced for parsing the field
                } // Else: just ignore because we are in quote
            }
            csr_index++;
            cursor_val = csv_line_str[csr_index];
            field_end_index = csr_index;
            field_end_index_moved = true;
        }
        return result;
    }

    function parse_csv_field(csv_field_str) {
        with_quote = (csv_field_str[0] === '"');

        if (with_quote) {
            csv_field_str = csv_field_str.substring(1, csv_field_str.length - 1); // remove the start and end quotes
            csv_field_str = csv_field_str.split('""').join('"'); // handle double quotes
        }
        return csv_field_str;
    }

    // Initial method: check the first newline character only
    function get_new_line_char(csv_str) {
        if (csv_str.indexOf('\r\n') > -1) {
            return '\r\n';
        } else {
            return '\n'
        }
    }