sv - npm Package Compare versions

Comparing version 0.3.2 to 0.3.3

bin/sv.js

@@ -17,8 +17,7 @@ #!/usr/bin/env node
       'Parser options:',
-      '  --peek 10         infer columns from first ten lines of input',
       '  --in-delimiter    field separator (inferred if unspecified)',
       '  --in-quotechar "  ',
-      // '  --escapechar \\  escape quotechars when quoted',
       '',
       'Stringifier options:',
+      '  --peek 10         infer columns from first ten objects of input',
       '  --out-delimiter , field separator',
@@ -50,3 +49,2 @@ '  --out-quotechar " marks beginning and end of fields containing delimiter',
   var parser_opts = {
-    peek: argv.peek,
     delimiter: argv['in-delimiter'],
@@ -58,2 +56,3 @@ quotechar: argv['in-quotechar'],
     quotechar: argv['out-quotechar'],
+    peek: argv.peek,
     filter: argv.filter,
@@ -65,3 +64,3 @@ omit: argv.omit,
 
-  // func: function (stream, filename, parser_opts, stringifier_opts, callback) { ... }
+  // func: function(stream, filename, parser_opts, stringifier_opts, callback) { ... }
   var func = argv.describe ? sv.describe : sv.transform;
@@ -68,0 +67,0 @@ var exit = function(err) {

example.js

@@ -32,3 +32,3 @@ var sv = require('./'); // require('sv') elsewhere
 
-  var parser = new sv.Parser({peek: 3});
+  var parser = new sv.Parser();
   parser.on('data', function(row) {
@@ -35,0 +35,0 @@ console.log(row);

inference.js

@@ -58,4 +58,5 @@ 'use strict'; /*jslint node: true, es5: true, indent: 2 */
   for (var candidate, j = 0; (candidate = candidates[j]); j++) {
-    if (counts[candidate] > 0)
+    if (counts[candidate] > 0) {
       return candidate;
+    }
   }
@@ -62,0 +63,0 @@ };

package.json

 {
   "name": "sv",
-  "version": "0.3.2",
+  "version": "0.3.3",
   "description": "Any separated values.",
@@ -22,3 +22,3 @@ "keywords": [
   },
-  "author": "Christopher Brown <[e-mail]>",
+  "author": "Christopher Brown <[e-mail]> (http://henrian.com)",
   "license": "MIT",
@@ -25,0 +25,0 @@ "dependencies": {

parser.js

@@ -7,14 +7,22 @@ 'use strict'; /*jslint node: true, es5: true, indent: 2 */
 
-/* Parser class
-  new Parser();
-  - `_bytes_buffer` is a buffer (of bytes) that have yet to be processed.
-  - `_cells_buffer` is a list of strings that have yet to be processed.
-  - `delimiter` is the field separator used for incoming strings.
-  - `columns` is an array of strings used as object keys.
-    * They are inferred if they are missing once the headers have been inferred.
-  - `missing` is the value we use for 'time' when we have
-    `columns = ['index', 'time']` and `write({index: 90})` is called.
-    at before inferring headers and flushing
-*/
 var Parser = module.exports = function(opts) {
+  /** new Parser(opts)
+
+  Options (`opts`):
+
+  - `encoding` for converting to strings.
+  - `missing` is the value we use for 'time' when we have `columns = ['index', 'time']` and `write({index: 90})` is called.
+  - `columns` is an array of strings used as object keys. Inferred by default (but after inferring `delimiter`)
+
+  - `delimiter` is the field separator used for incoming strings, ',' for csv, '\t' for tsv, etc.
+  - `quote` is the value that designates a string in which there might be delimiters to ignore. Defaults to '"'
+  - `escape` is the character that escapes special characters in a quoted field
+
+  Private values:
+
+  - `_bytes_buffer` is a buffer (of bytes) that have yet to be processed (and sent to output).
+  - `_cells_buffer` is a list of strings that have yet to be processed (and sent to output).
+  - etc.
+
+  */
   stream.Transform.call(this, {
@@ -24,7 +32,7 @@ decodeStrings: true, // Writable option, ensure _transform always gets a Buffer
   });
-  // this._readableState.objectMode = true; // default
+  // this._readableState.objectMode = true; // default, good
   // decodeStrings: true, dammit! ()
   // stream.Transform({decodeStrings: true}) is not honored if objectMode: true,
   // because objectMode: true (intended for the Readable) overrides the decodeStrings: true
-  // if this gets fixed, you can remove the setting below.
+  // if this gets fixed, you can remove the private field setting below.
   // Issue at https://github.com/joyent/node/issues/5580
@@ -34,15 +42,23 @@ this._writableState.objectMode = false;
   if (opts === undefined) opts = {};
-  this.missing = opts.missing || ''; // should be a string
-  this.delimiter = opts.delimiter;
+
+  // arbitrary settings (non-inferrable, but with sane & safe defaults)
+  this.encoding = opts.encoding;
+  this.missing_string = opts.missing || ''; // should be a string
   this.columns = opts.columns;
-  this.encoding = opts.encoding;
-  this.escapechar = opts.escapechar || '\\';
-  this.escapebyte = this.escapechar.charCodeAt(0);
-  this.quotechar = opts.quotechar;
-  if (this.quotechar) {
-    this.quotecharquotechar_regex = new RegExp(this.quotechar + this.quotechar, 'g');
-    this.escapequotechar_regex = new RegExp('\\\\' + this.quotechar, 'g');
-    this.quotebyte = this.quotechar.charCodeAt(0);
-  }
 
