keyang / node-csvtojson Goto Github PK

Is possible to load column arrays joining data from multiple lines in case if we have 1..n number of indexes to be loaded we can not have multiple columns defined, instead it is easier to have multiple lines of data.

Example:

name, age, employer[].name, employer[].startDate, employer[].endDate
Paul, 38, Company 1, 01-01-2010, 20-08-2011
, , Company 2, 21-08-2011, 21-01-2015

Thanks,
DGKM

csvtojson --ignoreEmpty true packages/slabs/data/tiledata.csv
Unknown boolean value  for parameter --ignoreEmpty.

I tried 1, 0, nothing and true / false.
what arguments does it expect?

I have a complex json structure to replicate.
It is basically something like this:

csv: [{
    levelA : {
    title: 'foo',
    levelB: [
      {
      title : 'bar',
      levelC: [
        {title : 'deep blue'},
        {title : 'deep red'},
        { ... }
        ]
      },
      { ... }
      ]
    }
  }, 
  { ... }
]

As you can see it is a deep nested structure with arrays of objects.
Do you think it would be possible to achieve this just using csvtojson parser keys or should I go for a custom parser?

i am using some code from your docs:

  //Converter Class
  var Converter = csvtojson.core.Converter;

  //CSV File Path or CSV String or Readable Stream Object
  var csvFileName = process.cwd()+'/sample_data/orders.csv';

  //new converter instance
  var csvConverter = new Converter();

  //end_parsed will be emitted once parsing finished
  csvConverter.on("end_parsed",function(jsonObj){
     console.log(jsonObj); //here is your result json object
  });

  //read from file
  csvConverter.from(csvFileName);

how can i change the delimiter of the orders.csv file? youre expecting , but it should be ;

tanks for your help

even in the csvtojson/bin directory, I have to use ~$node csvtojson to get a reaction, and I get this error:

module.js:340
throw err;
^
Error: Cannot find module '../libs/interfaces'
at Function.Module._resolveFilename (module.js:338:15)
at Function.Module._load (module.js:280:25)
at Module.require (module.js:364:17)
at require (module.js:380:17)
at Object. (/Users/jeffmjack/csvtojson/bin/csvtojson:3:13)
at Module._compile (module.js:456:26)
at Object.Module._extensions..js (module.js:474:10)
at Module.load (module.js:356:32)
at Function.Module._load (module.js:312:12)
at Function.Module.runMain (module.js:497:10)

Converting csv to json needs computing resource. it is good to have an option to allow the whole converting happen in another process so that it wont affect main process which always has busy.

This is my code to read the csv:

   var csvtojson = require("csvtojson").core.Converter,
   var fileStream = fs.createReadStream("my.csv");
    var csvConverter = new csvtojson({
      constructResult: true
    });
    fileStream.pipe(csvConverter);
    csvConverter.on("end_parsed", function(data) {
      next(null, data); //1
    });

The only problem I have here is that this works great for a small csv, but it fails to give any output for large csv . The data I receive in the 1 is an empty array.

You can get the sample csv here.
It's very common for me to parse very large csv to the tune of upto 20k or even 40k records.

I receive the error "Incomplete CSV file detected. Quotes does not match in pairs." on a tab-delimited file, where this entry includes a " to indicate inches.

Is there a way to escape it or otherwise convince it to continue past it?

Hi Guys,

I pulled in the latest code change which support multi cpu support. However somehow it seem to not work as expected. Here is the sample code I am using.

var data = fs.readFileSync(config.destination, 'utf8');
parse_csv(data)
.then(function (object) {
console.log(object)
}

function parse_csv_internal(data, onComplete){

var csvConverter = new Converter({constructResult:true, workerNum:4});
csvConverter.fromString(data,function(err,jsonObj) {
  if (err){
      log("Parse CSV Internal Function Error" + err);
      onComplete(err, null);
  } else {
      log('Parse CSV Internal function has finished!');
      onComplete(null, jsonObj);
  }
});

csvConverter.on("end_parsed", function(jsonObj) {
  //final result poped here as normal.
});

}

function parse_csv(csv_data) {
return new Promise(function (fulfill, reject) {
parse_csv_internal(csv_data, function(err, result) {
if (err == null) {
fulfill(result);
} else {
log('Parse CSV function has an error!');
return reject(err);
}
});
});
}

