Class: Flay

Inherits:
Object
  • Object
show all
Defined in:
lib/flay.rb,
lib/flay_erb.rb

Defined Under Namespace

Classes: Erubi, Item, Location

Constant Summary collapse

VERSION =

:nodoc:

"2.13.2"
DEFAULT_IGNORE =

so I can move this to flog wholesale

".flayignore"
MAX_NODE_SIZE =

:stopdoc:

10
MAX_AVG_MASS =

prevents exponential blowout

12

Instance Attribute Summary collapse

Class Method Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(option = {}) ⇒ Flay

Create a new instance of Flay with options.



165
166
167
168
169
170
171
172
173
 
# File 'lib/flay.rb', line 165

def initialize option = {}
  @option = Flay.default_options.merge option
  @hashes = Hash.new { |h,k| h[k] = [] }

  self.identical      = {}
  self.masses         = {}
  self.total          = 0
  self.mass_threshold = @option[:mass]
end

Instance Attribute Details

#hashes ⇒ Object (readonly)

Returns the value of attribute hashes.



159
160
161
 
# File 'lib/flay.rb', line 159

def hashes
  @hashes
end

#identical ⇒ Object

:stopdoc:



158
159
160
 
# File 'lib/flay.rb', line 158

def identical
  @identical
end

#mass_threshold ⇒ Object

:stopdoc:



158
159
160
 
# File 'lib/flay.rb', line 158

def mass_threshold
  @mass_threshold
end

#masses ⇒ Object

:stopdoc:



158
159
160
 
# File 'lib/flay.rb', line 158

def masses
  @masses
end

#option ⇒ Object (readonly)

Returns the value of attribute option.



159
160
161
 
# File 'lib/flay.rb', line 159

def option
  @option
end

#total ⇒ Object

:stopdoc:



158
159
160
 
# File 'lib/flay.rb', line 158

def total
  @total
end

Class Method Details

.default_options ⇒ Object

Returns the default options.



36
37
38
39
40
41
42
43
44
45
46
47
48
49
 
# File 'lib/flay.rb', line 36

def self.default_options
  {
    :diff    => false,
    :mass    => 16,
    :summary => false,
    :verbose => false,
    :number  => true,
    :timeout => 10,
    :liberal => false,
    :fuzzy   => false,
    :only    => nil,
    :filters => [],
  }
end

.load_plugins ⇒ Object

Loads all flay plugins. Files must be named “flay_*.rb”.



135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
 
# File 'lib/flay.rb', line 135

def self.load_plugins
  unless defined? @@plugins then
    @@plugins = []

    plugins = Gem.find_files("flay_*.rb").reject { |p| p =~ /flay_task/ }

    plugins.each do |plugin|
      plugin_name = File.basename(plugin, ".rb").sub(/^flay_/, "")
      next if @@plugins.include? plugin_name
      begin
        load plugin
        @@plugins << plugin_name
      rescue LoadError => e
        warn "error loading #{plugin.inspect}: #{e.message}. skipping..."
      end
    end
  end
  @@plugins
rescue => e
  warn "Error loading plugins: #{e}" if option[:verbose]
end

.parse_options(args = ARGV) ⇒ Object

Process options in args, defaulting to ARGV.



54
55
56
57
58
59
60
61
62
63
64
65
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
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
 
# File 'lib/flay.rb', line 54

def self.parse_options args = ARGV
  options = self.default_options

  OptionParser.new do |opts|
    opts.banner  = "flay [options] files_or_dirs"
    opts.version = Flay::VERSION

    opts.separator ""
    opts.separator "Specific options:"
    opts.separator ""

    opts.on("-h", "--help", "Display this help.") do
      puts opts
      exit
    end

    opts.on("-f", "--fuzzy [DIFF]", Integer,
            "Detect fuzzy (copy & paste) duplication (default 1).") do |n|
      options[:fuzzy] = n || 1
    end

    opts.on("-l", "--liberal", "Use a more liberal detection method.") do
      options[:liberal] = true
    end

    opts.on("-m", "--mass MASS", Integer,
            "Sets mass threshold (default = #{options[:mass]})") do |m|
      options[:mass] = m.to_i
    end

    opts.on("-#", "Don't number output (helps with diffs)") do |m|
      options[:number] = false
    end

