sv - npm Package Compare versions

Comparing version 0.2.0 to 0.2.2

index.js

 #!/usr/bin/env node
 'use strict'; /*jslint node: true, es5: true, indent: 2 */
 
+var os = require('os');
+var util = require('util');
+var stream = require('stream');
 var Parser = exports.Parser = require('./parser');
@@ -15,2 +18,46 @@ var Stringifier = exports.Stringifier = require('./stringifier');
 
+var JSONStringifier = function(opts) {
+  stream.Transform.call(this, {objectMode: true});
+  this._readableState.objectMode = false;
+};
+util.inherits(JSONStringifier, stream.Transform);
+JSONStringifier.prototype._transform = function(chunk, encoding, callback) {
+  this.push(JSON.stringify(chunk) + os.EOL, encoding);
+  callback();
+};
+
+var ObjectFilter = function(fields) {
+  // objects in, objects out
+  stream.Transform.call(this, {objectMode: true});
+  this.fields = {};
+  for (var i = 0, l = fields.length; i < l; i++) {
+    this.fields[fields[i]] = 1;
+  }
+};
+util.inherits(ObjectFilter, stream.Transform);
+ObjectFilter.prototype._transform = function(chunk, encoding, callback) {
+  var filtered = {};
+  for (var field in this.fields)
+    filtered[field] = chunk[field];
+  this.push(filtered, encoding);
+  callback();
+};
+
+var ObjectOmitter = function(fields) {
+  // objects in, objects out
+  stream.Transform.call(this, {objectMode: true});
+  this.fields = {};
+  for (var i = 0, l = fields.length; i < l; i++) {
+    this.fields[fields[i]] = 1;
+  }
+};
+util.inherits(ObjectOmitter, stream.Transform);
+ObjectOmitter.prototype._transform = function(chunk, encoding, callback) {
+  for (var field in this.fields)
+    delete chunk[field];
+  this.push(chunk, encoding);
+  callback();
+};
+
+
 if (require.main === module) {
@@ -21,19 +68,52 @@ var optimist = require('optimist')
       '',
-      '  argv will be passed directly to the Stringifier constructor.',
-      '  process.stdin will be set to utf8',
+      'Usage: <sprints.txt sv [options] > sprints.csv',
       '',
-      '  cat data.tsv | sv > data.csv'
-    ].join('\n'));
+      'Options:',
+      '  -p, --peek 10       infer columns from first ten lines of input',
+      '  -d, --delimiter ,   field separator',
+      '  -q, --quotechar "   mark beginning and end of fields containing delimiter',
+      '  -e, --escapechar \\  escape quotechars when quoted',
+      '  -j, --json          write one JSON object per row',
+      '      --filter a,b    keep only fields a and b in the results',
+      '      --omit c,d      leave out fields x and y from the results',
+      '                      (do not use filter and omit together)',
+      '',
+      'Only STDIN is supported, and it is coerced to utf8',
+    ].join('\n'))
+    .string('delimiter')
+    .boolean(['json',])
+    .alias({
+      p: 'peek',
+      d: 'delimiter',
+      q: 'quotechar',
+      e: 'escapechar',
+      j: 'json',
+    });
+  var argv = optimist.argv;
 
-  var parser = new Parser();
-  var stringifier = new Stringifier(optimist.argv);
-
-  if (process.stdin.isTTY) {
+  if (argv.help) {
     optimist.showHelp();
-    console.error("You must supply data via STDIN");
+    console.log('ARGV: ' + process.argv.join(' '));
   }
+  else if (process.stdin.isTTY) {
+    optimist.showHelp();
+    console.error('You must supply data via STDIN');
+  }
   else {
+    var stringifier = argv.json ? new JSONStringifier() : new Stringifier(argv);
+
     process.stdin.setEncoding('utf8');
-    process.stdin.pipe(parser).pipe(stringifier).pipe(process.stdout);
+    var parser = process.stdin.pipe(new Parser());
+    var filtered = null;
+    if (argv.filter) {
+      filtered = parser.pipe(new ObjectFilter(argv.filter.split(/,/g)));
+    }
+    else if (argv.omit) {
+      filtered = parser.pipe(new ObjectOmitter(argv.omit.split(/,/g)));
+    }
+    else {
+      filtered = parser;
+    }
+    filtered.pipe(stringifier).pipe(process.stdout);
   }
 }

package.json

 {
   "name": "sv",
-  "version": "0.2.0",
+  "version": "0.2.2",
   "description": "Any separated values.",
@@ -25,6 +25,6 @@ "keywords": [
   "dependencies": {
-    "optimist": "*"
+    "optimist": "0.5.2"
   },
   "devDependencies": {
-    "tap": "*"
+    "tap": "0.4.x"
   },
@@ -31,0 +31,0 @@ "scripts": {

