add node_modules to address #4364

34 files changed, 6891 insertions, 0 deletions

diff --git a/node_modules/glob-stream/LICENSE b/node_modules/glob-stream/LICENSE
new file mode 100755
index 000000000..4f482f9ba
--- /dev/null
+++ b/node_modules/glob-stream/LICENSE
@@ -0,0 +1,20 @@
+Copyright (c) 2013 Fractal <contact@wearefractal.com>
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
diff --git a/node_modules/glob-stream/README.md b/node_modules/glob-stream/README.md
new file mode 100644
index 000000000..b0dbd4544
--- /dev/null
+++ b/node_modules/glob-stream/README.md
@@ -0,0 +1,67 @@
+# glob-stream [![NPM version][npm-image]][npm-url] [![Build Status][travis-image]][travis-url] [![Coveralls Status][coveralls-image]][coveralls-url] [![Dependency Status][david-image]][david-url]
+
+
+## Information
+
+<table>
+<tr> 
+<td>Package</td><td>glob-stream</td>
+</tr>
+<tr>
+<td>Description</td>
+<td>File system globs as a stream</td>
+</tr>
+<tr>
+<td>Node Version</td>
+<td>>= 0.9</td>
+</tr>
+</table>
+
+This is a simple wrapper around node-glob to make it streamy.
+
+## Usage
+
+```javascript
+var gs = require('glob-stream');
+
+var stream = gs.create("./files/**/*.coffee", {options});
+
+stream.on('data', function(file){
+  // file has path, base, and cwd attrs
+});
+```
+
+You can pass any combination of globs. One caveat is that you can not only pass a glob negation, you must give it at least one positive glob so it knows where to start. All given must match for the file to be returned.
+
+### Options
+
+- cwd
+  - Default is `process.cwd()`
+- base
+  - Default is everything before a glob starts (see [glob2base](https://github.com/wearefractal/glob2base))
+- cwdbase
+  - Default is `false`
+  - When true it is the same as saying opt.base = opt.cwd
+
+This argument is passed directly to [node-glob](https://github.com/isaacs/node-glob) so check there for more options
+
+#### Glob
+
+```javascript
+var stream = gs.create(["./**/*.js", "!./node_modules/**/*.*"]);
+```
+
+[npm-url]: https://npmjs.org/package/glob-stream
+[npm-image]: https://badge.fury.io/js/glob-stream.png
+
+[travis-url]: https://travis-ci.org/wearefractal/glob-stream
+[travis-image]: https://travis-ci.org/wearefractal/glob-stream.png?branch=master
+
+[coveralls-url]: https://coveralls.io/r/wearefractal/glob-stream
+[coveralls-image]: https://coveralls.io/repos/wearefractal/glob-stream/badge.png
+
+[depstat-url]: https://david-dm.org/wearefractal/glob-stream
+[depstat-image]: https://david-dm.org/wearefractal/glob-stream.png
+
+[david-url]: https://david-dm.org/wearefractal/glob-stream
+[david-image]: https://david-dm.org/wearefractal/glob-stream.png?theme=shields.io
diff --git a/node_modules/glob-stream/index.js b/node_modules/glob-stream/index.js
new file mode 100644
index 000000000..0960c7c74
--- /dev/null
+++ b/node_modules/glob-stream/index.js
@@ -0,0 +1,117 @@
+/*jslint node: true */
+
+'use strict';
+
+var through2 = require('through2');
+var Combine = require('ordered-read-streams');
+var unique = require('unique-stream');
+
+var glob = require('glob');
+var minimatch = require('minimatch');
+var glob2base = require('glob2base');
+var path = require('path');
+
+var gs = {
+  // creates a stream for a single glob or filter
+  createStream: function(ourGlob, negatives, opt) {
+    if (!negatives) negatives = [];
+    if (!opt) opt = {};
+    if (typeof opt.cwd !== 'string') opt.cwd = process.cwd();
+    if (typeof opt.dot !== 'boolean') opt.dot = false;
+    if (typeof opt.silent !== 'boolean') opt.silent = true;
+    if (typeof opt.nonull !== 'boolean') opt.nonull = false;
+    if (typeof opt.cwdbase !== 'boolean') opt.cwdbase = false;
+    if (opt.cwdbase) opt.base = opt.cwd;
+
+    // remove path relativity to make globs make sense
+    ourGlob = unrelative(opt.cwd, ourGlob);
+    negatives = negatives.map(unrelative.bind(null, opt.cwd));
+
+    // create globbing stuff
+    var globber = new glob.Glob(ourGlob, opt);
+
+    // extract base path from glob
+    var basePath = opt.base ? opt.base : glob2base(globber);
+
+    // create stream and map events from globber to it
+    var stream = through2.obj(negatives.length ? filterNegatives : undefined);
+
+    globber.on('error', stream.emit.bind(stream, 'error'));
+    globber.on('end', function(/* some args here so can't use bind directly */){
+      stream.end();
+    });
+    globber.on('match', function(filename) {
+      stream.write({
+        cwd: opt.cwd,
+        base: basePath,
+        path: path.resolve(opt.cwd, filename)
+      });
+    });
+
+    return stream;
+
+    function filterNegatives(filename, enc, cb) {
+      var matcha = isMatch.bind(null, filename, opt);
+      if (negatives.every(matcha)) {
+        cb(null, filename); // pass
+      } else {
+        cb(); // ignore
+      }
+    }
+  },
+
+  // creates a stream for multiple globs or filters
+  create: function(globs, opt) {
+    if (!opt) opt = {};
+
+    // only one glob no need to aggregate
+    if (!Array.isArray(globs)) return gs.createStream(globs, null, opt);
+
+    var positives = globs.filter(isPositive);
+    var negatives = globs.filter(isNegative);
+
+    if (positives.length === 0) throw new Error("Missing positive glob");
+
+    // only one positive glob no need to aggregate
+    if (positives.length === 1) return gs.createStream(positives[0], negatives, opt);
+
+    // create all individual streams
+    var streams = positives.map(function(glob){
+      return gs.createStream(glob, negatives, opt);
+    });
+
+    // then just pipe them to a single unique stream and return it
+    var aggregate = new Combine(streams);
+    var uniqueStream = unique('path');
+
+    return aggregate.pipe(uniqueStream);
+  }
+};
+
+function isMatch(file, opt, pattern) {
+  if (typeof pattern === 'string') return minimatch(file.path, pattern, opt);
+  if (pattern instanceof RegExp) return pattern.test(file.path);
+  return true; // unknown glob type?
+}
+
+function isNegative(pattern) {
+  if (typeof pattern !== 'string') return true;
+  if (pattern[0] === '!') return true;
+  return false;
+}
+
+function isPositive(pattern) {
+  return !isNegative(pattern);
+}
+
+function unrelative(cwd, glob) {
+  var mod = '';
+  if (glob[0] === '!') {
+    mod = glob[0];
+    glob = glob.slice(1);
+  }
+  return mod+path.resolve(cwd, glob);
+}
+
+
+module.exports = gs;
diff --git a/node_modules/glob-stream/node_modules/glob/LICENSE b/node_modules/glob-stream/node_modules/glob/LICENSE
new file mode 100644
index 000000000..19129e315
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/glob/LICENSE
@@ -0,0 +1,15 @@
+The ISC License
+
+Copyright (c) Isaac Z. Schlueter and Contributors
+
+Permission to use, copy, modify, and/or distribute this software for any
+purpose with or without fee is hereby granted, provided that the above
+copyright notice and this permission notice appear in all copies.
+
+THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR
+IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
diff --git a/node_modules/glob-stream/node_modules/glob/README.md b/node_modules/glob-stream/node_modules/glob/README.md
new file mode 100644
index 000000000..258257ecb
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/glob/README.md
@@ -0,0 +1,369 @@
+[![Build Status](https://travis-ci.org/isaacs/node-glob.svg?branch=master)](https://travis-ci.org/isaacs/node-glob/) [![Dependency Status](https://david-dm.org/isaacs/node-glob.svg)](https://david-dm.org/isaacs/node-glob) [![devDependency Status](https://david-dm.org/isaacs/node-glob/dev-status.svg)](https://david-dm.org/isaacs/node-glob#info=devDependencies) [![optionalDependency Status](https://david-dm.org/isaacs/node-glob/optional-status.svg)](https://david-dm.org/isaacs/node-glob#info=optionalDependencies)
+
+# Glob
+
+Match files using the patterns the shell uses, like stars and stuff.
+
+This is a glob implementation in JavaScript.  It uses the `minimatch`
+library to do its matching.
+
+![](oh-my-glob.gif)
+
+## Usage
+
+```javascript
+var glob = require("glob")
+
+// options is optional
+glob("**/*.js", options, function (er, files) {
+  // files is an array of filenames.
+  // If the `nonull` option is set, and nothing
+  // was found, then files is ["**/*.js"]
+  // er is an error object or null.
+})
+```
+
+## Glob Primer
+
+"Globs" are the patterns you type when you do stuff like `ls *.js` on
+the command line, or put `build/*` in a `.gitignore` file.
+
+Before parsing the path part patterns, braced sections are expanded
+into a set.  Braced sections start with `{` and end with `}`, with any
+number of comma-delimited sections within.  Braced sections may contain
+slash characters, so `a{/b/c,bcd}` would expand into `a/b/c` and `abcd`.
+
+The following characters have special magic meaning when used in a
+path portion:
+
+* `*` Matches 0 or more characters in a single path portion
+* `?` Matches 1 character
+* `[...]` Matches a range of characters, similar to a RegExp range.
+  If the first character of the range is `!` or `^` then it matches
+  any character not in the range.
+* `!(pattern|pattern|pattern)` Matches anything that does not match
+  any of the patterns provided.
+* `?(pattern|pattern|pattern)` Matches zero or one occurrence of the
+  patterns provided.
+* `+(pattern|pattern|pattern)` Matches one or more occurrences of the
+  patterns provided.
+* `*(a|b|c)` Matches zero or more occurrences of the patterns provided
+* `@(pattern|pat*|pat?erN)` Matches exactly one of the patterns
+  provided
+* `**` If a "globstar" is alone in a path portion, then it matches
+  zero or more directories and subdirectories searching for matches.
+  It does not crawl symlinked directories.
+
+### Dots
+
+If a file or directory path portion has a `.` as the first character,
+then it will not match any glob pattern unless that pattern's
+corresponding path part also has a `.` as its first character.
+
+For example, the pattern `a/.*/c` would match the file at `a/.b/c`.
+However the pattern `a/*/c` would not, because `*` does not start with
+a dot character.
+
+You can make glob treat dots as normal characters by setting
+`dot:true` in the options.
+
+### Basename Matching
+
+If you set `matchBase:true` in the options, and the pattern has no
+slashes in it, then it will seek for any file anywhere in the tree
+with a matching basename.  For example, `*.js` would match
+`test/simple/basic.js`.
+
+### Negation
+
+The intent for negation would be for a pattern starting with `!` to
+match everything that *doesn't* match the supplied pattern.  However,
+the implementation is weird, and for the time being, this should be
+avoided.  The behavior will change or be deprecated in version 5.
+
+### Empty Sets
+
+If no matching files are found, then an empty array is returned.  This
+differs from the shell, where the pattern itself is returned.  For
+example:
+
+    $ echo a*s*d*f
+    a*s*d*f
+
+To get the bash-style behavior, set the `nonull:true` in the options.
+
+### See Also:
+
+* `man sh`
+* `man bash` (Search for "Pattern Matching")
+* `man 3 fnmatch`
+* `man 5 gitignore`
+* [minimatch documentation](https://github.com/isaacs/minimatch)
+
+## glob.hasMagic(pattern, [options])
+
+Returns `true` if there are any special characters in the pattern, and
+`false` otherwise.
+
+Note that the options affect the results.  If `noext:true` is set in
+the options object, then `+(a|b)` will not be considered a magic
+pattern.  If the pattern has a brace expansion, like `a/{b/c,x/y}`
+then that is considered magical, unless `nobrace:true` is set in the
+options.
+
+## glob(pattern, [options], cb)
+
+* `pattern` {String} Pattern to be matched
+* `options` {Object}
+* `cb` {Function}
+  * `err` {Error | null}
+  * `matches` {Array<String>} filenames found matching the pattern
+
+Perform an asynchronous glob search.
+
+## glob.sync(pattern, [options])
+
+* `pattern` {String} Pattern to be matched
+* `options` {Object}
+* return: {Array<String>} filenames found matching the pattern
+
+Perform a synchronous glob search.
+
+## Class: glob.Glob
+
+Create a Glob object by instantiating the `glob.Glob` class.
+
+```javascript
+var Glob = require("glob").Glob
+var mg = new Glob(pattern, options, cb)
+```
+
+It's an EventEmitter, and starts walking the filesystem to find matches
+immediately.
+
+### new glob.Glob(pattern, [options], [cb])
+
+* `pattern` {String} pattern to search for
+* `options` {Object}
+* `cb` {Function} Called when an error occurs, or matches are found
+  * `err` {Error | null}
+  * `matches` {Array<String>} filenames found matching the pattern
+
+Note that if the `sync` flag is set in the options, then matches will
+be immediately available on the `g.found` member.
+
+### Properties
+
+* `minimatch` The minimatch object that the glob uses.
+* `options` The options object passed in.
+* `aborted` Boolean which is set to true when calling `abort()`.  There
+  is no way at this time to continue a glob search after aborting, but
+  you can re-use the statCache to avoid having to duplicate syscalls.
+* `statCache` Collection of all the stat results the glob search
+  performed.
+* `cache` Convenience object.  Each field has the following possible
+  values:
+  * `false` - Path does not exist
+  * `true` - Path exists
+  * `'DIR'` - Path exists, and is not a directory
+  * `'FILE'` - Path exists, and is a directory
+  * `[file, entries, ...]` - Path exists, is a directory, and the
+    array value is the results of `fs.readdir`
+* `statCache` Cache of `fs.stat` results, to prevent statting the same
+  path multiple times.
+* `symlinks` A record of which paths are symbolic links, which is
+  relevant in resolving `**` patterns.
+* `realpathCache` An optional object which is passed to `fs.realpath`
+  to minimize unnecessary syscalls.  It is stored on the instantiated
+  Glob object, and may be re-used.
+
+### Events
+
+* `end` When the matching is finished, this is emitted with all the
+  matches found.  If the `nonull` option is set, and no match was found,
+  then the `matches` list contains the original pattern.  The matches
+  are sorted, unless the `nosort` flag is set.
+* `match` Every time a match is found, this is emitted with the matched.
+* `error` Emitted when an unexpected error is encountered, or whenever
+  any fs error occurs if `options.strict` is set.
+* `abort` When `abort()` is called, this event is raised.
+
+### Methods
+
+* `pause` Temporarily stop the search
+* `resume` Resume the search
+* `abort` Stop the search forever
+
+### Options
+
+All the options that can be passed to Minimatch can also be passed to
+Glob to change pattern matching behavior.  Also, some have been added,
+or have glob-specific ramifications.
+
+All options are false by default, unless otherwise noted.
+
+All options are added to the Glob object, as well.
+
+If you are running many `glob` operations, you can pass a Glob object
+as the `options` argument to a subsequent operation to shortcut some
+`stat` and `readdir` calls.  At the very least, you may pass in shared
+`symlinks`, `statCache`, `realpathCache`, and `cache` options, so that
+parallel glob operations will be sped up by sharing information about
+the filesystem.
+
+* `cwd` The current working directory in which to search.  Defaults
+  to `process.cwd()`.
+* `root` The place where patterns starting with `/` will be mounted
+  onto.  Defaults to `path.resolve(options.cwd, "/")` (`/` on Unix
+  systems, and `C:\` or some such on Windows.)
+* `dot` Include `.dot` files in normal matches and `globstar` matches.
+  Note that an explicit dot in a portion of the pattern will always
+  match dot files.
+* `nomount` By default, a pattern starting with a forward-slash will be
+  "mounted" onto the root setting, so that a valid filesystem path is
+  returned.  Set this flag to disable that behavior.
+* `mark` Add a `/` character to directory matches.  Note that this
+  requires additional stat calls.
+* `nosort` Don't sort the results.
+* `stat` Set to true to stat *all* results.  This reduces performance
+  somewhat, and is completely unnecessary, unless `readdir` is presumed
+  to be an untrustworthy indicator of file existence.
+* `silent` When an unusual error is encountered when attempting to
+  read a directory, a warning will be printed to stderr.  Set the
+  `silent` option to true to suppress these warnings.
+* `strict` When an unusual error is encountered when attempting to
+  read a directory, the process will just continue on in search of
+  other matches.  Set the `strict` option to raise an error in these
+  cases.
+* `cache` See `cache` property above.  Pass in a previously generated
+  cache object to save some fs calls.
+* `statCache` A cache of results of filesystem information, to prevent
+  unnecessary stat calls.  While it should not normally be necessary
+  to set this, you may pass the statCache from one glob() call to the
+  options object of another, if you know that the filesystem will not
+  change between calls.  (See "Race Conditions" below.)
+* `symlinks` A cache of known symbolic links.  You may pass in a
+  previously generated `symlinks` object to save `lstat` calls when
+  resolving `**` matches.
+* `sync` DEPRECATED: use `glob.sync(pattern, opts)` instead.
+* `nounique` In some cases, brace-expanded patterns can result in the
+  same file showing up multiple times in the result set.  By default,
+  this implementation prevents duplicates in the result set.  Set this
+  flag to disable that behavior.
+* `nonull` Set to never return an empty set, instead returning a set
+  containing the pattern itself.  This is the default in glob(3).
+* `debug` Set to enable debug logging in minimatch and glob.
+* `nobrace` Do not expand `{a,b}` and `{1..3}` brace sets.
+* `noglobstar` Do not match `**` against multiple filenames.  (Ie,
+  treat it as a normal `*` instead.)
+* `noext` Do not match `+(a|b)` "extglob" patterns.
+* `nocase` Perform a case-insensitive match.  Note: on
+  case-insensitive filesystems, non-magic patterns will match by
+  default, since `stat` and `readdir` will not raise errors.
+* `matchBase` Perform a basename-only match if the pattern does not
+  contain any slash characters.  That is, `*.js` would be treated as
+  equivalent to `**/*.js`, matching all js files in all directories.
+* `nonegate` Suppress `negate` behavior.  (See below.)
+* `nocomment` Suppress `comment` behavior.  (See below.)
+* `nonull` Return the pattern when no matches are found.
+* `nodir` Do not match directories, only files.  (Note: to match
+  *only* directories, simply put a `/` at the end of the pattern.)
+* `ignore` Add a pattern or an array of patterns to exclude matches.
+* `follow` Follow symlinked directories when expanding `**` patterns.
+  Note that this can result in a lot of duplicate references in the
+  presence of cyclic links.
+* `realpath` Set to true to call `fs.realpath` on all of the results.
+  In the case of a symlink that cannot be resolved, the full absolute
+  path to the matched entry is returned (though it will usually be a
+  broken symlink)
+
+## Comparisons to other fnmatch/glob implementations
+
+While strict compliance with the existing standards is a worthwhile
+goal, some discrepancies exist between node-glob and other
+implementations, and are intentional.
+
+If the pattern starts with a `!` character, then it is negated.  Set the
+`nonegate` flag to suppress this behavior, and treat leading `!`
+characters normally.  This is perhaps relevant if you wish to start the
+pattern with a negative extglob pattern like `!(a|B)`.  Multiple `!`
+characters at the start of a pattern will negate the pattern multiple
+times.
+
+If a pattern starts with `#`, then it is treated as a comment, and
+will not match anything.  Use `\#` to match a literal `#` at the
+start of a line, or set the `nocomment` flag to suppress this behavior.
+
+The double-star character `**` is supported by default, unless the
+`noglobstar` flag is set.  This is supported in the manner of bsdglob
+and bash 4.3, where `**` only has special significance if it is the only
+thing in a path part.  That is, `a/**/b` will match `a/x/y/b`, but
+`a/**b` will not.
+
+Note that symlinked directories are not crawled as part of a `**`,
+though their contents may match against subsequent portions of the
+pattern.  This prevents infinite loops and duplicates and the like.
+
+If an escaped pattern has no matches, and the `nonull` flag is set,
+then glob returns the pattern as-provided, rather than
+interpreting the character escapes.  For example,
+`glob.match([], "\\*a\\?")` will return `"\\*a\\?"` rather than
+`"*a?"`.  This is akin to setting the `nullglob` option in bash, except
+that it does not resolve escaped pattern characters.
+
+If brace expansion is not disabled, then it is performed before any
+other interpretation of the glob pattern.  Thus, a pattern like
+`+(a|{b),c)}`, which would not be valid in bash or zsh, is expanded
+**first** into the set of `+(a|b)` and `+(a|c)`, and those patterns are
+checked for validity.  Since those two are valid, matching proceeds.
+
+## Windows
+
+**Please only use forward-slashes in glob expressions.**
+
+Though windows uses either `/` or `\` as its path separator, only `/`
+characters are used by this glob implementation.  You must use
+forward-slashes **only** in glob expressions.  Back-slashes will always
+be interpreted as escape characters, not path separators.
+
+Results from absolute patterns such as `/foo/*` are mounted onto the
+root setting using `path.join`.  On windows, this will by default result
+in `/foo/*` matching `C:\foo\bar.txt`.
+
+## Race Conditions
+
+Glob searching, by its very nature, is susceptible to race conditions,
+since it relies on directory walking and such.
+
+As a result, it is possible that a file that exists when glob looks for
+it may have been deleted or modified by the time it returns the result.
+
+As part of its internal implementation, this program caches all stat
+and readdir calls that it makes, in order to cut down on system
+overhead.  However, this also makes it even more susceptible to races,
+especially if the cache or statCache objects are reused between glob
+calls.
+
+Users are thus advised not to use a glob result as a guarantee of
+filesystem state in the face of rapid changes.  For the vast majority
+of operations, this is never a problem.
+
+## Contributing
+
+Any change to behavior (including bugfixes) must come with a test.
+
+Patches that fail tests or reduce performance will be rejected.
+
+```
+# to run tests
+npm test
+
+# to re-generate test fixtures
+npm run test-regen
+
+# to benchmark against bash/zsh
+npm run bench
+
+# to profile javascript
+npm run prof
+```
diff --git a/node_modules/glob-stream/node_modules/glob/common.js b/node_modules/glob-stream/node_modules/glob/common.js
new file mode 100644
index 000000000..cd7c82448
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/glob/common.js
@@ -0,0 +1,237 @@
+exports.alphasort = alphasort
+exports.alphasorti = alphasorti
+exports.isAbsolute = process.platform === "win32" ? absWin : absUnix
+exports.setopts = setopts
+exports.ownProp = ownProp
+exports.makeAbs = makeAbs
+exports.finish = finish
+exports.mark = mark
+exports.isIgnored = isIgnored
+exports.childrenIgnored = childrenIgnored
+
+function ownProp (obj, field) {
+  return Object.prototype.hasOwnProperty.call(obj, field)
+}
+
+var path = require("path")
+var minimatch = require("minimatch")
+var Minimatch = minimatch.Minimatch
+
+function absWin (p) {
+  if (absUnix(p)) return true
+  // pull off the device/UNC bit from a windows path.
+  // from node's lib/path.js
+  var splitDeviceRe =
+      /^([a-zA-Z]:|[\\\/]{2}[^\\\/]+[\\\/]+[^\\\/]+)?([\\\/])?([\s\S]*?)$/
+  var result = splitDeviceRe.exec(p)
+  var device = result[1] || ''
+  var isUnc = device && device.charAt(1) !== ':'
+  var isAbsolute = !!result[2] || isUnc // UNC paths are always absolute
+
+  return isAbsolute
+}
+
+function absUnix (p) {
+  return p.charAt(0) === "/" || p === ""
+}
+
+function alphasorti (a, b) {
+  return a.toLowerCase().localeCompare(b.toLowerCase())
+}
+
+function alphasort (a, b) {
+  return a.localeCompare(b)
+}
+
+function setupIgnores (self, options) {
+  self.ignore = options.ignore || []
+
+  if (!Array.isArray(self.ignore))
+    self.ignore = [self.ignore]
+
+  if (self.ignore.length) {
+    self.ignore = self.ignore.map(ignoreMap)
+  }
+}
+
+function ignoreMap (pattern) {
+  var gmatcher = null
+  if (pattern.slice(-3) === '/**') {
+    var gpattern = pattern.replace(/(\/\*\*)+$/, '')
+    gmatcher = new Minimatch(gpattern, { nonegate: true })
+  }
+
+  return {
+    matcher: new Minimatch(pattern, { nonegate: true }),
+    gmatcher: gmatcher
+  }
+}
+
+function setopts (self, pattern, options) {
+  if (!options)
+    options = {}
+
+  // base-matching: just use globstar for that.
+  if (options.matchBase && -1 === pattern.indexOf("/")) {
+    if (options.noglobstar) {
+      throw new Error("base matching requires globstar")
+    }
+    pattern = "**/" + pattern
+  }
+
+  self.pattern = pattern
+  self.strict = options.strict !== false
+  self.realpath = !!options.realpath
+  self.realpathCache = options.realpathCache || Object.create(null)
+  self.follow = !!options.follow
+  self.dot = !!options.dot
+  self.mark = !!options.mark
+  self.nodir = !!options.nodir
+  if (self.nodir)
+    self.mark = true
+  self.sync = !!options.sync
+  self.nounique = !!options.nounique
+  self.nonull = !!options.nonull
+  self.nosort = !!options.nosort
+  self.nocase = !!options.nocase
+  self.stat = !!options.stat
+  self.noprocess = !!options.noprocess
+
+  self.maxLength = options.maxLength || Infinity
+  self.cache = options.cache || Object.create(null)
+  self.statCache = options.statCache || Object.create(null)
+  self.symlinks = options.symlinks || Object.create(null)
+
+  setupIgnores(self, options)
+
+  self.changedCwd = false
+  var cwd = process.cwd()
+  if (!ownProp(options, "cwd"))
+    self.cwd = cwd
+  else {
+    self.cwd = options.cwd
+    self.changedCwd = path.resolve(options.cwd) !== cwd
+  }
+
+  self.root = options.root || path.resolve(self.cwd, "/")
+  self.root = path.resolve(self.root)
+  if (process.platform === "win32")
+    self.root = self.root.replace(/\\/g, "/")
+
+  self.nomount = !!options.nomount
+
+  self.minimatch = new Minimatch(pattern, options)
+  self.options = self.minimatch.options
+}
+
+function finish (self) {
+  var nou = self.nounique
+  var all = nou ? [] : Object.create(null)
+
+  for (var i = 0, l = self.matches.length; i < l; i ++) {
+    var matches = self.matches[i]
+    if (!matches || Object.keys(matches).length === 0) {
+      if (self.nonull) {
+        // do like the shell, and spit out the literal glob
+        var literal = self.minimatch.globSet[i]
+        if (nou)
+          all.push(literal)
+        else
+          all[literal] = true
+      }
+    } else {
+      // had matches
+      var m = Object.keys(matches)
+      if (nou)
+        all.push.apply(all, m)
+      else
+        m.forEach(function (m) {
+          all[m] = true
+        })
+    }
+  }
+
+  if (!nou)
+    all = Object.keys(all)
+
+  if (!self.nosort)
+    all = all.sort(self.nocase ? alphasorti : alphasort)
+
+  // at *some* point we statted all of these
+  if (self.mark) {
+    for (var i = 0; i < all.length; i++) {
+      all[i] = self._mark(all[i])
+    }
+    if (self.nodir) {
+      all = all.filter(function (e) {
+        return !(/\/$/.test(e))
+      })
+    }
+  }
+
+  if (self.ignore.length)
+    all = all.filter(function(m) {
+      return !isIgnored(self, m)
+    })
+
+  self.found = all
+}
+
+function mark (self, p) {
+  var abs = makeAbs(self, p)
+  var c = self.cache[abs]
+  var m = p
+  if (c) {
+    var isDir = c === 'DIR' || Array.isArray(c)
+    var slash = p.slice(-1) === '/'
+
+    if (isDir && !slash)
+      m += '/'
+    else if (!isDir && slash)
+      m = m.slice(0, -1)
+
+    if (m !== p) {
+      var mabs = makeAbs(self, m)
+      self.statCache[mabs] = self.statCache[abs]
+      self.cache[mabs] = self.cache[abs]
+    }
+  }
+
+  return m
+}
+
+// lotta situps...