It takes like 20mins to finish because of
node 0.0 00:00.15 5 0 28 10M 0B 0B 9766 9766 sleeping *0[1] 0.00000 0.00000 501 5119 156 50 25 450 85 258 0 0 0.0 varun N/A N/A N/A N/A N/A N/A
9769 node 0.0 00:00.16 5 0 28 10M 0B 0B 9766 9766 sleeping *0[1] 0.00000 0.00000 501 5116 153 50 25 452 85 251 0 0 0.0 varun N/A N/A N/A N/A N/A N/A
9768 node 0.0 00:00.16 5 0 28 10M 0B 0B 9766 9766 sleeping *0[1] 0.00000 0.00000 501 5149 158 50 25 450 85 207 0 0 0.0 varun N/A N/A N/A N/A N/A N/A
9767 curl 0.0 00:00.05 2 0 38 1996K 0B 0B 9767 420 sleeping *0[4] 0.00000 0.00000 501 1335 122 119 46 1755+ 250 817+ 9 0 0.0 varun N/A N/A N/A N/A N/A N/A
9766 node 100.1 07:35.63 7/1 0 37

I see 4 threads but only one thread is doing work. Others are sleeping. What am I doing wrong here?

Hello,

Im trying to use your module to convert a csv file to json and send said json as the response to a request.

A few issues have occurred.

First, It seams csvtojson is creating duplicates of my csv. My csv currently has 4 lines including the header line (subject to change) and the output contains 8 json objects with 2 headers.

Second, I am submitting a request for data through a form. The first time I submit the request, I get data back but receive the error above. If I change a few parameters and submit again, my express server dies and spits out the error Error: CSV no longer writable

Is there any insight you can provide to solve this problem?

Thanks!

Thanks for adding the error event Listener from 0.4.3 to 0.4.4

However, my code stopped working when I upgraded to this new version. What I mean by that is whenever I upload a csv, my code gets stuck now.

Current code, for reference:

var form = new formidable.IncomingForm();
form.parse(req, function(err, fields, files) {
 console.log(files.file.type);
 if (!files || !files.file || files.file.size == '0') {
   res.send("CSV must not be empty.");
 }else if( files.file.type == 'text/csv'){
   var Converter = require("csvtojson").Converter;

  var csvFileName = files.file.path;
  var fileStream = fs.createReadStream(csvFileName);
  //new converter instance

  var csvConverter = new Converter({constructResult:true});

  csvConverter.on("record_parsed",function(resultRow,rawRow,rowIndex) {
    //console.log(resultRow); //here is your result json object
    fileStream.unpipe(csvConverter);
    console.log('manually close the file stream2');
    csvConverter.end();
  });

  //end_parsed will be emitted once parsing finished
  csvConverter.on("end_parsed",function(jsonObj) {
    res.json(200, jsonObj);
  });

  //read from file
  fileStream.pipe(csvConverter, { end: false });
} else {
  res.send("Not a valid type! Must be CSV format");
}
});

hi Keyang,

Is there a way to tell the parser how to convert a column type?
for example, i have this csv data:
name, timestamp, value
a.b.c.d.e.f, 1397434200, 1.82

yield this json:

[ { name: 'a.b.c.d.e.f', timestamp: '1397434200', value: '1.82' } ]

i would like that timestamp and value will be a number and not a string

thanks in advance,
Shay

I have been getting the following error while using this module tonight. The relevant code is below the error. The error comes when, after a converted page is rendered, i go to a second url to render another page.

Wed Jul 02 2014 23:14:36 GMT-0400 (Eastern Daylight Time): Node server stopped.

stream.js:94
      throw er; // Unhandled stream error in pipe.
           ^
Error: write after end
    at writeAfterEnd (_stream_writable.js:133:12)
    at csvAdv.Writable.write (_stream_writable.js:181:5)
    at Request.ondata (stream.js:51:26)
    at Request.emit (events.js:95:17)
    at IncomingMessage.<anonymous> (D:\repos\openshift-nodejs\node_modules\request\request.js:932:12)
    at IncomingMessage.emit (events.js:95:17)
    at IncomingMessage.<anonymous> (_stream_readable.js:748:14)
    at IncomingMessage.emit (events.js:92:17)
    at emitReadable_ (_stream_readable.js:410:10)
    at emitReadable (_stream_readable.js:406:5)

And the code...