    opts.on("-v", "--verbose", "Verbose. Show progress processing files.") do
      options[:verbose] = true
    end

    opts.on("-o", "--only NODE", String, "Only show matches on NODE type.") do |s|
      options[:only] = s.to_sym
    end

    opts.on("-d", "--diff", "Diff Mode. Display N-Way diff for ruby.") do
      options[:diff] = true
    end

    opts.on("-s", "--summary", "Summarize. Show flay score per file only.") do
      options[:summary] = true
    end

    opts.on("-t", "--timeout TIME", Integer,
            "Set the timeout. (default = #{options[:timeout]})") do |t|
      options[:timeout] = t.to_i
    end

    extensions = ["rb"] + Flay.load_plugins

    opts.separator ""
    opts.separator "Known extensions: #{extensions.join(", ")}"

    extensions.each do |meth|
      msg = "options_#{meth}"
      send msg, opts, options if self.respond_to?(msg)
    end

    begin
      opts.parse! args
    rescue => e
      abort "#{e}\n\n#{opts}"
    end
  end

  options
end

.run(args = ARGV) ⇒ Object 



21
22
23
24
25
26
27
28
29
30
31
 
# File 'lib/flay.rb', line 21

def self.run args = ARGV
  extensions = ["rb"] + Flay.load_plugins
  glob = "**/*.{#{extensions.join ","}}"

  expander = PathExpander.new args, glob
  files = expander.filter_files expander.process, DEFAULT_IGNORE

  flay = Flay.new Flay.parse_options args
  flay.process(*files)
  flay
end

Instance Method Details

#analyze(filter = nil) ⇒ Object

Prune, find identical nodes, and update masses.



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
 
# File 'lib/flay.rb', line 212

def analyze filter = nil
  self.prune

  self.hashes.each do |hash,nodes|
    identical[hash] = nodes[1..-1].all? { |n| n == nodes.first }
  end

  update_masses

  sorted = masses.sort_by { |h,m|
    exp = hashes[h].first
    [-m,
     exp.file,
     exp.line,
     exp.sexp_type.to_s]
  }

  sorted.map { |hash, mass|
    nodes = hashes[hash]

    next unless nodes.first.first == filter if filter

    same  = identical[hash]
    node  = nodes.first
    n     = nodes.size
    bonus = "*#{n}" if same

    locs = nodes.sort_by { |x| [x.file, x.line] }.each_with_index.map { |x, i|
      extra = :fuzzy if x.modified?
      Location[x.file, x.line, extra]
    }

    Item[hash, node.sexp_type, bonus, mass, locs]
  }.compact
end

#collapse_and_label(ary) ⇒ Object

:nodoc:



447
448
449
450
451
452
453
454
455
 
# File 'lib/flay.rb', line 447

def collapse_and_label ary # :nodoc:
  ary[0].zip(*ary[1..-1]).map { |lines|
    if lines.uniq.size == 1 then
      "   #{lines.first}"
    else
      lines.reject { |l| l.empty? }.map { |l| "#{l.group}: #{l}" }
    end
  }
end

#filter_sexp(exp) ⇒ Object

Before processing, filter any sexp’s that match against filters specified in option[:filters]. This changes the sexp itself.



279
280
281
282
283
284
285
286
 
# File 'lib/flay.rb', line 279

def filter_sexp exp
  exp.delete_if { |sexp|
    if Sexp === sexp then
      del = option[:filters].any? { |pattern| pattern.satisfy? sexp }
      del or (filter_sexp(sexp); false)
    end
  }
end

#n_way_diff(*data) ⇒ Object

Output an n-way diff from data. This is only used if –diff is given.



411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
 
# File 'lib/flay.rb', line 411

def n_way_diff *data
  comments = []
  codes    = []

  split_and_group(data).each do |subdata|
    n = subdata.find_index { |s| s !~ /^#/ }

    comment, code = subdata[0..n-1], subdata[n..-1]
    comment = [] if n == 0

    comments << comment
    codes    << code
  end

  comments = collapse_and_label pad_with_empty_strings comments
  codes    = collapse_and_label pad_with_empty_strings codes

  (comments + codes).flatten.join("\n")
end

#pad_with_empty_strings(ary) ⇒ Object

:nodoc:



441
442
443
444
445
 
# File 'lib/flay.rb', line 441

def pad_with_empty_strings ary # :nodoc:
  max = ary.map { |s| s.size }.max

  ary.map { |a| a + ([""] * (max - a.size)) }
end

#process(*files) ⇒ Object

Process any number of files.



