Class: StorageVisualizer

Inherits:
Object
  • Object
show all
Defined in:
lib/storage_visualizer.rb

Instance Attribute Summary collapse

Class Method Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(target_dir_passed = nil) ⇒ StorageVisualizer

Constructor



143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
 
# File 'lib/storage_visualizer.rb', line 143

def initialize(target_dir_passed = nil)

  if (target_dir_passed != nil)
    expanded = File.expand_path(target_dir_passed)
    # puts "Target dir: #{expanded}"
    if (Dir.exist?(expanded))
      self.target_dir = expanded
    else
      raise "Target directory does not exist: #{expanded}"
    end
  else
    # no target passed, use the user's home dir
    self.target_dir = File.expand_path('~')
  end
  
  # how much space is considered worthy of noting on the chart
  self.threshold_pct = 0.05
  self.diskhash = {}
  self.tree = []
  self.tree_formatted = ''
  self.dupe_counter = 0
end

Instance Attribute Details

#available ⇒ Object

Returns the value of attribute available.



125
126
127
 
# File 'lib/storage_visualizer.rb', line 125

def available
  @available
end

#available_gb ⇒ Object

Returns the value of attribute available_gb.



129
130
131
 
# File 'lib/storage_visualizer.rb', line 129

def available_gb
  @available_gb
end

#capacity ⇒ Object

disk Bytes



123
124
125
 
# File 'lib/storage_visualizer.rb', line 123

def capacity
  @capacity
end

#capacity_gb ⇒ Object

disk GB for display



127
128
129
 
# File 'lib/storage_visualizer.rb', line 127

def capacity_gb
  @capacity_gb
end

#dir_tree ⇒ Object

this is the root DirNode object



140
141
142
 
# File 'lib/storage_visualizer.rb', line 140

def dir_tree
  @dir_tree
end

#diskhash ⇒ Object

Returns the value of attribute diskhash.



134
135
136
 
# File 'lib/storage_visualizer.rb', line 134

def diskhash
  @diskhash
end

#dupe_counter ⇒ Object

Returns the value of attribute dupe_counter.



137
138
139
 
# File 'lib/storage_visualizer.rb', line 137

def dupe_counter
  @dupe_counter
end

#target_dir ⇒ Object

other



131
132
133
 
# File 'lib/storage_visualizer.rb', line 131

def target_dir
  @target_dir
end

#target_volume ⇒ Object

Returns the value of attribute target_volume.



136
137
138
 
# File 'lib/storage_visualizer.rb', line 136

def target_volume
  @target_volume
end

#threshold_pct ⇒ Object

Returns the value of attribute threshold_pct.



135
136
137
 
# File 'lib/storage_visualizer.rb', line 135

def threshold_pct
  @threshold_pct
end

#tree ⇒ Object

Returns the value of attribute tree.



132
133
134
 
# File 'lib/storage_visualizer.rb', line 132

def tree
  @tree
end

#tree_formatted ⇒ Object

Returns the value of attribute tree_formatted.



133
134
135
 
# File 'lib/storage_visualizer.rb', line 133

def tree_formatted
  @tree_formatted
end

#used ⇒ Object

Returns the value of attribute used.



124
125
126
 
# File 'lib/storage_visualizer.rb', line 124

def used
  @used
end

#used_gb ⇒ Object

Returns the value of attribute used_gb.



