I have an Array that I need to sort exactly like using order by in Oracle SQl.
If I have following Array:
var array = ['Ba12nes','Apfel','Banane','banane','abc','ABC','123','2', null,'ba998ne']
var.sort(compare);
I would like to have the following result
var array = ['abc','ABC','Apfel','banane','Banane','Ba12nes','ba998ne','123','2', null]
If the null values are somewhere else, I don't have a Problem with it.
My current solution, which does not help me ^^
function compare(a,b) {
if(a == null)
return -1;
if (b == null)
return 1;
if (a.toLowerCase() < b.toLowerCase())
return -1;
if (a.toLowerCase() > b.toLowerCase())
return 1;
return 0;
}
I do understand that i need a custom sorting function. And at the moment I am thinking that only a regular expression can solve the problem of sorting the string values in front of the numbers. But I am still not sure how to solve the Problem with lowercase letters in bevor Uppercase letters.
Iirc, Oracle implements a 3-tiered lexicographic sorting (but heed the advice of Alex Poole and check the NLS settings first):
First sort by base characters ignoring case and diacritics, digits come after letters in the collation sequence.
Second, on ties sort respecting diacritics, ignoring case.
Third, on ties sort by case.
You can emulate the behavior using javascript locale apis by mimicking each step in turn in a custom compare function, with the exception of the letter-digit inversion in the collation sequence.
Tackle the latter by identifying 10 contiguous code points that do not represent digits and that lie beyond the set of code points that may occur in the strings you are sorting. Map digits onto the the chosen code point range preserving order. When you sort, specify the Unicode collating extension 'direct' which means 'sorting by code point'. Remap after sorting.
In the PoC code below I have chosen some cyrillic characters.
function cmptiered(a,b) {
//
// aka oracle sort
//
return lc_base.compare(a, b) || lc_accent.compare(a, b) || lc_case.compare(a, b);
} // cmptiered
var lc_accent = new Intl.Collator('de', { sensitivity: 'accent' });
var lc_base = new Intl.Collator('de-DE-u-co-direct', { sensitivity: 'base' });
var lc_case = new Intl.Collator('de', { caseFirst: 'lower', sensitivity: 'variant' });
var array = ['Ba12nes','Apfel','Banane','banane','abc','ABC','123','2', null, 'ba998ne' ];
// Map onto substitute code blocks
array = array.map ( function ( item ) { return (item === null) ? null : item.replace ( /[0-9]/g, function (c) { return String.fromCharCode(c.charCodeAt(0) - "0".charCodeAt(0) + "\u0430".charCodeAt(0)); } ); } );
array.sort(cmptiered);
// Remap substitute code point
array = array.map ( function ( item ) { return (item === null) ? null : item.replace ( /[\u0430-\u0439]/g, function (c) { return String.fromCharCode(c.charCodeAt(0) - "\u0430".charCodeAt(0) + "0".charCodeAt(0)); } ); } );
Edit
Function cmptiered streamlined following Nina Scholz' comment.
This proposals feature sorting without use of Intl.Collator. The first solution works with direct sort and comparing the given values.
var array = ['Ba12nes', 'Apfel', 'Banane', 'banane', 'abc', 'ABC', '123', '2', null, 'ba998ne'];
array.sort(function (a, b) {
var i = 0;
if (a === null && b === null) { return 0; }
if (a === null) { return 1; }
if (b === null) { return -1; }
while (i < a.length && i < b.length && a[i].toLocaleLowerCase() === b[i].toLocaleLowerCase()) {
i++;
}
if (isFinite(a[i]) && isFinite(b[i])) { return a[i] - b[i]; }
if (isFinite(a[i])) { return 1; }
if (isFinite(b[i])) { return -1; }
return a.localeCompare(b);
});
document.write(JSON.stringify(array));
The second solution features a different approach, based on Sorting with map and a custom sort scheme which takes a new string. The string is build by this rules:
If the value is null take the string 'null'.
If a character is a decimal, takes the character with space paddded around, eg. if it is 9 take the string ' 9 '.