+function makeAbs (self, f) {
+  var abs = f
+  if (f.charAt(0) === '/') {
+    abs = path.join(self.root, f)
+  } else if (exports.isAbsolute(f)) {
+    abs = f
+  } else if (self.changedCwd) {
+    abs = path.resolve(self.cwd, f)
+  } else if (self.realpath) {
+    abs = path.resolve(f)
+  }
+  return abs
+}
+
+
+// Return true, if pattern ends with globstar '**', for the accompanying parent directory.
+// Ex:- If node_modules/** is the pattern, add 'node_modules' to ignore list along with it's contents
+function isIgnored (self, path) {
+  if (!self.ignore.length)
+    return false
+
+  return self.ignore.some(function(item) {
+    return item.matcher.match(path) || !!(item.gmatcher && item.gmatcher.match(path))
+  })
+}
+
+function childrenIgnored (self, path) {
+  if (!self.ignore.length)
+    return false
+
+  return self.ignore.some(function(item) {
+    return !!(item.gmatcher && item.gmatcher.match(path))
+  })
+}
diff --git a/node_modules/glob-stream/node_modules/glob/glob.js b/node_modules/glob-stream/node_modules/glob/glob.js
new file mode 100644
index 000000000..eac0693cc
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/glob/glob.js
@@ -0,0 +1,740 @@
+// Approach:
+//
+// 1. Get the minimatch set
+// 2. For each pattern in the set, PROCESS(pattern, false)
+// 3. Store matches per-set, then uniq them
+//
+// PROCESS(pattern, inGlobStar)
+// Get the first [n] items from pattern that are all strings
+// Join these together.  This is PREFIX.
+//   If there is no more remaining, then stat(PREFIX) and
+//   add to matches if it succeeds.  END.
+//
+// If inGlobStar and PREFIX is symlink and points to dir
+//   set ENTRIES = []
+// else readdir(PREFIX) as ENTRIES
+//   If fail, END
+//
+// with ENTRIES
+//   If pattern[n] is GLOBSTAR
+//     // handle the case where the globstar match is empty
+//     // by pruning it out, and testing the resulting pattern
+//     PROCESS(pattern[0..n] + pattern[n+1 .. $], false)
+//     // handle other cases.
+//     for ENTRY in ENTRIES (not dotfiles)
+//       // attach globstar + tail onto the entry
+//       // Mark that this entry is a globstar match
+//       PROCESS(pattern[0..n] + ENTRY + pattern[n .. $], true)
+//
+//   else // not globstar
+//     for ENTRY in ENTRIES (not dotfiles, unless pattern[n] is dot)
+//       Test ENTRY against pattern[n]
+//       If fails, continue
+//       If passes, PROCESS(pattern[0..n] + item + pattern[n+1 .. $])
+//
+// Caveat:
+//   Cache all stats and readdirs results to minimize syscall.  Since all
+//   we ever care about is existence and directory-ness, we can just keep
+//   `true` for files, and [children,...] for directories, or `false` for
+//   things that don't exist.
+
+module.exports = glob
+
+var fs = require('fs')
+var minimatch = require('minimatch')
+var Minimatch = minimatch.Minimatch
+var inherits = require('inherits')
+var EE = require('events').EventEmitter
+var path = require('path')
+var assert = require('assert')
+var globSync = require('./sync.js')
+var common = require('./common.js')
+var alphasort = common.alphasort
+var alphasorti = common.alphasorti
+var isAbsolute = common.isAbsolute
+var setopts = common.setopts
+var ownProp = common.ownProp
+var inflight = require('inflight')
+var util = require('util')
+var childrenIgnored = common.childrenIgnored
+
+var once = require('once')
+
+function glob (pattern, options, cb) {
+  if (typeof options === 'function') cb = options, options = {}
+  if (!options) options = {}
+
+  if (options.sync) {
+    if (cb)
+      throw new TypeError('callback provided to sync glob')
+    return globSync(pattern, options)
+  }
+
+  return new Glob(pattern, options, cb)
+}
+
+glob.sync = globSync
+var GlobSync = glob.GlobSync = globSync.GlobSync
+
+// old api surface
+glob.glob = glob
+
+glob.hasMagic = function (pattern, options_) {
+  var options = util._extend({}, options_)
+  options.noprocess = true
+
+  var g = new Glob(pattern, options)
+  var set = g.minimatch.set
+  if (set.length > 1)
+    return true
+
+  for (var j = 0; j < set[0].length; j++) {
+    if (typeof set[0][j] !== 'string')
+      return true
+  }
+
+  return false
+}
+
+glob.Glob = Glob
+inherits(Glob, EE)
+function Glob (pattern, options, cb) {
+  if (typeof options === 'function') {
+    cb = options
+    options = null
+  }
+
+  if (options && options.sync) {
+    if (cb)
+      throw new TypeError('callback provided to sync glob')
+    return new GlobSync(pattern, options)
+  }
+
+  if (!(this instanceof Glob))
+    return new Glob(pattern, options, cb)
+
+  setopts(this, pattern, options)
+  this._didRealPath = false
+
+  // process each pattern in the minimatch set
+  var n = this.minimatch.set.length
+
+  // The matches are stored as {<filename>: true,...} so that
+  // duplicates are automagically pruned.
+  // Later, we do an Object.keys() on these.
+  // Keep them as a list so we can fill in when nonull is set.
+  this.matches = new Array(n)
+
+  if (typeof cb === 'function') {
+    cb = once(cb)
+    this.on('error', cb)
+    this.on('end', function (matches) {
+      cb(null, matches)
+    })
+  }
+
+  var self = this
+  var n = this.minimatch.set.length
+  this._processing = 0
+  this.matches = new Array(n)
+
+  this._emitQueue = []
+  this._processQueue = []
+  this.paused = false
+
+  if (this.noprocess)
+    return this
+
+  if (n === 0)
+    return done()
+
+  for (var i = 0; i < n; i ++) {
+    this._process(this.minimatch.set[i], i, false, done)
+  }
+
+  function done () {
+    --self._processing
+    if (self._processing <= 0)
+      self._finish()
+  }
+}
+
+Glob.prototype._finish = function () {
+  assert(this instanceof Glob)
+  if (this.aborted)
+    return
+
+  if (this.realpath && !this._didRealpath)
+    return this._realpath()
+
+  common.finish(this)
+  this.emit('end', this.found)
+}
+
+Glob.prototype._realpath = function () {
+  if (this._didRealpath)
+    return
+
+  this._didRealpath = true
+
+  var n = this.matches.length
+  if (n === 0)
+    return this._finish()
+
+  var self = this
+  for (var i = 0; i < this.matches.length; i++)
+    this._realpathSet(i, next)
+
+  function next () {
+    if (--n === 0)
+      self._finish()
+  }
+}
+
+Glob.prototype._realpathSet = function (index, cb) {
+  var matchset = this.matches[index]
+  if (!matchset)
+    return cb()
+
+  var found = Object.keys(matchset)
+  var self = this
+  var n = found.length
+
+  if (n === 0)
+    return cb()
+
+  var set = this.matches[index] = Object.create(null)
+  found.forEach(function (p, i) {
+    // If there's a problem with the stat, then it means that
+    // one or more of the links in the realpath couldn't be
+    // resolved.  just return the abs value in that case.
+    p = self._makeAbs(p)
+    fs.realpath(p, self.realpathCache, function (er, real) {
+      if (!er)
+        set[real] = true
+      else if (er.syscall === 'stat')
+        set[p] = true
+      else
+        self.emit('error', er) // srsly wtf right here
+
+      if (--n === 0) {
+        self.matches[index] = set
+        cb()
+      }
+    })
+  })
+}
+
+Glob.prototype._mark = function (p) {
+  return common.mark(this, p)
+}
+
+Glob.prototype._makeAbs = function (f) {
+  return common.makeAbs(this, f)
+}
+
+Glob.prototype.abort = function () {
+  this.aborted = true
+  this.emit('abort')
+}
+
+Glob.prototype.pause = function () {
+  if (!this.paused) {
+    this.paused = true
+    this.emit('pause')
+  }
+}
+
+Glob.prototype.resume = function () {
+  if (this.paused) {
+    this.emit('resume')
+    this.paused = false
+    if (this._emitQueue.length) {
+      var eq = this._emitQueue.slice(0)
+      this._emitQueue.length = 0
+      for (var i = 0; i < eq.length; i ++) {
+        var e = eq[i]
+        this._emitMatch(e[0], e[1])
+      }
+    }
+    if (this._processQueue.length) {
+      var pq = this._processQueue.slice(0)
+      this._processQueue.length = 0
+      for (var i = 0; i < pq.length; i ++) {
+        var p = pq[i]
+        this._processing--
+        this._process(p[0], p[1], p[2], p[3])
+      }
+    }
+  }
+}
+
+Glob.prototype._process = function (pattern, index, inGlobStar, cb) {
+  assert(this instanceof Glob)
+  assert(typeof cb === 'function')
+
+  if (this.aborted)
+    return
+
+  this._processing++
+  if (this.paused) {
+    this._processQueue.push([pattern, index, inGlobStar, cb])
+    return
+  }
+
+  //console.error('PROCESS %d', this._processing, pattern)
+
+  // Get the first [n] parts of pattern that are all strings.
+  var n = 0
+  while (typeof pattern[n] === 'string') {
+    n ++
+  }
+  // now n is the index of the first one that is *not* a string.
+
+  // see if there's anything else
+  var prefix
+  switch (n) {
+    // if not, then this is rather simple
+    case pattern.length:
+      this._processSimple(pattern.join('/'), index, cb)
+      return
+
+    case 0:
+      // pattern *starts* with some non-trivial item.
+      // going to readdir(cwd), but not include the prefix in matches.
+      prefix = null
+      break
+
+    default:
+      // pattern has some string bits in the front.
+      // whatever it starts with, whether that's 'absolute' like /foo/bar,
+      // or 'relative' like '../baz'
+      prefix = pattern.slice(0, n).join('/')
+      break
+  }
+
+  var remain = pattern.slice(n)
+
+  // get the list of entries.
+  var read
+  if (prefix === null)
+    read = '.'
+  else if (isAbsolute(prefix) || isAbsolute(pattern.join('/'))) {
+    if (!prefix || !isAbsolute(prefix))
+      prefix = '/' + prefix
+    read = prefix
+  } else
+    read = prefix
+
+  var abs = this._makeAbs(read)
+
+  //if ignored, skip _processing
+  if (childrenIgnored(this, read))
+    return cb()
+
+  var isGlobStar = remain[0] === minimatch.GLOBSTAR
+  if (isGlobStar)
+    this._processGlobStar(prefix, read, abs, remain, index, inGlobStar, cb)
+  else
+    this._processReaddir(prefix, read, abs, remain, index, inGlobStar, cb)
+}
+
+Glob.prototype._processReaddir = function (prefix, read, abs, remain, index, inGlobStar, cb) {
+  var self = this
+  this._readdir(abs, inGlobStar, function (er, entries) {
+    return self._processReaddir2(prefix, read, abs, remain, index, inGlobStar, entries, cb)
+  })
+}
+
+Glob.prototype._processReaddir2 = function (prefix, read, abs, remain, index, inGlobStar, entries, cb) {
+
+  // if the abs isn't a dir, then nothing can match!
+  if (!entries)
+    return cb()
+
+  // It will only match dot entries if it starts with a dot, or if
+  // dot is set.  Stuff like @(.foo|.bar) isn't allowed.
+  var pn = remain[0]
+  var negate = !!this.minimatch.negate
+  var rawGlob = pn._glob
+  var dotOk = this.dot || rawGlob.charAt(0) === '.'
+
+  var matchedEntries = []
+  for (var i = 0; i < entries.length; i++) {
+    var e = entries[i]
+    if (e.charAt(0) !== '.' || dotOk) {
+      var m
+      if (negate && !prefix) {
+        m = !e.match(pn)
+      } else {
+        m = e.match(pn)
+      }
+      if (m)
+        matchedEntries.push(e)
+    }
+  }
+
+  //console.error('prd2', prefix, entries, remain[0]._glob, matchedEntries)
+
+  var len = matchedEntries.length
+  // If there are no matched entries, then nothing matches.
+  if (len === 0)
+    return cb()
+
+  // if this is the last remaining pattern bit, then no need for
+  // an additional stat *unless* the user has specified mark or
+  // stat explicitly.  We know they exist, since readdir returned
+  // them.
+
+  if (remain.length === 1 && !this.mark && !this.stat) {
+    if (!this.matches[index])
+      this.matches[index] = Object.create(null)
+
+    for (var i = 0; i < len; i ++) {
+      var e = matchedEntries[i]
+      if (prefix) {
+        if (prefix !== '/')
+          e = prefix + '/' + e
+        else
+          e = prefix + e
+      }
+
+      if (e.charAt(0) === '/' && !this.nomount) {
+        e = path.join(this.root, e)
+      }
+      this._emitMatch(index, e)
+    }
+    // This was the last one, and no stats were needed
+    return cb()
+  }
+
+  // now test all matched entries as stand-ins for that part
+  // of the pattern.
+  remain.shift()
+  for (var i = 0; i < len; i ++) {
+    var e = matchedEntries[i]
+    var newPattern
+    if (prefix) {
+      if (prefix !== '/')
+        e = prefix + '/' + e
+      else
+        e = prefix + e
+    }
+    this._process([e].concat(remain), index, inGlobStar, cb)
+  }
+  cb()
+}
+
+Glob.prototype._emitMatch = function (index, e) {
+  if (this.aborted)
+    return
+
+  if (this.matches[index][e])
+    return
+
+  if (this.paused) {
+    this._emitQueue.push([index, e])
+    return
+  }
+
+  var abs = this._makeAbs(e)
+
+  if (this.nodir) {
+    var c = this.cache[abs]
+    if (c === 'DIR' || Array.isArray(c))
+      return
+  }
+
+  if (this.mark)
+    e = this._mark(e)
+
+  this.matches[index][e] = true
+
+  var st = this.statCache[abs]
+  if (st)
+    this.emit('stat', e, st)
+
+  this.emit('match', e)
+}
+
+Glob.prototype._readdirInGlobStar = function (abs, cb) {
+  if (this.aborted)
+    return
+
+  // follow all symlinked directories forever
+  // just proceed as if this is a non-globstar situation
+  if (this.follow)
+    return this._readdir(abs, false, cb)
+
+  var lstatkey = 'lstat\0' + abs
+  var self = this
+  var lstatcb = inflight(lstatkey, lstatcb_)
+
+  if (lstatcb)
+    fs.lstat(abs, lstatcb)
+
+  function lstatcb_ (er, lstat) {
+    if (er)
+      return cb()
+
+    var isSym = lstat.isSymbolicLink()
+    self.symlinks[abs] = isSym
+
+    // If it's not a symlink or a dir, then it's definitely a regular file.
+    // don't bother doing a readdir in that case.
+    if (!isSym && !lstat.isDirectory()) {
+      self.cache[abs] = 'FILE'
+      cb()
+    } else
+      self._readdir(abs, false, cb)
+  }
+}
+
+Glob.prototype._readdir = function (abs, inGlobStar, cb) {
+  if (this.aborted)
+    return
+
+  cb = inflight('readdir\0'+abs+'\0'+inGlobStar, cb)
+  if (!cb)
+    return
+
+  //console.error('RD %j %j', +inGlobStar, abs)
+  if (inGlobStar && !ownProp(this.symlinks, abs))
+    return this._readdirInGlobStar(abs, cb)
+
+  if (ownProp(this.cache, abs)) {
+    var c = this.cache[abs]
+    if (!c || c === 'FILE')
+      return cb()
+
+    if (Array.isArray(c))
+      return cb(null, c)
+  }
+
+  var self = this
+  fs.readdir(abs, readdirCb(this, abs, cb))
+}
+
+function readdirCb (self, abs, cb) {
+  return function (er, entries) {
+    if (er)
+      self._readdirError(abs, er, cb)
+    else
+      self._readdirEntries(abs, entries, cb)
+  }
+}
+
+Glob.prototype._readdirEntries = function (abs, entries, cb) {
+  if (this.aborted)
+    return
+
+  // if we haven't asked to stat everything, then just
+  // assume that everything in there exists, so we can avoid
+  // having to stat it a second time.
+  if (!this.mark && !this.stat) {
+    for (var i = 0; i < entries.length; i ++) {
+      var e = entries[i]
+      if (abs === '/')
+        e = abs + e
+      else
+        e = abs + '/' + e
+      this.cache[e] = true
+    }
+  }
+
+  this.cache[abs] = entries
+  return cb(null, entries)
+}
+
+Glob.prototype._readdirError = function (f, er, cb) {
+  if (this.aborted)
+    return
+
+  // handle errors, and cache the information
+  switch (er.code) {
+    case 'ENOTDIR': // totally normal. means it *does* exist.
+      this.cache[this._makeAbs(f)] = 'FILE'
+      break
+
+    case 'ENOENT': // not terribly unusual
+    case 'ELOOP':
+    case 'ENAMETOOLONG':
+    case 'UNKNOWN':
+      this.cache[this._makeAbs(f)] = false
+      break
+
+    default: // some unusual error.  Treat as failure.
+      this.cache[this._makeAbs(f)] = false
+      if (this.strict) return this.emit('error', er)
+      if (!this.silent) console.error('glob error', er)
+      break
+  }
+  return cb()
+}
+
+Glob.prototype._processGlobStar = function (prefix, read, abs, remain, index, inGlobStar, cb) {
+  var self = this
+  this._readdir(abs, inGlobStar, function (er, entries) {
+    self._processGlobStar2(prefix, read, abs, remain, index, inGlobStar, entries, cb)
+  })
+}
+
+
+Glob.prototype._processGlobStar2 = function (prefix, read, abs, remain, index, inGlobStar, entries, cb) {
+  //console.error('pgs2', prefix, remain[0], entries)
+
+  // no entries means not a dir, so it can never have matches
+  // foo.txt/** doesn't match foo.txt
+  if (!entries)
+    return cb()
+
+  // test without the globstar, and with every child both below
+  // and replacing the globstar.
+  var remainWithoutGlobStar = remain.slice(1)
+  var gspref = prefix ? [ prefix ] : []
+  var noGlobStar = gspref.concat(remainWithoutGlobStar)
+
+  // the noGlobStar pattern exits the inGlobStar state
+  this._process(noGlobStar, index, false, cb)
+
+  var isSym = this.symlinks[abs]
+  var len = entries.length
+
+  // If it's a symlink, and we're in a globstar, then stop
+  if (isSym && inGlobStar)
+    return cb()
+
+  for (var i = 0; i < len; i++) {
+    var e = entries[i]
+    if (e.charAt(0) === '.' && !this.dot)
+      continue
+
+    // these two cases enter the inGlobStar state
+    var instead = gspref.concat(entries[i], remainWithoutGlobStar)
+    this._process(instead, index, true, cb)
+
+    var below = gspref.concat(entries[i], remain)
+    this._process(below, index, true, cb)
+  }
+
+  cb()
+}
+
+Glob.prototype._processSimple = function (prefix, index, cb) {
+  // XXX review this.  Shouldn't it be doing the mounting etc
+  // before doing stat?  kinda weird?
+  var self = this
+  this._stat(prefix, function (er, exists) {
+    self._processSimple2(prefix, index, er, exists, cb)
+  })
+}
+Glob.prototype._processSimple2 = function (prefix, index, er, exists, cb) {
+
+  //console.error('ps2', prefix, exists)
+
+  if (!this.matches[index])
+    this.matches[index] = Object.create(null)
+
+  // If it doesn't exist, then just mark the lack of results
+  if (!exists)
+    return cb()
+
+  if (prefix && isAbsolute(prefix) && !this.nomount) {
+    var trail = /[\/\\]$/.test(prefix)
+    if (prefix.charAt(0) === '/') {
+      prefix = path.join(this.root, prefix)
+    } else {
+      prefix = path.resolve(this.root, prefix)
+      if (trail)
+        prefix += '/'
+    }
+  }
+
+  if (process.platform === 'win32')
+    prefix = prefix.replace(/\\/g, '/')
+
+  // Mark this as a match
+  this._emitMatch(index, prefix)
+  cb()
+}
+
+// Returns either 'DIR', 'FILE', or false
+Glob.prototype._stat = function (f, cb) {
+  var abs = this._makeAbs(f)
+  var needDir = f.slice(-1) === '/'
+
+  if (f.length > this.maxLength)
+    return cb()
+
+  if (!this.stat && ownProp(this.cache, abs)) {
+    var c = this.cache[abs]
+
+    if (Array.isArray(c))
+      c = 'DIR'
+
+    // It exists, but maybe not how we need it
+    if (!needDir || c === 'DIR')
+      return cb(null, c)
+
+    if (needDir && c === 'FILE')
+      return cb()
+
+    // otherwise we have to stat, because maybe c=true
+    // if we know it exists, but not what it is.
+  }
+
+  var exists
+  var stat = this.statCache[abs]
+  if (stat !== undefined) {
+    if (stat === false)
+      return cb(null, stat)
+    else {
+      var type = stat.isDirectory() ? 'DIR' : 'FILE'
+      if (needDir && type === 'FILE')
+        return cb()
+      else
+        return cb(null, type, stat)
+    }
+  }
+
+  var self = this
+  var statcb = inflight('stat\0' + abs, lstatcb_)
+  if (statcb)
+    fs.lstat(abs, statcb)
+
+  function lstatcb_ (er, lstat) {
+    if (lstat && lstat.isSymbolicLink()) {
+      // If it's a symlink, then treat it as the target, unless
+      // the target does not exist, then treat it as a file.
+      return fs.stat(abs, function (er, stat) {
+        if (er)
+          self._stat2(f, abs, null, lstat, cb)
+        else
+          self._stat2(f, abs, er, stat, cb)
+      })
+    } else {
+      self._stat2(f, abs, er, lstat, cb)
+    }
+  }
+}
+
+Glob.prototype._stat2 = function (f, abs, er, stat, cb) {
+  if (er) {
+    this.statCache[abs] = false
+    return cb()
+  }
+
+  var needDir = f.slice(-1) === '/'
+  this.statCache[abs] = stat
+
+  if (abs.slice(-1) === '/' && !stat.isDirectory())
+    return cb(null, false, stat)
+
+  var c = stat.isDirectory() ? 'DIR' : 'FILE'
+  this.cache[abs] = this.cache[abs] || c
+
+  if (needDir && c !== 'DIR')
+    return cb()
+
+  return cb(null, c, stat)
+}
diff --git a/node_modules/glob-stream/node_modules/glob/package.json b/node_modules/glob-stream/node_modules/glob/package.json
new file mode 100644
index 000000000..887bad528
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/glob/package.json
@@ -0,0 +1,105 @@
+{
+  "_args": [
+    [
+      {
+        "raw": "glob@^4.3.1",
+        "scope": null,
+        "escapedName": "glob",
+        "name": "glob",
+        "rawSpec": "^4.3.1",
+        "spec": ">=4.3.1 <5.0.0",
+        "type": "range"
+      },
+      "/home/dold/repos/taler/wallet-webex/node_modules/glob-stream"
+    ]
+  ],
+  "_from": "glob@>=4.3.1 <5.0.0",
+  "_id": "glob@4.5.3",
+  "_inCache": true,
+  "_location": "/glob-stream/glob",
+  "_nodeVersion": "1.4.2",
+  "_npmUser": {
+    "name": "isaacs",
+    "email": "i@izs.me"
+  },
+  "_npmVersion": "2.7.1",
+  "_phantomChildren": {},
+  "_requested": {
+    "raw": "glob@^4.3.1",
+    "scope": null,
+    "escapedName": "glob",
+    "name": "glob",
+    "rawSpec": "^4.3.1",
+    "spec": ">=4.3.1 <5.0.0",
+    "type": "range"
+  },
+  "_requiredBy": [
+    "/glob-stream"
+  ],
+  "_resolved": "https://registry.npmjs.org/glob/-/glob-4.5.3.tgz",
+  "_shasum": "c6cb73d3226c1efef04de3c56d012f03377ee15f",
+  "_shrinkwrap": null,
+  "_spec": "glob@^4.3.1",
+  "_where": "/home/dold/repos/taler/wallet-webex/node_modules/glob-stream",
+  "author": {
+    "name": "Isaac Z. Schlueter",
+    "email": "i@izs.me",
+    "url": "http://blog.izs.me/"
+  },
+  "bugs": {
+    "url": "https://github.com/isaacs/node-glob/issues"
+  },
+  "dependencies": {
+    "inflight": "^1.0.4",
+    "inherits": "2",
+    "minimatch": "^2.0.1",
+    "once": "^1.3.0"
+  },
+  "description": "a little globber",
+  "devDependencies": {
+    "mkdirp": "0",
+    "rimraf": "^2.2.8",
+    "tap": "^0.5.0",
+    "tick": "0.0.6"
+  },
+  "directories": {},
+  "dist": {
+    "shasum": "c6cb73d3226c1efef04de3c56d012f03377ee15f",
+    "tarball": "https://registry.npmjs.org/glob/-/glob-4.5.3.tgz"
+  },
+  "engines": {
+    "node": "*"
+  },
+  "files": [
+    "glob.js",
+    "sync.js",
+    "common.js"
+  ],
+  "gitHead": "a4e461ab59a837eee80a4d8dbdbf5ae1054a646f",
+  "homepage": "https://github.com/isaacs/node-glob",
+  "license": "ISC",
+  "main": "glob.js",
+  "maintainers": [
+    {
+      "name": "isaacs",
+      "email": "i@izs.me"
+    }
+  ],
+  "name": "glob",
+  "optionalDependencies": {},
+  "readme": "ERROR: No README data found!",
+  "repository": {
+    "type": "git",
+    "url": "git://github.com/isaacs/node-glob.git"
+  },
+  "scripts": {
+    "bench": "bash benchmark.sh",
+    "benchclean": "bash benchclean.sh",
+    "prepublish": "npm run benchclean",
+    "prof": "bash prof.sh && cat profile.txt",
+    "profclean": "rm -f v8.log profile.txt",
+    "test": "npm run profclean && tap test/*.js",
+    "test-regen": "npm run profclean && TEST_REGEN=1 node test/00-setup.js"
+  },
+  "version": "4.5.3"
+}
diff --git a/node_modules/glob-stream/node_modules/glob/sync.js b/node_modules/glob-stream/node_modules/glob/sync.js
new file mode 100644
index 000000000..f4f5e36d4
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/glob/sync.js
@@ -0,0 +1,457 @@
+module.exports = globSync
+globSync.GlobSync = GlobSync
+
+var fs = require('fs')
+var minimatch = require('minimatch')
+var Minimatch = minimatch.Minimatch
+var Glob = require('./glob.js').Glob
+var util = require('util')
+var path = require('path')
+var assert = require('assert')
+var common = require('./common.js')
+var alphasort = common.alphasort
+var alphasorti = common.alphasorti
+var isAbsolute = common.isAbsolute
+var setopts = common.setopts
+var ownProp = common.ownProp
+var childrenIgnored = common.childrenIgnored
+
+function globSync (pattern, options) {
+  if (typeof options === 'function' || arguments.length === 3)
+    throw new TypeError('callback provided to sync glob\n'+
+                        'See: https://github.com/isaacs/node-glob/issues/167')
+
+  return new GlobSync(pattern, options).found
+}
+
+function GlobSync (pattern, options) {
+  if (!pattern)
+    throw new Error('must provide pattern')
+
+  if (typeof options === 'function' || arguments.length === 3)
+    throw new TypeError('callback provided to sync glob\n'+
+                        'See: https://github.com/isaacs/node-glob/issues/167')
+
+  if (!(this instanceof GlobSync))
+    return new GlobSync(pattern, options)
+
+  setopts(this, pattern, options)
+
+  if (this.noprocess)
+    return this
+
+  var n = this.minimatch.set.length
+  this.matches = new Array(n)
+  for (var i = 0; i < n; i ++) {
+    this._process(this.minimatch.set[i], i, false)
+  }
+  this._finish()
+}
+
+GlobSync.prototype._finish = function () {
+  assert(this instanceof GlobSync)
+  if (this.realpath) {
+    var self = this
+    this.matches.forEach(function (matchset, index) {
+      var set = self.matches[index] = Object.create(null)
+      for (var p in matchset) {
+        try {
+          p = self._makeAbs(p)
+          var real = fs.realpathSync(p, this.realpathCache)
+          set[real] = true
+        } catch (er) {
+          if (er.syscall === 'stat')
+            set[self._makeAbs(p)] = true
+          else
+            throw er
+        }
+      }
+    })
+  }
+  common.finish(this)
+}
+
+
+GlobSync.prototype._process = function (pattern, index, inGlobStar) {
+  assert(this instanceof GlobSync)
+
+  // Get the first [n] parts of pattern that are all strings.