128
129
130
 
# File 'lib/storage_visualizer.rb', line 128

def used_gb
  @used_gb
end

Class Method Details

.install ⇒ Object 



66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
 
# File 'lib/storage_visualizer.rb', line 66

def self.install
  # This function installs a copy & symlink into /usr/local/bin, so the utility can simply be run by typing `storage_visualizer`
  # To install for command line use type:
  # git clone https://github.com/teecay/StorageVisualizer.git && ./StorageVisualizer/storage_visualizer.rb --install
  # To install for gem usage type:
  # gem install storage_visualizer


  script_src_path = File.expand_path(__FILE__) # File.expand_path('./StorageVisualizer/lib/storage_visualizer.rb')
  script_dest_path = '/usr/local/bin/storage_visualizer.rb'
  symlink_path = '/usr/local/bin/storage_visualizer'


  if (!File.exist?(script_src_path))
    raise "Error: file does not exist: #{script_src_path}"
  end


  if (File.exist?(script_dest_path))
    puts "Removing old installed script"
    File.delete(script_dest_path)
  end


  if (File.exist?(symlink_path))
    puts "Removing old symlink"
    File.delete(symlink_path)
  end

  cp_cmd = "cp -f #{script_src_path} #{script_dest_path}"
  puts "Copying script into place: #{cp_cmd}"
  `#{cp_cmd}`

  ln_cmd = "ln -s #{script_dest_path} #{symlink_path}"
  puts "Installing: #{ln_cmd}"
  `#{ln_cmd}`

  chmod_cmd = "chmod ugo+x #{symlink_path}"
  puts "Setting permissions: #{chmod_cmd}"
  `#{chmod_cmd}`

  puts "Installation is complete, run `storage_visualizer -h` for help"
end

.print_usage ⇒ Object

Static



51
52
53
54
55
56
57
58
59
60
61
62
 
# File 'lib/storage_visualizer.rb', line 51

def self.print_usage
  puts "\nThis tool helps visualize which directories are occupying the most storage. Any directory that occupies more than 5% of disk space is added to a visual hierarchichal storage report in the form of a Google Sankey diagram. The storage data is gathered using the linux `du` utility. It has been tested on Mac OSX, should work on linux systems, will not work on Windows. Run as sudo if analyzing otherwise inaccessible directories. May take a while to run\n"
  puts "\nCommand line usage: \n\t[sudo] storage_visualizer[.rb] [directory to visualize (default ~/) | -h (help) -i | --install (install to /usr/local/bin)]\n\n"
  puts "API usage: "
  puts "\tgem install storage_visualizer"
  puts "\t'require storage_visualizer'"
  puts "\tsv = StorageVisualizer.new('[directory to visualize, ~/ by default]')"
  puts "\tsv.run()\n\n"
  puts "A report will be created in the current directory named as such: StorageReport_2015_05_25-17_19_30.html"
  puts "Status messages are printed to STDOUT"
  puts "\n\n"
end

Instance Method Details

#analyze_dirs(dir_to_analyze, parent) ⇒ Object

Crawl the dirs recursively, beginning with the target dir



393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
 
# File 'lib/storage_visualizer.rb', line 393

def analyze_dirs(dir_to_analyze, parent)


  # bootstrap case
  # don't create an entry for the root because there's nothing to link to yet, scan the subdirs
  if (dir_to_analyze == self.target_volume)
    # puts "Dir to analyze is the target volume"
    # run on all child dirs, not this dir
    Dir.entries(dir_to_analyze).reject {|d| d.start_with?('.')}.each do |name|
      # puts "\tentry: >#{file}<"
      full_path = File.join(dir_to_analyze, name)
      if (Dir.exist?(full_path) && !File.symlink?(full_path))
        # puts "Contender: >#{full_path}<"
        analyze_dirs(full_path, self.dir_tree)
      end
    end
    return
  end

  # use "P" to help prevent following any symlinks
  cmd = "du -sxkP \"#{dir_to_analyze}\""
  puts "\trunning #{cmd}"
  output = `#{cmd}`.strip().split("\t")
  # puts "Du output:"
  # pp output
  size = output[0].to_i 
  size_gb = "#{'%.0f' % (size.to_f / 1024 / 1024)}"
  # puts "Size: #{size}\nCapacity: #{self.diskhash['/']['capacity']}"