178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
 
# File 'lib/flay.rb', line 178

def process(*files) # TODO: rename from process - should act as SexpProcessor
  files.each do |file|
    warn "Processing #{file}" if option[:verbose]

    ext = File.extname(file).sub(/^\./, "")
    ext = "rb" if ext.nil? || ext.empty?
    msg = "process_#{ext}"

    unless respond_to? msg then
      warn "  Unknown file type: #{ext}, defaulting to ruby"
      msg = "process_rb"
    end

    begin
      sexp = begin
               send msg, file
             rescue => e
               warn "  #{e.message.strip}"
               warn "  skipping #{file}"
               nil
             end

      next unless sexp

      process_sexp sexp
    rescue SyntaxError => e
      warn "  skipping #{file}: #{e.message}"
    end
  end
end

#process_erb(file) ⇒ Object

Process erb and parse the result. Returns the sexp of the parsed ruby.



10
11
12
13
14
15
16
17
18
19
20
21
 
# File 'lib/flay_erb.rb', line 10

def process_erb file
  erb = File.read file

  ruby = Erubi.new(erb).src

  begin
    RubyParser.new.process(ruby, file)
  rescue => e
    warn ruby if option[:verbose]
    raise e
  end
end

#process_fuzzy(node, difference) ⇒ Object

Process “fuzzy” matches for node. A fuzzy match is a subset of node up to difference elements less than the original.



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
 
# File 'lib/flay.rb', line 311

def process_fuzzy node, difference
  return unless node.has_code?

  avg_mass = node.mass / node.size
  return if node.size > MAX_NODE_SIZE or avg_mass > MAX_AVG_MASS

  tmpl, code = node.split_code
  tmpl.modified = true

  (code.size - 1).downto(code.size - difference) do |n|
    code.combination(n).each do |subcode|
      new_node = tmpl + subcode

      next unless new_node.any? { |sub| Sexp === sub }
      next if new_node.mass < self.mass_threshold

      # they're already structurally similar, don"t bother adding another
      next if self.hashes[new_node.structural_hash].any? { |sub|
        sub.file == new_node.file and sub.line == new_node.line
      }

      self.hashes[new_node.structural_hash] << new_node
    end
  end
end

#process_rb(file) ⇒ Object

Parse a ruby file and return the sexp.

– TODO: change the system and rename this to parse_rb.



267
268
269
270
271
272
273
 
# File 'lib/flay.rb', line 267

def process_rb file
  begin
    RubyParser.new.process(File.binread(file), file, option[:timeout])
  rescue Timeout::Error
    warn "TIMEOUT parsing #{file}. Skipping."
  end
end

#process_sexp(pt) ⇒ Object

Process a sexp pt.



291
292
293
294
295
296
297
298
299
300
 
# File 'lib/flay.rb', line 291

def process_sexp pt
  filter_sexp(pt).deep_each do |node|
    next :skip if node.none? { |sub| Sexp === sub }
    next :skip if node.mass < self.mass_threshold

    self.hashes[node.structural_hash] << node

    process_fuzzy node, option[:fuzzy] if option[:fuzzy]
  end
end

#prune ⇒ Object

Prunes nodes that aren’t relevant to analysis or are already covered by another node. Also deletes nodes based on the :filters option.



342
343
344
345
346
347
348
349
350
351
352
 
# File 'lib/flay.rb', line 342

def prune
  # prune trees that aren't duped at all, or are too small
  self.hashes.delete_if { |_,nodes| nodes.size == 1 }
  self.hashes.delete_if { |_,nodes| nodes.all?(&:modified?) }

  if option[:liberal] then
    prune_liberally
  else
    prune_conservatively
  end
end

#prune_conservatively ⇒ Object

Conservative prune. Remove any bucket that is known to contain a subnode element of a node in another bucket.