Otherwise for every other character take two spaces and the character itself, like ' o'.
The new build string is used with a a.value.localeCompare(b.value).
Here are the strings with the mapped values:
' B a 1 2 n e s'
' A p f e l'
' B a n a n e'
' b a n a n e'
' a b c'
' A B C'
' 1 2 3 '
' 2 '
'null'
' b a 9 9 8 n e'
sorted, it became
' a b c'
' A B C'
' A p f e l'
' b a n a n e'
' B a n a n e'
' B a 1 2 n e s'
' b a 9 9 8 n e'
' 1 2 3 '
' 2 '
'null'
var array = ['Ba12nes', 'Apfel', 'Banane', 'banane', 'abc', 'ABC', '123', '2', null, 'ba998ne'],
mapped = array.map(function (el, i) {
var j, o = { index: i, value: '' };
if (el === null) {
o.value = 'null';
return o;
}
for (j = 0; j < el.length; j++) {
o.value += /\d/.test(el[j]) ? ' ' + el[j] + ' ' : ' ' + el[j];
}
return o;
});
mapped.sort(function (a, b) {
return a.value.localeCompare(b.value);
});
var result = mapped.map(function (el) {
return array[el.index];
});
document.write(JSON.stringify(result));
A simple head on solution that works at least for english & russian (mimicking NLS_SORT=RUSSIAN) and doesn't rely on fancy things like Intl.Collator, locales and options that don't exist for IE<11.
function compareStringOracle(str1, str2) {
if (str1 == null && str2 != null)
return 1;
else if (str1 != null && str2 == null)
return -1;
else if (str1 == null && str2 == null)
return 0;
else {
return compareStringCaseInsensitiveDigitsLast(str1, str2) ||
/* upper case wins between otherwise equal values, which can be checked with
a simple binary comparison (at least for eng & rus) */
((str1 < str2) ? -1 : (str1 > str2) ? 1 : 0);
}
}
function compareStringCaseInsensitiveDigitsLast(str1, str2) {
for (var i = 0; i < str1.length; ++i) {
if (i === str2.length)
return 1;
// toLocaleLowerCase is unnecessary for eng & rus
var c1 = str1.charAt(i).toLowerCase();
var c2 = str2.charAt(i).toLowerCase();
var d1 = "0" <= c1 && c1 <= "9";
var d2 = "0" <= c2 && c2 <= "9";
if (!d1 && d2)
return -1;
else if (d1 && !d2)
return 1;
else if (c1 !== c2)
return (c1 < c2) ? -1 : (c1 > c2) ? 1 : 0;
}
if (str1.length < str2.length)
return -1;
else
return 0;
}
I have a string that contains a create query, and I'm trying to split it based on commas. Unfortunately, some of the lines have commas in them surrounded by parenthesis.
Example:
dbo.Display_Test1.Column,
CASE WHEN CHARINDEX(' To ', Test3) > 0 AND CHARINDEX('XVR', Test3) = 0 THEN LEFT(Test3, (CHARINDEX(' To ', Test3) - 12)) ELSE Test3 END AS Test3,
dbo.Display_Test2.Column,
ISNULL((CASE WHEN [2-Display-Test4].[Total Number] > 0 AND [1-Display-Test5].SumOfNumber = 0 THEN 0 ELSE (([2-Display-Test4].[Total Number] * 1000) / [1-Display-Test5].SumOfNumber) END), 0) AS Test6,
I want to split my string based on commas that aren't inside of (possibly multiple) parenthesis. For reference, my example should be split where the line breaks are (although my string doesn't have the line breaks in it).
I've tried a number of different solutions, but none work quite right:
/(?:\(*[^()]*\)|[^,])+/g works on lines 1,3, and 4, but fails on line 2. It breaks up the line into multiple matches.
/((?:[^,(]+|(\((?:[^()]+)|$1\)))+)/g works on lines 1,2, and 3, but fails on line 4. It also breaks up the line into multiple matches.