  # Occupancy as a fraction of total space
  # occupancy = (size.to_f / self.capacity.to_f)

  # Occupancy as a fraction of USED space
  occupancy = (size.to_f / self.used.to_f)

  occupancy_pct = "#{'%.0f' % (occupancy * 100)}"
  capacity_gb = "#{'%.0f' % (self.capacity.to_f / 1024 / 1024)}"
  
  # if this dir contains more than 5% of disk space, add it to the tree

  if (dir_to_analyze == self.target_dir)
    # puts "Dir to analyze is the target dir, space used outside this dir.."
    # account for space used outside of target dir
    other_space = self.used - size
    other_space_gb = "#{'%.0f' % (other_space / 1024 / 1024)}"
    parent.children.push(DirNode.new(parent, self.target_volume, 'Other Space', other_space_gb))
  end
  
  
  if (occupancy > self.threshold_pct)
    # puts "Dir contains more than 5% of disk space: #{dir_to_analyze} \n\tsize:\t#{size_gb} / \ncapacity:\t#{capacity_gb} = #{occupancy_pct}%"
    puts "Dir contains more than 5% of used disk space: #{dir_to_analyze} \n\tsize:\t\t#{size_gb} / \n\toccupancy:\t#{self.used_gb} = #{occupancy_pct}% of used space"

    # puts "Dir to analyze (#{dir_to_analyze}) is not the target dir (#{self.target_dir})"
    dirs = dir_to_analyze.split('/')
    
    short_dir = dirs.pop().gsub("'","\\\\'")
    full_parent = dirs.join('/')
    if (dir_to_analyze == self.target_dir || full_parent == self.target_volume)
      # puts "Either this dir is the target dir, or the parent is the target volume, make parent the full target volume"
      short_parent = self.target_volume.gsub("'","\\\\'")
    else
      # puts "Neither this dir or parent is the target dir, making parent short"
      short_parent = dirs.pop().gsub("'","\\\\'")
    end
    

    this_node = DirNode.new(parent, dir_to_analyze, short_dir, size_gb)
    parent.children.push(this_node)

    # run on all child dirs
    Dir.entries(dir_to_analyze).reject {|d| d.start_with?('.')}.each do |name|
      full_path = File.join(dir_to_analyze, name)
      # don't follow any symlinks
      if (Dir.exist?(full_path) && !File.symlink?(full_path))
        # puts "Contender: >#{full_path}<"
        analyze_dirs(full_path, this_node)
      end
    end
    
  end # occupancy > threshold
    
end

#format_data_for_the_chart ⇒ Object 



168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
 
# File 'lib/storage_visualizer.rb', line 168

def format_data_for_the_chart

  # Build the list of nodes
  nodes = []
  nodes.push(self.dir_tree)
  comparison_list = []
  while true
    if (nodes.length == 0)
      break
    end
    node = nodes.shift
    comparison_list.push(node)
    nodes.concat(node.children)
  end
  

  # format the data for the chart
  working_string = "[\n"
  comparison_list.each_with_index do |entry, index|
    if (entry.parent == nil)
      next
    end
    if(index == comparison_list.length - 1)
      # this is the next to last element, it gets no comma
      working_string << "[ '#{entry.parent.dir_short}', '#{entry.dir_short}', #{entry.size_gb} ]\n"
    else
      # mind the comma
      working_string << "[ '#{entry.parent.dir_short}', '#{entry.dir_short}', #{entry.size_gb} ],\n"
    end
  end
  working_string << "]\n"
  self.tree_formatted = working_string
  
end

#get_basic_disk_info ⇒ Object 



294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
 
# File 'lib/storage_visualizer.rb', line 294

def get_basic_disk_info
  # df -l gets info about locally-mounted filesystems
  output = `df -k`
  
  # OSX:
  #   Filesystem                          1024-blocks  Used      Available   Capacity   iused     ifree       %iused  Mounted on
  #   /dev/disk1                          975912960    349150592 626506368   36%        87351646  156626592   36%     /
  #   localhost:/QwnJE6UBvlR1EvqouX6gMM   975912960    975912960         0   100%        0         0          100%    /Volumes/MobileBackups
  