+  // special demarcating characters
+  // 1. delimiter
+  this._delimiter_byte = opts.delimiter ? opts.delimiter.charCodeAt(0) : null;
+
+  // 2. quote
+  this._quote_string = opts.quote || '"';
+  this._quote_byte = this._quote_string.charCodeAt(0);
+  this._quotequote_regex = new RegExp(this._quote_string + this._quote_string, 'g');
+
+  // 3. escape
+  var escape_string = opts.escape || '\\';
+  this._escape_byte = escape_string.charCodeAt(0);
+  this._escapequote_regex = new RegExp('\\' + escape_string + this._quote_string, 'g');
+
+  // private storage
   this._bytes_buffer = new Buffer(0);
@@ -59,3 +75,3 @@ this._cells_buffer = [];
   else {
-    this.push(inference.zip(this.columns, cells, this.missing));
+    this.push(inference.zip(this.columns, cells, this.missing_string));
   }
@@ -68,5 +84,5 @@ };
 
-  if (!this.delimiter) {
+  if (!this._delimiter_byte) {
     // should we wait for some minimum amount of data?
-    this.delimiter = inference.delimiter(buffer);
+    this._delimiter_byte = inference.delimiter(buffer);
   }
@@ -88,14 +104,14 @@
     // if we are on an escape char, simply skip over it (++) and the (default)
-    if (!eos && buffer[i] == this.escapebyte) {
+    if (!eos && buffer[i] == this._escape_byte) {
       // excel is bizarre. An escape before a quotebyte doesn't count,
       //   so we only increment if the next character is not a quotebyte
       // unless we are not inside quotes, in which case we do skip over it.
-      if (!inside_quote || buffer[i+1] !== this.quotebyte) {
+      if (!inside_quote || buffer[i+1] !== this._quote_byte) {
         i++;
       }
     }
-    else if (!eos && buffer[i] === this.quotebyte && inside_quote) {
+    else if (!eos && buffer[i] === this._quote_byte && inside_quote) {
       // if we are inside, and on a "
       // handle excel dialect: double quotebyte => single literal quotebyte
-      if (buffer[i+1] === this.quotebyte) {
+      if (buffer[i+1] === this._quote_byte) {
         // double quotebyte
@@ -113,3 +129,3 @@ // we just advance over it for now, so that we can put this back on the buffer, if needed.
     }
-    else if (!eos && buffer[i] === this.quotebyte && !inside_quote && i == start) {
+    else if (!eos && buffer[i] === this._quote_byte && !inside_quote && i == start) {
       // if we are not already inside, and on a "
@@ -124,3 +140,3 @@ inside_quote = true;
       // OR, we push a new cell whenever we hit a delimiter (say, tab) and are not inside a quote
-      (!inside_quote && (buffer[i] == 13 || buffer[i] == 10 || buffer[i] == this.delimiter))
+      (!inside_quote && (buffer[i] == 13 || buffer[i] == 10 || buffer[i] == this._delimiter_byte))
       ) {
@@ -135,4 +151,4 @@
         var trimmed_cell = buffer.toString(this.encoding, start + 1, i - 1);
-        if (this.quotecharquotechar_regex) {
-          trimmed_cell = trimmed_cell.replace(this.quotecharquotechar_regex, this.quotechar);
+        if (this._quotequote_regex) {
+          trimmed_cell = trimmed_cell.replace(this._quotequote_regex, this._quote_string);
         }
@@ -145,4 +161,4 @@ // is this good enough?
         var cell = buffer.toString(this.encoding, start, i);
-        if (this.escapequotechar_regex) {
-          cell = cell.replace(this.escapequotechar_regex, this.quotechar);
+        if (this._escapequote_regex) {
+          cell = cell.replace(this._escapequote_regex, this._quote_string);
         }
@@ -155,3 +171,3 @@ cells.push(cell);
       // we flush the row, also, if we are at the end and this is the final chunk
-      if (eos || (buffer[i] != this.delimiter)) {
+      if (eos || (buffer[i] != this._delimiter_byte)) {
         // add these cells to the emit queue
@@ -158,0 +174,0 @@ this._row(cells);

stringifier.js

@@ -10,3 +10,3 @@ 'use strict'; /*jslint node: true, es5: true, indent: 2 */
   - `peek` is an integer (or undefined / null) describing how many rows we
-    should peek at before inferring headers and flushing.
+    should look at before inferring headers and flushing.
   - `columns` is an array of strings once the headers have been inferred
@@ -13,0 +13,0 @@ - `encoding` is the encoding that the stream's read function will use.