I can't quite seem to make it work. Any help is appreciated.
One solution possible : ( note : we remove the separations comas )
var str = "dbo.Display_Test1.Column, CASE WHEN CHARINDEX(' To ', Test3) > 0 AND CHARINDEX('XVR', Test3) = 0 THEN LEFT(Test3, (CHARINDEX(' To ', Test3) - 12)) ELSE Test3 END AS Test3, dbo.Display_Test2.Column, ISNULL((CASE WHEN [2-Display-Test4].[Total Number] > 0 AND [1-Display-Test5].SumOfNumber = 0 THEN 0 ELSE (([2-Display-Test4].[Total Number] * 1000) / [1-Display-Test5].SumOfNumber) END), 0) AS Test6,";
// if the string don't end by a coma , we add it.
var strArr = str.replace(/,*\s*$/ , ',').split('');
var res;
res = strArr.reduce(function( trans , charValue ){
if(charValue === '(') {
trans.deep++;
}
if(charValue === ')') {
trans.deep--;
}
if( trans.deep === 0){
if(charValue===',') {
trans.arr.push( trans.str);
trans.str = '';
}else{
trans.str += charValue;
}
}else{
trans.str += charValue;
}
return trans;
}, { arr : [] , str : '' ,deep : 0});
document.write('<pre>' + JSON.stringify(res.arr , null , ' ') + '</pre>');
Edit :
As suggested by Thriggle in the comments i've added the case where the string don't end by a coma.
changed variables name .
Of course if we want to work with unlimited nesting of parentheses then it would be impossible to avoid a loop. Still for some bounded nesting we do can write a regular expression capturing it.
For up to 1 parentheses level it could be
/((?:[^(),]|\([^()]*\))+)/g
For up to 2 parentheses level (probably it's your case)
/((?:[^(),]|\((?:[^()]|\([^()]*\))*\))+)/g
And so on, recursively substituting [^()]* with (?:[^()]|\([^()]*\))*
Regex won't get you where you want and still allow arbitrary complexity.
One approach is to iterate through each character in the string, keeping track of when you hit open and close parentheses, and only splitting on commas when the numbers of open and close parentheses cancel each other out (so you know you're not within a parenthetical statement).
function splitQuery(input){
var arr = [];
var lastStart = 0;
var open = 0;
for(var i = 0, len = input.length; i < len; i++){
var curr = input[i];
if(curr === "("){open += 1;}
else if(curr === ")"){ open = open < 1 ? 0 :open -=1;}
else if(curr === ","){
if(open === 0){
arr.push(input.substring(lastStart,i));
lastStart = i+1;
}
}else if(i+1 === len){
arr.push(input.substring(lastStart,i+1));
}
}
return arr;
}
Just be aware that this approach can be expensive (from a performance perspective) when dealing with especially large strings.
Here's a working example:
var query = "dbo.Display_Test1.Column, CASE WHEN CHARINDEX(' To ', Test3) > 0 AND CHARINDEX('XVR', Test3) = 0 THEN LEFT(Test3, (CHARINDEX(' To ', Test3) - 12)) ELSE Test3 END AS Test3, dbo.Display_Test2.Column, ISNULL((CASE WHEN [2-Display-Test4].[Total Number] > 0 AND [1-Display-Test5].SumOfNumber = 0 THEN 0 ELSE (([2-Display-Test4].[Total Number] * 1000) / [1-Display-Test5].SumOfNumber) END), 0) AS Test6,";
var split = splitQuery(query);
var output = document.getElementById("output")
for(var el in split){
output.insertAdjacentHTML("beforeend",el + ": "+ split[el]+"<hr/>");
}
function splitQuery(input){
var arr = [];
var lastStart = 0;
var open = 0;
for(var i = 0, len = input.length; i < len; i++){
var curr = input[i];
if(curr === "("){open += 1;}
else if(curr === ")"){ open = open < 1 ? 0 :open -=1;}
else if(curr === ","){
if(open === 0){
arr.push(input.substring(lastStart,i));
lastStart = i+1;
}
}else if(i+1 === len){
arr.push(input.substring(lastStart,i+1));
}
}
return arr;
}
<div id="output"></div>
Is there a reg exp or function that will convert camel case, css and underscore to human readable format? It does not need to support non-humans at this time. Sorry aliens. :(
Examples:
helloWorld -> "Hello World"
hello-world -> "Hello World"
hello_world -> "Hello World"
Split by non-words; capitalize; join:
function toCapitalizedWords(name) {
var words = name.match(/[A-Za-z][a-z]*/g) || [];
return words.map(capitalize).join(" ");
}
function capitalize(word) {
return word.charAt(0).toUpperCase() + word.substring(1);
}
Extract all words with a regular expression. Capitalize them. Then, join them with spaces.