  # CentOS:
  #   Filesystem     1K-blocks    Used Available Use% Mounted on
  #   /dev/xvda1      82436764 3447996  78888520   5% /
  #   devtmpfs        15434608      56  15434552   1% /dev
  #   tmpfs           15443804       0  15443804   0% /dev/shm
  
  # Ubuntu:
  #   Filesystem     1K-blocks   Used Available Use% Mounted on
  #   /dev/xvda1      30832636 797568  28676532   3% /
  #   none                   4      0         4   0% /sys/fs/cgroup
  #   udev             3835900     12   3835888   1% /dev
  #   tmpfs             769376    188    769188   1% /run
  #   none                5120      0      5120   0% /run/lock
  #   none             3846876      0   3846876   0% /run/shm
  #   none              102400      0    102400   0% /run/user
  #   /dev/xvdb       30824956  45124  29207352   1% /mnt

  # Populate disk info into a hash of hashes
  # {"/"=>
  #   {"capacity"=>498876809216, "used"=>434777001984, "available"=>63837663232},
  #  "/Volumes/MobileBackups"=>
  #   {"capacity"=>498876809216, "used"=>498876809216, "available"=>0}
  # }

  # get each mount's capacity & utilization
  output.lines.each_with_index do |line, index|
    if (index == 0)
      # skip the header line
      next
    end
    cols = line.split
    # ["Filesystem", "1024-blocks", "Used", "Available", "Capacity", "iused", "ifree", "%iused", "Mounted", "on"]
    # line: ["/dev/disk1", "974368768", "849157528", "124699240", "88%", "106208689", "15587405", "87%", "/"]
    
    if cols.length == 9
      # OSX
      self.diskhash[cols[8]] = {
        'capacity' => (cols[1].to_i ).to_i,
        'used' => (cols[2].to_i ).to_i,
        'available' => (cols[3].to_i ).to_i
      }
    elsif cols.length == 6
      # Ubuntu & CentOS
      self.diskhash[cols[5]] = {
        'capacity' => (cols[1].to_i ).to_i,
        'used' => (cols[2].to_i ).to_i,
        'available' => (cols[3].to_i ).to_i
      }
    else
      raise "Reported disk utilization not understood"
    end
  end

  # puts "Disk mount info:"
  # pp diskhash


  # find the (self.)target_volume 
  # look through diskhash keys, to find the one that most matches target_dir
  val_of_min = 1000
  # puts "Determining which volume contains the target directory.."
  self.diskhash.keys.each do |volume|
    result = self.target_dir.gsub(volume, '')
    diskhash['match_amt'] = result.length
    # puts "Considering:\t#{volume}, \t closeness: #{result.length}, \t (#{result})"
    if (result.length < val_of_min)
      # puts "Candidate: #{volume}"
      val_of_min = result.length
      self.target_volume = volume
    end
  end    
  
  puts "Target volume is #{self.target_volume}"
  

  self.capacity   = self.diskhash[self.target_volume]['capacity']
  self.used       = self.diskhash[self.target_volume]['used']
  self.available  = self.diskhash[self.target_volume]['available']

  self.capacity_gb  = "#{'%.0f' % (self.capacity.to_i / 1024 / 1024)}"
  self.used_gb      = "#{'%.0f' % (self.used.to_i / 1024 / 1024)}"
  self.available_gb = "#{'%.0f' % (self.available.to_i / 1024 / 1024)}"

  self.dir_tree = DirNode.new(nil, self.target_volume, self.target_volume, self.capacity)
  self.dir_tree.children.push(DirNode.new(self.dir_tree, 'Free Space', 'Free Space', self.available_gb))

end

#run ⇒ Object 



519
520
521
522
523
524
525
526
 
# File 'lib/storage_visualizer.rb', line 519

def run
  self.get_basic_disk_info
  self.analyze_dirs(self.target_dir, self.dir_tree)
  self.traverse_tree_and_remove_duplicates
  self.format_data_for_the_chart
  self.write_storage_report
  
