This is an event-based dBase file parser for very efficiently reading data from *.dbf
files.
- ES5
- uses
iconv-lite
to support multiple encodings - uses Node streams
- accepts
stdin
- can be used as standalone app to convert
.dbf
to.csv
To get started, simply install the module using npm:
$ npm install @episage/dbf-parser
and then require
it:
var Parser = require('@episage/dbf-parser');
This is the main interface for reading data from dBase files.
- stream
readableStream
- stream fromprocess.stdin
orfs.createReadStream
or other - string
encoding
- the character encoding to use (default =utf-8
, usesiconv-lite
under the hood)
Creates a new parser and attaches it to the specified filename.
var fs = require('fs');
var Parser = require('@episage/dbf-parser');
var parser = Parser(fs.createReadStream('/path/to/my/dbase/file.dbf'));
- event
String
- The event name to listen for (see below for details) - listener
Function
- The callback to bind to the event
- header
Header
The header object as parsed from the dBase file
This event is emitted once the header has been parsed from the dBase file
- record
Object
An object representing the record that has been found
The record object will have a key for each field within the record, named after the field. It is trimmed (leading and trailing) of any blank characters (dBase files use \x20 for padding).
In addition to the fields, the object contains two special keys:
- @sequenceNumber
Number
indicates the order in which it was extracted - @deleted
Boolean
whether this record has been deleted or not
This object may look like:
{
"@sequenceNumber": 123,
"@deleted": false,
"firstName": "John",
"lastName": "Smith"
}
This event is emitted when the parser reached end of the stream or end of DBF database.
The following code example illustrates a very simple usage for this module:
var fs = require('fs');
var Parser = require('@episage/dbf-parser');
var parser = Parser(fs.createReadStream('/path/to/my/dbase/file.dbf'));
parser.on('header', (h) => {
console.log('dBase file header has been parsed');
console.log(h);
});
parser.on('record', (record) => {
console.log('Name: ' + record.firstName + ' ' + record.lastName); // Name: John Smith
});
The parser also supports a command-line interface (CLI) for converting DBF files to CSV. You can invoke it as follows:
$ dbf-parser < /path/to/file.dbf
This will write the converted rows to stdout
.
$ dbf-parser < /path/to/file.dbf > file.csv
For more help information on using the command line options, use the integrated help:
$ dbf-parser --help
On MacBook Pro (Retina, 13-inch, Early 2015):
iconv-lite
- ~5MB/s - currently implemented versionBuffer.toString()
- ~10MB/s
Tests are written in Mocha using Chai BDD for the expectations. Data on San Francisco zip codes was used as a reference test file - downloaded from SF OpenData and included in the ./test/fixtures/bayarea_zipcodes.dbf
file within the repository.
Some of the texts and inspiration for the rewrite were taken from https://github.com/abstractvector/node-dbf