var converter = require('csvtojson').core.Converter;
var express   = require('express');
var request   = require('request');
var router    = express.Router();

// Don't save everything to memory. This facilitates large CSV's
var csvConverter = new converter({constructResult:false});

/* 
 *  GET home page.
 */
router.get('/', function(req, res) {
  res.render('index');
});


/*
 *  URL's to pull data from
 */
var aSchool= require('./schools/aSchool');

router.get('/aSchool/bulletin/:term', function(req, res) {
  var term = req.params.term;
  if ( aSchool.validTerms.indexOf(term) >= 0 ) {
    res.writeHead(200, {"Content-Type": "application/json"});
    var url = '';
    switch (term) {
      case 'fall':
        var url = aSchool.bulletin + aSchool.fall;
        break;
      case 'spring':
        var url = westga.bulletin + westga.spring;
        break;
      case 'summer':
        var url = westga.bulletin + westga.summer;
        break;
      default:
        console.log('Something went wrong...');
    }
    console.log(url);
    request.get(url).pipe(csvConverter).pipe(res);
  } else {
    res.status(404);
  }
});


module.exports = router;

• NodeJS version: v0.10.22
• csvtojson version: 0.3.8 (also 0.3.11)
• Mac OS version: 10.9.4
• The exception is thrown when the application receives empty csv file. The exception is as follows:

undefined:2
]
^
SyntaxError: Unexpected token ]
at Object.parse (native)
at Result.getBuffer (/Users/trangtungn/Working/Omnia_Git/om_file_uploader/node_modules/csvtojson/libs/core/Result.js:19:15)
at csvAdv. (/Users/trangtungn/Working/Omnia_Git/om_file_uploader/node_modules/csvtojson/libs/core/init_onend.js:8:70)
at csvAdv.EventEmitter.emit (events.js:117:20)
at _stream_readable.js:920:16
at process._tickDomainCallback (node.js:459:13)

• Looking at the source code (function csvAdv.prototype._flush(cb)) and found that the Converter only pushes eol and ] whenever the buffer length is 0; [ is not included which causes the issue.
• By catching zero buffer length case as below, I can run the app without the issue:

csvAdv.prototype._flush = function(cb) {
if (this._buffer.length != 0) { //emit last line
this.emit("record", this._buffer, this.rowIndex++, true);
this.push(eol + "]");
} else { // incase of _buffer length is 0, add [, eol and ]
this.push("[" + eol + "]");
}
cb();
};

Hi how i can use the default parser but avoid the representation of dot(.) in a header as a nested JSON? or the only options is to create a custom parser?

Thanks

The command line conversion doesn't create a .json file, it simply streams the csv data in the terminal.

This module is not working with windows platform please help

Hi Keyang

The current csvtojson does not support invalid CSV file conversion error handle. (See http://stackoverflow.com/questions/30028453/how-to-handle-incomplete-csv-file-detected-quotes-does-not-match-in-pairs)

Is it possible to let csvtojson emit invalid csv (e.g. a binary file with fake.csv name) when it detects the error.

Currently nodejs will report "Incomplete CSV file detected. Quotes does not match in pairs" message. It would be nice to emit "invalid csv" message at this error so we can catch the error and handle it.

I know we should have our separate CSV validation code but most of other CSV nodejs module already have validation and conversion built in. I really want to use your module as it provides convenience as in convert csv to json but it could be better by including this "emit invalid csv file message" feature instead of just print out a message to the console.

Thanks.

@Keyang I'm getting a weird issue where quoted csv text like
"Many types of citrus are available, including oranges, white grapefruit, tangelos, and lemons."
when converted into json looks like
"notes":"Many types of citrus are availableincluding orangeswhite grapefruittangelos,and lemons."
The object coming back from csvtojson has several, but not all, of the non-delineating commas removed and whitespace randomly trimmed. If I set trim: false I don't have the issue with the whitespace being removed but the issue with the commas being removed is still there. Any ideas? csvtojson is being used here https://github.com/azgs/az-agriculture/blob/master/data/csv2json.js#L85. Thanks.

Is there a command line option to ignore fields in a record that do not have a value?
For example the following csv file

col1, col2, col3
d1,,d3

would become

{
  col1:"d1",
  col3:"d3"
}

I'm particularly interested in this feature for use with arrays

col1, col2[], col2[]
d1,,d3

would become

{
  col1: "d1",
  col2: ["d3"]
}

Hi, first of all thank you for the great lib.