+  var n = 0
+  while (typeof pattern[n] === 'string') {
+    n ++
+  }
+  // now n is the index of the first one that is *not* a string.
+
+  // See if there's anything else
+  var prefix
+  switch (n) {
+    // if not, then this is rather simple
+    case pattern.length:
+      this._processSimple(pattern.join('/'), index)
+      return
+
+    case 0:
+      // pattern *starts* with some non-trivial item.
+      // going to readdir(cwd), but not include the prefix in matches.
+      prefix = null
+      break
+
+    default:
+      // pattern has some string bits in the front.
+      // whatever it starts with, whether that's 'absolute' like /foo/bar,
+      // or 'relative' like '../baz'
+      prefix = pattern.slice(0, n).join('/')
+      break
+  }
+
+  var remain = pattern.slice(n)
+
+  // get the list of entries.
+  var read
+  if (prefix === null)
+    read = '.'
+  else if (isAbsolute(prefix) || isAbsolute(pattern.join('/'))) {
+    if (!prefix || !isAbsolute(prefix))
+      prefix = '/' + prefix
+    read = prefix
+  } else
+    read = prefix
+
+  var abs = this._makeAbs(read)
+
+  //if ignored, skip processing
+  if (childrenIgnored(this, read))
+    return
+
+  var isGlobStar = remain[0] === minimatch.GLOBSTAR
+  if (isGlobStar)
+    this._processGlobStar(prefix, read, abs, remain, index, inGlobStar)
+  else
+    this._processReaddir(prefix, read, abs, remain, index, inGlobStar)
+}
+
+
+GlobSync.prototype._processReaddir = function (prefix, read, abs, remain, index, inGlobStar) {
+  var entries = this._readdir(abs, inGlobStar)
+
+  // if the abs isn't a dir, then nothing can match!
+  if (!entries)
+    return
+
+  // It will only match dot entries if it starts with a dot, or if
+  // dot is set.  Stuff like @(.foo|.bar) isn't allowed.
+  var pn = remain[0]
+  var negate = !!this.minimatch.negate
+  var rawGlob = pn._glob
+  var dotOk = this.dot || rawGlob.charAt(0) === '.'
+
+  var matchedEntries = []
+  for (var i = 0; i < entries.length; i++) {
+    var e = entries[i]
+    if (e.charAt(0) !== '.' || dotOk) {
+      var m
+      if (negate && !prefix) {
+        m = !e.match(pn)
+      } else {
+        m = e.match(pn)
+      }
+      if (m)
+        matchedEntries.push(e)
+    }
+  }
+
+  var len = matchedEntries.length
+  // If there are no matched entries, then nothing matches.
+  if (len === 0)
+    return
+
+  // if this is the last remaining pattern bit, then no need for
+  // an additional stat *unless* the user has specified mark or
+  // stat explicitly.  We know they exist, since readdir returned
+  // them.
+
+  if (remain.length === 1 && !this.mark && !this.stat) {
+    if (!this.matches[index])
+      this.matches[index] = Object.create(null)
+
+    for (var i = 0; i < len; i ++) {
+      var e = matchedEntries[i]
+      if (prefix) {
+        if (prefix.slice(-1) !== '/')
+          e = prefix + '/' + e
+        else
+          e = prefix + e
+      }
+
+      if (e.charAt(0) === '/' && !this.nomount) {
+        e = path.join(this.root, e)
+      }
+      this.matches[index][e] = true
+    }
+    // This was the last one, and no stats were needed
+    return
+  }
+
+  // now test all matched entries as stand-ins for that part
+  // of the pattern.
+  remain.shift()
+  for (var i = 0; i < len; i ++) {
+    var e = matchedEntries[i]
+    var newPattern
+    if (prefix)
+      newPattern = [prefix, e]
+    else
+      newPattern = [e]
+    this._process(newPattern.concat(remain), index, inGlobStar)
+  }
+}
+
+
+GlobSync.prototype._emitMatch = function (index, e) {
+  var abs = this._makeAbs(e)
+  if (this.mark)
+    e = this._mark(e)
+
+  if (this.matches[index][e])
+    return
+
+  if (this.nodir) {
+    var c = this.cache[this._makeAbs(e)]
+    if (c === 'DIR' || Array.isArray(c))
+      return
+  }
+
+  this.matches[index][e] = true
+  if (this.stat)
+    this._stat(e)
+}
+
+
+GlobSync.prototype._readdirInGlobStar = function (abs) {
+  // follow all symlinked directories forever
+  // just proceed as if this is a non-globstar situation
+  if (this.follow)
+    return this._readdir(abs, false)
+
+  var entries
+  var lstat
+  var stat
+  try {
+    lstat = fs.lstatSync(abs)
+  } catch (er) {
+    // lstat failed, doesn't exist
+    return null
+  }
+
+  var isSym = lstat.isSymbolicLink()
+  this.symlinks[abs] = isSym
+
+  // If it's not a symlink or a dir, then it's definitely a regular file.
+  // don't bother doing a readdir in that case.
+  if (!isSym && !lstat.isDirectory())
+    this.cache[abs] = 'FILE'
+  else
+    entries = this._readdir(abs, false)
+
+  return entries
+}
+
+GlobSync.prototype._readdir = function (abs, inGlobStar) {
+  var entries
+
+  if (inGlobStar && !ownProp(this.symlinks, abs))
+    return this._readdirInGlobStar(abs)
+
+  if (ownProp(this.cache, abs)) {
+    var c = this.cache[abs]
+    if (!c || c === 'FILE')
+      return null
+
+    if (Array.isArray(c))
+      return c
+  }
+
+  try {
+    return this._readdirEntries(abs, fs.readdirSync(abs))
+  } catch (er) {
+    this._readdirError(abs, er)
+    return null
+  }
+}
+
+GlobSync.prototype._readdirEntries = function (abs, entries) {
+  // if we haven't asked to stat everything, then just
+  // assume that everything in there exists, so we can avoid
+  // having to stat it a second time.
+  if (!this.mark && !this.stat) {
+    for (var i = 0; i < entries.length; i ++) {
+      var e = entries[i]
+      if (abs === '/')
+        e = abs + e
+      else
+        e = abs + '/' + e
+      this.cache[e] = true
+    }
+  }
+
+  this.cache[abs] = entries
+
+  // mark and cache dir-ness
+  return entries
+}
+
+GlobSync.prototype._readdirError = function (f, er) {
+  // handle errors, and cache the information
+  switch (er.code) {
+    case 'ENOTDIR': // totally normal. means it *does* exist.
+      this.cache[this._makeAbs(f)] = 'FILE'
+      break
+
+    case 'ENOENT': // not terribly unusual
+    case 'ELOOP':
+    case 'ENAMETOOLONG':
+    case 'UNKNOWN':
+      this.cache[this._makeAbs(f)] = false
+      break
+
+    default: // some unusual error.  Treat as failure.
+      this.cache[this._makeAbs(f)] = false
+      if (this.strict) throw er
+      if (!this.silent) console.error('glob error', er)
+      break
+  }
+}
+
+GlobSync.prototype._processGlobStar = function (prefix, read, abs, remain, index, inGlobStar) {
+
+  var entries = this._readdir(abs, inGlobStar)
+
+  // no entries means not a dir, so it can never have matches
+  // foo.txt/** doesn't match foo.txt
+  if (!entries)
+    return
+
+  // test without the globstar, and with every child both below
+  // and replacing the globstar.
+  var remainWithoutGlobStar = remain.slice(1)
+  var gspref = prefix ? [ prefix ] : []
+  var noGlobStar = gspref.concat(remainWithoutGlobStar)
+
+  // the noGlobStar pattern exits the inGlobStar state
+  this._process(noGlobStar, index, false)
+
+  var len = entries.length
+  var isSym = this.symlinks[abs]
+
+  // If it's a symlink, and we're in a globstar, then stop
+  if (isSym && inGlobStar)
+    return
+
+  for (var i = 0; i < len; i++) {
+    var e = entries[i]
+    if (e.charAt(0) === '.' && !this.dot)
+      continue
+
+    // these two cases enter the inGlobStar state
+    var instead = gspref.concat(entries[i], remainWithoutGlobStar)
+    this._process(instead, index, true)
+
+    var below = gspref.concat(entries[i], remain)
+    this._process(below, index, true)
+  }
+}
+
+GlobSync.prototype._processSimple = function (prefix, index) {
+  // XXX review this.  Shouldn't it be doing the mounting etc
+  // before doing stat?  kinda weird?
+  var exists = this._stat(prefix)
+
+  if (!this.matches[index])
+    this.matches[index] = Object.create(null)
+
+  // If it doesn't exist, then just mark the lack of results
+  if (!exists)
+    return
+
+  if (prefix && isAbsolute(prefix) && !this.nomount) {
+    var trail = /[\/\\]$/.test(prefix)
+    if (prefix.charAt(0) === '/') {
+      prefix = path.join(this.root, prefix)
+    } else {
+      prefix = path.resolve(this.root, prefix)
+      if (trail)
+        prefix += '/'
+    }
+  }
+
+  if (process.platform === 'win32')
+    prefix = prefix.replace(/\\/g, '/')
+
+  // Mark this as a match
+  this.matches[index][prefix] = true
+}
+
+// Returns either 'DIR', 'FILE', or false
+GlobSync.prototype._stat = function (f) {
+  var abs = this._makeAbs(f)
+  var needDir = f.slice(-1) === '/'
+
+  if (f.length > this.maxLength)
+    return false
+
+  if (!this.stat && ownProp(this.cache, abs)) {
+    var c = this.cache[abs]
+
+    if (Array.isArray(c))
+      c = 'DIR'
+
+    // It exists, but maybe not how we need it
+    if (!needDir || c === 'DIR')
+      return c
+
+    if (needDir && c === 'FILE')
+      return false
+
+    // otherwise we have to stat, because maybe c=true
+    // if we know it exists, but not what it is.
+  }
+
+  var exists
+  var stat = this.statCache[abs]
+  if (!stat) {
+    var lstat
+    try {
+      lstat = fs.lstatSync(abs)
+    } catch (er) {
+      return false
+    }
+
+    if (lstat.isSymbolicLink()) {
+      try {
+        stat = fs.statSync(abs)
+      } catch (er) {
+        stat = lstat
+      }
+    } else {
+      stat = lstat
+    }
+  }
+
+  this.statCache[abs] = stat
+
+  var c = stat.isDirectory() ? 'DIR' : 'FILE'
+  this.cache[abs] = this.cache[abs] || c
+
+  if (needDir && c !== 'DIR')
+    return false
+
+  return c
+}
+
+GlobSync.prototype._mark = function (p) {
+  return common.mark(this, p)
+}
+
+GlobSync.prototype._makeAbs = function (f) {
+  return common.makeAbs(this, f)
+}
diff --git a/node_modules/glob-stream/node_modules/minimatch/LICENSE b/node_modules/glob-stream/node_modules/minimatch/LICENSE
new file mode 100644
index 000000000..19129e315
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/minimatch/LICENSE
@@ -0,0 +1,15 @@
+The ISC License
+
+Copyright (c) Isaac Z. Schlueter and Contributors
+
+Permission to use, copy, modify, and/or distribute this software for any
+purpose with or without fee is hereby granted, provided that the above
+copyright notice and this permission notice appear in all copies.
+
+THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR
+IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
diff --git a/node_modules/glob-stream/node_modules/minimatch/README.md b/node_modules/glob-stream/node_modules/minimatch/README.md
new file mode 100644
index 000000000..d458bc2e0
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/minimatch/README.md
@@ -0,0 +1,216 @@
+# minimatch
+
+A minimal matching utility.
+
+[![Build Status](https://secure.travis-ci.org/isaacs/minimatch.png)](http://travis-ci.org/isaacs/minimatch)
+
+
+This is the matching library used internally by npm.
+
+It works by converting glob expressions into JavaScript `RegExp`
+objects.
+
+## Usage
+
+```javascript
+var minimatch = require("minimatch")
+
+minimatch("bar.foo", "*.foo") // true!
+minimatch("bar.foo", "*.bar") // false!
+minimatch("bar.foo", "*.+(bar|foo)", { debug: true }) // true, and noisy!
+```
+
+## Features
+
+Supports these glob features:
+
+* Brace Expansion
+* Extended glob matching
+* "Globstar" `**` matching
+
+See:
+
+* `man sh`
+* `man bash`
+* `man 3 fnmatch`
+* `man 5 gitignore`
+
+## Minimatch Class
+
+Create a minimatch object by instanting the `minimatch.Minimatch` class.
+
+```javascript
+var Minimatch = require("minimatch").Minimatch
+var mm = new Minimatch(pattern, options)
+```
+
+### Properties
+
+* `pattern` The original pattern the minimatch object represents.
+* `options` The options supplied to the constructor.
+* `set` A 2-dimensional array of regexp or string expressions.
+  Each row in the
+  array corresponds to a brace-expanded pattern.  Each item in the row
+  corresponds to a single path-part.  For example, the pattern
+  `{a,b/c}/d` would expand to a set of patterns like:
+
+        [ [ a, d ]
+        , [ b, c, d ] ]
+
+    If a portion of the pattern doesn't have any "magic" in it
+    (that is, it's something like `"foo"` rather than `fo*o?`), then it
+    will be left as a string rather than converted to a regular
+    expression.
+
+* `regexp` Created by the `makeRe` method.  A single regular expression
+  expressing the entire pattern.  This is useful in cases where you wish
+  to use the pattern somewhat like `fnmatch(3)` with `FNM_PATH` enabled.
+* `negate` True if the pattern is negated.
+* `comment` True if the pattern is a comment.
+* `empty` True if the pattern is `""`.
+
+### Methods
+
+* `makeRe` Generate the `regexp` member if necessary, and return it.
+  Will return `false` if the pattern is invalid.
+* `match(fname)` Return true if the filename matches the pattern, or
+  false otherwise.
+* `matchOne(fileArray, patternArray, partial)` Take a `/`-split
+  filename, and match it against a single row in the `regExpSet`.  This
+  method is mainly for internal use, but is exposed so that it can be
+  used by a glob-walker that needs to avoid excessive filesystem calls.
+
+All other methods are internal, and will be called as necessary.
+
+## Functions
+
+The top-level exported function has a `cache` property, which is an LRU
+cache set to store 100 items.  So, calling these methods repeatedly
+with the same pattern and options will use the same Minimatch object,
+saving the cost of parsing it multiple times.
+
+### minimatch(path, pattern, options)
+
+Main export.  Tests a path against the pattern using the options.
+
+```javascript
+var isJS = minimatch(file, "*.js", { matchBase: true })
+```
+
+### minimatch.filter(pattern, options)
+
+Returns a function that tests its
+supplied argument, suitable for use with `Array.filter`.  Example:
+
+```javascript
+var javascripts = fileList.filter(minimatch.filter("*.js", {matchBase: true}))
+```
+
+### minimatch.match(list, pattern, options)
+
+Match against the list of
+files, in the style of fnmatch or glob.  If nothing is matched, and
+options.nonull is set, then return a list containing the pattern itself.
+
+```javascript
+var javascripts = minimatch.match(fileList, "*.js", {matchBase: true}))
+```
+
+### minimatch.makeRe(pattern, options)
+
+Make a regular expression object from the pattern.
+
+## Options
+
+All options are `false` by default.
+
+### debug
+
+Dump a ton of stuff to stderr.
+
+### nobrace
+
+Do not expand `{a,b}` and `{1..3}` brace sets.
+
+### noglobstar
+
+Disable `**` matching against multiple folder names.
+
+### dot
+
+Allow patterns to match filenames starting with a period, even if
+the pattern does not explicitly have a period in that spot.
+
+Note that by default, `a/**/b` will **not** match `a/.d/b`, unless `dot`
+is set.
+
+### noext
+
+Disable "extglob" style patterns like `+(a|b)`.
+
+### nocase
+
+Perform a case-insensitive match.
+
+### nonull
+
+When a match is not found by `minimatch.match`, return a list containing
+the pattern itself if this option is set.  When not set, an empty list
+is returned if there are no matches.
+
+### matchBase
+
+If set, then patterns without slashes will be matched
+against the basename of the path if it contains slashes.  For example,
+`a?b` would match the path `/xyz/123/acb`, but not `/xyz/acb/123`.
+
+### nocomment
+
+Suppress the behavior of treating `#` at the start of a pattern as a
+comment.
+
+### nonegate
+
+Suppress the behavior of treating a leading `!` character as negation.
+
+### flipNegate
+
+Returns from negate expressions the same as if they were not negated.
+(Ie, true on a hit, false on a miss.)
+
+
+## Comparisons to other fnmatch/glob implementations
+
+While strict compliance with the existing standards is a worthwhile
+goal, some discrepancies exist between minimatch and other
+implementations, and are intentional.
+
+If the pattern starts with a `!` character, then it is negated.  Set the
+`nonegate` flag to suppress this behavior, and treat leading `!`
+characters normally.  This is perhaps relevant if you wish to start the
+pattern with a negative extglob pattern like `!(a|B)`.  Multiple `!`
+characters at the start of a pattern will negate the pattern multiple
+times.
+
+If a pattern starts with `#`, then it is treated as a comment, and
+will not match anything.  Use `\#` to match a literal `#` at the
+start of a line, or set the `nocomment` flag to suppress this behavior.
+
+The double-star character `**` is supported by default, unless the
+`noglobstar` flag is set.  This is supported in the manner of bsdglob
+and bash 4.1, where `**` only has special significance if it is the only
+thing in a path part.  That is, `a/**/b` will match `a/x/y/b`, but
+`a/**b` will not.
+
+If an escaped pattern has no matches, and the `nonull` flag is set,
+then minimatch.match returns the pattern as-provided, rather than
+interpreting the character escapes.  For example,
+`minimatch.match([], "\\*a\\?")` will return `"\\*a\\?"` rather than
+`"*a?"`.  This is akin to setting the `nullglob` option in bash, except
+that it does not resolve escaped pattern characters.
+
+If brace expansion is not disabled, then it is performed before any
+other interpretation of the glob pattern.  Thus, a pattern like
+`+(a|{b),c)}`, which would not be valid in bash or zsh, is expanded
+**first** into the set of `+(a|b)` and `+(a|c)`, and those patterns are
+checked for validity.  Since those two are valid, matching proceeds.
diff --git a/node_modules/glob-stream/node_modules/minimatch/browser.js b/node_modules/glob-stream/node_modules/minimatch/browser.js
new file mode 100644
index 000000000..7d0515920
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/minimatch/browser.js
@@ -0,0 +1,1159 @@
+(function(f){if(typeof exports==="object"&&typeof module!=="undefined"){module.exports=f()}else if(typeof define==="function"&&define.amd){define([],f)}else{var g;if(typeof window!=="undefined"){g=window}else if(typeof global!=="undefined"){g=global}else if(typeof self!=="undefined"){g=self}else{g=this}g.minimatch = f()}})(function(){var define,module,exports;return (function e(t,n,r){function s(o,u){if(!n[o]){if(!t[o]){var a=typeof require=="function"&&require;if(!u&&a)return a(o,!0);if(i)return i(o,!0);var f=new Error("Cannot find module '"+o+"'");throw f.code="MODULE_NOT_FOUND",f}var l=n[o]={exports:{}};t[o][0].call(l.exports,function(e){var n=t[o][1][e];return s(n?n:e)},l,l.exports,e,t,n,r)}return n[o].exports}var i=typeof require=="function"&&require;for(var o=0;o<r.length;o++)s(r[o]);return s})({1:[function(require,module,exports){
+module.exports = minimatch
+minimatch.Minimatch = Minimatch
+
+var path = { sep: '/' }
+try {
+  path = require('path')
+} catch (er) {}
+
+var GLOBSTAR = minimatch.GLOBSTAR = Minimatch.GLOBSTAR = {}
+var expand = require('brace-expansion')
+
+// any single thing other than /
+// don't need to escape / when using new RegExp()
+var qmark = '[^/]'
+
+// * => any number of characters
+var star = qmark + '*?'
+
+// ** when dots are allowed.  Anything goes, except .. and .
+// not (^ or / followed by one or two dots followed by $ or /),
+// followed by anything, any number of times.
+var twoStarDot = '(?:(?!(?:\\\/|^)(?:\\.{1,2})($|\\\/)).)*?'
+
+// not a ^ or / followed by a dot,
+// followed by anything, any number of times.
+var twoStarNoDot = '(?:(?!(?:\\\/|^)\\.).)*?'
+
+// characters that need to be escaped in RegExp.
+var reSpecials = charSet('().*{}+?[]^$\\!')
+
+// "abc" -> { a:true, b:true, c:true }
+function charSet (s) {
+  return s.split('').reduce(function (set, c) {
+    set[c] = true
+    return set
+  }, {})
+}
+
+// normalizes slashes.
+var slashSplit = /\/+/
+
+minimatch.filter = filter
+function filter (pattern, options) {
+  options = options || {}
+  return function (p, i, list) {
+    return minimatch(p, pattern, options)
+  }
+}
+
+function ext (a, b) {
+  a = a || {}
+  b = b || {}
+  var t = {}
+  Object.keys(b).forEach(function (k) {
+    t[k] = b[k]
+  })
+  Object.keys(a).forEach(function (k) {
+    t[k] = a[k]
+  })
+  return t
+}
+
+minimatch.defaults = function (def) {
+  if (!def || !Object.keys(def).length) return minimatch
+
+  var orig = minimatch
+
+  var m = function minimatch (p, pattern, options) {
+    return orig.minimatch(p, pattern, ext(def, options))
+  }
+
+  m.Minimatch = function Minimatch (pattern, options) {
+    return new orig.Minimatch(pattern, ext(def, options))
+  }
+
+  return m
+}
+
+Minimatch.defaults = function (def) {
+  if (!def || !Object.keys(def).length) return Minimatch
+  return minimatch.defaults(def).Minimatch
+}
+
+function minimatch (p, pattern, options) {
+  if (typeof pattern !== 'string') {
+    throw new TypeError('glob pattern string required')
+  }
+
+  if (!options) options = {}
+
+  // shortcut: comments match nothing.
+  if (!options.nocomment && pattern.charAt(0) === '#') {
+    return false
+  }
+
+  // "" only matches ""
+  if (pattern.trim() === '') return p === ''
+
+  return new Minimatch(pattern, options).match(p)
+}
+
+function Minimatch (pattern, options) {
+  if (!(this instanceof Minimatch)) {
+    return new Minimatch(pattern, options)
+  }
+
+  if (typeof pattern !== 'string') {
+    throw new TypeError('glob pattern string required')
+  }
+
+  if (!options) options = {}
+  pattern = pattern.trim()
+
+  // windows support: need to use /, not \
+  if (path.sep !== '/') {
+    pattern = pattern.split(path.sep).join('/')
+  }
+
+  this.options = options
+  this.set = []
+  this.pattern = pattern
+  this.regexp = null
+  this.negate = false
+  this.comment = false
+  this.empty = false
+
+  // make the set of regexps etc.
+  this.make()
+}
+
+Minimatch.prototype.debug = function () {}
+
+Minimatch.prototype.make = make
+function make () {
+  // don't do it more than once.
+  if (this._made) return
+
+  var pattern = this.pattern
+  var options = this.options
+
+  // empty patterns and comments match nothing.
+  if (!options.nocomment && pattern.charAt(0) === '#') {
+    this.comment = true
+    return
+  }
+  if (!pattern) {
+    this.empty = true
+    return
+  }
+
+  // step 1: figure out negation, etc.
+  this.parseNegate()
+
+  // step 2: expand braces
+  var set = this.globSet = this.braceExpand()
+
+  if (options.debug) this.debug = console.error
+
+  this.debug(this.pattern, set)
+
+  // step 3: now we have a set, so turn each one into a series of path-portion
+  // matching patterns.
+  // These will be regexps, except in the case of "**", which is
+  // set to the GLOBSTAR object for globstar behavior,
+  // and will not contain any / characters
+  set = this.globParts = set.map(function (s) {
+    return s.split(slashSplit)
+  })
+
+  this.debug(this.pattern, set)
+
+  // glob --> regexps
+  set = set.map(function (s, si, set) {
+    return s.map(this.parse, this)
+  }, this)
+
+  this.debug(this.pattern, set)
+
+  // filter out everything that didn't compile properly.
+  set = set.filter(function (s) {
+    return s.indexOf(false) === -1
+  })
+
+  this.debug(this.pattern, set)
+
+  this.set = set
+}
+
+Minimatch.prototype.parseNegate = parseNegate
+function parseNegate () {
+  var pattern = this.pattern
+  var negate = false
+  var options = this.options
+  var negateOffset = 0
+
+  if (options.nonegate) return
+
+  for (var i = 0, l = pattern.length
+    ; i < l && pattern.charAt(i) === '!'