358
359
360
361
362
363
364
365
366
367
368
369
370
 
# File 'lib/flay.rb', line 358

def prune_conservatively
  hashes_to_prune = {}

  # extract all subtree hashes from all nodes
  self.hashes.values.each do |nodes|
    nodes.first.all_structural_subhashes.each do |h|
      hashes_to_prune[h] = true
    end
  end

  # nuke subtrees so we show the biggest matching tree possible
  self.hashes.delete_if { |h,_| hashes_to_prune[h] }
end

#prune_liberally ⇒ Object

Liberal prune. Remove any element from a bucket that is known to be a subnode of another node. Removed by identity.



376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
 
# File 'lib/flay.rb', line 376

def prune_liberally
  update_masses

  hashes_to_prune = Hash.new { |h,k| h[k] = [] }

  # record each subtree by subhash, but skip if subtree mass > parent mass
  self.hashes.values.each do |nodes|
    nodes.each do |node|
      tophash  = node.structural_hash
      topscore = self.masses[tophash]

      node.deep_each do |subnode|
        subhash  = subnode.structural_hash
        subscore = self.masses[subhash]

        next if subscore and subscore > topscore

        hashes_to_prune[subhash] << subnode
      end
    end
  end

  # nuke only individual items by object identity
  self.hashes.each do |h,v|
    v.delete_eql hashes_to_prune[h]
  end

  # nuke buckets we happened to fully empty
  self.hashes.delete_if { |k,v| v.size <= 1 }
end

#report(io = $stdout) ⇒ Object

Output the report. Duh.



477
478
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
516
517
518
519
520
521
522
 
# File 'lib/flay.rb', line 477

def report io = $stdout
  only = option[:only]

  data = analyze only

  io.puts "Total score (lower is better) = #{self.total}"

  if option[:summary] then
    io.puts

    self.summary.sort_by { |_,v| -v }.each do |file, score|
      io.puts "%8.2f: %s" % [score, file]
    end

    return
  end

  data.each_with_index do |item, count|
    prefix = "%d) " % (count + 1) if option[:number]

    match = item.identical? ? "IDENTICAL" : "Similar"

    io.puts
    io.puts "%s%s code found in %p (mass%s = %d)" %
      [prefix, match, item.name, item.bonus, item.mass]

    item.locations.each_with_index do |loc, i|
      loc_prefix = "%s: " % (?A.ord + i).chr if option[:diff]
      extra = " (FUZZY)" if loc.fuzzy?
      io.puts "  %s%s:%d%s" % [loc_prefix, loc.file, loc.line, extra]
    end

    if option[:diff] then
      io.puts

      nodes = hashes[item.structural_hash]

      sources = nodes.map do |s|
        msg = "sexp_to_#{File.extname(s.file).sub(/./, "")}"
        self.respond_to?(msg) ? self.send(msg, s) : sexp_to_rb(s)
      end

      io.puts n_way_diff(*sources)
    end
  end
end

#sexp_to_rb(sexp) ⇒ Object 



524
525
526
527
528
529
530
531
532
 
# File 'lib/flay.rb', line 524

def sexp_to_rb sexp
  begin
    require "ruby2ruby"
  rescue LoadError
    return "ruby2ruby is required for diff"
  end
  @r2r ||= Ruby2Ruby.new
  @r2r.process sexp.deep_clone
end

#split_and_group(ary) ⇒ Object

:nodoc:



431
432
433
434
435
436
437
438
439
 
# File 'lib/flay.rb', line 431

def split_and_group ary # :nodoc:
  ary.each_with_index.map { |s, i|
    c = (?A.ord + i).chr
    s.scan(/^.*/).map { |s2|
      s2.group = c
      s2
    }
  }
end

#summary ⇒ Object

Calculate summary scores on a per-file basis. For –summary.



460
461
462
463
464
465
466
467
468
469
470
471
472
 
# File 'lib/flay.rb', line 460

def summary
  score = Hash.new 0

  masses.each do |hash, mass|
    sexps = hashes[hash]
    mass_per_file = mass.to_f / sexps.size
    sexps.each do |sexp|
      score[sexp.file] += mass_per_file
    end
  end

  score
end

#update_masses ⇒ Object

Reset total and recalculate the masses for all nodes in hashes.



251
252
253
254
255
256
257
258
259
 
# File 'lib/flay.rb', line 251

def update_masses
  self.total = 0
  masses.clear
  self.hashes.each do |hash, nodes|
    masses[hash] = nodes.first.mass * nodes.size
    masses[hash] *= (nodes.size) if identical[hash]
    self.total += masses[hash]
  end
end