stream-dbf

A dBase DBF file parser that outputs a node stream.

stream-dbf uses the logic from abstractvector's node-dbf in a Transform stream. It's written in plain JavaScript.

npm install stream-dbf

Usage

Create a new instance of the parser by specifying the file path of the dBase file, and optionally any of the options flags (described below).

var DBF = require('stream-dbf');
var parser = new DBF(fileName, [options]);

options

• parseTypes - default: true - If false, integers and floats will be returned as strings.
• recAsArray - default: false - If true, stream will emit arrays instead objects (this slightly improve throughput).
• encoding - default: "utf-8" - It is passed to the iconv-lite to decode the field names and values in order to support encodings like GBK etc.
• lowercase - default: false - If true, the field names will be converted to lowercase.
• withMeta - default: true - If true, each record will contain two meta fields named "@sequenceNumber" and "@deleted".

parser.stream

Attach standard stream listeners to this object to access the records.

var stream = parser.stream;
stream.on('readable', function() {
  var record = stream.read();
  // do something with the record
});
stream.on('end', function() {
  console.log('finished');
});

You can also use the stream in flowing mode by attaching a data event listener.

var stream = parser.stream;
stream.on('data', function(record) {
  // do something with the record
});
stream.on('end', function() {
  console.log('finished');
});

Lastly, you can also pipe the stream like you would any other readable stream.

var stream = parser.stream;
var writableStream = somehowGetWritableStream();
stream.pipe(writableStream);

parser.header

Returns the header object, which contains information like the modified date, number of records, and a list of the fields and their types and lengths.

parser.header.fields

Array object which contains information about fields. Every item is an object with next fields: name, type, displacement, length, decimalPlaces, indexFlag.

parser.header.fields.raw

If need, field value can be returned as raw buffer for custom parsing (e.g. convert encodings). To enable this behavior you need set raw property to true:

var DBF = require('stream-dbf');
var parser = new DBF(fileName, [options]);
parser.header.fields[1].raw = true;

Array mode

If recAsArray option is enabled emitted arrays will have parser.header.fields.length+2 items. Zero number item will be sequenceNumber and first item will be a deleted flag.

For searching field number by you can use getFieldNo(name[, case_sensitivity]) function:

var DBF = require('stream-dbf');
var parser = new DBF(fileName, {'recAsArray': true});
var idxName = parser.getFieldNo("Name");
parser.stream.on('data', function(record) {
  console.log("Name: " + record[idxName]);
});