Initial pass at jawk. A surprising amount of functionality is working\!

.gitignore

@@ -12,3 +12,5 @@ logs
 results
 
 npm-debug.log
+
+node_modules

example/BBS-list

@@ -0,0 +1,11 @@
+     aardvark     555-5553     1200/300          B
+     alpo-net     555-3412     2400/1200/300     A
+     barfly       555-7685     1200/300          A
+     bites        555-1675     2400/1200/300     A
+     camelot      555-0542     300               C
+     core         555-2912     1200/300          C
+     fooey        555-1234     2400/1200/300     B
+     foot         555-6699     1200/300          B
+     macfoo       555-6480     1200/300          A
+     sdace        555-3430     2400/1200/300     A
+     sabafoo      555-2127     1200/300          C

example/inventory-shipped

@@ -0,0 +1,17 @@
+     Jan  13  25  15 115
+     Feb  15  32  24 226
+     Mar  15  24  34 228
+     Apr  31  52  63 420
+     May  16  34  29 208
+     Jun  31  42  75 492
+     Jul  24  34  67 436
+     Aug  15  34  47 316
+     Sep  13  55  37 277
+     Oct  29  54  68 525
+     Nov  20  87  82 577
+     Dec  17  35  61 401
+
+     Jan  21  36  64 620
+     Feb  26  58  80 652
+     Mar  24  75  70 495
+     Apr  21  70  74 514

example/test.jawk

@@ -0,0 +1,5 @@
+BEGIN { var barfl = 0; foo = 0; print('starting up!'); }
+/barfl/ { print($0 + ' (barfl=' + (++barfl) + ')') }		# this does a bunch of nothing!
+/foo/ { print($0 + ' (foo=' + (++foo) + ')') }
+{ print( $0 +' ('+$1+','+$2+','+$3+','+$4+','+$5+','+$6+')' ) }
+END { print( 'total barfl: ' + barfl + '\ntotal foo: ' + foo ) }

jawk.js

@@ -0,0 +1,96 @@
+var fs = require('fs'),
+	vm = require('vm'),
+	readlines = require('./lib/readlines').readlines;
+
+var opts = require('nomnom')
+	.option( 'file', {
+		abbr: 'f',
+		flag: false,
+		help: 'Script file'
+	})
+	.option( 'verbose', {
+		abbr: 'v',
+		flag: true,
+		help: 'Verbose output'
+	})
+	.parse();
+
+var program;
+var inputFname;
+var inputEncoding = 'utf8';
+
+if( opts.file ) {
+	program = fs.readFileSync( opts.file, 'utf8' ).replace(/\r/g,'');
+	// TODO: handle case of console input
+	inputFname = opts[0];
+} else {
+	program = opts[0].replace(/\r/g,'');
+	// TODO: handle case of console input
+	inputFname = opts[1];
+}
+
+// note that for functions below, this!=jawkContext, so we have to use jawkContext explicitly.  this is a little
+// fragile, but it works, so i'm leaving it for now....
+var jawkContext = vm.createContext({
+	RS: '\\s+',
+	NR: 0,
+	print: function(s) { console.log( s===undefined ? jawkContext.$0 : s ); },
+	length: function(x) { return x===undefined ? jawkContext.$0.length : x.length; }
+});
+
+var begin, end,
+	rules = [];
+
+var lineRegex = /^(.*?)\s*(\{.*?\})?\s*(#.*)?$/;
+
+var programLineNumber = 0;
+program.split('\n').forEach( function(line) {
+	programLineNumber++;
+	var m = line.match( lineRegex );
+	if( !m ) return;
+	// note that rules can consist only of a pattern (with no action; the action defaults to printing the whole input line)
+	if( m[2]===undefined ) m[2] = 'print($0)';
+	// construct the rule; note that we remove the brackets from the action
+	var rule = { line: line, lineNumber: programLineNumber, pattern: m[1], action: m[2].replace(/^\{(.*)\}$/,'$1').trim(), comment: m[3] };
+	if( rule.pattern.trim() === 'BEGIN' ) {
+		begin = rule;
+	} else if( rule.pattern.trim() === 'END' ) {
+		end = rule;
+	} else {
+		// the rule may already have slashes around it, so we remove those
+		rule.regex = new RegExp( rule.pattern.replace(/^\/?(.*?)\/?$/,'$1') );
+		rules.push( rule );
+	}
+});
+
+function executeAction( rule, context ) {
+	try {
+		vm.runInContext( rule.action, context );
+	} catch( ex ) {
+		console.log( 'invalid jawk rule:' );
+		console.log( rule.lineNumber + ': ' + rule.line );
+		if( opts.verbose ) console.log( ex );
+		process.exit( 1 );
+	}
+}
+
+if( begin ) executeAction( begin, jawkContext );
+
+readlines( inputFname, 'utf8', function(line) {
+	jawkContext.NR++;
+	for( var i=0; i<rules.length; i++ ) {
+		if( line.match( rules[i].regex ) ) {
+			jawkContext.$0 = line;
+			// build record fields
+			var fields = line.split( new RegExp( jawkContext.RS ) );
+			jawkContext.NR = fields.length;
+			var j;
+			for( j=0; j<fields.length; j++ ) jawkContext['$'+(j+1)] = fields[j];
+			for( j=fields.length; j<1000; j++ ) jawkContext['$'+(j+1)] = '';
+			executeAction( rules[i], jawkContext );
+			continue;		// once we've found a matching rule, we're done; move to the next input line
+		}
+	}
+});
+
+if( end ) executeAction( end, jawkContext );