parser.js

@@ -82,5 +82,10 @@ 'use strict'; /*jslint node: true, es5: true, indent: 2 */
   for (var i = 0; i < end; i++) {
+    // if we are on an escape char, simply skip over it (++) and the (default)
     if (buffer[i] === this.escapechar) {
-      i++;
+      // excel is bizarre. An escape before a quotechar doesn't count.
+      if (buffer[i+1] !== this.quotechar) {
+        i++;
+      }
     }
+    // if we are outside quoting and on a "
     else if (!inside && buffer[i] === this.quotechar) {
@@ -90,10 +95,23 @@ inside = true;
     }
+    // if we are inside quoting and on a "
     else if (inside && buffer[i] === this.quotechar) {
-      inside = false;
-      cells.push(buffer.toString(this.encoding, start, i));
-      // assume that an end quotechar is always followed by a delimiter
-      // advance so that buffer[i] == '\t'
+      // handle excel dialect: double quotechar => single literal quotechar
+      if (buffer[i+1] === this.quotechar) {
+        // double quotechar
+        // `inside` remains true
+        // we need to collapse out the current index. this might be optimized somehow
+        // buffer.copy(targetBuffer, [targetStart], [sourceStart], [sourceEnd])#
+        buffer.copy(buffer, i, i+1);
+        end--;
+      }
+      else {
+        // otherwise, assume that an end quotechar is always followed by a delimiter.
+        // advance so that buffer[i] == '\t'
+        inside = false;
+        cells.push(buffer.toString(this.encoding, start, i));
+        start = i + 2;
+      }
       i++;
-      start = i + 1;
     }
+    // otherwise we just wait for the delimiter
     else if (!inside && buffer[i] === this.delimiter) {
@@ -106,3 +124,3 @@ cells.push(buffer.toString(this.encoding, start, i));
     // we may have consumed the last field, already, if it was quoted.
-    cells.push(buffer.toString(this.encoding, start));
+    cells.push(buffer.toString(this.encoding, start, end));
   }
@@ -120,3 +138,2 @@
 Parser.prototype._flush = function(callback) {
-  // console.log('_flush');
   // if there was a trailing newline, this._buffer.length = 0
@@ -123,0 +140,0 @@ if (this._buffer && this._buffer.length)

stringifier.js

@@ -12,7 +12,9 @@ 'use strict'; /*jslint node: true, es5: true, indent: 2 */
     if (typeof(obj) !== 'string' && !util.isArray(obj)) {
-      var keys = Object.keys(obj);
-      for (var key, k = 0; (key = keys[k]); k++) {
-        if (!(key in seen)) {
-          columns.push(key);
-          seen[key] = 1;
+      for (var key in obj) {
+        if (obj.hasOwnProperty(key)) {
+          // maybe should also check that obj[key] != null
+          if (!(key in seen)) {
+            columns.push(key);
+            seen[key] = 1;
+          }
         }
@@ -19,0 +21,0 @@ }

test/basic.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');
@@ -62,1 +62,20 @@ var test = require('tap').test;
 });
+
+test('excel parser', function (t) {
+  var input = [
+    'index  name  time',
+    '1  "chris ""breezy"" brown" 1:18',
+    '2  "stephen" 1:16',
+  ].join('\n');
+
+  var rows = [];
+  var parser = new sv.Parser();
+  parser.on('data', function(obj) {
+    rows.push(obj);
+  });
+  parser.end(input, function() {
+    t.equal(rows.length, 2, 'There should be two rows.');
+    t.equal(rows[0].name, 'chris "breezy" brown', 'The paired double quotes should be interpreted as just one double quote.');
+    t.end();
+  });
+});

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

Wildcard dependency

Quality

Package has a dependency with a floating version range. This can cause issues if the dependency publishes a new major version.

Found 1 instance in 1 package

Major refactor

Supply chain risk

Package has recently undergone a major refactor. It may be unstable or indicate significant internal changes. Use caution when updating to versions that include significant changes.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

23170

increased by20.16%

Number of package files

11

increased by10%

Lines of code

557

increased by25.17%

Number of medium quality alerts

0

decreased by-100%

Number of medium supply chain risk alerts

0

decreased by-100%

Dependency changes

• + Addedoptimist@0.5.2(transitive)

• - Removedminimist@0.0.10(transitive)

• - Removedoptimist@0.6.1(transitive)

• Updatedoptimist@0.5.2