I have an issue to parse a CSV using the streaming API because it does not return a JSON array. I don't know if it is the "normal" behavior but here an example to explain my problem (with Node v0.12):

The original CSV file:

annee;jour;date;b_1;b_2;devise;
2015029;LUNDI   ;09/03/2015;35;31;eur;
2015028;SAMEDI  ;07/03/2015;48;9;eur;

The code to parse the CSV:

var rs = require('fs').createReadStream('source.csv');
var ws = require('fs').createWriteStream('destination.json');

 rs.pipe(new Converter({ constuctResult: false, delimiter: ';', trim: true })).pipe(ws);

The result file content:

{"annee":2015029,"jour":"LUNDI","date":"09/03/2015","b_1":35,"b_2":31,"devise":"eur","":""}
{"annee":2015028,"jour":"SAMEDI","date":"07/03/2015","b_1":48,"b_2":9,"devise":"eur","":""}

I would expected to have this json object:

[
  {"annee":2015029,"jour":"LUNDI","date":"09/03/2015","b_1":35,"b_2":31,"devise":"eur","":""},
  {"annee":2015028,"jour":"SAMEDI","date":"07/03/2015","b_1":48,"b_2":9,"devise":"eur","":""}
]

It works fine if I use the end_parsed event instead, but not with the stream and the pipe API.

Thank you for your help

Note: I don't know why by it adds me an empty field in the JSON result too

Brief summary:

CSV record is

... ,"[""Neil"", ""Bill"", ""Carl"", ""Richard"", ""Linus""]", ...

What's parsed is

..."[\"\"Neil\"\", \"\"Bill\"\", \"\"Carl\"\", \"\"Richard\"\", \"\"Linus\"\"]\", ...

In this case, the comma at the end is not treated as a delimiter...

Which means that the column is not delimited correctly, many columns run together, and the entire json result is incorrect. Seems to be an issue with determining quote escaping.

When piping a fs ReadStream object into the csvtojson Converter with the following large CSV file (5291 lines), only the first 4259 lines trigger record_parsed events. The Converter seems to stop after reaching a certain number of bytes.

Example js:
https://gist.github.com/Adamanthil/f31eb8314ce2d05a131a

Sample csv file:
https://gist.github.com/Adamanthil/f5747f4c07787c918887

The following file:

Date, Type, Description, Value, Balance, Account Name, Account Number

11/04/2012,DPC,"'CALL REF.NO. 1778 , FROM A/C 12345678",50.00,149.47,"'Some Business Ltd","'123456-24438749",

Reasons:

• The first line endings are CRLF (\r\n)
• The CSV starts with a blank line (removing the first line solves the issue).

When creating a .csv generated by the OSX version of Excel, you have to save the file specifically as "Windows Commas Separated" .csv file to get it to parse, otherwise, the converter will return an empty array.

When saving it as the normal, "default" .csv type, it won't parse.

var Converter       = require('csvtojson').Converter;
var converter = new Converter({constructResult:true});
converter.on("end_parsed",function(json){
    if(json.length>0){
        report.json = json;
        defer.resolve(json);
    }else{
        //this happens when a mac .csv is sent
        defer.reject({message:'.csv could not be parsed. Please create a "Windows Comma Separated" .csv file'});
    }
});
fs.createReadStream(fileLocation).pipe(converter);

I have seen examples where I saw that {constructResult:false} will help reduce memory consumption. While I understand that, I have 16GB of ram and I can effort to put the result in memory and do some processing. My main issue is that it took 24 minutes to finish this work. While I understand the load, 24 minutes sounds really high!!! Can this take advantage of multi cpu cores for faster processing?

• csvtojson version: 0.3.11
• NodeJS version: v0.10.22
• Mac OS version: 10.9.4

Incase csv file is empty, this.eol will not be set that leads to _flush function throws the following exception:
undefined:1
undefined]
^
SyntaxError: Unexpected token u
at Object.parse (native)
at Result.getBuffer (/Users/trangtungn/Working/Omnia_Git/om_file_uploader/node_modules/csvtojson/libs/core/Result.js:19:15)
at csvAdv. (/Users/trangtungn/Working/Omnia_Git/om_file_uploader/node_modules/csvtojson/libs/core/init_onend.js:8:70)
at csvAdv.EventEmitter.emit (events.js:117:20)
at _stream_readable.js:920:16
at process._tickDomainCallback (node.js:459:13)

By initiallizing this.eol, I can get through the problem.