Example regexp:
/^[a-z]+|[A-Z][a-z]*/g
/ ^[a-z]+ // 1 or more lowercase letters at the beginning of the string
| // OR
[A-Z][a-z]* // a capital letter followed by zero or more lowercase letters
/g // global, match all instances
Example function:
var camelCaseToWords = function(str){
return str.match(/^[a-z]+|[A-Z][a-z]*/g).map(function(x){
return x[0].toUpperCase() + x.substr(1).toLowerCase();
}).join(' ');
};
camelCaseToWords('camelCaseString');
// Camel Case String
camelCaseToWords('thisIsATest');
// This Is A Test
Here is the ActionScript version based on the idea from Ricks C example code. For JavaScript version remove the strong typing. For example, change var value:String to var value. Basically remove any declaration that starts with a semicolon, :String, :int, etc.
/**
* Changes camel case to a human readable format. So helloWorld, hello-world and hello_world becomes "Hello World".
* */
public static function prettifyCamelCase(value:String=""):String {
var output:String = "";
var len:int = value.length;
var char:String;
for (var i:int;i<len;i++) {
char = value.charAt(i);
if (i==0) {
output += char.toUpperCase();
}
else if (char !== char.toLowerCase() && char === char.toUpperCase()) {
output += " " + char;
}
else if (char == "-" || char == "_") {
output += " ";
}
else {
output += char;
}
}
return output;
}
JavaScript version:
/**
* Changes camel case to a human readable format. So helloWorld, hello-world and hello_world becomes "Hello World".
* */
function prettifyCamelCase(str) {
var output = "";
var len = str.length;
var char;
for (var i=0 ; i<len ; i++) {
char = str.charAt(i);
if (i==0) {
output += char.toUpperCase();
}
else if (char !== char.toLowerCase() && char === char.toUpperCase()) {
output += " " + char;
}
else if (char == "-" || char == "_") {
output += " ";
}
else {
output += char;
}
}
return output;
}
You can use a replacement function for String.replace, e.g.
function capitalize(s) {
return s[0].toUpperCase() + s.slice(1);
}
function replacer1(match, p1, p2, p3, offset, s) {
return p1 + capitalize(p2) + ' ' + p3;
}
var s1 = "helloWorld";
var r1 = s1.replace(/(^|[^a-z])([a-z]+)([A-Z])/, replacer1);
console.log(r1);
hello-world and hello_world work similar.
See JSFiddle
I don't know if there is already a built in method to do this but you could loop through the string and every time you see a character that you want to split on do so.
In your case something like:
my_str = 'helloWorld';
returnString = '';
for(var i = 0; i < my_str.length; i++) {
if(i == 0) {
returnString += (my_str[i] + 32); // capitalize the first character
}
else if(my_str[i] > 'A' || my_str[i] < 'Z') {
returnString += ' ' + my_str[i]; // add a space
}
else if(my_str[i] == '-' || my_str[i] == '_') {
returnString += ' ';
}
else {
returnString += my_string[i];
}
}
return returnString;
Edit:
After the numerous comments I have come to realize that I put up some broken code :P
Here is a tested version of it:
my_str = 'helloWorld';
function readable(str) {
// and this was a mistake about javascript/actionscript being able to capitalize
// by adding 32
returnString = str[0].toUpperCase();
for(var i = 1; i < str.length; i++) {
// my mistakes here were that it needs to be between BOTH 'A' and 'Z' inclusive
if(str[i] >= 'A' && str[i] <= 'Z') {
returnString += ' ' + str[i];
}
else if(str[i] == '-' || str[i] == '_') {
returnString += ' ';
}
else {
returnString += str[i];
}
}
return returnString;
}
Non elegant one liner using regex replaces with functions.