+    ; i++) {
+    negate = !negate
+    negateOffset++
+  }
+
+  if (negateOffset) this.pattern = pattern.substr(negateOffset)
+  this.negate = negate
+}
+
+// Brace expansion:
+// a{b,c}d -> abd acd
+// a{b,}c -> abc ac
+// a{0..3}d -> a0d a1d a2d a3d
+// a{b,c{d,e}f}g -> abg acdfg acefg
+// a{b,c}d{e,f}g -> abdeg acdeg abdeg abdfg
+//
+// Invalid sets are not expanded.
+// a{2..}b -> a{2..}b
+// a{b}c -> a{b}c
+minimatch.braceExpand = function (pattern, options) {
+  return braceExpand(pattern, options)
+}
+
+Minimatch.prototype.braceExpand = braceExpand
+
+function braceExpand (pattern, options) {
+  if (!options) {
+    if (this instanceof Minimatch) {
+      options = this.options
+    } else {
+      options = {}
+    }
+  }
+
+  pattern = typeof pattern === 'undefined'
+    ? this.pattern : pattern
+
+  if (typeof pattern === 'undefined') {
+    throw new Error('undefined pattern')
+  }
+
+  if (options.nobrace ||
+    !pattern.match(/\{.*\}/)) {
+    // shortcut. no need to expand.
+    return [pattern]
+  }
+
+  return expand(pattern)
+}
+
+// parse a component of the expanded set.
+// At this point, no pattern may contain "/" in it
+// so we're going to return a 2d array, where each entry is the full
+// pattern, split on '/', and then turned into a regular expression.
+// A regexp is made at the end which joins each array with an
+// escaped /, and another full one which joins each regexp with |.
+//
+// Following the lead of Bash 4.1, note that "**" only has special meaning
+// when it is the *only* thing in a path portion.  Otherwise, any series
+// of * is equivalent to a single *.  Globstar behavior is enabled by
+// default, and can be disabled by setting options.noglobstar.
+Minimatch.prototype.parse = parse
+var SUBPARSE = {}
+function parse (pattern, isSub) {
+  var options = this.options
+
+  // shortcuts
+  if (!options.noglobstar && pattern === '**') return GLOBSTAR
+  if (pattern === '') return ''
+
+  var re = ''
+  var hasMagic = !!options.nocase
+  var escaping = false
+  // ? => one single character
+  var patternListStack = []
+  var negativeLists = []
+  var plType
+  var stateChar
+  var inClass = false
+  var reClassStart = -1
+  var classStart = -1
+  // . and .. never match anything that doesn't start with .,
+  // even when options.dot is set.
+  var patternStart = pattern.charAt(0) === '.' ? '' // anything
+  // not (start or / followed by . or .. followed by / or end)
+  : options.dot ? '(?!(?:^|\\\/)\\.{1,2}(?:$|\\\/))'
+  : '(?!\\.)'
+  var self = this
+
+  function clearStateChar () {
+    if (stateChar) {
+      // we had some state-tracking character
+      // that wasn't consumed by this pass.
+      switch (stateChar) {
+        case '*':
+          re += star
+          hasMagic = true
+        break
+        case '?':
+          re += qmark
+          hasMagic = true
+        break
+        default:
+          re += '\\' + stateChar
+        break
+      }
+      self.debug('clearStateChar %j %j', stateChar, re)
+      stateChar = false
+    }
+  }
+
+  for (var i = 0, len = pattern.length, c
+    ; (i < len) && (c = pattern.charAt(i))
+    ; i++) {
+    this.debug('%s\t%s %s %j', pattern, i, re, c)
+
+    // skip over any that are escaped.
+    if (escaping && reSpecials[c]) {
+      re += '\\' + c
+      escaping = false
+      continue
+    }
+
+    switch (c) {
+      case '/':
+        // completely not allowed, even escaped.
+        // Should already be path-split by now.
+        return false
+
+      case '\\':
+        clearStateChar()
+        escaping = true
+      continue
+
+      // the various stateChar values
+      // for the "extglob" stuff.
+      case '?':
+      case '*':
+      case '+':
+      case '@':
+      case '!':
+        this.debug('%s\t%s %s %j <-- stateChar', pattern, i, re, c)
+
+        // all of those are literals inside a class, except that
+        // the glob [!a] means [^a] in regexp
+        if (inClass) {
+          this.debug('  in class')
+          if (c === '!' && i === classStart + 1) c = '^'
+          re += c
+          continue
+        }
+
+        // if we already have a stateChar, then it means
+        // that there was something like ** or +? in there.
+        // Handle the stateChar, then proceed with this one.
+        self.debug('call clearStateChar %j', stateChar)
+        clearStateChar()
+        stateChar = c
+        // if extglob is disabled, then +(asdf|foo) isn't a thing.
+        // just clear the statechar *now*, rather than even diving into
+        // the patternList stuff.
+        if (options.noext) clearStateChar()
+      continue
+
+      case '(':
+        if (inClass) {
+          re += '('
+          continue
+        }
+
+        if (!stateChar) {
+          re += '\\('
+          continue
+        }
+
+        plType = stateChar
+        patternListStack.push({
+          type: plType,
+          start: i - 1,
+          reStart: re.length
+        })
+        // negation is (?:(?!js)[^/]*)
+        re += stateChar === '!' ? '(?:(?!(?:' : '(?:'
+        this.debug('plType %j %j', stateChar, re)
+        stateChar = false
+      continue
+
+      case ')':
+        if (inClass || !patternListStack.length) {
+          re += '\\)'
+          continue
+        }
+
+        clearStateChar()
+        hasMagic = true
+        re += ')'
+        var pl = patternListStack.pop()
+        plType = pl.type
+        // negation is (?:(?!js)[^/]*)
+        // The others are (?:<pattern>)<type>
+        switch (plType) {
+          case '!':
+            negativeLists.push(pl)
+            re += ')[^/]*?)'
+            pl.reEnd = re.length
+            break
+          case '?':
+          case '+':
+          case '*':
+            re += plType
+            break
+          case '@': break // the default anyway
+        }
+      continue
+
+      case '|':
+        if (inClass || !patternListStack.length || escaping) {
+          re += '\\|'
+          escaping = false
+          continue
+        }
+
+        clearStateChar()
+        re += '|'
+      continue
+
+      // these are mostly the same in regexp and glob
+      case '[':
+        // swallow any state-tracking char before the [
+        clearStateChar()
+
+        if (inClass) {
+          re += '\\' + c
+          continue
+        }
+
+        inClass = true
+        classStart = i
+        reClassStart = re.length
+        re += c
+      continue
+
+      case ']':
+        //  a right bracket shall lose its special
+        //  meaning and represent itself in
+        //  a bracket expression if it occurs
+        //  first in the list.  -- POSIX.2 2.8.3.2
+        if (i === classStart + 1 || !inClass) {
+          re += '\\' + c
+          escaping = false
+          continue
+        }
+
+        // handle the case where we left a class open.
+        // "[z-a]" is valid, equivalent to "\[z-a\]"
+        if (inClass) {
+          // split where the last [ was, make sure we don't have
+          // an invalid re. if so, re-walk the contents of the
+          // would-be class to re-translate any characters that
+          // were passed through as-is
+          // TODO: It would probably be faster to determine this
+          // without a try/catch and a new RegExp, but it's tricky
+          // to do safely.  For now, this is safe and works.
+          var cs = pattern.substring(classStart + 1, i)
+          try {
+            RegExp('[' + cs + ']')
+          } catch (er) {
+            // not a valid class!
+            var sp = this.parse(cs, SUBPARSE)
+            re = re.substr(0, reClassStart) + '\\[' + sp[0] + '\\]'
+            hasMagic = hasMagic || sp[1]
+            inClass = false
+            continue
+          }
+        }
+
+        // finish up the class.
+        hasMagic = true
+        inClass = false
+        re += c
+      continue
+
+      default:
+        // swallow any state char that wasn't consumed
+        clearStateChar()
+
+        if (escaping) {
+          // no need
+          escaping = false
+        } else if (reSpecials[c]
+          && !(c === '^' && inClass)) {
+          re += '\\'
+        }
+
+        re += c
+
+    } // switch
+  } // for
+
+  // handle the case where we left a class open.
+  // "[abc" is valid, equivalent to "\[abc"
+  if (inClass) {
+    // split where the last [ was, and escape it
+    // this is a huge pita.  We now have to re-walk
+    // the contents of the would-be class to re-translate
+    // any characters that were passed through as-is
+    cs = pattern.substr(classStart + 1)
+    sp = this.parse(cs, SUBPARSE)
+    re = re.substr(0, reClassStart) + '\\[' + sp[0]
+    hasMagic = hasMagic || sp[1]
+  }
+
+  // handle the case where we had a +( thing at the *end*
+  // of the pattern.
+  // each pattern list stack adds 3 chars, and we need to go through
+  // and escape any | chars that were passed through as-is for the regexp.
+  // Go through and escape them, taking care not to double-escape any
+  // | chars that were already escaped.
+  for (pl = patternListStack.pop(); pl; pl = patternListStack.pop()) {
+    var tail = re.slice(pl.reStart + 3)
+    // maybe some even number of \, then maybe 1 \, followed by a |
+    tail = tail.replace(/((?:\\{2})*)(\\?)\|/g, function (_, $1, $2) {
+      if (!$2) {
+        // the | isn't already escaped, so escape it.
+        $2 = '\\'
+      }
+
+      // need to escape all those slashes *again*, without escaping the
+      // one that we need for escaping the | character.  As it works out,
+      // escaping an even number of slashes can be done by simply repeating
+      // it exactly after itself.  That's why this trick works.
+      //
+      // I am sorry that you have to see this.
+      return $1 + $1 + $2 + '|'
+    })
+
+    this.debug('tail=%j\n   %s', tail, tail)
+    var t = pl.type === '*' ? star
+      : pl.type === '?' ? qmark
+      : '\\' + pl.type
+
+    hasMagic = true
+    re = re.slice(0, pl.reStart) + t + '\\(' + tail
+  }
+
+  // handle trailing things that only matter at the very end.
+  clearStateChar()
+  if (escaping) {
+    // trailing \\
+    re += '\\\\'
+  }
+
+  // only need to apply the nodot start if the re starts with
+  // something that could conceivably capture a dot
+  var addPatternStart = false
+  switch (re.charAt(0)) {
+    case '.':
+    case '[':
+    case '(': addPatternStart = true
+  }
+
+  // Hack to work around lack of negative lookbehind in JS
+  // A pattern like: *.!(x).!(y|z) needs to ensure that a name
+  // like 'a.xyz.yz' doesn't match.  So, the first negative
+  // lookahead, has to look ALL the way ahead, to the end of
+  // the pattern.
+  for (var n = negativeLists.length - 1; n > -1; n--) {
+    var nl = negativeLists[n]
+
+    var nlBefore = re.slice(0, nl.reStart)
+    var nlFirst = re.slice(nl.reStart, nl.reEnd - 8)
+    var nlLast = re.slice(nl.reEnd - 8, nl.reEnd)
+    var nlAfter = re.slice(nl.reEnd)
+
+    nlLast += nlAfter
+
+    // Handle nested stuff like *(*.js|!(*.json)), where open parens
+    // mean that we should *not* include the ) in the bit that is considered
+    // "after" the negated section.
+    var openParensBefore = nlBefore.split('(').length - 1
+    var cleanAfter = nlAfter
+    for (i = 0; i < openParensBefore; i++) {
+      cleanAfter = cleanAfter.replace(/\)[+*?]?/, '')
+    }
+    nlAfter = cleanAfter
+
+    var dollar = ''
+    if (nlAfter === '' && isSub !== SUBPARSE) {
+      dollar = '$'
+    }
+    var newRe = nlBefore + nlFirst + nlAfter + dollar + nlLast
+    re = newRe
+  }
+
+  // if the re is not "" at this point, then we need to make sure
+  // it doesn't match against an empty path part.
+  // Otherwise a/* will match a/, which it should not.
+  if (re !== '' && hasMagic) {
+    re = '(?=.)' + re
+  }
+
+  if (addPatternStart) {
+    re = patternStart + re
+  }
+
+  // parsing just a piece of a larger pattern.
+  if (isSub === SUBPARSE) {
+    return [re, hasMagic]
+  }
+
+  // skip the regexp for non-magical patterns
+  // unescape anything in it, though, so that it'll be
+  // an exact match against a file etc.
+  if (!hasMagic) {
+    return globUnescape(pattern)
+  }
+
+  var flags = options.nocase ? 'i' : ''
+  var regExp = new RegExp('^' + re + '$', flags)
+
+  regExp._glob = pattern
+  regExp._src = re
+
+  return regExp
+}
+
+minimatch.makeRe = function (pattern, options) {
+  return new Minimatch(pattern, options || {}).makeRe()
+}
+
+Minimatch.prototype.makeRe = makeRe
+function makeRe () {
+  if (this.regexp || this.regexp === false) return this.regexp
+
+  // at this point, this.set is a 2d array of partial
+  // pattern strings, or "**".
+  //
+  // It's better to use .match().  This function shouldn't
+  // be used, really, but it's pretty convenient sometimes,
+  // when you just want to work with a regex.
+  var set = this.set
+
+  if (!set.length) {
+    this.regexp = false
+    return this.regexp
+  }
+  var options = this.options
+
+  var twoStar = options.noglobstar ? star
+    : options.dot ? twoStarDot
+    : twoStarNoDot
+  var flags = options.nocase ? 'i' : ''
+
+  var re = set.map(function (pattern) {
+    return pattern.map(function (p) {
+      return (p === GLOBSTAR) ? twoStar
+      : (typeof p === 'string') ? regExpEscape(p)
+      : p._src
+    }).join('\\\/')
+  }).join('|')
+
+  // must match entire pattern
+  // ending in a * or ** will make it less strict.
+  re = '^(?:' + re + ')$'
+
+  // can match anything, as long as it's not this.
+  if (this.negate) re = '^(?!' + re + ').*$'
+
+  try {
+    this.regexp = new RegExp(re, flags)
+  } catch (ex) {
+    this.regexp = false
+  }
+  return this.regexp
+}
+
+minimatch.match = function (list, pattern, options) {
+  options = options || {}
+  var mm = new Minimatch(pattern, options)
+  list = list.filter(function (f) {
+    return mm.match(f)
+  })
+  if (mm.options.nonull && !list.length) {
+    list.push(pattern)
+  }
+  return list
+}
+
+Minimatch.prototype.match = match
+function match (f, partial) {
+  this.debug('match', f, this.pattern)
+  // short-circuit in the case of busted things.
+  // comments, etc.
+  if (this.comment) return false
+  if (this.empty) return f === ''
+
+  if (f === '/' && partial) return true
+
+  var options = this.options
+
+  // windows: need to use /, not \
+  if (path.sep !== '/') {
+    f = f.split(path.sep).join('/')
+  }
+
+  // treat the test path as a set of pathparts.
+  f = f.split(slashSplit)
+  this.debug(this.pattern, 'split', f)
+
+  // just ONE of the pattern sets in this.set needs to match
+  // in order for it to be valid.  If negating, then just one
+  // match means that we have failed.
+  // Either way, return on the first hit.
+
+  var set = this.set
+  this.debug(this.pattern, 'set', set)
+
+  // Find the basename of the path by looking for the last non-empty segment
+  var filename
+  var i
+  for (i = f.length - 1; i >= 0; i--) {
+    filename = f[i]
+    if (filename) break
+  }
+
+  for (i = 0; i < set.length; i++) {
+    var pattern = set[i]
+    var file = f
+    if (options.matchBase && pattern.length === 1) {
+      file = [filename]
+    }
+    var hit = this.matchOne(file, pattern, partial)
+    if (hit) {
+      if (options.flipNegate) return true
+      return !this.negate
+    }
+  }
+
+  // didn't get any hits.  this is success if it's a negative
+  // pattern, failure otherwise.
+  if (options.flipNegate) return false
+  return this.negate
+}
+
+// set partial to true to test if, for example,
+// "/a/b" matches the start of "/*/b/*/d"
+// Partial means, if you run out of file before you run
+// out of pattern, then that's fine, as long as all
+// the parts match.
+Minimatch.prototype.matchOne = function (file, pattern, partial) {
+  var options = this.options
+
+  this.debug('matchOne',
+    { 'this': this, file: file, pattern: pattern })
+
+  this.debug('matchOne', file.length, pattern.length)
+
+  for (var fi = 0,
+      pi = 0,
+      fl = file.length,
+      pl = pattern.length
+      ; (fi < fl) && (pi < pl)
+      ; fi++, pi++) {
+    this.debug('matchOne loop')
+    var p = pattern[pi]
+    var f = file[fi]
+
+    this.debug(pattern, p, f)
+
+    // should be impossible.
+    // some invalid regexp stuff in the set.
+    if (p === false) return false
+
+    if (p === GLOBSTAR) {
+      this.debug('GLOBSTAR', [pattern, p, f])
+
+      // "**"
+      // a/**/b/**/c would match the following:
+      // a/b/x/y/z/c
+      // a/x/y/z/b/c
+      // a/b/x/b/x/c
+      // a/b/c
+      // To do this, take the rest of the pattern after
+      // the **, and see if it would match the file remainder.
+      // If so, return success.
+      // If not, the ** "swallows" a segment, and try again.
+      // This is recursively awful.
+      //
+      // a/**/b/**/c matching a/b/x/y/z/c
+      // - a matches a
+      // - doublestar
+      //   - matchOne(b/x/y/z/c, b/**/c)
+      //     - b matches b
+      //     - doublestar
+      //       - matchOne(x/y/z/c, c) -> no
+      //       - matchOne(y/z/c, c) -> no
+      //       - matchOne(z/c, c) -> no
+      //       - matchOne(c, c) yes, hit
+      var fr = fi
+      var pr = pi + 1
+      if (pr === pl) {
+        this.debug('** at the end')
+        // a ** at the end will just swallow the rest.
+        // We have found a match.
+        // however, it will not swallow /.x, unless
+        // options.dot is set.
+        // . and .. are *never* matched by **, for explosively
+        // exponential reasons.
+        for (; fi < fl; fi++) {
+          if (file[fi] === '.' || file[fi] === '..' ||
+            (!options.dot && file[fi].charAt(0) === '.')) return false
+        }
+        return true
+      }
+
+      // ok, let's see if we can swallow whatever we can.
+      while (fr < fl) {
+        var swallowee = file[fr]
+
+        this.debug('\nglobstar while', file, fr, pattern, pr, swallowee)
+
+        // XXX remove this slice.  Just pass the start index.
+        if (this.matchOne(file.slice(fr), pattern.slice(pr), partial)) {
+          this.debug('globstar found match!', fr, fl, swallowee)
+          // found a match.
+          return true
+        } else {
+          // can't swallow "." or ".." ever.
+          // can only swallow ".foo" when explicitly asked.
+          if (swallowee === '.' || swallowee === '..' ||
+            (!options.dot && swallowee.charAt(0) === '.')) {
+            this.debug('dot detected!', file, fr, pattern, pr)
+            break
+          }
+
+          // ** swallows a segment, and continue.
+          this.debug('globstar swallow a segment, and continue')
+          fr++
+        }
+      }
+
+      // no match was found.
+      // However, in partial mode, we can't say this is necessarily over.
+      // If there's more *pattern* left, then
+      if (partial) {
+        // ran out of file
+        this.debug('\n>>> no match, partial?', file, fr, pattern, pr)
+        if (fr === fl) return true
+      }
+      return false
+    }
+
+    // something other than **
+    // non-magic patterns just have to match exactly
+    // patterns with magic have been turned into regexps.
+    var hit
+    if (typeof p === 'string') {
+      if (options.nocase) {
+        hit = f.toLowerCase() === p.toLowerCase()
+      } else {
+        hit = f === p
+      }
+      this.debug('string match', p, f, hit)
+    } else {
+      hit = f.match(p)
+      this.debug('pattern match', p, f, hit)
+    }
+
+    if (!hit) return false
+  }
+
+  // Note: ending in / means that we'll get a final ""
+  // at the end of the pattern.  This can only match a
+  // corresponding "" at the end of the file.
+  // If the file ends in /, then it can only match a
+  // a pattern that ends in /, unless the pattern just
+  // doesn't have any more for it. But, a/b/ should *not*
+  // match "a/b/*", even though "" matches against the
+  // [^/]*? pattern, except in partial mode, where it might
+  // simply not be reached yet.
+  // However, a/b/ should still satisfy a/*
+
+  // now either we fell off the end of the pattern, or we're done.
+  if (fi === fl && pi === pl) {
+    // ran out of pattern and filename at the same time.
+    // an exact hit!
+    return true
+  } else if (fi === fl) {
+    // ran out of file, but still had pattern left.
+    // this is ok if we're doing the match as part of
+    // a glob fs traversal.
+    return partial
+  } else if (pi === pl) {
+    // ran out of pattern, still have file left.
+    // this is only acceptable if we're on the very last
+    // empty segment of a file with a trailing slash.
+    // a/* should match a/b/
+    var emptyFileEnd = (fi === fl - 1) && (file[fi] === '')
+    return emptyFileEnd
+  }
+
+  // should be unreachable.
+  throw new Error('wtf?')
+}
+
+// replace stuff like \* with *
+function globUnescape (s) {
+  return s.replace(/\\(.)/g, '$1')
+}
+
+function regExpEscape (s) {
+  return s.replace(/[-[\]{}()*+?.,\\^$|#\s]/g, '\\$&')
+}
+
+},{"brace-expansion":2,"path":undefined}],2:[function(require,module,exports){
+var concatMap = require('concat-map');
+var balanced = require('balanced-match');
+
+module.exports = expandTop;
+
+var escSlash = '\0SLASH'+Math.random()+'\0';
+var escOpen = '\0OPEN'+Math.random()+'\0';
+var escClose = '\0CLOSE'+Math.random()+'\0';
+var escComma = '\0COMMA'+Math.random()+'\0';
+var escPeriod = '\0PERIOD'+Math.random()+'\0';
+
+function numeric(str) {
+  return parseInt(str, 10) == str
+    ? parseInt(str, 10)
+    : str.charCodeAt(0);
+}
+
+function escapeBraces(str) {
+  return str.split('\\\\').join(escSlash)
+            .split('\\{').join(escOpen)
+            .split('\\}').join(escClose)
+            .split('\\,').join(escComma)
+            .split('\\.').join(escPeriod);
+}
+
+function unescapeBraces(str) {
+  return str.split(escSlash).join('\\')
+            .split(escOpen).join('{')
+            .split(escClose).join('}')
+            .split(escComma).join(',')
+            .split(escPeriod).join('.');
+}
+
+
+// Basically just str.split(","), but handling cases
+// where we have nested braced sections, which should be
+// treated as individual members, like {a,{b,c},d}
+function parseCommaParts(str) {
+  if (!str)
+    return [''];
+
+  var parts = [];
+  var m = balanced('{', '}', str);
+
+  if (!m)
+    return str.split(',');
+
+  var pre = m.pre;
+  var body = m.body;
+  var post = m.post;
+  var p = pre.split(',');
+
+  p[p.length-1] += '{' + body + '}';
+  var postParts = parseCommaParts(post);
+  if (post.length) {
+    p[p.length-1] += postParts.shift();
+    p.push.apply(p, postParts);
+  }
+
+  parts.push.apply(parts, p);
+
+  return parts;
+}
+
+function expandTop(str) {
+  if (!str)
+    return [];
+
+  var expansions = expand(escapeBraces(str));
+  return expansions.filter(identity).map(unescapeBraces);
+}
+
+function identity(e) {
+  return e;
+}
+
+function embrace(str) {
+  return '{' + str + '}';
+}
+function isPadded(el) {
+  return /^-?0\d/.test(el);
+}
+
+function lte(i, y) {
+  return i <= y;
+}
+function gte(i, y) {
+  return i >= y;
+}
+
+function expand(str) {
+  var expansions = [];
+
+  var m = balanced('{', '}', str);
+  if (!m || /\$$/.test(m.pre)) return [str];
+
+  var isNumericSequence = /^-?\d+\.\.-?\d+(?:\.\.-?\d+)?$/.test(m.body);
+  var isAlphaSequence = /^[a-zA-Z]\.\.[a-zA-Z](?:\.\.-?\d+)?$/.test(m.body);
+  var isSequence = isNumericSequence || isAlphaSequence;
+  var isOptions = /^(.*,)+(.+)?$/.test(m.body);
+  if (!isSequence && !isOptions) {
+    // {a},b}
+    if (m.post.match(/,.*}/)) {
+      str = m.pre + '{' + m.body + escClose + m.post;
+      return expand(str);
+    }
+    return [str];
+  }
+
+  var n;
+  if (isSequence) {
+    n = m.body.split(/\.\./);
+  } else {
+    n = parseCommaParts(m.body);
+    if (n.length === 1) {
+      // x{{a,b}}y ==> x{a}y x{b}y
+      n = expand(n[0]).map(embrace);
+      if (n.length === 1) {
+        var post = m.post.length
+          ? expand(m.post)
+          : [''];
+        return post.map(function(p) {
+          return m.pre + n[0] + p;
+        });
+      }
+    }
+  }
+
+  // at this point, n is the parts, and we know it's not a comma set
+  // with a single entry.
+
+  // no need to expand pre, since it is guaranteed to be free of brace-sets
+  var pre = m.pre;
+  var post = m.post.length
+    ? expand(m.post)
+    : [''];
+
+  var N;
+
+  if (isSequence) {
+    var x = numeric(n[0]);
+    var y = numeric(n[1]);
+    var width = Math.max(n[0].length, n[1].length)
+    var incr = n.length == 3
+      ? Math.abs(numeric(n[2]))
+      : 1;
+    var test = lte;
+    var reverse = y < x;
+    if (reverse) {
+      incr *= -1;
+      test = gte;
+    }
+    var pad = n.some(isPadded);
+
+    N = [];
+
+    for (var i = x; test(i, y); i += incr) {
+      var c;
+      if (isAlphaSequence) {
+        c = String.fromCharCode(i);
+        if (c === '\\')
+          c = '';
+      } else {
+        c = String(i);
+        if (pad) {
+          var need = width - c.length;
+          if (need > 0) {
+            var z = new Array(need + 1).join('0');
+            if (i < 0)
+              c = '-' + z + c.slice(1);
+            else
+              c = z + c;
+          }
+        }
+      }
+      N.push(c);
+    }
+  } else {
+    N = concatMap(n, function(el) { return expand(el) });
+  }
+
+  for (var j = 0; j < N.length; j++) {
+    for (var k = 0; k < post.length; k++) {
+      expansions.push([pre, N[j], post[k]].join(''))
+    }
+  }
+
+  return expansions;
+}
+
+
+},{"balanced-match":3,"concat-map":4}],3:[function(require,module,exports){
+module.exports = balanced;
+function balanced(a, b, str) {
+  var bal = 0;
+  var m = {};
+  var ended = false;
+
+  for (var i = 0; i < str.length; i++) {
+    if (a == str.substr(i, a.length)) {
+      if (!('start' in m)) m.start = i;
+      bal++;
+    }
+    else if (b == str.substr(i, b.length) && 'start' in m) {
+      ended = true;
+      bal--;
+      if (!bal) {
+        m.end = i;
+        m.pre = str.substr(0, m.start);
+        m.body = (m.end - m.start > 1)
+          ? str.substring(m.start + a.length, m.end)
+          : '';
+        m.post = str.slice(m.end + b.length);
+        return m;
+      }
+    }
+  }
+
+  // if we opened more than we closed, find the one we closed
+  if (bal && ended) {
+    var start = m.start + a.length;
+    m = balanced(a, b, str.substr(start));
+    if (m) {
+      m.start += start;
+      m.end += start;
+      m.pre = str.slice(0, start) + m.pre;
+    }
+    return m;
+  }
+}
+
+},{}],4:[function(require,module,exports){
+module.exports = function (xs, fn) {
+    var res = [];
+    for (var i = 0; i < xs.length; i++) {
+        var x = fn(xs[i], i);
+        if (Array.isArray(x)) res.push.apply(res, x);
+        else res.push(x);
+    }
+    return res;
+};
+
+},{}]},{},[1])(1)
+});
\ No newline at end of file
diff --git a/node_modules/glob-stream/node_modules/minimatch/minimatch.js b/node_modules/glob-stream/node_modules/minimatch/minimatch.js
new file mode 100644
index 000000000..ec4c05c57
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/minimatch/minimatch.js
@@ -0,0 +1,912 @@
+module.exports = minimatch
+minimatch.Minimatch = Minimatch
+
+var path = { sep: '/' }
+try {
+  path = require('path')
+} catch (er) {}
+
+var GLOBSTAR = minimatch.GLOBSTAR = Minimatch.GLOBSTAR = {}
+var expand = require('brace-expansion')
+
+// any single thing other than /
+// don't need to escape / when using new RegExp()
+var qmark = '[^/]'
+
+// * => any number of characters
+var star = qmark + '*?'