lib/readlines.js

@@ -0,0 +1,42 @@
+var fs = require('fs');
+
+exports.readlines = function( inputFname, inputEncoding, processLine ) {
+
+	try {
+		var fd = fs.openSync( inputFname, 'r' );
+
+		var bufSize = 4096,
+			buf = new Buffer( bufSize ),
+			bufStr,
+			remainder = '',
+			newlineIdx,
+			bytesRead = 0;
+
+		do {
+			bytesRead = fs.readSync( fd, buf, 0, bufSize );
+			// get a string representation of the buffer, and remove any pesky CRs (lines in Windows files end in CRLF, not just LF)
+			bufStr = buf.toString(inputEncoding,0,bytesRead).replace(/\r/g,'');
+			// we split on LFs; if there are any full lines contained within the buffer, they will be processed
+			var lines = bufStr.split('\n');
+			if( lines.length > 1 ) {
+				// process first line (which starts with the last buffer's remainder)
+				processLine( remainder + lines[0] );
+				remainder = '';
+				// process everything but the last line (which will become the remainder)...
+				for( var i=1; i<lines.length-1; i++ ) {
+					// process line
+					processLine( lines[i] );
+				}
+			}
+			// the last line becomes the remainder
+			remainder += lines[lines.length-1];
+		} while( bytesRead > 0 );
+		// if the line isn't terminated with a line ending, there will be a remainder
+		if( remainder.length ) processLine( remainder );
+
+	} finally {
+		if( fd ) fs.close( fd );
+	}
+
+}
+

package.json

@@ -0,0 +1,25 @@
+{
+	"name": "jawk",
+	"description": "Interpreter for jawk, an awk-like data proccessing language with JavaScript syntax.",
+	"author": "Ethan Brown <[email protected]>",
+	"contributors": [
+		"Ethan Brown <[email protected]>"
+	],
+	"version": "0.0.1beta",
+	"license": "GPL",
+	"dependencies": {
+		"nomnom": "~1.6"
+	}
+	"keywords": [
+		"awk",
+		"data-driven",
+		"interpreter",
+		"language"
+	],
+	"repository": "https://github.com/Jammerwoch/jawk.git",
+	"main": "jawk",
+	"bin" : {
+		"jawk" : "./jawk.js"
+	}
+}
+