replace 1 - upper case first letter and remove _-
replace 2 - add space between lower case letters and upper case letters
var titleCase = s => s
.replace(/(^|[_-])([a-z])/g, (a, b, c) => c.toUpperCase())
.replace(/([a-z])([A-Z])/g, (a, b, c) => `${b} ${c}`);
console.log(titleCase("helloWorld"));
console.log(titleCase("hello-world"));
console.log(titleCase("hello_world"));
If using a library is an option, Lodash's startCase or lowerCase might be useful:
https://lodash.com/docs/#startCase
https://lodash.com/docs/#lowerCase
const result = _.chain("hello-world")
.snakeCase()
.split("_")
.map(w => _.capitalize(w))
.join(" ")
.value()
console.log(result)
<script src="
https://cdn.jsdelivr.net/npm/lodash#4.17.21/lodash.min.js
"></script>
Where could I find some JavaScript code to parse CSV data?
You can use the CSVToArray() function mentioned in this blog entry.
<script type="text/javascript">
// ref: http://stackoverflow.com/a/1293163/2343
// This will parse a delimited string into an array of
// arrays. The default delimiter is the comma, but this
// can be overriden in the second argument.
function CSVToArray( strData, strDelimiter ){
// Check to see if the delimiter is defined. If not,
// then default to comma.
strDelimiter = (strDelimiter || ",");
// Create a regular expression to parse the CSV values.
var objPattern = new RegExp(
(
// Delimiters.
"(\\" + strDelimiter + "|\\r?\\n|\\r|^)" +
// Quoted fields.
"(?:\"([^\"]*(?:\"\"[^\"]*)*)\"|" +
// Standard fields.
"([^\"\\" + strDelimiter + "\\r\\n]*))"
),
"gi"
);
// Create an array to hold our data. Give the array
// a default empty first row.
var arrData = [[]];
// Create an array to hold our individual pattern
// matching groups.
var arrMatches = null;
// Keep looping over the regular expression matches
// until we can no longer find a match.
while (arrMatches = objPattern.exec( strData )){
// Get the delimiter that was found.
var strMatchedDelimiter = arrMatches[ 1 ];
// Check to see if the given delimiter has a length
// (is not the start of string) and if it matches
// field delimiter. If id does not, then we know
// that this delimiter is a row delimiter.
if (
strMatchedDelimiter.length &&
strMatchedDelimiter !== strDelimiter
){
// Since we have reached a new row of data,
// add an empty row to our data array.
arrData.push( [] );
}
var strMatchedValue;
// Now that we have our delimiter out of the way,
// let's check to see which kind of value we
// captured (quoted or unquoted).
if (arrMatches[ 2 ]){
// We found a quoted value. When we capture
// this value, unescape any double quotes.
strMatchedValue = arrMatches[ 2 ].replace(
new RegExp( "\"\"", "g" ),
"\""
);
} else {
// We found a non-quoted value.
strMatchedValue = arrMatches[ 3 ];
}
// Now that we have our value string, let's add
// it to the data array.
arrData[ arrData.length - 1 ].push( strMatchedValue );
}
// Return the parsed data.
return( arrData );
}
</script>
jQuery-CSV
It's a jQuery plugin designed to work as an end-to-end solution for parsing CSV into JavaScript data. It handles every single edge case presented in RFC 4180, as well as some that pop up for Excel/Google spreadsheet exports (i.e., mostly involving null values) that the specification is missing.