+
+// ** when dots are allowed.  Anything goes, except .. and .
+// not (^ or / followed by one or two dots followed by $ or /),
+// followed by anything, any number of times.
+var twoStarDot = '(?:(?!(?:\\\/|^)(?:\\.{1,2})($|\\\/)).)*?'
+
+// not a ^ or / followed by a dot,
+// followed by anything, any number of times.
+var twoStarNoDot = '(?:(?!(?:\\\/|^)\\.).)*?'
+
+// characters that need to be escaped in RegExp.
+var reSpecials = charSet('().*{}+?[]^$\\!')
+
+// "abc" -> { a:true, b:true, c:true }
+function charSet (s) {
+  return s.split('').reduce(function (set, c) {
+    set[c] = true
+    return set
+  }, {})
+}
+
+// normalizes slashes.
+var slashSplit = /\/+/
+
+minimatch.filter = filter
+function filter (pattern, options) {
+  options = options || {}
+  return function (p, i, list) {
+    return minimatch(p, pattern, options)
+  }
+}
+
+function ext (a, b) {
+  a = a || {}
+  b = b || {}
+  var t = {}
+  Object.keys(b).forEach(function (k) {
+    t[k] = b[k]
+  })
+  Object.keys(a).forEach(function (k) {
+    t[k] = a[k]
+  })
+  return t
+}
+
+minimatch.defaults = function (def) {
+  if (!def || !Object.keys(def).length) return minimatch
+
+  var orig = minimatch
+
+  var m = function minimatch (p, pattern, options) {
+    return orig.minimatch(p, pattern, ext(def, options))
+  }
+
+  m.Minimatch = function Minimatch (pattern, options) {
+    return new orig.Minimatch(pattern, ext(def, options))
+  }
+
+  return m
+}
+
+Minimatch.defaults = function (def) {
+  if (!def || !Object.keys(def).length) return Minimatch
+  return minimatch.defaults(def).Minimatch
+}
+
+function minimatch (p, pattern, options) {
+  if (typeof pattern !== 'string') {
+    throw new TypeError('glob pattern string required')
+  }
+
+  if (!options) options = {}
+
+  // shortcut: comments match nothing.
+  if (!options.nocomment && pattern.charAt(0) === '#') {
+    return false
+  }
+
+  // "" only matches ""
+  if (pattern.trim() === '') return p === ''
+
+  return new Minimatch(pattern, options).match(p)
+}
+
+function Minimatch (pattern, options) {
+  if (!(this instanceof Minimatch)) {
+    return new Minimatch(pattern, options)
+  }
+
+  if (typeof pattern !== 'string') {
+    throw new TypeError('glob pattern string required')
+  }
+
+  if (!options) options = {}
+  pattern = pattern.trim()
+
+  // windows support: need to use /, not \
+  if (path.sep !== '/') {
+    pattern = pattern.split(path.sep).join('/')
+  }
+
+  this.options = options
+  this.set = []
+  this.pattern = pattern
+  this.regexp = null
+  this.negate = false
+  this.comment = false
+  this.empty = false
+
+  // make the set of regexps etc.
+  this.make()
+}
+
+Minimatch.prototype.debug = function () {}
+
+Minimatch.prototype.make = make
+function make () {
+  // don't do it more than once.
+  if (this._made) return
+
+  var pattern = this.pattern
+  var options = this.options
+
+  // empty patterns and comments match nothing.
+  if (!options.nocomment && pattern.charAt(0) === '#') {
+    this.comment = true
+    return
+  }
+  if (!pattern) {
+    this.empty = true
+    return
+  }
+
+  // step 1: figure out negation, etc.
+  this.parseNegate()
+
+  // step 2: expand braces
+  var set = this.globSet = this.braceExpand()
+
+  if (options.debug) this.debug = console.error
+
+  this.debug(this.pattern, set)
+
+  // step 3: now we have a set, so turn each one into a series of path-portion
+  // matching patterns.
+  // These will be regexps, except in the case of "**", which is
+  // set to the GLOBSTAR object for globstar behavior,
+  // and will not contain any / characters
+  set = this.globParts = set.map(function (s) {
+    return s.split(slashSplit)
+  })
+
+  this.debug(this.pattern, set)
+
+  // glob --> regexps
+  set = set.map(function (s, si, set) {
+    return s.map(this.parse, this)
+  }, this)
+
+  this.debug(this.pattern, set)
+
+  // filter out everything that didn't compile properly.
+  set = set.filter(function (s) {
+    return s.indexOf(false) === -1
+  })
+
+  this.debug(this.pattern, set)
+
+  this.set = set
+}
+
+Minimatch.prototype.parseNegate = parseNegate
+function parseNegate () {
+  var pattern = this.pattern
+  var negate = false
+  var options = this.options
+  var negateOffset = 0
+
+  if (options.nonegate) return
+
+  for (var i = 0, l = pattern.length
+    ; i < l && pattern.charAt(i) === '!'
+    ; i++) {
+    negate = !negate
+    negateOffset++
+  }
+
+  if (negateOffset) this.pattern = pattern.substr(negateOffset)
+  this.negate = negate
+}
+
+// Brace expansion:
+// a{b,c}d -> abd acd
+// a{b,}c -> abc ac
+// a{0..3}d -> a0d a1d a2d a3d
+// a{b,c{d,e}f}g -> abg acdfg acefg
+// a{b,c}d{e,f}g -> abdeg acdeg abdeg abdfg
+//
+// Invalid sets are not expanded.
+// a{2..}b -> a{2..}b
+// a{b}c -> a{b}c
+minimatch.braceExpand = function (pattern, options) {
+  return braceExpand(pattern, options)
+}
+
+Minimatch.prototype.braceExpand = braceExpand
+
+function braceExpand (pattern, options) {
+  if (!options) {
+    if (this instanceof Minimatch) {
+      options = this.options
+    } else {
+      options = {}
+    }
+  }
+
+  pattern = typeof pattern === 'undefined'
+    ? this.pattern : pattern
+
+  if (typeof pattern === 'undefined') {
+    throw new Error('undefined pattern')
+  }
+
+  if (options.nobrace ||
+    !pattern.match(/\{.*\}/)) {
+    // shortcut. no need to expand.
+    return [pattern]
+  }
+
+  return expand(pattern)
+}
+
+// parse a component of the expanded set.
+// At this point, no pattern may contain "/" in it
+// so we're going to return a 2d array, where each entry is the full
+// pattern, split on '/', and then turned into a regular expression.
+// A regexp is made at the end which joins each array with an
+// escaped /, and another full one which joins each regexp with |.
+//
+// Following the lead of Bash 4.1, note that "**" only has special meaning
+// when it is the *only* thing in a path portion.  Otherwise, any series
+// of * is equivalent to a single *.  Globstar behavior is enabled by
+// default, and can be disabled by setting options.noglobstar.
+Minimatch.prototype.parse = parse
+var SUBPARSE = {}
+function parse (pattern, isSub) {
+  var options = this.options
+
+  // shortcuts
+  if (!options.noglobstar && pattern === '**') return GLOBSTAR
+  if (pattern === '') return ''
+
+  var re = ''
+  var hasMagic = !!options.nocase
+  var escaping = false
+  // ? => one single character
+  var patternListStack = []
+  var negativeLists = []
+  var plType
+  var stateChar
+  var inClass = false
+  var reClassStart = -1
+  var classStart = -1
+  // . and .. never match anything that doesn't start with .,
+  // even when options.dot is set.
+  var patternStart = pattern.charAt(0) === '.' ? '' // anything
+  // not (start or / followed by . or .. followed by / or end)
+  : options.dot ? '(?!(?:^|\\\/)\\.{1,2}(?:$|\\\/))'
+  : '(?!\\.)'
+  var self = this
+
+  function clearStateChar () {
+    if (stateChar) {
+      // we had some state-tracking character
+      // that wasn't consumed by this pass.
+      switch (stateChar) {
+        case '*':
+          re += star
+          hasMagic = true
+        break
+        case '?':
+          re += qmark
+          hasMagic = true
+        break
+        default:
+          re += '\\' + stateChar
+        break
+      }
+      self.debug('clearStateChar %j %j', stateChar, re)
+      stateChar = false
+    }
+  }
+
+  for (var i = 0, len = pattern.length, c
+    ; (i < len) && (c = pattern.charAt(i))
+    ; i++) {
+    this.debug('%s\t%s %s %j', pattern, i, re, c)
+
+    // skip over any that are escaped.
+    if (escaping && reSpecials[c]) {
+      re += '\\' + c
+      escaping = false
+      continue
+    }
+
+    switch (c) {
+      case '/':
+        // completely not allowed, even escaped.
+        // Should already be path-split by now.
+        return false
+
+      case '\\':
+        clearStateChar()
+        escaping = true
+      continue
+
+      // the various stateChar values
+      // for the "extglob" stuff.
+      case '?':
+      case '*':
+      case '+':
+      case '@':
+      case '!':
+        this.debug('%s\t%s %s %j <-- stateChar', pattern, i, re, c)
+
+        // all of those are literals inside a class, except that
+        // the glob [!a] means [^a] in regexp
+        if (inClass) {
+          this.debug('  in class')
+          if (c === '!' && i === classStart + 1) c = '^'
+          re += c
+          continue
+        }
+
+        // if we already have a stateChar, then it means
+        // that there was something like ** or +? in there.
+        // Handle the stateChar, then proceed with this one.
+        self.debug('call clearStateChar %j', stateChar)
+        clearStateChar()
+        stateChar = c
+        // if extglob is disabled, then +(asdf|foo) isn't a thing.
+        // just clear the statechar *now*, rather than even diving into
+        // the patternList stuff.
+        if (options.noext) clearStateChar()
+      continue
+
+      case '(':
+        if (inClass) {
+          re += '('
+          continue
+        }
+
+        if (!stateChar) {
+          re += '\\('
+          continue
+        }
+
+        plType = stateChar
+        patternListStack.push({
+          type: plType,
+          start: i - 1,
+          reStart: re.length
+        })
+        // negation is (?:(?!js)[^/]*)
+        re += stateChar === '!' ? '(?:(?!(?:' : '(?:'
+        this.debug('plType %j %j', stateChar, re)
+        stateChar = false
+      continue
+
+      case ')':
+        if (inClass || !patternListStack.length) {
+          re += '\\)'
+          continue
+        }
+
+        clearStateChar()
+        hasMagic = true
+        re += ')'
+        var pl = patternListStack.pop()
+        plType = pl.type
+        // negation is (?:(?!js)[^/]*)
+        // The others are (?:<pattern>)<type>
+        switch (plType) {
+          case '!':
+            negativeLists.push(pl)
+            re += ')[^/]*?)'
+            pl.reEnd = re.length
+            break
+          case '?':
+          case '+':
+          case '*':
+            re += plType
+            break
+          case '@': break // the default anyway
+        }
+      continue
+
+      case '|':
+        if (inClass || !patternListStack.length || escaping) {
+          re += '\\|'
+          escaping = false
+          continue
+        }
+
+        clearStateChar()
+        re += '|'
+      continue
+
+      // these are mostly the same in regexp and glob
+      case '[':
+        // swallow any state-tracking char before the [
+        clearStateChar()
+
+        if (inClass) {
+          re += '\\' + c
+          continue
+        }
+
+        inClass = true
+        classStart = i
+        reClassStart = re.length
+        re += c
+      continue
+
+      case ']':
+        //  a right bracket shall lose its special
+        //  meaning and represent itself in
+        //  a bracket expression if it occurs
+        //  first in the list.  -- POSIX.2 2.8.3.2
+        if (i === classStart + 1 || !inClass) {
+          re += '\\' + c
+          escaping = false
+          continue
+        }
+
+        // handle the case where we left a class open.
+        // "[z-a]" is valid, equivalent to "\[z-a\]"
+        if (inClass) {
+          // split where the last [ was, make sure we don't have
+          // an invalid re. if so, re-walk the contents of the
+          // would-be class to re-translate any characters that
+          // were passed through as-is
+          // TODO: It would probably be faster to determine this
+          // without a try/catch and a new RegExp, but it's tricky
+          // to do safely.  For now, this is safe and works.
+          var cs = pattern.substring(classStart + 1, i)
+          try {
+            RegExp('[' + cs + ']')
+          } catch (er) {
+            // not a valid class!
+            var sp = this.parse(cs, SUBPARSE)
+            re = re.substr(0, reClassStart) + '\\[' + sp[0] + '\\]'
+            hasMagic = hasMagic || sp[1]
+            inClass = false
+            continue
+          }
+        }
+
+        // finish up the class.
+        hasMagic = true
+        inClass = false
+        re += c
+      continue
+
+      default:
+        // swallow any state char that wasn't consumed
+        clearStateChar()
+
+        if (escaping) {
+          // no need
+          escaping = false
+        } else if (reSpecials[c]
+          && !(c === '^' && inClass)) {
+          re += '\\'
+        }
+
+        re += c
+
+    } // switch
+  } // for
+
+  // handle the case where we left a class open.
+  // "[abc" is valid, equivalent to "\[abc"
+  if (inClass) {
+    // split where the last [ was, and escape it
+    // this is a huge pita.  We now have to re-walk
+    // the contents of the would-be class to re-translate
+    // any characters that were passed through as-is
+    cs = pattern.substr(classStart + 1)
+    sp = this.parse(cs, SUBPARSE)
+    re = re.substr(0, reClassStart) + '\\[' + sp[0]
+    hasMagic = hasMagic || sp[1]
+  }
+
+  // handle the case where we had a +( thing at the *end*
+  // of the pattern.
+  // each pattern list stack adds 3 chars, and we need to go through
+  // and escape any | chars that were passed through as-is for the regexp.
+  // Go through and escape them, taking care not to double-escape any
+  // | chars that were already escaped.
+  for (pl = patternListStack.pop(); pl; pl = patternListStack.pop()) {
+    var tail = re.slice(pl.reStart + 3)
+    // maybe some even number of \, then maybe 1 \, followed by a |
+    tail = tail.replace(/((?:\\{2})*)(\\?)\|/g, function (_, $1, $2) {
+      if (!$2) {
+        // the | isn't already escaped, so escape it.
+        $2 = '\\'
+      }
+
+      // need to escape all those slashes *again*, without escaping the
+      // one that we need for escaping the | character.  As it works out,
+      // escaping an even number of slashes can be done by simply repeating
+      // it exactly after itself.  That's why this trick works.
+      //
+      // I am sorry that you have to see this.
+      return $1 + $1 + $2 + '|'
+    })
+
+    this.debug('tail=%j\n   %s', tail, tail)
+    var t = pl.type === '*' ? star
+      : pl.type === '?' ? qmark
+      : '\\' + pl.type
+
+    hasMagic = true
+    re = re.slice(0, pl.reStart) + t + '\\(' + tail
+  }
+
+  // handle trailing things that only matter at the very end.
+  clearStateChar()
+  if (escaping) {
+    // trailing \\
+    re += '\\\\'
+  }
+
+  // only need to apply the nodot start if the re starts with
+  // something that could conceivably capture a dot
+  var addPatternStart = false
+  switch (re.charAt(0)) {
+    case '.':
+    case '[':
+    case '(': addPatternStart = true
+  }
+
+  // Hack to work around lack of negative lookbehind in JS
+  // A pattern like: *.!(x).!(y|z) needs to ensure that a name
+  // like 'a.xyz.yz' doesn't match.  So, the first negative
+  // lookahead, has to look ALL the way ahead, to the end of
+  // the pattern.
+  for (var n = negativeLists.length - 1; n > -1; n--) {
+    var nl = negativeLists[n]
+
+    var nlBefore = re.slice(0, nl.reStart)
+    var nlFirst = re.slice(nl.reStart, nl.reEnd - 8)
+    var nlLast = re.slice(nl.reEnd - 8, nl.reEnd)
+    var nlAfter = re.slice(nl.reEnd)
+
+    nlLast += nlAfter
+
+    // Handle nested stuff like *(*.js|!(*.json)), where open parens
+    // mean that we should *not* include the ) in the bit that is considered
+    // "after" the negated section.
+    var openParensBefore = nlBefore.split('(').length - 1
+    var cleanAfter = nlAfter
+    for (i = 0; i < openParensBefore; i++) {
+      cleanAfter = cleanAfter.replace(/\)[+*?]?/, '')
+    }
+    nlAfter = cleanAfter
+
+    var dollar = ''
+    if (nlAfter === '' && isSub !== SUBPARSE) {
+      dollar = '$'
+    }
+    var newRe = nlBefore + nlFirst + nlAfter + dollar + nlLast
+    re = newRe
+  }
+
+  // if the re is not "" at this point, then we need to make sure
+  // it doesn't match against an empty path part.
+  // Otherwise a/* will match a/, which it should not.
+  if (re !== '' && hasMagic) {
+    re = '(?=.)' + re
+  }
+
+  if (addPatternStart) {
+    re = patternStart + re
+  }
+
+  // parsing just a piece of a larger pattern.
+  if (isSub === SUBPARSE) {
+    return [re, hasMagic]
+  }
+
+  // skip the regexp for non-magical patterns
+  // unescape anything in it, though, so that it'll be
+  // an exact match against a file etc.
+  if (!hasMagic) {
+    return globUnescape(pattern)
+  }
+
+  var flags = options.nocase ? 'i' : ''
+  var regExp = new RegExp('^' + re + '$', flags)
+
+  regExp._glob = pattern
+  regExp._src = re
+
+  return regExp
+}
+
+minimatch.makeRe = function (pattern, options) {
+  return new Minimatch(pattern, options || {}).makeRe()
+}
+
+Minimatch.prototype.makeRe = makeRe
+function makeRe () {
+  if (this.regexp || this.regexp === false) return this.regexp
+
+  // at this point, this.set is a 2d array of partial
+  // pattern strings, or "**".
+  //
+  // It's better to use .match().  This function shouldn't
+  // be used, really, but it's pretty convenient sometimes,
+  // when you just want to work with a regex.
+  var set = this.set
+
+  if (!set.length) {
+    this.regexp = false
+    return this.regexp
+  }
+  var options = this.options
+
+  var twoStar = options.noglobstar ? star
+    : options.dot ? twoStarDot
+    : twoStarNoDot
+  var flags = options.nocase ? 'i' : ''
+
+  var re = set.map(function (pattern) {
+    return pattern.map(function (p) {
+      return (p === GLOBSTAR) ? twoStar
+      : (typeof p === 'string') ? regExpEscape(p)
+      : p._src
+    }).join('\\\/')
+  }).join('|')
+
+  // must match entire pattern
+  // ending in a * or ** will make it less strict.
+  re = '^(?:' + re + ')$'
+
+  // can match anything, as long as it's not this.
+  if (this.negate) re = '^(?!' + re + ').*$'
+
+  try {
+    this.regexp = new RegExp(re, flags)
+  } catch (ex) {
+    this.regexp = false
+  }
+  return this.regexp
+}
+
+minimatch.match = function (list, pattern, options) {
+  options = options || {}
+  var mm = new Minimatch(pattern, options)
+  list = list.filter(function (f) {
+    return mm.match(f)
+  })
+  if (mm.options.nonull && !list.length) {
+    list.push(pattern)
+  }
+  return list
+}
+
+Minimatch.prototype.match = match
+function match (f, partial) {
+  this.debug('match', f, this.pattern)
+  // short-circuit in the case of busted things.
+  // comments, etc.
+  if (this.comment) return false
+  if (this.empty) return f === ''
+
+  if (f === '/' && partial) return true
+
+  var options = this.options
+
+  // windows: need to use /, not \
+  if (path.sep !== '/') {
+    f = f.split(path.sep).join('/')
+  }
+
+  // treat the test path as a set of pathparts.
+  f = f.split(slashSplit)
+  this.debug(this.pattern, 'split', f)
+
+  // just ONE of the pattern sets in this.set needs to match
+  // in order for it to be valid.  If negating, then just one
+  // match means that we have failed.
+  // Either way, return on the first hit.
+
+  var set = this.set
+  this.debug(this.pattern, 'set', set)
+
+  // Find the basename of the path by looking for the last non-empty segment
+  var filename
+  var i
+  for (i = f.length - 1; i >= 0; i--) {
+    filename = f[i]
+    if (filename) break
+  }
+
+  for (i = 0; i < set.length; i++) {
+    var pattern = set[i]
+    var file = f
+    if (options.matchBase && pattern.length === 1) {
+      file = [filename]
+    }
+    var hit = this.matchOne(file, pattern, partial)
+    if (hit) {
+      if (options.flipNegate) return true
+      return !this.negate
+    }
+  }
+
+  // didn't get any hits.  this is success if it's a negative
+  // pattern, failure otherwise.
+  if (options.flipNegate) return false
+  return this.negate
+}
+
+// set partial to true to test if, for example,
+// "/a/b" matches the start of "/*/b/*/d"
+// Partial means, if you run out of file before you run
+// out of pattern, then that's fine, as long as all
+// the parts match.
+Minimatch.prototype.matchOne = function (file, pattern, partial) {
+  var options = this.options
+
+  this.debug('matchOne',
+    { 'this': this, file: file, pattern: pattern })
+
+  this.debug('matchOne', file.length, pattern.length)
+
+  for (var fi = 0,
+      pi = 0,
+      fl = file.length,
+      pl = pattern.length
+      ; (fi < fl) && (pi < pl)
+      ; fi++, pi++) {
+    this.debug('matchOne loop')
+    var p = pattern[pi]
+    var f = file[fi]
+
+    this.debug(pattern, p, f)
+
+    // should be impossible.
+    // some invalid regexp stuff in the set.
+    if (p === false) return false
+
+    if (p === GLOBSTAR) {
+      this.debug('GLOBSTAR', [pattern, p, f])
+
+      // "**"
+      // a/**/b/**/c would match the following:
+      // a/b/x/y/z/c
+      // a/x/y/z/b/c
+      // a/b/x/b/x/c
+      // a/b/c
+      // To do this, take the rest of the pattern after
+      // the **, and see if it would match the file remainder.
+      // If so, return success.
+      // If not, the ** "swallows" a segment, and try again.
+      // This is recursively awful.
+      //
+      // a/**/b/**/c matching a/b/x/y/z/c
+      // - a matches a
+      // - doublestar
+      //   - matchOne(b/x/y/z/c, b/**/c)
+      //     - b matches b
+      //     - doublestar
+      //       - matchOne(x/y/z/c, c) -> no
+      //       - matchOne(y/z/c, c) -> no
+      //       - matchOne(z/c, c) -> no
+      //       - matchOne(c, c) yes, hit
+      var fr = fi
+      var pr = pi + 1
+      if (pr === pl) {
+        this.debug('** at the end')
+        // a ** at the end will just swallow the rest.
+        // We have found a match.
+        // however, it will not swallow /.x, unless
+        // options.dot is set.
+        // . and .. are *never* matched by **, for explosively
+        // exponential reasons.
+        for (; fi < fl; fi++) {
+          if (file[fi] === '.' || file[fi] === '..' ||
+            (!options.dot && file[fi].charAt(0) === '.')) return false
+        }
+        return true
+      }
+
+      // ok, let's see if we can swallow whatever we can.
+      while (fr < fl) {
+        var swallowee = file[fr]
+
+        this.debug('\nglobstar while', file, fr, pattern, pr, swallowee)
+
+        // XXX remove this slice.  Just pass the start index.
+        if (this.matchOne(file.slice(fr), pattern.slice(pr), partial)) {
+          this.debug('globstar found match!', fr, fl, swallowee)
+          // found a match.
+          return true
+        } else {
+          // can't swallow "." or ".." ever.
+          // can only swallow ".foo" when explicitly asked.
+          if (swallowee === '.' || swallowee === '..' ||
+            (!options.dot && swallowee.charAt(0) === '.')) {
+            this.debug('dot detected!', file, fr, pattern, pr)
+            break
+          }
+
+          // ** swallows a segment, and continue.
+          this.debug('globstar swallow a segment, and continue')
+          fr++
+        }
+      }
+
+      // no match was found.
+      // However, in partial mode, we can't say this is necessarily over.
+      // If there's more *pattern* left, then
+      if (partial) {
+        // ran out of file
+        this.debug('\n>>> no match, partial?', file, fr, pattern, pr)
+        if (fr === fl) return true
+      }
+      return false
+    }
+
+    // something other than **
+    // non-magic patterns just have to match exactly
+    // patterns with magic have been turned into regexps.
+    var hit
+    if (typeof p === 'string') {
+      if (options.nocase) {
+        hit = f.toLowerCase() === p.toLowerCase()
+      } else {
+        hit = f === p
+      }
+      this.debug('string match', p, f, hit)
+    } else {
+      hit = f.match(p)
+      this.debug('pattern match', p, f, hit)
+    }
+
+    if (!hit) return false
+  }
+
+  // Note: ending in / means that we'll get a final ""
+  // at the end of the pattern.  This can only match a
+  // corresponding "" at the end of the file.
+  // If the file ends in /, then it can only match a
+  // a pattern that ends in /, unless the pattern just
+  // doesn't have any more for it. But, a/b/ should *not*
+  // match "a/b/*", even though "" matches against the
+  // [^/]*? pattern, except in partial mode, where it might
+  // simply not be reached yet.
+  // However, a/b/ should still satisfy a/*
+
+  // now either we fell off the end of the pattern, or we're done.
+  if (fi === fl && pi === pl) {
+    // ran out of pattern and filename at the same time.
+    // an exact hit!
+    return true
+  } else if (fi === fl) {
+    // ran out of file, but still had pattern left.