end

#traverse_tree_and_remove_duplicates ⇒ Object 



479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
 
# File 'lib/storage_visualizer.rb', line 479

def traverse_tree_and_remove_duplicates
  puts "\nHandling duplicate entries.."
  nodes = []
  nodes.push(self.dir_tree)
  comparison_list = []
  while true
    if (nodes.length == 0)
      break
    end
    
    node = nodes.shift
    comparison_list.push(node)
    # pp node
    if node.parent == nil
      # puts "\tparent: no parent \n\tdir:    #{node.dir_name} \n\tshort:  #{node.dir_short} \n\tsize:   #{node.size_gb}"
    else 
      # puts "\tparent: #{node.parent.dir_short.to_s} \n\tdir:    #{node.dir_name} \n\tshort:  #{node.dir_short} \n\tsize:   #{node.size_gb}"
    end
    nodes.concat(node.children)
  end
  # puts "Done building node list"
  
  
  
  for i in 0..comparison_list.length do
    for j in 0..comparison_list.length do
      if (comparison_list[i] != nil && comparison_list[j] != nil)
        if (i != j && comparison_list[i].dir_short == comparison_list[j].dir_short)
          puts "\t#{comparison_list[i].dir_short} is the same as #{comparison_list[j].dir_short}, changing to #{comparison_list[j].dir_short}*"
          comparison_list[j].dir_short = "#{comparison_list[j].dir_short}*"
        end
      end
    end
  end
  puts "Duplicate handling complete"
  
end

#write_storage_report ⇒ Object 



205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
 
# File 'lib/storage_visualizer.rb', line 205

def write_storage_report

  the_html = %q|<html>
  <body>
  <script type="text/javascript"
        src="https://www.google.com/jsapi?autoload={'modules':[{'name':'visualization','version':'1.1','packages':['sankey']}]}">
  </script>
  
  <style>
    td
      {
        font-family:sans-serif;
        font-size:8pt;
      }

    .bigger
        {
          font-family:sans-serif;
          font-size:10pt;
          font-weight:bold
        }
  
  </style>

  <div class="table">
    <div class="bigger">Storage Report</div>
    <table>
      <tr>
        <td style="text-align:right">Disk Capacity:</td><td>| + self.capacity_gb + %q| GB</td>
      </tr>
      <tr>
        <td style="text-align:right">Disk Used:</td><td>| + self.used_gb + %q| GB</td>
      </tr>
      <tr>
        <td style="text-align:right">Free Space:</td><td>| + self.available_gb + %q| GB</td>
      </tr>
    </table>
        
  </div>


  <div id="sankey_multiple" style="width: 900px; height: 300px;"></div>

  <script type="text/javascript">

  google.setOnLoadCallback(drawChart);
     function drawChart() {
      var data = new google.visualization.DataTable();
      data.addColumn('string', 'From');
      data.addColumn('string', 'To');
      data.addColumn('number', 'Size (GB)');
      data.addRows( | + self.tree_formatted + %q|);

      // Set chart options
      var options = {
      
            width: 1000,
            sankey: {
              iterations: 32,
              node: { label: { 
                        fontName: 'Arial',
                        fontSize: 10,
                        color: '#871b47',
                        bold: false,
                        italic: true } } }, 
          };
          
          
          
      // Instantiate and draw our chart, passing in some options.
      var chart = new google.visualization.Sankey(document.getElementById('sankey_multiple'));
      chart.draw(data, options);
     }
     
     
  </script>
  </body>
  </html>|
  
  
  filename = DateTime.now.strftime("./StorageReport_%Y_%m_%d-%H_%M_%S.html")
  puts "Writing html file #{filename}"
  f = File.open(filename, 'w+')
  f.write(the_html)
  f.close
  
end