Example:
track,artist,album,year
Dangerous,'Busta Rhymes','When Disaster Strikes',1997
// Calling this
music = $.csv.toArrays(csv)
// Outputs...
[
["track", "artist", "album", "year"],
["Dangerous", "Busta Rhymes", "When Disaster Strikes", "1997"]
]
console.log(music[1][2]) // Outputs: 'When Disaster Strikes'
Update:
Oh yeah, I should also probably mention that it's completely configurable.
music = $.csv.toArrays(csv, {
delimiter: "'", // Sets a custom value delimiter character
separator: ';', // Sets a custom field separator character
});
Update 2:
It now works with jQuery on Node.js too. So you have the option of doing either client-side or server-side parsing with the same library.
Update 3:
Since the Google Code shutdown, jquery-csv has been migrated to GitHub.
Disclaimer: I am also the author of jQuery-CSV.
Here's an extremely simple CSV parser that handles quoted fields with commas, new lines, and escaped double quotation marks. There's no splitting or regular expression. It scans the input string 1-2 characters at a time and builds an array.
Test it at http://jsfiddle.net/vHKYH/.
function parseCSV(str) {
var arr = [];
var quote = false; // 'true' means we're inside a quoted field
// Iterate over each character, keep track of current row and column (of the returned array)
for (var row = 0, col = 0, c = 0; c < str.length; c++) {
var cc = str[c], nc = str[c+1]; // Current character, next character
arr[row] = arr[row] || []; // Create a new row if necessary
arr[row][col] = arr[row][col] || ''; // Create a new column (start with empty string) if necessary
// If the current character is a quotation mark, and we're inside a
// quoted field, and the next character is also a quotation mark,
// add a quotation mark to the current column and skip the next character
if (cc == '"' && quote && nc == '"') { arr[row][col] += cc; ++c; continue; }
// If it's just one quotation mark, begin/end quoted field
if (cc == '"') { quote = !quote; continue; }
// If it's a comma and we're not in a quoted field, move on to the next column
if (cc == ',' && !quote) { ++col; continue; }
// If it's a newline (CRLF) and we're not in a quoted field, skip the next character
// and move on to the next row and move to column 0 of that new row
if (cc == '\r' && nc == '\n' && !quote) { ++row; col = 0; ++c; continue; }
// If it's a newline (LF or CR) and we're not in a quoted field,
// move on to the next row and move to column 0 of that new row
if (cc == '\n' && !quote) { ++row; col = 0; continue; }
if (cc == '\r' && !quote) { ++row; col = 0; continue; }
// Otherwise, append the current character to the current column
arr[row][col] += cc;
}
return arr;
}
I have an implementation as part of a spreadsheet project.
This code is not yet tested thoroughly, but anyone is welcome to use it.
As some of the answers noted though, your implementation can be much simpler if you actually have DSV or TSV file, as they disallow the use of the record and field separators in the values. CSV, on the other hand, can actually have commas and newlines inside a field, which breaks most regular expression and split-based approaches.
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;
}
};
csvToArray v1.3
A compact (645 bytes), but compliant function to convert a CSV string into a 2D array, conforming to the RFC4180 standard.
https://code.google.com/archive/p/csv-to-array/downloads
Common Usage: jQuery
$.ajax({
url: "test.csv",
dataType: 'text',
cache: false
}).done(function(csvAsString){
csvAsArray=csvAsString.csvToArray();
});
Common usage: JavaScript
csvAsArray = csvAsString.csvToArray();
Override field separator
csvAsArray = csvAsString.csvToArray("|");
Override record separator
csvAsArray = csvAsString.csvToArray("", "#");
Override Skip Header
csvAsArray = csvAsString.csvToArray("", "", 1);
Override all
csvAsArray = csvAsString.csvToArray("|", "#", 1);
Here's my PEG(.js) grammar that seems to do ok at RFC 4180 (i.e. it handles the examples at http://en.wikipedia.org/wiki/Comma-separated_values):
start
= [\n\r]* first:line rest:([\n\r]+ data:line { return data; })* [\n\r]* { rest.unshift(first); return rest; }
line
= first:field rest:("," text:field { return text; })*
& { return !!first || rest.length; } // ignore blank lines
{ rest.unshift(first); return rest; }
field
= '"' text:char* '"' { return text.join(''); }
/ text:[^\n\r,]* { return text.join(''); }
char
= '"' '"' { return '"'; }
/ [^"]
Try it out at http://jsfiddle.net/knvzk/10 or http://pegjs.majda.cz/online. Download the generated parser at https://gist.github.com/3362830.