Is it possible, in the current state of this module, to omit a specific column form CSV before getting the JSON result.

e.g

id, name, age, address
 1, john,  25, 120 road ave

I want to omit address before converting the data to JSON.

i.e. something like this

"addresses":[
{"type":"work", "address_line1":"val1", "address_line2":"val2"},
{"type"home","address_line1":"val3", "address_line2":"val4" }
]

Is that possible now? or does it require a custom parser to be written?

Thanks! Seems like a promising solution.

Hello Keyang, thanks for the great libray.

I m not able to understand how can i pass a simple csv string and get the json out put.

can you please show an example on to how to pass raw csv data to the convertor?

I am getting the error "Cannot call method 'parse' of undefined" when the CSV rows have trailing commas. For example "field1,field2 \n value1,value2, \n value1,value2, \n"

Thanks,

Nathan

I am trying to use csv2json from the command line to convert a csv file to json. I am doing the following: csvtojson test.csv>file.json. The entire file is converted but the program never exits; it rather hangs.

[deploy@shippinghub ~]$ csvtojson test.csv>file.json
^C[deploy@shippinghub ~]$

As you can see I had to do a ctrl+C to force it to quit. I think this is a bug. it should exit and give control back to shell after the file is converted.

Please advise.

Hi!
I'm trying to convert a CSV to JSON.
Text value BEFORE to convert:

• "O FUTEBOL.¿"
• " emoção"

The same value AFTER convertion:

• "O FUTEBOL.\ufffd"
• " emo\ufffd\ufffdo"

There's an option to disable code/encode special chars?

Is worth to say that all special chars (ç,ã,õ,í,¿, etc) are being converted to the unicode char \ufffd that is equal to 'REPLACEMENT CHARACTER'.
More info: http://www.fileformat.info/info/unicode/char/0fffd/index.htm

Can you help me?

HI there, it would be good if you enable a custom line ending character to be passed in the array of params when creating the Converter.

how do i specify a tab delim at the cmd line?
i tried

csvtojson --delimiter='\t' data.tsv
csvtojson --delimiter='¥t' data.tsv

also the | seems to get taken as a pipe in the cmd:

csvtojson --delimiter=| data/jgram/comments.tsv
zsh: permission denied: data/jgram/comments.tsv

I'm using "csvtojson": "^0.4.3",
When parsing a tsv string, I was only getting 7 out of 30 records. (A small MWS GetReportList)
When I set workerNum to 4 then it returned all 30 records.
My tsv string is only 12 KB. It fails silently with a partial result.
I could readStream.pipe(csvConverter).pipe(writeStream) but it is always a small file (<100k).

Questions, What is the best way to detect a partial result? I am concerned that I could have the same issue with workerNum:4 in production. I won't always have the number of expected records to test against. Should I just use pipe writeStream all the time? - Thanks for the help.

code:
function convertTsvToJson(val){
var deferred = Q.defer();
var CSVConverter = require("csvtojson").Converter;
var converter = new CSVConverter({
workerNum:4,
delimiter:"\t",
});

converter.fromString(val, function(err, result){
if (err) {
deferred.reject(err);
} else {
if (result === undefined || _.isEmpty(result) ) {
deferred.reject('Issue parsing tsv, result is invalid');
} else {
console.log('resultJsonLength', result.length);
deferred.resolve(result);
}
}
});
return deferred.promise;
}

Is there a reason to emit 'record_parsed' instead of the 'data' event from the stream API?

This is an issue when trying to pipe parsed records.

Hello !

I must convert a .csv file written in French that contain special characters.
When I do that all the special characters are replaced by "�".
What can I do ?

Thank you !

What is the syntax for disabling the checkType logic via the command line? I have a large file to convert, with a large String field, however random lines are being detected as numeric. This is throwing off the import logic in the next app I'm using. My goal is to have all fields treated as strings so I may convert them in the next phase of my processing.

Hi @Keyang

This looks like it could be a really useful library for me. But it's not working as expected with objects that are nested more than two levels.

Take a very simple example. Here is a CSV with a single header and single value.

leve1.level2.level3
hello

According to the docs it should produce the following JSON:

[ { leve1: { level2: { level3: "hello" } } } ]

However, it is actually producing:

[ { leve1: { level2: [Object] } } ]

Can you take a look at this - I can't really see in your code where it's going wrong.