+    // this is ok if we're doing the match as part of
+    // a glob fs traversal.
+    return partial
+  } else if (pi === pl) {
+    // ran out of pattern, still have file left.
+    // this is only acceptable if we're on the very last
+    // empty segment of a file with a trailing slash.
+    // a/* should match a/b/
+    var emptyFileEnd = (fi === fl - 1) && (file[fi] === '')
+    return emptyFileEnd
+  }
+
+  // should be unreachable.
+  throw new Error('wtf?')
+}
+
+// replace stuff like \* with *
+function globUnescape (s) {
+  return s.replace(/\\(.)/g, '$1')
+}
+
+function regExpEscape (s) {
+  return s.replace(/[-[\]{}()*+?.,\\^$|#\s]/g, '\\$&')
+}
diff --git a/node_modules/glob-stream/node_modules/minimatch/package.json b/node_modules/glob-stream/node_modules/minimatch/package.json
new file mode 100644
index 000000000..9e01fca0a
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/minimatch/package.json
@@ -0,0 +1,98 @@
+{
+  "_args": [
+    [
+      {
+        "raw": "minimatch@^2.0.1",
+        "scope": null,
+        "escapedName": "minimatch",
+        "name": "minimatch",
+        "rawSpec": "^2.0.1",
+        "spec": ">=2.0.1 <3.0.0",
+        "type": "range"
+      },
+      "/home/dold/repos/taler/wallet-webex/node_modules/glob-stream"
+    ]
+  ],
+  "_from": "minimatch@>=2.0.1 <3.0.0",
+  "_id": "minimatch@2.0.10",
+  "_inCache": true,
+  "_location": "/glob-stream/minimatch",
+  "_nodeVersion": "2.2.1",
+  "_npmUser": {
+    "name": "isaacs",
+    "email": "isaacs@npmjs.com"
+  },
+  "_npmVersion": "3.1.0",
+  "_phantomChildren": {},
+  "_requested": {
+    "raw": "minimatch@^2.0.1",
+    "scope": null,
+    "escapedName": "minimatch",
+    "name": "minimatch",
+    "rawSpec": "^2.0.1",
+    "spec": ">=2.0.1 <3.0.0",
+    "type": "range"
+  },
+  "_requiredBy": [
+    "/glob-stream",
+    "/glob-stream/glob"
+  ],
+  "_resolved": "https://registry.npmjs.org/minimatch/-/minimatch-2.0.10.tgz",
+  "_shasum": "8d087c39c6b38c001b97fca7ce6d0e1e80afbac7",
+  "_shrinkwrap": null,
+  "_spec": "minimatch@^2.0.1",
+  "_where": "/home/dold/repos/taler/wallet-webex/node_modules/glob-stream",
+  "author": {
+    "name": "Isaac Z. Schlueter",
+    "email": "i@izs.me",
+    "url": "http://blog.izs.me"
+  },
+  "bugs": {
+    "url": "https://github.com/isaacs/minimatch/issues"
+  },
+  "dependencies": {
+    "brace-expansion": "^1.0.0"
+  },
+  "deprecated": "Please update to minimatch 3.0.2 or higher to avoid a RegExp DoS issue",
+  "description": "a glob matcher in javascript",
+  "devDependencies": {
+    "browserify": "^9.0.3",
+    "standard": "^3.7.2",
+    "tap": "^1.2.0"
+  },
+  "directories": {},
+  "dist": {
+    "shasum": "8d087c39c6b38c001b97fca7ce6d0e1e80afbac7",
+    "tarball": "https://registry.npmjs.org/minimatch/-/minimatch-2.0.10.tgz"
+  },
+  "engines": {
+    "node": "*"
+  },
+  "files": [
+    "minimatch.js",
+    "browser.js"
+  ],
+  "gitHead": "6afb85f0c324b321f76a38df81891e562693e257",
+  "homepage": "https://github.com/isaacs/minimatch#readme",
+  "license": "ISC",
+  "main": "minimatch.js",
+  "maintainers": [
+    {
+      "name": "isaacs",
+      "email": "i@izs.me"
+    }
+  ],
+  "name": "minimatch",
+  "optionalDependencies": {},
+  "readme": "ERROR: No README data found!",
+  "repository": {
+    "type": "git",
+    "url": "git://github.com/isaacs/minimatch.git"
+  },
+  "scripts": {
+    "posttest": "standard minimatch.js test/*.js",
+    "prepublish": "browserify -o browser.js -e minimatch.js -s minimatch --bare",
+    "test": "tap test/*.js"
+  },
+  "version": "2.0.10"
+}
diff --git a/node_modules/glob-stream/node_modules/readable-stream/.npmignore b/node_modules/glob-stream/node_modules/readable-stream/.npmignore
new file mode 100644
index 000000000..38344f87a
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/readable-stream/.npmignore
@@ -0,0 +1,5 @@
+build/
+test/
+examples/
+fs.js
+zlib.js
\ No newline at end of file
diff --git a/node_modules/glob-stream/node_modules/readable-stream/LICENSE b/node_modules/glob-stream/node_modules/readable-stream/LICENSE
new file mode 100644
index 000000000..e3d4e695a
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/readable-stream/LICENSE
@@ -0,0 +1,18 @@
+Copyright Joyent, Inc. and other Node contributors. All rights reserved.
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to
+deal in the Software without restriction, including without limitation the
+rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+sell copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in
+all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+IN THE SOFTWARE.
diff --git a/node_modules/glob-stream/node_modules/readable-stream/README.md b/node_modules/glob-stream/node_modules/readable-stream/README.md
new file mode 100644
index 000000000..3fb3e8023
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/readable-stream/README.md
@@ -0,0 +1,15 @@
+# readable-stream
+
+***Node-core streams for userland***
+
+[![NPM](https://nodei.co/npm/readable-stream.png?downloads=true&downloadRank=true)](https://nodei.co/npm/readable-stream/)
+[![NPM](https://nodei.co/npm-dl/readable-stream.png?&months=6&height=3)](https://nodei.co/npm/readable-stream/)
+
+This package is a mirror of the Streams2 and Streams3 implementations in Node-core.
+
+If you want to guarantee a stable streams base, regardless of what version of Node you, or the users of your libraries are using, use **readable-stream** *only* and avoid the *"stream"* module in Node-core.
+
+**readable-stream** comes in two major versions, v1.0.x and v1.1.x. The former tracks the Streams2 implementation in Node 0.10, including bug-fixes and minor improvements as they are added. The latter tracks Streams3 as it develops in Node 0.11; we will likely see a v1.2.x branch for Node 0.12.
+
+**readable-stream** uses proper patch-level versioning so if you pin to `"~1.0.0"` you’ll get the latest Node 0.10 Streams2 implementation, including any fixes and minor non-breaking improvements. The patch-level versions of 1.0.x and 1.1.x should mirror the patch-level versions of Node-core releases. You should prefer the **1.0.x** releases for now and when you’re ready to start using Streams3, pin to `"~1.1.0"`
+
diff --git a/node_modules/glob-stream/node_modules/readable-stream/duplex.js b/node_modules/glob-stream/node_modules/readable-stream/duplex.js
new file mode 100644
index 000000000..ca807af87
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/readable-stream/duplex.js
@@ -0,0 +1 @@
+module.exports = require("./lib/_stream_duplex.js")
diff --git a/node_modules/glob-stream/node_modules/readable-stream/lib/_stream_duplex.js b/node_modules/glob-stream/node_modules/readable-stream/lib/_stream_duplex.js
new file mode 100644
index 000000000..b513d61a9
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/readable-stream/lib/_stream_duplex.js
@@ -0,0 +1,89 @@
+// Copyright Joyent, Inc. and other Node contributors.
+//
+// Permission is hereby granted, free of charge, to any person obtaining a
+// copy of this software and associated documentation files (the
+// "Software"), to deal in the Software without restriction, including
+// without limitation the rights to use, copy, modify, merge, publish,
+// distribute, sublicense, and/or sell copies of the Software, and to permit
+// persons to whom the Software is furnished to do so, subject to the
+// following conditions:
+//
+// The above copyright notice and this permission notice shall be included
+// in all copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+// OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+// MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN
+// NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
+// DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+// OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
+// USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+// a duplex stream is just a stream that is both readable and writable.
+// Since JS doesn't have multiple prototypal inheritance, this class
+// prototypally inherits from Readable, and then parasitically from
+// Writable.
+
+module.exports = Duplex;
+
+/*<replacement>*/
+var objectKeys = Object.keys || function (obj) {
+  var keys = [];
+  for (var key in obj) keys.push(key);
+  return keys;
+}
+/*</replacement>*/
+
+
+/*<replacement>*/
+var util = require('core-util-is');
+util.inherits = require('inherits');
+/*</replacement>*/
+
+var Readable = require('./_stream_readable');
+var Writable = require('./_stream_writable');
+
+util.inherits(Duplex, Readable);
+
+forEach(objectKeys(Writable.prototype), function(method) {
+  if (!Duplex.prototype[method])
+    Duplex.prototype[method] = Writable.prototype[method];
+});
+
+function Duplex(options) {
+  if (!(this instanceof Duplex))
+    return new Duplex(options);
+
+  Readable.call(this, options);
+  Writable.call(this, options);
+
+  if (options && options.readable === false)
+    this.readable = false;
+
+  if (options && options.writable === false)
+    this.writable = false;
+
+  this.allowHalfOpen = true;
+  if (options && options.allowHalfOpen === false)
+    this.allowHalfOpen = false;
+
+  this.once('end', onend);
+}
+
+// the no-half-open enforcer
+function onend() {
+  // if we allow half-open state, or if the writable side ended,
+  // then we're ok.
+  if (this.allowHalfOpen || this._writableState.ended)
+    return;
+
+  // no more data can be written.
+  // But allow more writes to happen in this tick.
+  process.nextTick(this.end.bind(this));
+}
+
+function forEach (xs, f) {
+  for (var i = 0, l = xs.length; i < l; i++) {
+    f(xs[i], i);
+  }
+}
diff --git a/node_modules/glob-stream/node_modules/readable-stream/lib/_stream_passthrough.js b/node_modules/glob-stream/node_modules/readable-stream/lib/_stream_passthrough.js
new file mode 100644
index 000000000..895ca50a1
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/readable-stream/lib/_stream_passthrough.js
@@ -0,0 +1,46 @@
+// Copyright Joyent, Inc. and other Node contributors.
+//
+// Permission is hereby granted, free of charge, to any person obtaining a
+// copy of this software and associated documentation files (the
+// "Software"), to deal in the Software without restriction, including
+// without limitation the rights to use, copy, modify, merge, publish,
+// distribute, sublicense, and/or sell copies of the Software, and to permit
+// persons to whom the Software is furnished to do so, subject to the
+// following conditions:
+//
+// The above copyright notice and this permission notice shall be included
+// in all copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+// OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+// MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN
+// NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
+// DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+// OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
+// USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+// a passthrough stream.
+// basically just the most minimal sort of Transform stream.
+// Every written chunk gets output as-is.
+
+module.exports = PassThrough;
+
+var Transform = require('./_stream_transform');
+
+/*<replacement>*/
+var util = require('core-util-is');
+util.inherits = require('inherits');
+/*</replacement>*/
+
+util.inherits(PassThrough, Transform);
+
+function PassThrough(options) {
+  if (!(this instanceof PassThrough))
+    return new PassThrough(options);
+
+  Transform.call(this, options);
+}
+
+PassThrough.prototype._transform = function(chunk, encoding, cb) {
+  cb(null, chunk);
+};
diff --git a/node_modules/glob-stream/node_modules/readable-stream/lib/_stream_readable.js b/node_modules/glob-stream/node_modules/readable-stream/lib/_stream_readable.js
new file mode 100644
index 000000000..630722099
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/readable-stream/lib/_stream_readable.js
@@ -0,0 +1,982 @@
+// Copyright Joyent, Inc. and other Node contributors.
+//
+// Permission is hereby granted, free of charge, to any person obtaining a
+// copy of this software and associated documentation files (the
+// "Software"), to deal in the Software without restriction, including
+// without limitation the rights to use, copy, modify, merge, publish,
+// distribute, sublicense, and/or sell copies of the Software, and to permit
+// persons to whom the Software is furnished to do so, subject to the
+// following conditions:
+//
+// The above copyright notice and this permission notice shall be included
+// in all copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+// OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+// MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN
+// NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
+// DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+// OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
+// USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+module.exports = Readable;
+
+/*<replacement>*/
+var isArray = require('isarray');
+/*</replacement>*/
+
+
+/*<replacement>*/
+var Buffer = require('buffer').Buffer;
+/*</replacement>*/
+
+Readable.ReadableState = ReadableState;
+
+var EE = require('events').EventEmitter;
+
+/*<replacement>*/
+if (!EE.listenerCount) EE.listenerCount = function(emitter, type) {
+  return emitter.listeners(type).length;
+};
+/*</replacement>*/
+
+var Stream = require('stream');
+
+/*<replacement>*/
+var util = require('core-util-is');
+util.inherits = require('inherits');
+/*</replacement>*/
+
+var StringDecoder;
+
+util.inherits(Readable, Stream);
+
+function ReadableState(options, stream) {
+  options = options || {};
+
+  // the point at which it stops calling _read() to fill the buffer
+  // Note: 0 is a valid value, means "don't call _read preemptively ever"
+  var hwm = options.highWaterMark;
+  this.highWaterMark = (hwm || hwm === 0) ? hwm : 16 * 1024;
+
+  // cast to ints.
+  this.highWaterMark = ~~this.highWaterMark;
+
+  this.buffer = [];
+  this.length = 0;
+  this.pipes = null;
+  this.pipesCount = 0;
+  this.flowing = false;
+  this.ended = false;
+  this.endEmitted = false;
+  this.reading = false;
+
+  // In streams that never have any data, and do push(null) right away,
+  // the consumer can miss the 'end' event if they do some I/O before
+  // consuming the stream.  So, we don't emit('end') until some reading
+  // happens.
+  this.calledRead = false;
+
+  // a flag to be able to tell if the onwrite cb is called immediately,
+  // or on a later tick.  We set this to true at first, becuase any
+  // actions that shouldn't happen until "later" should generally also
+  // not happen before the first write call.
+  this.sync = true;
+
+  // whenever we return null, then we set a flag to say
+  // that we're awaiting a 'readable' event emission.
+  this.needReadable = false;
+  this.emittedReadable = false;
+  this.readableListening = false;
+
+
+  // object stream flag. Used to make read(n) ignore n and to
+  // make all the buffer merging and length checks go away
+  this.objectMode = !!options.objectMode;
+
+  // Crypto is kind of old and crusty.  Historically, its default string
+  // encoding is 'binary' so we have to make this configurable.
+  // Everything else in the universe uses 'utf8', though.
+  this.defaultEncoding = options.defaultEncoding || 'utf8';
+
+  // when piping, we only care about 'readable' events that happen
+  // after read()ing all the bytes and not getting any pushback.
+  this.ranOut = false;
+
+  // the number of writers that are awaiting a drain event in .pipe()s
+  this.awaitDrain = 0;
+
+  // if true, a maybeReadMore has been scheduled
+  this.readingMore = false;
+
+  this.decoder = null;
+  this.encoding = null;
+  if (options.encoding) {
+    if (!StringDecoder)
+      StringDecoder = require('string_decoder/').StringDecoder;
+    this.decoder = new StringDecoder(options.encoding);
+    this.encoding = options.encoding;
+  }
+}
+
+function Readable(options) {
+  if (!(this instanceof Readable))
+    return new Readable(options);
+
+  this._readableState = new ReadableState(options, this);
+
+  // legacy
+  this.readable = true;
+
+  Stream.call(this);
+}
+
+// Manually shove something into the read() buffer.
+// This returns true if the highWaterMark has not been hit yet,
+// similar to how Writable.write() returns true if you should
+// write() some more.
+Readable.prototype.push = function(chunk, encoding) {
+  var state = this._readableState;
+
+  if (typeof chunk === 'string' && !state.objectMode) {
+    encoding = encoding || state.defaultEncoding;
+    if (encoding !== state.encoding) {
+      chunk = new Buffer(chunk, encoding);
+      encoding = '';
+    }
+  }
+
+  return readableAddChunk(this, state, chunk, encoding, false);
+};
+
+// Unshift should *always* be something directly out of read()
+Readable.prototype.unshift = function(chunk) {
+  var state = this._readableState;
+  return readableAddChunk(this, state, chunk, '', true);
+};
+
+function readableAddChunk(stream, state, chunk, encoding, addToFront) {
+  var er = chunkInvalid(state, chunk);
+  if (er) {
+    stream.emit('error', er);
+  } else if (chunk === null || chunk === undefined) {
+    state.reading = false;
+    if (!state.ended)
+      onEofChunk(stream, state);
+  } else if (state.objectMode || chunk && chunk.length > 0) {
+    if (state.ended && !addToFront) {
+      var e = new Error('stream.push() after EOF');
+      stream.emit('error', e);
+    } else if (state.endEmitted && addToFront) {
+      var e = new Error('stream.unshift() after end event');
+      stream.emit('error', e);
+    } else {
+      if (state.decoder && !addToFront && !encoding)
+        chunk = state.decoder.write(chunk);
+
+      // update the buffer info.
+      state.length += state.objectMode ? 1 : chunk.length;
+      if (addToFront) {
+        state.buffer.unshift(chunk);
+      } else {
+        state.reading = false;
+        state.buffer.push(chunk);
+      }
+
+      if (state.needReadable)
+        emitReadable(stream);
+
+      maybeReadMore(stream, state);
+    }
+  } else if (!addToFront) {
+    state.reading = false;
+  }
+
+  return needMoreData(state);
+}
+
+
+
+// if it's past the high water mark, we can push in some more.
+// Also, if we have no data yet, we can stand some
+// more bytes.  This is to work around cases where hwm=0,
+// such as the repl.  Also, if the push() triggered a
+// readable event, and the user called read(largeNumber) such that
+// needReadable was set, then we ought to push more, so that another
+// 'readable' event will be triggered.
+function needMoreData(state) {
+  return !state.ended &&
+         (state.needReadable ||
+          state.length < state.highWaterMark ||
+          state.length === 0);
+}
+
+// backwards compatibility.
+Readable.prototype.setEncoding = function(enc) {
+  if (!StringDecoder)
+    StringDecoder = require('string_decoder/').StringDecoder;
+  this._readableState.decoder = new StringDecoder(enc);
+  this._readableState.encoding = enc;
+};
+
+// Don't raise the hwm > 128MB
+var MAX_HWM = 0x800000;
+function roundUpToNextPowerOf2(n) {
+  if (n >= MAX_HWM) {
+    n = MAX_HWM;
+  } else {
+    // Get the next highest power of 2
+    n--;
+    for (var p = 1; p < 32; p <<= 1) n |= n >> p;
+    n++;
+  }
+  return n;
+}
+
+function howMuchToRead(n, state) {
+  if (state.length === 0 && state.ended)
+    return 0;
+
+  if (state.objectMode)
+    return n === 0 ? 0 : 1;
+
+  if (n === null || isNaN(n)) {
+    // only flow one buffer at a time
+    if (state.flowing && state.buffer.length)
+      return state.buffer[0].length;
+    else
+      return state.length;
+  }
+
+  if (n <= 0)
+    return 0;
+
+  // If we're asking for more than the target buffer level,
+  // then raise the water mark.  Bump up to the next highest
+  // power of 2, to prevent increasing it excessively in tiny
+  // amounts.
+  if (n > state.highWaterMark)
+    state.highWaterMark = roundUpToNextPowerOf2(n);
+
+  // don't have that much.  return null, unless we've ended.
+  if (n > state.length) {
+    if (!state.ended) {
+      state.needReadable = true;
+      return 0;
+    } else
+      return state.length;
+  }
+
+  return n;
+}
+
+// you can override either this method, or the async _read(n) below.
+Readable.prototype.read = function(n) {
+  var state = this._readableState;
+  state.calledRead = true;
+  var nOrig = n;
+  var ret;
+
+  if (typeof n !== 'number' || n > 0)
+    state.emittedReadable = false;
+
+  // if we're doing read(0) to trigger a readable event, but we
+  // already have a bunch of data in the buffer, then just trigger
+  // the 'readable' event and move on.
+  if (n === 0 &&
+      state.needReadable &&
+      (state.length >= state.highWaterMark || state.ended)) {
+    emitReadable(this);
+    return null;
+  }
+
+  n = howMuchToRead(n, state);
+
+  // if we've ended, and we're now clear, then finish it up.
+  if (n === 0 && state.ended) {
+    ret = null;
+
+    // In cases where the decoder did not receive enough data
+    // to produce a full chunk, then immediately received an
+    // EOF, state.buffer will contain [<Buffer >, <Buffer 00 ...>].
+    // howMuchToRead will see this and coerce the amount to
+    // read to zero (because it's looking at the length of the
+    // first <Buffer > in state.buffer), and we'll end up here.
+    //
+    // This can only happen via state.decoder -- no other venue
+    // exists for pushing a zero-length chunk into state.buffer
+    // and triggering this behavior. In this case, we return our
+    // remaining data and end the stream, if appropriate.
+    if (state.length > 0 && state.decoder) {
+      ret = fromList(n, state);
+      state.length -= ret.length;
+    }
+
+    if (state.length === 0)
+      endReadable(this);
+
+    return ret;
+  }
+
+  // All the actual chunk generation logic needs to be
+  // *below* the call to _read.  The reason is that in certain
+  // synthetic stream cases, such as passthrough streams, _read
+  // may be a completely synchronous operation which may change
+  // the state of the read buffer, providing enough data when
+  // before there was *not* enough.
+  //
+  // So, the steps are:
+  // 1. Figure out what the state of things will be after we do
+  // a read from the buffer.
+  //
+  // 2. If that resulting state will trigger a _read, then call _read.
+  // Note that this may be asynchronous, or synchronous.  Yes, it is
+  // deeply ugly to write APIs this way, but that still doesn't mean
+  // that the Readable class should behave improperly, as streams are
+  // designed to be sync/async agnostic.
+  // Take note if the _read call is sync or async (ie, if the read call
+  // has returned yet), so that we know whether or not it's safe to emit
+  // 'readable' etc.
+  //
+  // 3. Actually pull the requested chunks out of the buffer and return.
+
+  // if we need a readable event, then we need to do some reading.
+  var doRead = state.needReadable;
+
+  // if we currently have less than the highWaterMark, then also read some
+  if (state.length - n <= state.highWaterMark)
+    doRead = true;
+
+  // however, if we've ended, then there's no point, and if we're already
+  // reading, then it's unnecessary.
+  if (state.ended || state.reading)
+    doRead = false;
+
+  if (doRead) {
+    state.reading = true;
+    state.sync = true;
+    // if the length is currently zero, then we *need* a readable event.
+    if (state.length === 0)
+      state.needReadable = true;
+    // call internal read method
+    this._read(state.highWaterMark);
+    state.sync = false;
+  }
+
+  // If _read called its callback synchronously, then `reading`
+  // will be false, and we need to re-evaluate how much data we
+  // can return to the user.
+  if (doRead && !state.reading)
+    n = howMuchToRead(nOrig, state);
+
+  if (n > 0)
+    ret = fromList(n, state);
+  else
+    ret = null;
+
+  if (ret === null) {
+    state.needReadable = true;
+    n = 0;
+  }
+
+  state.length -= n;
+
+  // If we have nothing in the buffer, then we want to know
+  // as soon as we *do* get something into the buffer.
+  if (state.length === 0 && !state.ended)
+    state.needReadable = true;
+
+  // If we happened to read() exactly the remaining amount in the
+  // buffer, and the EOF has been seen at this point, then make sure
+  // that we emit 'end' on the very next tick.
+  if (state.ended && !state.endEmitted && state.length === 0)
+    endReadable(this);
+
+  return ret;
+};
+
+function chunkInvalid(state, chunk) {
+  var er = null;
+  if (!Buffer.isBuffer(chunk) &&
+      'string' !== typeof chunk &&
+      chunk !== null &&
+      chunk !== undefined &&
+      !state.objectMode) {
+    er = new TypeError('Invalid non-string/buffer chunk');
+  }
+  return er;
+}
+
+
+function onEofChunk(stream, state) {
+  if (state.decoder && !state.ended) {
+    var chunk = state.decoder.end();
+    if (chunk && chunk.length) {
+      state.buffer.push(chunk);
+      state.length += state.objectMode ? 1 : chunk.length;
+    }
+  }
+  state.ended = true;
+
+  // if we've ended and we have some data left, then emit
+  // 'readable' now to make sure it gets picked up.
+  if (state.length > 0)
+    emitReadable(stream);
+  else
+    endReadable(stream);
+}
+
+// Don't emit readable right away in sync mode, because this can trigger
+// another read() call => stack overflow.  This way, it might trigger
+// a nextTick recursion warning, but that's not so bad.
+function emitReadable(stream) {
+  var state = stream._readableState;
+  state.needReadable = false;
+  if (state.emittedReadable)
+    return;
+
+  state.emittedReadable = true;
+  if (state.sync)
+    process.nextTick(function() {
+      emitReadable_(stream);
+    });
+  else
+    emitReadable_(stream);
+}
+
+function emitReadable_(stream) {
+  stream.emit('readable');
+}
+
+
+// at this point, the user has presumably seen the 'readable' event,
+// and called read() to consume some data.  that may have triggered
+// in turn another _read(n) call, in which case reading = true if
+// it's in progress.
+// However, if we're not ended, or reading, and the length < hwm,
+// then go ahead and try to read some more preemptively.
+function maybeReadMore(stream, state) {
+  if (!state.readingMore) {
+    state.readingMore = true;
+    process.nextTick(function() {
+      maybeReadMore_(stream, state);
+    });
+  }
+}
+
+function maybeReadMore_(stream, state) {
+  var len = state.length;
+  while (!state.reading && !state.flowing && !state.ended &&
+         state.length < state.highWaterMark) {
+    stream.read(0);
+    if (len === state.length)
+      // didn't get any data, stop spinning.
+      break;
+    else
+      len = state.length;
+  }
+  state.readingMore = false;
+}
+
+// abstract method.  to be overridden in specific implementation classes.
+// call cb(er, data) where data is <= n in length.
+// for virtual (non-string, non-buffer) streams, "length" is somewhat
+// arbitrary, and perhaps not very meaningful.
+Readable.prototype._read = function(n) {
+  this.emit('error', new Error('not implemented'));
+};
+
+Readable.prototype.pipe = function(dest, pipeOpts) {
+  var src = this;
+  var state = this._readableState;
+
+  switch (state.pipesCount) {
+    case 0:
+      state.pipes = dest;
+      break;
+    case 1:
+      state.pipes = [state.pipes, dest];
+      break;
+    default:
+      state.pipes.push(dest);
+      break;
+  }
+  state.pipesCount += 1;
+
+  var doEnd = (!pipeOpts || pipeOpts.end !== false) &&
+              dest !== process.stdout &&
+              dest !== process.stderr;
+
+  var endFn = doEnd ? onend : cleanup;
+  if (state.endEmitted)
+    process.nextTick(endFn);
+  else
+    src.once('end', endFn);
+
+  dest.on('unpipe', onunpipe);
+  function onunpipe(readable) {
+    if (readable !== src) return;
+    cleanup();
+  }
+
+  function onend() {
+    dest.end();
+  }
+
+  // when the dest drains, it reduces the awaitDrain counter
+  // on the source.  This would be more elegant with a .once()
+  // handler in flow(), but adding and removing repeatedly is
+  // too slow.