Here's another solution. This uses:
a coarse global regular expression for splitting the CSV string (which includes surrounding quotes and trailing commas)
fine-grained regular expression for cleaning up the surrounding quotes and trailing commas
also, has type correction differentiating strings, numbers, boolean values and null values
For the following input string:
"This is\, a value",Hello,4,-123,3.1415,'This is also\, possible',true,
The code outputs:
[
"This is, a value",
"Hello",
4,
-123,
3.1415,
"This is also, possible",
true,
null
]
Here's my implementation of parseCSVLine() in a runnable code snippet:
function parseCSVLine(text) {
return text.match( /\s*(\"[^"]*\"|'[^']*'|[^,]*)\s*(,|$)/g ).map( function (text) {
let m;
if (m = text.match(/^\s*,?$/)) return null; // null value
if (m = text.match(/^\s*\"([^"]*)\"\s*,?$/)) return m[1]; // Double Quoted Text
if (m = text.match(/^\s*'([^']*)'\s*,?$/)) return m[1]; // Single Quoted Text
if (m = text.match(/^\s*(true|false)\s*,?$/)) return m[1] === "true"; // Boolean
if (m = text.match(/^\s*((?:\+|\-)?\d+)\s*,?$/)) return parseInt(m[1]); // Integer Number
if (m = text.match(/^\s*((?:\+|\-)?\d*\.\d*)\s*,?$/)) return parseFloat(m[1]); // Floating Number
if (m = text.match(/^\s*(.*?)\s*,?$/)) return m[1]; // Unquoted Text
return text;
} );
}
let data = `"This is\, a value",Hello,4,-123,3.1415,'This is also\, possible',true,`;
let obj = parseCSVLine(data);
console.log( JSON.stringify( obj, undefined, 2 ) );
Here's my simple vanilla JavaScript code:
let a = 'one,two,"three, but with a comma",four,"five, with ""quotes"" in it.."'
console.log(splitQuotes(a))
function splitQuotes(line) {
if(line.indexOf('"') < 0)
return line.split(',')
let result = [], cell = '', quote = false;
for(let i = 0; i < line.length; i++) {
char = line[i]
if(char == '"' && line[i+1] == '"') {
cell += char
i++
} else if(char == '"') {
quote = !quote;
} else if(!quote && char == ',') {
result.push(cell)
cell = ''
} else {
cell += char
}
if ( i == line.length-1 && cell) {
result.push(cell)
}
}
return result
}
I'm not sure why I couldn't get Kirtan's example to work for me. It seemed to be failing on empty fields or maybe fields with trailing commas...
This one seems to handle both.
I did not write the parser code, just a wrapper around the parser function to make this work for a file. See attribution.
var Strings = {
/**
* Wrapped CSV line parser
* #param s String delimited CSV string
* #param sep Separator override
* #attribution: http://www.greywyvern.com/?post=258 (comments closed on blog :( )
*/
parseCSV : function(s,sep) {
// http://stackoverflow.com/questions/1155678/javascript-string-newline-character
var universalNewline = /\r\n|\r|\n/g;
var a = s.split(universalNewline);
for(var i in a){
for (var f = a[i].split(sep = sep || ","), x = f.length - 1, tl; x >= 0; x--) {
if (f[x].replace(/"\s+$/, '"').charAt(f[x].length - 1) == '"') {
if ((tl = f[x].replace(/^\s+"/, '"')).length > 1 && tl.charAt(0) == '"') {
f[x] = f[x].replace(/^\s*"|"\s*$/g, '').replace(/""/g, '"');
} else if (x) {
f.splice(x - 1, 2, [f[x - 1], f[x]].join(sep));
} else f = f.shift().split(sep).concat(f);
} else f[x].replace(/""/g, '"');
} a[i] = f;
}
return a;
}
}
Regular expressions to the rescue! These few lines of code handle properly quoted fields with embedded commas, quotes, and newlines based on the RFC 4180 standard.