test/basic.js

 'use strict'; /*jslint node: true, es5: true, indent: 2 */
 var fs = require('fs');
 var tap = require('tap');
+var streaming = require('streaming');
 
 var sv = require('..');
 
-tap.test('import', function (t) {
+tap.test('import', function(t) {
   t.ok(sv !== undefined, 'sv should load from the current directory');
@@ -12,3 +13,3 @@ t.end();
 
-tap.test('parser', function (t) {
+tap.test('parser', function(t) {
   var input = [
@@ -23,8 +24,4 @@ 'index	name	time',
 
-  var rows = [];
-  var parser = new sv.Parser({quotechar: '"'});
-  parser.on('data', function(obj) {
-    rows.push(obj);
-  });
-  parser.end(input, function() {
+  var parser = new sv.Parser();
+  streaming.readToEnd(parser, function(err, rows) {
     t.ok(rows[2], 'There should be a third row');
@@ -34,5 +31,6 @@ t.equal(rows[2].name, 'lewis', 'The name attribute of the third row should be "lewis"');
   });
+  parser.end(input);
 });
 
-tap.test('stringify', function (t) {
+tap.test('stringify', function(t) {
   var expected = [
@@ -49,8 +47,4 @@ 'index,name,time',
   var stringifier = new sv.Stringifier({peek: 2, missing: 'NA'});
-  var string = '';
-  stringifier.on('data', function(chunk) {
-    string += chunk.toString();
-  });
-  stringifier.on('end', function() {
-    t.equal(string, expected, 'Stringify output should equal expected.');
+  streaming.readToEnd(stringifier, function(err, chunks) {
+    t.equal(chunks.join(''), expected, 'Stringify output should equal expected.');
     t.end();
@@ -57,0 +51,0 @@ });

test/quotes.js

@@ -7,3 +7,3 @@ 'use strict'; /*jslint node: true, es5: true, indent: 2 */
 
-tap.test('excel dialect parser', function (t) {
+tap.test('excel dialect parser', function(t) {
   var input = [
@@ -18,3 +18,3 @@ 'index\tname\ttime',
   var rows = [];
-  var parser = new sv.Parser({quotechar: '"'});
+  var parser = new sv.Parser();
   parser.on('data', function(obj) {
@@ -33,3 +33,3 @@ rows.push(obj);
 
-tap.test('quoted newlines', function (t) {
+tap.test('quoted newlines', function(t) {
   var input = [
@@ -42,3 +42,3 @@ 'index  name  time',
   var rows = [];
-  var parser = new sv.Parser({quotechar: '"'});
+  var parser = new sv.Parser();
   parser.on('data', function(obj) {
@@ -45,0 +45,0 @@ rows.push(obj);

test/throughput.js

@@ -1,2 +0,2 @@
-'use strict'; /*jslint node: true, es5: true, indent: 2, multistr: true */
+'use strict'; /*jslint node: true, es5: true, indent: 2 */
 var fs = require('fs');
@@ -7,14 +7,14 @@ var tap = require('tap');
 
-var input = [
-  { index: '1', name: 'chris', time: '1:29' },
-  { index: '2', name: 'daniel', time: '1:17' },
-  { index: '3', name: 'lewis', time: '1:30' },
-  { index: '4', name: 'stephen', time: '1:16' },
-  { index: '5', name: 'larry', time: '1:31' },
-];
+tap.test('passthrough', function(t) {
+  var input = [
+    { index: '1', name: 'chris', time: '1:29' },
+    { index: '2', name: 'daniel', time: '1:17' },
+    { index: '3', name: 'lewis', time: '1:30' },
+    { index: '4', name: 'stephen', time: '1:16' },
+    { index: '5', name: 'larry', time: '1:31' },
+  ];
 
-tap.test('passthrough', function(t) {
   var stringifier = new sv.Stringifier({peek: 2, missing: 'NA'});
-  var parser = new sv.Parser({encoding: stringifier.encoding, quotechar: '"'});
 
+  var parser = stringifier.pipe(new sv.Parser());
   streaming.readToEnd(parser, function(err, output) {
@@ -25,3 +25,2 @@ t.equivalent(output, input, 'Throughput should be transparent.');
 
-  stringifier.pipe(parser);
   input.forEach(function(record) {
@@ -34,3 +33,3 @@ stringifier.write(record);
 tap.test('filter', function(t) {
-  var parser = new sv.Parser({quotechar: '"'});
+  var parser = new sv.Parser();
   var filter = new streaming.Filter(['index', 'name']);
@@ -58,3 +57,3 @@ parser.pipe(filter);
 tap.test('omitter', function(t) {
-  var parser = new sv.Parser({quotechar: '"'});
+  var parser = new sv.Parser();
   var omitter = new streaming.Omitter(['index', 'name']);
@@ -61,0 +60,0 @@ parser.pipe(omitter);

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

45023

increased by21.67%

Number of package files

25

increased by56.25%

Lines of code

1007

increased by13.91%

Worsened metrics

Number of low supply chain risk alerts

9

increased by12.5%

No dependency changes