+  var ondrain = pipeOnDrain(src);
+  dest.on('drain', ondrain);
+
+  function cleanup() {
+    // cleanup event handlers once the pipe is broken
+    dest.removeListener('close', onclose);
+    dest.removeListener('finish', onfinish);
+    dest.removeListener('drain', ondrain);
+    dest.removeListener('error', onerror);
+    dest.removeListener('unpipe', onunpipe);
+    src.removeListener('end', onend);
+    src.removeListener('end', cleanup);
+
+    // if the reader is waiting for a drain event from this
+    // specific writer, then it would cause it to never start
+    // flowing again.
+    // So, if this is awaiting a drain, then we just call it now.
+    // If we don't know, then assume that we are waiting for one.
+    if (!dest._writableState || dest._writableState.needDrain)
+      ondrain();
+  }
+
+  // if the dest has an error, then stop piping into it.
+  // however, don't suppress the throwing behavior for this.
+  function onerror(er) {
+    unpipe();
+    dest.removeListener('error', onerror);
+    if (EE.listenerCount(dest, 'error') === 0)
+      dest.emit('error', er);
+  }
+  // This is a brutally ugly hack to make sure that our error handler
+  // is attached before any userland ones.  NEVER DO THIS.
+  if (!dest._events || !dest._events.error)
+    dest.on('error', onerror);
+  else if (isArray(dest._events.error))
+    dest._events.error.unshift(onerror);
+  else
+    dest._events.error = [onerror, dest._events.error];
+
+
+
+  // Both close and finish should trigger unpipe, but only once.
+  function onclose() {
+    dest.removeListener('finish', onfinish);
+    unpipe();
+  }
+  dest.once('close', onclose);
+  function onfinish() {
+    dest.removeListener('close', onclose);
+    unpipe();
+  }
+  dest.once('finish', onfinish);
+
+  function unpipe() {
+    src.unpipe(dest);
+  }
+
+  // tell the dest that it's being piped to
+  dest.emit('pipe', src);
+
+  // start the flow if it hasn't been started already.
+  if (!state.flowing) {
+    // the handler that waits for readable events after all
+    // the data gets sucked out in flow.
+    // This would be easier to follow with a .once() handler
+    // in flow(), but that is too slow.
+    this.on('readable', pipeOnReadable);
+
+    state.flowing = true;
+    process.nextTick(function() {
+      flow(src);
+    });
+  }
+
+  return dest;
+};
+
+function pipeOnDrain(src) {
+  return function() {
+    var dest = this;
+    var state = src._readableState;
+    state.awaitDrain--;
+    if (state.awaitDrain === 0)
+      flow(src);
+  };
+}
+
+function flow(src) {
+  var state = src._readableState;
+  var chunk;
+  state.awaitDrain = 0;
+
+  function write(dest, i, list) {
+    var written = dest.write(chunk);
+    if (false === written) {
+      state.awaitDrain++;
+    }
+  }
+
+  while (state.pipesCount && null !== (chunk = src.read())) {
+
+    if (state.pipesCount === 1)
+      write(state.pipes, 0, null);
+    else
+      forEach(state.pipes, write);
+
+    src.emit('data', chunk);
+
+    // if anyone needs a drain, then we have to wait for that.
+    if (state.awaitDrain > 0)
+      return;
+  }
+
+  // if every destination was unpiped, either before entering this
+  // function, or in the while loop, then stop flowing.
+  //
+  // NB: This is a pretty rare edge case.
+  if (state.pipesCount === 0) {
+    state.flowing = false;
+
+    // if there were data event listeners added, then switch to old mode.
+    if (EE.listenerCount(src, 'data') > 0)
+      emitDataEvents(src);
+    return;
+  }
+
+  // at this point, no one needed a drain, so we just ran out of data
+  // on the next readable event, start it over again.
+  state.ranOut = true;
+}
+
+function pipeOnReadable() {
+  if (this._readableState.ranOut) {
+    this._readableState.ranOut = false;
+    flow(this);
+  }
+}
+
+
+Readable.prototype.unpipe = function(dest) {
+  var state = this._readableState;
+
+  // if we're not piping anywhere, then do nothing.
+  if (state.pipesCount === 0)
+    return this;
+
+  // just one destination.  most common case.
+  if (state.pipesCount === 1) {
+    // passed in one, but it's not the right one.
+    if (dest && dest !== state.pipes)
+      return this;
+
+    if (!dest)
+      dest = state.pipes;
+
+    // got a match.
+    state.pipes = null;
+    state.pipesCount = 0;
+    this.removeListener('readable', pipeOnReadable);
+    state.flowing = false;
+    if (dest)
+      dest.emit('unpipe', this);
+    return this;
+  }
+
+  // slow case. multiple pipe destinations.
+
+  if (!dest) {
+    // remove all.
+    var dests = state.pipes;
+    var len = state.pipesCount;
+    state.pipes = null;
+    state.pipesCount = 0;
+    this.removeListener('readable', pipeOnReadable);
+    state.flowing = false;
+
+    for (var i = 0; i < len; i++)
+      dests[i].emit('unpipe', this);
+    return this;
+  }
+
+  // try to find the right one.
+  var i = indexOf(state.pipes, dest);
+  if (i === -1)
+    return this;
+
+  state.pipes.splice(i, 1);
+  state.pipesCount -= 1;
+  if (state.pipesCount === 1)
+    state.pipes = state.pipes[0];
+
+  dest.emit('unpipe', this);
+
+  return this;
+};
+
+// set up data events if they are asked for
+// Ensure readable listeners eventually get something
+Readable.prototype.on = function(ev, fn) {
+  var res = Stream.prototype.on.call(this, ev, fn);
+
+  if (ev === 'data' && !this._readableState.flowing)
+    emitDataEvents(this);
+
+  if (ev === 'readable' && this.readable) {
+    var state = this._readableState;
+    if (!state.readableListening) {
+      state.readableListening = true;
+      state.emittedReadable = false;
+      state.needReadable = true;
+      if (!state.reading) {
+        this.read(0);
+      } else if (state.length) {
+        emitReadable(this, state);
+      }
+    }
+  }
+
+  return res;
+};
+Readable.prototype.addListener = Readable.prototype.on;
+
+// pause() and resume() are remnants of the legacy readable stream API
+// If the user uses them, then switch into old mode.
+Readable.prototype.resume = function() {
+  emitDataEvents(this);
+  this.read(0);
+  this.emit('resume');
+};
+
+Readable.prototype.pause = function() {
+  emitDataEvents(this, true);
+  this.emit('pause');
+};
+
+function emitDataEvents(stream, startPaused) {
+  var state = stream._readableState;
+
+  if (state.flowing) {
+    // https://github.com/isaacs/readable-stream/issues/16
+    throw new Error('Cannot switch to old mode now.');
+  }
+
+  var paused = startPaused || false;
+  var readable = false;
+
+  // convert to an old-style stream.
+  stream.readable = true;
+  stream.pipe = Stream.prototype.pipe;
+  stream.on = stream.addListener = Stream.prototype.on;
+
+  stream.on('readable', function() {
+    readable = true;
+
+    var c;
+    while (!paused && (null !== (c = stream.read())))
+      stream.emit('data', c);
+
+    if (c === null) {
+      readable = false;
+      stream._readableState.needReadable = true;
+    }
+  });
+
+  stream.pause = function() {
+    paused = true;
+    this.emit('pause');
+  };
+
+  stream.resume = function() {
+    paused = false;
+    if (readable)
+      process.nextTick(function() {
+        stream.emit('readable');
+      });
+    else
+      this.read(0);
+    this.emit('resume');
+  };
+
+  // now make it start, just in case it hadn't already.
+  stream.emit('readable');
+}
+
+// wrap an old-style stream as the async data source.
+// This is *not* part of the readable stream interface.
+// It is an ugly unfortunate mess of history.
+Readable.prototype.wrap = function(stream) {
+  var state = this._readableState;
+  var paused = false;
+
+  var self = this;
+  stream.on('end', function() {
+    if (state.decoder && !state.ended) {
+      var chunk = state.decoder.end();
+      if (chunk && chunk.length)
+        self.push(chunk);
+    }
+
+    self.push(null);
+  });
+
+  stream.on('data', function(chunk) {
+    if (state.decoder)
+      chunk = state.decoder.write(chunk);
+
+    // don't skip over falsy values in objectMode
+    //if (state.objectMode && util.isNullOrUndefined(chunk))
+    if (state.objectMode && (chunk === null || chunk === undefined))
+      return;
+    else if (!state.objectMode && (!chunk || !chunk.length))
+      return;
+
+    var ret = self.push(chunk);
+    if (!ret) {
+      paused = true;
+      stream.pause();
+    }
+  });
+
+  // proxy all the other methods.
+  // important when wrapping filters and duplexes.
+  for (var i in stream) {
+    if (typeof stream[i] === 'function' &&
+        typeof this[i] === 'undefined') {
+      this[i] = function(method) { return function() {
+        return stream[method].apply(stream, arguments);
+      }}(i);
+    }
+  }
+
+  // proxy certain important events.
+  var events = ['error', 'close', 'destroy', 'pause', 'resume'];
+  forEach(events, function(ev) {
+    stream.on(ev, self.emit.bind(self, ev));
+  });
+
+  // when we try to consume some more bytes, simply unpause the
+  // underlying stream.
+  self._read = function(n) {
+    if (paused) {
+      paused = false;
+      stream.resume();
+    }
+  };
+
+  return self;
+};
+
+
+
+// exposed for testing purposes only.
+Readable._fromList = fromList;
+
+// Pluck off n bytes from an array of buffers.
+// Length is the combined lengths of all the buffers in the list.
+function fromList(n, state) {
+  var list = state.buffer;
+  var length = state.length;
+  var stringMode = !!state.decoder;
+  var objectMode = !!state.objectMode;
+  var ret;
+
+  // nothing in the list, definitely empty.
+  if (list.length === 0)
+    return null;
+
+  if (length === 0)
+    ret = null;
+  else if (objectMode)
+    ret = list.shift();
+  else if (!n || n >= length) {
+    // read it all, truncate the array.
+    if (stringMode)
+      ret = list.join('');
+    else
+      ret = Buffer.concat(list, length);
+    list.length = 0;
+  } else {
+    // read just some of it.
+    if (n < list[0].length) {
+      // just take a part of the first list item.
+      // slice is the same for buffers and strings.
+      var buf = list[0];
+      ret = buf.slice(0, n);
+      list[0] = buf.slice(n);
+    } else if (n === list[0].length) {
+      // first list is a perfect match
+      ret = list.shift();
+    } else {
+      // complex case.
+      // we have enough to cover it, but it spans past the first buffer.
+      if (stringMode)
+        ret = '';
+      else
+        ret = new Buffer(n);
+
+      var c = 0;
+      for (var i = 0, l = list.length; i < l && c < n; i++) {
+        var buf = list[0];
+        var cpy = Math.min(n - c, buf.length);
+
+        if (stringMode)
+          ret += buf.slice(0, cpy);
+        else
+          buf.copy(ret, c, 0, cpy);
+
+        if (cpy < buf.length)
+          list[0] = buf.slice(cpy);
+        else
+          list.shift();
+
+        c += cpy;
+      }
+    }
+  }
+
+  return ret;
+}
+
+function endReadable(stream) {
+  var state = stream._readableState;
+
+  // If we get here before consuming all the bytes, then that is a
+  // bug in node.  Should never happen.
+  if (state.length > 0)
+    throw new Error('endReadable called on non-empty stream');
+
+  if (!state.endEmitted && state.calledRead) {
+    state.ended = true;
+    process.nextTick(function() {
+      // Check that we didn't get one last unshift.
+      if (!state.endEmitted && state.length === 0) {
+        state.endEmitted = true;
+        stream.readable = false;
+        stream.emit('end');
+      }
+    });
+  }
+}
+
+function forEach (xs, f) {
+  for (var i = 0, l = xs.length; i < l; i++) {
+    f(xs[i], i);
+  }
+}
+
+function indexOf (xs, x) {
+  for (var i = 0, l = xs.length; i < l; i++) {
+    if (xs[i] === x) return i;
+  }
+  return -1;
+}
diff --git a/node_modules/glob-stream/node_modules/readable-stream/lib/_stream_transform.js b/node_modules/glob-stream/node_modules/readable-stream/lib/_stream_transform.js
new file mode 100644
index 000000000..eb188df3e
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/readable-stream/lib/_stream_transform.js
@@ -0,0 +1,210 @@
+// Copyright Joyent, Inc. and other Node contributors.
+//
+// Permission is hereby granted, free of charge, to any person obtaining a
+// copy of this software and associated documentation files (the
+// "Software"), to deal in the Software without restriction, including
+// without limitation the rights to use, copy, modify, merge, publish,
+// distribute, sublicense, and/or sell copies of the Software, and to permit
+// persons to whom the Software is furnished to do so, subject to the
+// following conditions:
+//
+// The above copyright notice and this permission notice shall be included
+// in all copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+// OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+// MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN
+// NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
+// DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+// OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
+// USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+
+// a transform stream is a readable/writable stream where you do
+// something with the data.  Sometimes it's called a "filter",
+// but that's not a great name for it, since that implies a thing where
+// some bits pass through, and others are simply ignored.  (That would
+// be a valid example of a transform, of course.)
+//
+// While the output is causally related to the input, it's not a
+// necessarily symmetric or synchronous transformation.  For example,
+// a zlib stream might take multiple plain-text writes(), and then
+// emit a single compressed chunk some time in the future.
+//
+// Here's how this works:
+//
+// The Transform stream has all the aspects of the readable and writable
+// stream classes.  When you write(chunk), that calls _write(chunk,cb)
+// internally, and returns false if there's a lot of pending writes
+// buffered up.  When you call read(), that calls _read(n) until
+// there's enough pending readable data buffered up.
+//
+// In a transform stream, the written data is placed in a buffer.  When
+// _read(n) is called, it transforms the queued up data, calling the
+// buffered _write cb's as it consumes chunks.  If consuming a single
+// written chunk would result in multiple output chunks, then the first
+// outputted bit calls the readcb, and subsequent chunks just go into
+// the read buffer, and will cause it to emit 'readable' if necessary.
+//
+// This way, back-pressure is actually determined by the reading side,
+// since _read has to be called to start processing a new chunk.  However,
+// a pathological inflate type of transform can cause excessive buffering
+// here.  For example, imagine a stream where every byte of input is
+// interpreted as an integer from 0-255, and then results in that many
+// bytes of output.  Writing the 4 bytes {ff,ff,ff,ff} would result in
+// 1kb of data being output.  In this case, you could write a very small
+// amount of input, and end up with a very large amount of output.  In
+// such a pathological inflating mechanism, there'd be no way to tell
+// the system to stop doing the transform.  A single 4MB write could
+// cause the system to run out of memory.
+//
+// However, even in such a pathological case, only a single written chunk
+// would be consumed, and then the rest would wait (un-transformed) until
+// the results of the previous transformed chunk were consumed.
+
+module.exports = Transform;
+
+var Duplex = require('./_stream_duplex');
+
+/*<replacement>*/
+var util = require('core-util-is');
+util.inherits = require('inherits');
+/*</replacement>*/
+
+util.inherits(Transform, Duplex);
+
+
+function TransformState(options, stream) {
+  this.afterTransform = function(er, data) {
+    return afterTransform(stream, er, data);
+  };
+
+  this.needTransform = false;
+  this.transforming = false;
+  this.writecb = null;
+  this.writechunk = null;
+}
+
+function afterTransform(stream, er, data) {
+  var ts = stream._transformState;
+  ts.transforming = false;
+
+  var cb = ts.writecb;
+
+  if (!cb)
+    return stream.emit('error', new Error('no writecb in Transform class'));
+
+  ts.writechunk = null;
+  ts.writecb = null;
+
+  if (data !== null && data !== undefined)
+    stream.push(data);
+
+  if (cb)
+    cb(er);
+
+  var rs = stream._readableState;
+  rs.reading = false;
+  if (rs.needReadable || rs.length < rs.highWaterMark) {
+    stream._read(rs.highWaterMark);
+  }
+}
+
+
+function Transform(options) {
+  if (!(this instanceof Transform))
+    return new Transform(options);
+
+  Duplex.call(this, options);
+
+  var ts = this._transformState = new TransformState(options, this);
+
+  // when the writable side finishes, then flush out anything remaining.
+  var stream = this;
+
+  // start out asking for a readable event once data is transformed.
+  this._readableState.needReadable = true;
+
+  // we have implemented the _read method, and done the other things
+  // that Readable wants before the first _read call, so unset the
+  // sync guard flag.
+  this._readableState.sync = false;
+
+  this.once('finish', function() {
+    if ('function' === typeof this._flush)
+      this._flush(function(er) {
+        done(stream, er);
+      });
+    else
+      done(stream);
+  });
+}
+
+Transform.prototype.push = function(chunk, encoding) {
+  this._transformState.needTransform = false;
+  return Duplex.prototype.push.call(this, chunk, encoding);
+};
+
+// This is the part where you do stuff!
+// override this function in implementation classes.
+// 'chunk' is an input chunk.
+//
+// Call `push(newChunk)` to pass along transformed output
+// to the readable side.  You may call 'push' zero or more times.
+//
+// Call `cb(err)` when you are done with this chunk.  If you pass
+// an error, then that'll put the hurt on the whole operation.  If you
+// never call cb(), then you'll never get another chunk.
+Transform.prototype._transform = function(chunk, encoding, cb) {
+  throw new Error('not implemented');
+};
+
+Transform.prototype._write = function(chunk, encoding, cb) {
+  var ts = this._transformState;
+  ts.writecb = cb;
+  ts.writechunk = chunk;
+  ts.writeencoding = encoding;
+  if (!ts.transforming) {
+    var rs = this._readableState;
+    if (ts.needTransform ||
+        rs.needReadable ||
+        rs.length < rs.highWaterMark)
+      this._read(rs.highWaterMark);
+  }
+};
+
+// Doesn't matter what the args are here.
+// _transform does all the work.
+// That we got here means that the readable side wants more data.
+Transform.prototype._read = function(n) {
+  var ts = this._transformState;
+
+  if (ts.writechunk !== null && ts.writecb && !ts.transforming) {
+    ts.transforming = true;
+    this._transform(ts.writechunk, ts.writeencoding, ts.afterTransform);
+  } else {
+    // mark that we need a transform, so that any data that comes in
+    // will get processed, now that we've asked for it.
+    ts.needTransform = true;
+  }
+};
+
+
+function done(stream, er) {
+  if (er)
+    return stream.emit('error', er);
+
+  // if there's nothing in the write buffer, then that means
+  // that nothing more will ever be provided
+  var ws = stream._writableState;
+  var rs = stream._readableState;
+  var ts = stream._transformState;
+
+  if (ws.length)
+    throw new Error('calling transform done when ws.length != 0');
+
+  if (ts.transforming)
+    throw new Error('calling transform done when still transforming');
+
+  return stream.push(null);
+}
diff --git a/node_modules/glob-stream/node_modules/readable-stream/lib/_stream_writable.js b/node_modules/glob-stream/node_modules/readable-stream/lib/_stream_writable.js
new file mode 100644
index 000000000..4bdaa4fa4
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/readable-stream/lib/_stream_writable.js
@@ -0,0 +1,386 @@
+// Copyright Joyent, Inc. and other Node contributors.
+//
+// Permission is hereby granted, free of charge, to any person obtaining a
+// copy of this software and associated documentation files (the
+// "Software"), to deal in the Software without restriction, including
+// without limitation the rights to use, copy, modify, merge, publish,
+// distribute, sublicense, and/or sell copies of the Software, and to permit
+// persons to whom the Software is furnished to do so, subject to the
+// following conditions:
+//
+// The above copyright notice and this permission notice shall be included
+// in all copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+// OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+// MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN
+// NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
+// DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+// OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
+// USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+// A bit simpler than readable streams.
+// Implement an async ._write(chunk, cb), and it'll handle all
+// the drain event emission and buffering.
+
+module.exports = Writable;
+
+/*<replacement>*/
+var Buffer = require('buffer').Buffer;
+/*</replacement>*/
+
+Writable.WritableState = WritableState;
+
+
+/*<replacement>*/
+var util = require('core-util-is');
+util.inherits = require('inherits');
+/*</replacement>*/
+
+var Stream = require('stream');
+
+util.inherits(Writable, Stream);
+
+function WriteReq(chunk, encoding, cb) {
+  this.chunk = chunk;
+  this.encoding = encoding;
+  this.callback = cb;
+}
+
+function WritableState(options, stream) {
+  options = options || {};
+
+  // the point at which write() starts returning false
+  // Note: 0 is a valid value, means that we always return false if
+  // the entire buffer is not flushed immediately on write()
+  var hwm = options.highWaterMark;
+  this.highWaterMark = (hwm || hwm === 0) ? hwm : 16 * 1024;
+
+  // object stream flag to indicate whether or not this stream
+  // contains buffers or objects.
+  this.objectMode = !!options.objectMode;
+
+  // cast to ints.
+  this.highWaterMark = ~~this.highWaterMark;
+
+  this.needDrain = false;
+  // at the start of calling end()
+  this.ending = false;
+  // when end() has been called, and returned
+  this.ended = false;
+  // when 'finish' is emitted
+  this.finished = false;
+
+  // should we decode strings into buffers before passing to _write?
+  // this is here so that some node-core streams can optimize string
+  // handling at a lower level.
+  var noDecode = options.decodeStrings === false;
+  this.decodeStrings = !noDecode;
+
+  // Crypto is kind of old and crusty.  Historically, its default string
+  // encoding is 'binary' so we have to make this configurable.
+  // Everything else in the universe uses 'utf8', though.
+  this.defaultEncoding = options.defaultEncoding || 'utf8';
+
+  // not an actual buffer we keep track of, but a measurement
+  // of how much we're waiting to get pushed to some underlying
+  // socket or file.
+  this.length = 0;
+
+  // a flag to see when we're in the middle of a write.
+  this.writing = false;
+
+  // a flag to be able to tell if the onwrite cb is called immediately,
+  // or on a later tick.  We set this to true at first, becuase any
+  // actions that shouldn't happen until "later" should generally also
+  // not happen before the first write call.
+  this.sync = true;
+
+  // a flag to know if we're processing previously buffered items, which
+  // may call the _write() callback in the same tick, so that we don't
+  // end up in an overlapped onwrite situation.
+  this.bufferProcessing = false;
+
+  // the callback that's passed to _write(chunk,cb)
+  this.onwrite = function(er) {
+    onwrite(stream, er);
+  };
+
+  // the callback that the user supplies to write(chunk,encoding,cb)
+  this.writecb = null;
+
+  // the amount that is being written when _write is called.
+  this.writelen = 0;
+
+  this.buffer = [];
+
+  // True if the error was already emitted and should not be thrown again
+  this.errorEmitted = false;
+}
+
+function Writable(options) {
+  var Duplex = require('./_stream_duplex');
+
+  // Writable ctor is applied to Duplexes, though they're not
+  // instanceof Writable, they're instanceof Readable.
+  if (!(this instanceof Writable) && !(this instanceof Duplex))
+    return new Writable(options);
+
+  this._writableState = new WritableState(options, this);
+
+  // legacy.
+  this.writable = true;
+
+  Stream.call(this);
+}
+
+// Otherwise people can pipe Writable streams, which is just wrong.
+Writable.prototype.pipe = function() {
+  this.emit('error', new Error('Cannot pipe. Not readable.'));
+};
+
+
+function writeAfterEnd(stream, state, cb) {
+  var er = new Error('write after end');
+  // TODO: defer error events consistently everywhere, not just the cb
+  stream.emit('error', er);
+  process.nextTick(function() {
+    cb(er);
+  });
+}
+
+// If we get something that is not a buffer, string, null, or undefined,
+// and we're not in objectMode, then that's an error.
+// Otherwise stream chunks are all considered to be of length=1, and the
+// watermarks determine how many objects to keep in the buffer, rather than
+// how many bytes or characters.
+function validChunk(stream, state, chunk, cb) {
+  var valid = true;
+  if (!Buffer.isBuffer(chunk) &&
+      'string' !== typeof chunk &&
+      chunk !== null &&
+      chunk !== undefined &&
+      !state.objectMode) {
+    var er = new TypeError('Invalid non-string/buffer chunk');
+    stream.emit('error', er);
+    process.nextTick(function() {
+      cb(er);
+    });
+    valid = false;
+  }
+  return valid;
+}
+
+Writable.prototype.write = function(chunk, encoding, cb) {
+  var state = this._writableState;
+  var ret = false;
+
+  if (typeof encoding === 'function') {
+    cb = encoding;
+    encoding = null;
+  }
+
+  if (Buffer.isBuffer(chunk))
+    encoding = 'buffer';
+  else if (!encoding)
+    encoding = state.defaultEncoding;
+
+  if (typeof cb !== 'function')
+    cb = function() {};
+
+  if (state.ended)
+    writeAfterEnd(this, state, cb);
+  else if (validChunk(this, state, chunk, cb))
+    ret = writeOrBuffer(this, state, chunk, encoding, cb);
+
+  return ret;
+};
+
+function decodeChunk(state, chunk, encoding) {
+  if (!state.objectMode &&
+      state.decodeStrings !== false &&
+      typeof chunk === 'string') {
+    chunk = new Buffer(chunk, encoding);
+  }
+  return chunk;
+}
+
+// if we're already writing something, then just put this
+// in the queue, and wait our turn.  Otherwise, call _write
+// If we return false, then we need a drain event, so set that flag.
+function writeOrBuffer(stream, state, chunk, encoding, cb) {
+  chunk = decodeChunk(state, chunk, encoding);
+  if (Buffer.isBuffer(chunk))
+    encoding = 'buffer';
+  var len = state.objectMode ? 1 : chunk.length;
+
+  state.length += len;
+
+  var ret = state.length < state.highWaterMark;
+  // we must ensure that previous needDrain will not be reset to false.
+  if (!ret)
+    state.needDrain = true;
+
+  if (state.writing)
+    state.buffer.push(new WriteReq(chunk, encoding, cb));
+  else
+    doWrite(stream, state, len, chunk, encoding, cb);
+
+  return ret;
+}
+
+function doWrite(stream, state, len, chunk, encoding, cb) {
+  state.writelen = len;
+  state.writecb = cb;
+  state.writing = true;
+  state.sync = true;
+  stream._write(chunk, encoding, state.onwrite);
+  state.sync = false;
+}
+
+function onwriteError(stream, state, sync, er, cb) {
+  if (sync)
+    process.nextTick(function() {
+      cb(er);
+    });
+  else
+    cb(er);
+
+  stream._writableState.errorEmitted = true;
+  stream.emit('error', er);
+}
+
+function onwriteStateUpdate(state) {
+  state.writing = false;
+  state.writecb = null;
+  state.length -= state.writelen;
+  state.writelen = 0;
+}
+
+function onwrite(stream, er) {
+  var state = stream._writableState;
+  var sync = state.sync;
+  var cb = state.writecb;
+
+  onwriteStateUpdate(state);
+
+  if (er)
+    onwriteError(stream, state, sync, er, cb);
+  else {
+    // Check if we're actually ready to finish, but don't emit yet
+    var finished = needFinish(stream, state);
+
+    if (!finished && !state.bufferProcessing && state.buffer.length)
+      clearBuffer(stream, state);
+
+    if (sync) {
+      process.nextTick(function() {
+        afterWrite(stream, state, finished, cb);
+      });
+    } else {
+      afterWrite(stream, state, finished, cb);
+    }
+  }
+}
+
+function afterWrite(stream, state, finished, cb) {
+  if (!finished)
+    onwriteDrain(stream, state);
+  cb();
+  if (finished)
+    finishMaybe(stream, state);
+}
+
+// Must force callback to be called on nextTick, so that we don't
+// emit 'drain' before the write() consumer gets the 'false' return
+// value, and has a chance to attach a 'drain' listener.