Thanks,

James

Here is an example file where it fails (generates tons of partial objects when there are newlines).

https://gist.github.com/wizonesolutions/5d9d9d880cb2002951db

Works pretty well otherwise. I might just try to fix it.

var fs = require("fs");
var Converter = require("csvtojson").Converter;
var csvFileName="./Sacramentorealestatetransactions.csv"
// Source http://samplecsvs.s3.amazonaws.com/Sacramentorealestatetransactions.csv

self = module.exports ={
    getCsvAsJSON: function (callback) {
    //Converter Class
    var fileStream = fs.createReadStream(csvFileName);
    //new converter instance
    var converter = new Converter({ constructResult: true });
    //end_parsed will be emitted once parsing finished
    converter.on("end_parsed", function (jsonObj) {
        console.log(jsonObj); //here is your result json object
        callback(jsonObj);
    });
    //read from file
    fileStream.pipe(converter);
   }

}

self.getCsvAsJSON(function(){})

Output I get is always empty
version: latest

You can see Demo here: http://www.convertcsv.com/csv-to-json.htm

This is an input data read from CSV :

productId,classification_id,mfgDate,gross_price,retail_price,color,size
CC102-PDMI-001,eClass_5.1.3,10/3/2014,12,40,green,40
CC200-009-001,eClass_5.1.3,11/3/2014,5,3,blue,38
CC200-070-001,eClass_5.1.3,10/4/2014,15,13,red,45
CC200-099,eClass_5.1.3,10/5/2014,20,17,orange,28
CC200-100,eClass_5.1.3,10/3/2014,5,4,black,32

And expected resulted JSON:
[
{
"FIELD1":"productId",
"FIELD2":"classification_id",
"FIELD3":"mfgDate",
"FIELD4":"gross_price",
"FIELD5":"retail_price",
"FIELD6":"color",
"FIELD7":"size"
},
{
"FIELD1":"CC102-PDMI-001",
"FIELD2":"eClass_5.1.3",
"FIELD3":"10/3/2014",
"FIELD4":"12",
"FIELD5":"40",
"FIELD6":"green",
"FIELD7":"40"
},
{
"FIELD1":"CC200-009-001",
"FIELD2":"eClass_5.1.3",
"FIELD3":"11/3/2014",
"FIELD4":"5",
"FIELD5":"3",
"FIELD6":"blue",
"FIELD7":"38"
},
{
"FIELD1":"CC200-070-001",
"FIELD2":"eClass_5.1.3",
"FIELD3":"10/4/2014",
"FIELD4":"15",
"FIELD5":"13",
"FIELD6":"red",
"FIELD7":"45"
},
{
"FIELD1":"CC200-099",
"FIELD2":"eClass_5.1.3",
"FIELD3":"10/5/2014",
"FIELD4":"20",
"FIELD5":"17",
"FIELD6":"orange",
"FIELD7":"28"
},
{
"FIELD1":"CC200-100",
"FIELD2":"eClass_5.1.3",
"FIELD3":"10/3/2014",
"FIELD4":"5",
"FIELD5":"4",
"FIELD6":"black",
"FIELD7":"32"
}
]

How to filter only a desired set of columns during CSV to JSON conversion.
So I have columnA, columnB, ColumnC.
And I would like to display only ColumncC in the JSON file.

How to go about it.

Raising an issue as I didn't find any mailing list. Thanks~

How can we efficiently map headers to field names for outputting a JSON object with the desired keys rather than what comes from the CSV file as-is?

So going from:

a, b, c
1, 2, 3

to:

{
aAlt: 1,
bAlt: 2,
cAlt: 3
}

If there is a CSV where there are fields containing line breaks, the csvtojson will parse each line as if it is a new entry in the csv. It doesn't respect the line breaks as part of the row value even though it is within the quotation marks (").

For example

day,month,description
"1","12","something that works"
"3","05","something
that<hr>
doesn't"

the above should only parse out into two objects, instead it will generate 4.

(This is a question so, please label it as that)

Hi, let me thank you for this repo, it is great!

I am using it a lot and I was wondering if I am managing large files using {constructResult:false} and then I need to upload the entire json to memory right after the json file is complete, it would be the same if I use {constructResult:true} and having the json directly in memory? Does that make any sense?

Thanks!