function parseCsv(data, fieldSep, newLine) {
fieldSep = fieldSep || ',';
newLine = newLine || '\n';
var nSep = '\x1D';
var qSep = '\x1E';
var cSep = '\x1F';
var nSepRe = new RegExp(nSep, 'g');
var qSepRe = new RegExp(qSep, 'g');
var cSepRe = new RegExp(cSep, 'g');
var fieldRe = new RegExp('(?<=(^|[' + fieldSep + '\\n]))"(|[\\s\\S]+?(?<![^"]"))"(?=($|[' + fieldSep + '\\n]))', 'g');
var grid = [];
data.replace(/\r/g, '').replace(/\n+$/, '').replace(fieldRe, function(match, p1, p2) {
return p2.replace(/\n/g, nSep).replace(/""/g, qSep).replace(/,/g, cSep);
}).split(/\n/).forEach(function(line) {
var row = line.split(fieldSep).map(function(cell) {
return cell.replace(nSepRe, newLine).replace(qSepRe, '"').replace(cSepRe, ',');
});
grid.push(row);
});
return grid;
}
const csv = 'A1,B1,C1\n"A ""2""","B, 2","C\n2"';
const separator = ','; // field separator, default: ','
const newline = ' <br /> '; // newline representation in case a field contains newlines, default: '\n'
var grid = parseCsv(csv, separator, newline);
// expected: [ [ 'A1', 'B1', 'C1' ], [ 'A "2"', 'B, 2', 'C <br /> 2' ] ]
You don't need a parser-generator such as lex/yacc. The regular expression handles RFC 4180 properly thanks to positive lookbehind, negative lookbehind, and positive lookahead.
Clone/download code at https://github.com/peterthoeny/parse-csv-js
Just throwing this out there.. I recently ran into the need to parse CSV columns with Javascript, and I opted for my own simple solution. It works for my needs, and may help someone else.
const csvString = '"Some text, some text",,"",true,false,"more text","more,text, more, text ",true';
const parseCSV = text => {
const lines = text.split('\n');
const output = [];
lines.forEach(line => {
line = line.trim();
if (line.length === 0) return;
const skipIndexes = {};
const columns = line.split(',');
output.push(columns.reduce((result, item, index) => {
if (skipIndexes[index]) return result;
if (item.startsWith('"') && !item.endsWith('"')) {
while (!columns[index + 1].endsWith('"')) {
index++;
item += `,${columns[index]}`;
skipIndexes[index] = true;
}
index++;
skipIndexes[index] = true;
item += `,${columns[index]}`;
}
result.push(item);
return result;
}, []));
});
return output;
};
console.log(parseCSV(csvString));
Personally I like to use deno std library since most modules are officially compatible with the browser
The problem is that the std is in typescript but official solution might happen in the future https://github.com/denoland/deno_std/issues/641 https://github.com/denoland/dotland/issues/1728
For now there is an actively maintained on the fly transpiler https://bundle.deno.dev/
so you can use it simply like this
<script type="module">
import { parse } from "https://bundle.deno.dev/https://deno.land/std#0.126.0/encoding/csv.ts"
console.log(await parse("a,b,c\n1,2,3"))
</script>
I have constructed this JavaScript script to parse a CSV in string to array object. I find it better to break down the whole CSV into lines, fields and process them accordingly. I think that it will make it easy for you to change the code to suit your need.
//
//
// 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'
}
}
Just use .split(','):
var str = "How are you doing today?";
var n = str.split(" ");