JarCastingpapa-parse 
<!-- download and parse into JSON -->
<papa-parse auto
url="https://rawgit.com/PolymerVis/papa-parse/master/demo/MOCK_DATA.csv"
rows="{{rows}}"></papa-parse>
<!-- render as a table -->
<h4>Parsed CSV table:</h4>
<table>
<template is="dom-repeat" items="[[rows]]" as="row">
<tr>
<template is="dom-repeat" items="[[row]]">
<td>[[item]]</td>
</template>
</tr>
</template>
</table>
<!-- unparse JSON into string -->
<papa-unparse header
json-array="[[rows]]"
csv-str="{{csvStr}}"></papa-unparse>
<!-- csv string output -->
<h4>Csv string</h4>
<pre>[[csvStr]]</pre>
Installation
bower install --save PolymerVis/papa-parse
Documentation and demos
More details @ webcomponents.org.
Disclaimers
PolymerVis is a personal project and is NOT in any way affliated with Papaparse, Polymer or Google.
papa-parse
papa-parse is a Polymer 2.0 element to parse CSV files into JSON object(s) with Papa parse.
papa-parse can download and parse a csv file via url, or from raw csv strings via raw, and File object via file. If the auto flag is set, papa-parse will automatically start the job, otherwise a manual call to the function start will be needed.
Parse from URL.
<papa-parse auto url="MOCK_DATA.csv" rows="{{rows}}"></papa-parse>
Parse from raw csv String.
<papa-parse auto raw="[[SomeCsvString]]" rows="{{rows}}"></papa-parse>
Parse from FileReader
<input id="selectfile" type="file"></input>
<papa-parse auto file="[[file]]" rows="{{rows}}"></papa-parse>
this.$.selectfile
.addEventListener('change', function(e) {
this.file = e.target.files[0];
});
There are 2 possible outputs for rows - an array of array of values, or an array of objects. An array of array is returned if the header flag is not set, and an array of objects otherwise, where each object is a map comprising of the column name and its corresponding value for the row (e.g. {col1: value1, col2: value2}).
Handling big files
Parsing is synchronous and may block the UI thread (i.e. freeze the screen) if the csv file is big. Parsing with a web-worker via worker flag is recommended for big file. However, note that papaparse.min.js should not be bundled during the build process if web-worker is required as the web-worker needs to load the script. You will also need to correctly reference the location of the script via the script-path attribute.
You should not use relative path if parsing from url, as the web-worker may load from an incorrect path.
<papa-parse auto header worker stream
script-path="../../papaparse/papaparse.min.js"
fields="{{fields}}"
url="https://some.domain/some.csv"
on-stream="handleRecord"></papa-parse>
function handleRecord(e, {data, meta, errors}) {
console.log(data);
}
Streaming via stream flag is also recommended for big file (stream and worker are not dependent on each other). However, when stream flag is set, rows attribute will only return an empty array as none of the records will be persisted. Instead, papa-parse will emit an stream event of the form {{data: Array[]|Object[], meta: Object, errors: Array, parser: Object}} for each record parsed.
<button onclick="javascript:start()">[[buttonLabel]]</button>
<papa-parse id="parser"
stream
url="MOCK_DATA.csv"
rows="{{rows}}"
on-stream="handleRecord"></papa-parse>
function start() {
app.$.parser.start();
}
function handleRecord(e, {data, meta, errors}) {
alert(JSON.stringify(data));
app.$.parser.pause();
}
papa-unparse
papa-unparse is a Polymer 2.0 element to convert JSON arrays or objects into CSV strings with Papa parse.
Basic usage
For array of objects, e.g. [{col1: 1, col2: 2}, {col1: 3, col2: 4}]
<papa-unparse header json-array="[[rows]]"></papa-unparse>
For array of arrays, e.g. [[1,2], [3,4]]
<papa-unparse json-array="[[rows]]" fields="[[fields]]"></papa-unparse>
Download as CSV file
<papa-unparse header json-array="[[rows]]"></papa-unparse>
<button onclick="javascript:download">Download</button>
function download() {
document.querySelector('papa-unparse').downloadCsv('somefile.csv');
}
Other options
<papa-unparse
header
quotes
quoteChar="'"
delimiter="|"
newline="\n"
json-array="[[rows]]"></papa-unparse>