+function onwriteDrain(stream, state) {
+  if (state.length === 0 && state.needDrain) {
+    state.needDrain = false;
+    stream.emit('drain');
+  }
+}
+
+
+// if there's something in the buffer waiting, then process it
+function clearBuffer(stream, state) {
+  state.bufferProcessing = true;
+
+  for (var c = 0; c < state.buffer.length; c++) {
+    var entry = state.buffer[c];
+    var chunk = entry.chunk;
+    var encoding = entry.encoding;
+    var cb = entry.callback;
+    var len = state.objectMode ? 1 : chunk.length;
+
+    doWrite(stream, state, len, chunk, encoding, cb);
+
+    // if we didn't call the onwrite immediately, then
+    // it means that we need to wait until it does.
+    // also, that means that the chunk and cb are currently
+    // being processed, so move the buffer counter past them.
+    if (state.writing) {
+      c++;
+      break;
+    }
+  }
+
+  state.bufferProcessing = false;
+  if (c < state.buffer.length)
+    state.buffer = state.buffer.slice(c);
+  else
+    state.buffer.length = 0;
+}
+
+Writable.prototype._write = function(chunk, encoding, cb) {
+  cb(new Error('not implemented'));
+};
+
+Writable.prototype.end = function(chunk, encoding, cb) {
+  var state = this._writableState;
+
+  if (typeof chunk === 'function') {
+    cb = chunk;
+    chunk = null;
+    encoding = null;
+  } else if (typeof encoding === 'function') {
+    cb = encoding;
+    encoding = null;
+  }
+
+  if (typeof chunk !== 'undefined' && chunk !== null)
+    this.write(chunk, encoding);
+
+  // ignore unnecessary end() calls.
+  if (!state.ending && !state.finished)
+    endWritable(this, state, cb);
+};
+
+
+function needFinish(stream, state) {
+  return (state.ending &&
+          state.length === 0 &&
+          !state.finished &&
+          !state.writing);
+}
+
+function finishMaybe(stream, state) {
+  var need = needFinish(stream, state);
+  if (need) {
+    state.finished = true;
+    stream.emit('finish');
+  }
+  return need;
+}
+
+function endWritable(stream, state, cb) {
+  state.ending = true;
+  finishMaybe(stream, state);
+  if (cb) {
+    if (state.finished)
+      process.nextTick(cb);
+    else
+      stream.once('finish', cb);
+  }
+  state.ended = true;
+}
diff --git a/node_modules/glob-stream/node_modules/readable-stream/package.json b/node_modules/glob-stream/node_modules/readable-stream/package.json
new file mode 100644
index 000000000..e28c8250f
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/readable-stream/package.json
@@ -0,0 +1,112 @@
+{
+  "_args": [
+    [
+      {
+        "raw": "readable-stream@>=1.0.33-1 <1.1.0-0",
+        "scope": null,
+        "escapedName": "readable-stream",
+        "name": "readable-stream",
+        "rawSpec": ">=1.0.33-1 <1.1.0-0",
+        "spec": ">=1.0.33-1 <1.1.0-0",
+        "type": "range"
+      },
+      "/home/dold/repos/taler/wallet-webex/node_modules/glob-stream/node_modules/through2"
+    ]
+  ],
+  "_from": "readable-stream@>=1.0.33-1 <1.1.0-0",
+  "_id": "readable-stream@1.0.34",
+  "_inCache": true,
+  "_location": "/glob-stream/readable-stream",
+  "_nodeVersion": "5.10.1",
+  "_npmOperationalInternal": {
+    "host": "packages-12-west.internal.npmjs.com",
+    "tmp": "tmp/readable-stream-1.0.34.tgz_1460562521506_0.019665231462568045"
+  },
+  "_npmUser": {
+    "name": "cwmma",
+    "email": "calvin.metcalf@gmail.com"
+  },
+  "_npmVersion": "3.8.3",
+  "_phantomChildren": {},
+  "_requested": {
+    "raw": "readable-stream@>=1.0.33-1 <1.1.0-0",
+    "scope": null,
+    "escapedName": "readable-stream",
+    "name": "readable-stream",
+    "rawSpec": ">=1.0.33-1 <1.1.0-0",
+    "spec": ">=1.0.33-1 <1.1.0-0",
+    "type": "range"
+  },
+  "_requiredBy": [
+    "/glob-stream/through2"
+  ],
+  "_resolved": "https://registry.npmjs.org/readable-stream/-/readable-stream-1.0.34.tgz",
+  "_shasum": "125820e34bc842d2f2aaafafe4c2916ee32c157c",
+  "_shrinkwrap": null,
+  "_spec": "readable-stream@>=1.0.33-1 <1.1.0-0",
+  "_where": "/home/dold/repos/taler/wallet-webex/node_modules/glob-stream/node_modules/through2",
+  "author": {
+    "name": "Isaac Z. Schlueter",
+    "email": "i@izs.me",
+    "url": "http://blog.izs.me/"
+  },
+  "browser": {
+    "util": false
+  },
+  "bugs": {
+    "url": "https://github.com/isaacs/readable-stream/issues"
+  },
+  "dependencies": {
+    "core-util-is": "~1.0.0",
+    "inherits": "~2.0.1",
+    "isarray": "0.0.1",
+    "string_decoder": "~0.10.x"
+  },
+  "description": "Streams2, a user-land copy of the stream library from Node.js v0.10.x",
+  "devDependencies": {
+    "tap": "~0.2.6"
+  },
+  "directories": {},
+  "dist": {
+    "shasum": "125820e34bc842d2f2aaafafe4c2916ee32c157c",
+    "tarball": "https://registry.npmjs.org/readable-stream/-/readable-stream-1.0.34.tgz"
+  },
+  "gitHead": "1227c7b66deedb1dc5284a89425854d5f7ad9576",
+  "homepage": "https://github.com/isaacs/readable-stream#readme",
+  "keywords": [
+    "readable",
+    "stream",
+    "pipe"
+  ],
+  "license": "MIT",
+  "main": "readable.js",
+  "maintainers": [
+    {
+      "name": "isaacs",
+      "email": "isaacs@npmjs.com"
+    },
+    {
+      "name": "tootallnate",
+      "email": "nathan@tootallnate.net"
+    },
+    {
+      "name": "rvagg",
+      "email": "rod@vagg.org"
+    },
+    {
+      "name": "cwmma",
+      "email": "calvin.metcalf@gmail.com"
+    }
+  ],
+  "name": "readable-stream",
+  "optionalDependencies": {},
+  "readme": "ERROR: No README data found!",
+  "repository": {
+    "type": "git",
+    "url": "git://github.com/isaacs/readable-stream.git"
+  },
+  "scripts": {
+    "test": "tap test/simple/*.js"
+  },
+  "version": "1.0.34"
+}
diff --git a/node_modules/glob-stream/node_modules/readable-stream/passthrough.js b/node_modules/glob-stream/node_modules/readable-stream/passthrough.js
new file mode 100644
index 000000000..27e8d8a55
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/readable-stream/passthrough.js
@@ -0,0 +1 @@
+module.exports = require("./lib/_stream_passthrough.js")
diff --git a/node_modules/glob-stream/node_modules/readable-stream/readable.js b/node_modules/glob-stream/node_modules/readable-stream/readable.js
new file mode 100644
index 000000000..26511e87b
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/readable-stream/readable.js
@@ -0,0 +1,11 @@
+var Stream = require('stream'); // hack to fix a circular dependency issue when used with browserify
+exports = module.exports = require('./lib/_stream_readable.js');
+exports.Stream = Stream;
+exports.Readable = exports;
+exports.Writable = require('./lib/_stream_writable.js');
+exports.Duplex = require('./lib/_stream_duplex.js');
+exports.Transform = require('./lib/_stream_transform.js');
+exports.PassThrough = require('./lib/_stream_passthrough.js');
+if (!process.browser && process.env.READABLE_STREAM === 'disable') {
+  module.exports = require('stream');
+}
diff --git a/node_modules/glob-stream/node_modules/readable-stream/transform.js b/node_modules/glob-stream/node_modules/readable-stream/transform.js
new file mode 100644
index 000000000..5d482f078
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/readable-stream/transform.js
@@ -0,0 +1 @@
+module.exports = require("./lib/_stream_transform.js")
diff --git a/node_modules/glob-stream/node_modules/readable-stream/writable.js b/node_modules/glob-stream/node_modules/readable-stream/writable.js
new file mode 100644
index 000000000..e1e9efdf3
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/readable-stream/writable.js
@@ -0,0 +1 @@
+module.exports = require("./lib/_stream_writable.js")
diff --git a/node_modules/glob-stream/node_modules/through2/.npmignore b/node_modules/glob-stream/node_modules/through2/.npmignore
new file mode 100644
index 000000000..1e1dcab34
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/through2/.npmignore
@@ -0,0 +1,3 @@
+test
+.jshintrc
+.travis.yml
\ No newline at end of file
diff --git a/node_modules/glob-stream/node_modules/through2/LICENSE b/node_modules/glob-stream/node_modules/through2/LICENSE
new file mode 100644
index 000000000..f6a0029de
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/through2/LICENSE
@@ -0,0 +1,39 @@
+Copyright 2013, Rod Vagg (the "Original Author")
+All rights reserved.
+
+MIT +no-false-attribs License
+
+Permission is hereby granted, free of charge, to any person
+obtaining a copy of this software and associated documentation
+files (the "Software"), to deal in the Software without
+restriction, including without limitation the rights to use,
+copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the
+Software is furnished to do so, subject to the following
+conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+Distributions of all or part of the Software intended to be used
+by the recipients as they would use the unmodified Software,
+containing modifications that substantially alter, remove, or
+disable functionality of the Software, outside of the documented
+configuration mechanisms provided by the Software, shall be
+modified such that the Original Author's bug reporting email
+addresses and urls are either replaced with the contact information
+of the parties responsible for the changes, or removed entirely.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+OTHER DEALINGS IN THE SOFTWARE.
+
+
+Except where noted, this license applies to any and all software
+programs and associated documentation files created by the
+Original Author, when distributed with the Software.
\ No newline at end of file
diff --git a/node_modules/glob-stream/node_modules/through2/README.md b/node_modules/glob-stream/node_modules/through2/README.md
new file mode 100644
index 000000000..11259a5f7
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/through2/README.md
@@ -0,0 +1,132 @@
+# through2
+
+[![NPM](https://nodei.co/npm/through2.png?downloads&downloadRank)](https://nodei.co/npm/through2/)
+
+**A tiny wrapper around Node streams.Transform (Streams2) to avoid explicit subclassing noise**
+
+Inspired by [Dominic Tarr](https://github.com/dominictarr)'s [through](https://github.com/dominictarr/through) in that it's so much easier to make a stream out of a function than it is to set up the prototype chain properly: `through(function (chunk) { ... })`.
+
+Note: A **Streams3** version of through2 is available in npm with the tag `"1.0"` rather than `"latest"` so an `npm install through2` will get you the current Streams2 version (version number is 0.x.x). To use a Streams3 version use `npm install through2@1` to fetch the latest version 1.x.x. More information about Streams2 vs Streams3 and recommendations see the article **[Why I don't use Node's core 'stream' module](http://r.va.gg/2014/06/why-i-dont-use-nodes-core-stream-module.html)**.
+
+```js
+fs.createReadStream('ex.txt')
+  .pipe(through2(function (chunk, enc, callback) {
+    for (var i = 0; i < chunk.length; i++)
+      if (chunk[i] == 97)
+        chunk[i] = 122 // swap 'a' for 'z'
+
+    this.push(chunk)
+
+    callback()
+   }))
+  .pipe(fs.createWriteStream('out.txt'))
+```
+
+Or object streams:
+
+```js
+var all = []
+
+fs.createReadStream('data.csv')
+  .pipe(csv2())
+  .pipe(through2.obj(function (chunk, enc, callback) {
+    var data = {
+        name    : chunk[0]
+      , address : chunk[3]
+      , phone   : chunk[10]
+    }
+    this.push(data)
+
+    callback()
+  }))
+  .on('data', function (data) {
+    all.push(data)
+  })
+  .on('end', function () {
+    doSomethingSpecial(all)
+  })
+```
+
+Note that `through2.obj(fn)` is a convenience wrapper around `through2({ objectMode: true }, fn)`.
+
+## API
+
+<b><code>through2([ options, ] [ transformFunction ] [, flushFunction ])</code></b>
+
+Consult the **[stream.Transform](http://nodejs.org/docs/latest/api/stream.html#stream_class_stream_transform)** documentation for the exact rules of the `transformFunction` (i.e. `this._transform`) and the optional `flushFunction` (i.e. `this._flush`).
+
+### options
+
+The options argument is optional and is passed straight through to `stream.Transform`. So you can use `objectMode:true` if you are processing non-binary streams (or just use `through2.obj()`).
+
+The `options` argument is first, unlike standard convention, because if I'm passing in an anonymous function then I'd prefer for the options argument to not get lost at the end of the call:
+
+```js
+fs.createReadStream('/tmp/important.dat')
+  .pipe(through2({ objectMode: true, allowHalfOpen: false },
+    function (chunk, enc, cb) {
+      cb(null, 'wut?') // note we can use the second argument on the callback
+                       // to provide data as an alternative to this.push('wut?')
+    }
+  )
+  .pipe(fs.createWriteStream('/tmp/wut.txt'))
+```
+
+### transformFunction
+
+The `transformFunction` must have the following signature: `function (chunk, encoding, callback) {}`. A minimal implementation should call the `callback` function to indicate that the transformation is done, even if that transformation means discarding the chunk.
+
+To queue a new chunk, call `this.push(chunk)`&mdash;this can be called as many times as required before the `callback()` if you have multiple pieces to send on.
+
+Alternatively, you may use `callback(err, chunk)` as shorthand for emitting a single chunk or an error.
+
+If you **do not provide a `transformFunction`** then you will get a simple pass-through stream.
+
+### flushFunction
+
+The optional `flushFunction` is provided as the last argument (2nd or 3rd, depending on whether you've supplied options) is called just prior to the stream ending. Can be used to finish up any processing that may be in progress.
+
+```js
+fs.createReadStream('/tmp/important.dat')
+  .pipe(through2(
+    function (chunk, enc, cb) { cb(null, chunk) }, // transform is a noop
+    function (cb) { // flush function
+      this.push('tacking on an extra buffer to the end');
+      cb();
+    }
+  ))
+  .pipe(fs.createWriteStream('/tmp/wut.txt'));
+```
+
+<b><code>through2.ctor([ options, ] transformFunction[, flushFunction ])</code></b>
+
+Instead of returning a `stream.Transform` instance, `through2.ctor()` returns a **constructor** for a custom Transform. This is useful when you want to use the same transform logic in multiple instances.
+
+```js
+var FToC = through2.ctor({objectMode: true}, function (record, encoding, callback) {
+  if (record.temp != null && record.unit = "F") {
+    record.temp = ( ( record.temp - 32 ) * 5 ) / 9
+    record.unit = "C"
+  }
+  this.push(record)
+  callback()
+})
+
+// Create instances of FToC like so:
+var converter = new FToC()
+// Or:
+var converter = FToC()
+// Or specify/override options when you instantiate, if you prefer:
+var converter = FToC({objectMode: true})
+```
+
+## See Also
+
+  - [through2-map](https://github.com/brycebaril/through2-map) - Array.prototype.map analog for streams.
+  - [through2-filter](https://github.com/brycebaril/through2-filter) - Array.prototype.filter analog for streams.
+  - [through2-reduce](https://github.com/brycebaril/through2-reduce) - Array.prototype.reduce analog for streams.
+  - [through2-spy](https://github.com/brycebaril/through2-spy) - Wrapper for simple stream.PassThrough spies.
+
+## License
+
+**through2** is Copyright (c) 2013 Rod Vagg [@rvagg](https://twitter.com/rvagg) and licenced under the MIT licence. All rights not explicitly granted in the MIT license are reserved. See the included LICENSE file for more details.
diff --git a/node_modules/glob-stream/node_modules/through2/package.json b/node_modules/glob-stream/node_modules/through2/package.json
new file mode 100644
index 000000000..56e31b7a7
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/through2/package.json
@@ -0,0 +1,98 @@
+{
+  "_args": [
+    [
+      {
+        "raw": "through2@^0.6.1",
+        "scope": null,
+        "escapedName": "through2",
+        "name": "through2",
+        "rawSpec": "^0.6.1",
+        "spec": ">=0.6.1 <0.7.0",
+        "type": "range"
+      },
+      "/home/dold/repos/taler/wallet-webex/node_modules/glob-stream"
+    ]
+  ],
+  "_from": "through2@>=0.6.1 <0.7.0",
+  "_id": "through2@0.6.5",
+  "_inCache": true,
+  "_location": "/glob-stream/through2",
+  "_npmUser": {
+    "name": "bryce",
+    "email": "bryce@ravenwall.com"
+  },
+  "_npmVersion": "1.4.28",
+  "_phantomChildren": {},
+  "_requested": {
+    "raw": "through2@^0.6.1",
+    "scope": null,
+    "escapedName": "through2",
+    "name": "through2",
+    "rawSpec": "^0.6.1",
+    "spec": ">=0.6.1 <0.7.0",
+    "type": "range"
+  },
+  "_requiredBy": [
+    "/glob-stream"
+  ],
+  "_resolved": "https://registry.npmjs.org/through2/-/through2-0.6.5.tgz",
+  "_shasum": "41ab9c67b29d57209071410e1d7a7a968cd3ad48",
+  "_shrinkwrap": null,
+  "_spec": "through2@^0.6.1",
+  "_where": "/home/dold/repos/taler/wallet-webex/node_modules/glob-stream",
+  "author": {
+    "name": "Rod Vagg",
+    "email": "r@va.gg",
+    "url": "https://github.com/rvagg"
+  },
+  "bugs": {
+    "url": "https://github.com/rvagg/through2/issues"
+  },
+  "dependencies": {
+    "readable-stream": ">=1.0.33-1 <1.1.0-0",
+    "xtend": ">=4.0.0 <4.1.0-0"
+  },
+  "description": "A tiny wrapper around Node streams2 Transform to avoid explicit subclassing noise",
+  "devDependencies": {
+    "bl": ">=0.9.0 <0.10.0-0",
+    "stream-spigot": ">=3.0.4 <3.1.0-0",
+    "tape": ">=2.14.0 <2.15.0-0"
+  },
+  "directories": {},
+  "dist": {
+    "shasum": "41ab9c67b29d57209071410e1d7a7a968cd3ad48",
+    "tarball": "https://registry.npmjs.org/through2/-/through2-0.6.5.tgz"
+  },
+  "gitHead": "ba4a87875f2c82323c10023e36f4ae4b386c1bf8",
+  "homepage": "https://github.com/rvagg/through2",
+  "keywords": [
+    "stream",
+    "streams2",
+    "through",
+    "transform"
+  ],
+  "license": "MIT",
+  "main": "through2.js",
+  "maintainers": [
+    {
+      "name": "rvagg",
+      "email": "rod@vagg.org"
+    },
+    {
+      "name": "bryce",
+      "email": "bryce@ravenwall.com"
+    }
+  ],
+  "name": "through2",
+  "optionalDependencies": {},
+  "readme": "ERROR: No README data found!",
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/rvagg/through2.git"
+  },
+  "scripts": {
+    "test": "node test/test.js",
+    "test-local": "brtapsauce-local test/basic-test.js"
+  },
+  "version": "0.6.5"
+}
diff --git a/node_modules/glob-stream/node_modules/through2/through2.js b/node_modules/glob-stream/node_modules/through2/through2.js
new file mode 100644
index 000000000..5b7a880e8
--- /dev/null
+++ b/node_modules/glob-stream/node_modules/through2/through2.js
@@ -0,0 +1,96 @@
+var Transform = require('readable-stream/transform')
+  , inherits  = require('util').inherits
+  , xtend     = require('xtend')
+
+function DestroyableTransform(opts) {
+  Transform.call(this, opts)
+  this._destroyed = false
+}
+
+inherits(DestroyableTransform, Transform)
+
+DestroyableTransform.prototype.destroy = function(err) {
+  if (this._destroyed) return
+  this._destroyed = true
+  
+  var self = this
+  process.nextTick(function() {
+    if (err)
+      self.emit('error', err)
+    self.emit('close')
+  })
+}
+
+// a noop _transform function
+function noop (chunk, enc, callback) {
+  callback(null, chunk)
+}
+
+
+// create a new export function, used by both the main export and
+// the .ctor export, contains common logic for dealing with arguments
+function through2 (construct) {
+  return function (options, transform, flush) {
+    if (typeof options == 'function') {
+      flush     = transform
+      transform = options
+      options   = {}
+    }
+
+    if (typeof transform != 'function')
+      transform = noop
+
+    if (typeof flush != 'function')
+      flush = null
+
+    return construct(options, transform, flush)
+  }
+}
+
+
+// main export, just make me a transform stream!
+module.exports = through2(function (options, transform, flush) {
+  var t2 = new DestroyableTransform(options)
+
+  t2._transform = transform
+
+  if (flush)
+    t2._flush = flush
+
+  return t2
+})
+
+
+// make me a reusable prototype that I can `new`, or implicitly `new`
+// with a constructor call
+module.exports.ctor = through2(function (options, transform, flush) {
+  function Through2 (override) {
+    if (!(this instanceof Through2))
+      return new Through2(override)
+
+    this.options = xtend(options, override)
+
+    DestroyableTransform.call(this, this.options)
+  }
+
+  inherits(Through2, DestroyableTransform)
+
+  Through2.prototype._transform = transform
+
+  if (flush)
+    Through2.prototype._flush = flush
+
+  return Through2
+})
+
+
+module.exports.obj = through2(function (options, transform, flush) {
+  var t2 = new DestroyableTransform(xtend({ objectMode: true, highWaterMark: 16 }, options))
+
+  t2._transform = transform
+
+  if (flush)
+    t2._flush = flush
+
+  return t2
+})
diff --git a/node_modules/glob-stream/package.json b/node_modules/glob-stream/package.json
new file mode 100644
index 000000000..d801bf2a6
--- /dev/null
+++ b/node_modules/glob-stream/package.json
@@ -0,0 +1,118 @@
+{
+  "_args": [
+    [
+      {
+        "raw": "glob-stream@^3.1.5",
+        "scope": null,
+        "escapedName": "glob-stream",
+        "name": "glob-stream",
+        "rawSpec": "^3.1.5",
+        "spec": ">=3.1.5 <4.0.0",
+        "type": "range"
+      },
+      "/home/dold/repos/taler/wallet-webex/node_modules/gulp/node_modules/vinyl-fs"
+    ]
+  ],
+  "_from": "glob-stream@>=3.1.5 <4.0.0",
+  "_id": "glob-stream@3.1.18",
+  "_inCache": true,
+  "_location": "/glob-stream",
+  "_nodeVersion": "0.10.33",
+  "_npmUser": {
+    "name": "fractal",
+    "email": "contact@wearefractal.com"
+  },
+  "_npmVersion": "2.1.6",
+  "_phantomChildren": {
+    "brace-expansion": "1.1.6",
+    "core-util-is": "1.0.2",
+    "inflight": "1.0.5",
+    "inherits": "2.0.3",
+    "isarray": "0.0.1",
+    "once": "1.4.0",
+    "string_decoder": "0.10.31",
+    "xtend": "4.0.1"
+  },
+  "_requested": {
+    "raw": "glob-stream@^3.1.5",
+    "scope": null,
+    "escapedName": "glob-stream",
+    "name": "glob-stream",
+    "rawSpec": "^3.1.5",
+    "spec": ">=3.1.5 <4.0.0",
+    "type": "range"
+  },
+  "_requiredBy": [
+    "/gulp/vinyl-fs"
+  ],
+  "_resolved": "https://registry.npmjs.org/glob-stream/-/glob-stream-3.1.18.tgz",
+  "_shasum": "9170a5f12b790306fdfe598f313f8f7954fd143b",
+  "_shrinkwrap": null,
+  "_spec": "glob-stream@^3.1.5",
+  "_where": "/home/dold/repos/taler/wallet-webex/node_modules/gulp/node_modules/vinyl-fs",
+  "author": {
+    "name": "Fractal",
+    "email": "contact@wearefractal.com",
+    "url": "http://wearefractal.com/"
+  },
+  "bugs": {
+    "url": "https://github.com/wearefractal/glob-stream/issues"
+  },
+  "dependencies": {
+    "glob": "^4.3.1",
+    "glob2base": "^0.0.12",
+    "minimatch": "^2.0.1",
+    "ordered-read-streams": "^0.1.0",
+    "through2": "^0.6.1",
+    "unique-stream": "^1.0.0"
+  },
+  "description": "File system globs as a stream",
+  "devDependencies": {
+    "coveralls": "^2.11.2",
+    "istanbul": "^0.3.0",
+    "istanbul-coveralls": "^1.0.1",
+    "jshint": "^2.5.10",
+    "mocha": "^2.0.0",
+    "mocha-lcov-reporter": "0.0.1",
+    "rimraf": "^2.2.5",
+    "should": "^4.3.0"
+  },
+  "directories": {},
+  "dist": {
+    "shasum": "9170a5f12b790306fdfe598f313f8f7954fd143b",
+    "tarball": "https://registry.npmjs.org/glob-stream/-/glob-stream-3.1.18.tgz"
+  },
+  "engines": {
+    "node": ">= 0.9"
+  },
+  "files": [
+    "index.js"
+  ],
+  "gitHead": "472b98e7a0a747a3c72454337def65cebc4fb78e",
+  "homepage": "http://github.com/wearefractal/glob-stream",
+  "licenses": [
+    {
+      "type": "MIT",
+      "url": "http://github.com/wearefractal/glob-stream/raw/master/LICENSE"
+    }
+  ],
+  "main": "./index.js",
+  "maintainers": [
+    {
+      "name": "fractal",
+      "email": "contact@wearefractal.com"
+    }
+  ],
+  "name": "glob-stream",
+  "optionalDependencies": {},
+  "readme": "ERROR: No README data found!",
+  "repository": {
+    "type": "git",
+    "url": "git://github.com/wearefractal/glob-stream.git"
+  },
+  "scripts": {
+    "coveralls": "istanbul cover _mocha --report lcovonly -- -R spec && istanbul-coveralls",
+    "test": "mocha --reporter spec && jshint"
+  },
+  "version": "3.1.18"
+}