Class: GitIgnoreSpec

Inherits:
RegexSpec show all
Defined in:
lib/puppet/vendor/pathspec/lib/pathspec/gitignorespec.rb

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(pattern) ⇒ GitIgnoreSpec


8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
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
128
129
130
131
132
133
134
135
136
137
138
139
140
141
# File 'lib/puppet/vendor/pathspec/lib/pathspec/gitignorespec.rb', line 8

def initialize(pattern)
  pattern = pattern.strip unless pattern.nil?

  # A pattern starting with a hash ('#') serves as a comment
  # (neither includes nor excludes files). Escape the hash with a
  # back-slash to match a literal hash (i.e., '\#').
  if pattern.start_with?('#')
    @regex = nil
    @inclusive = nil

  # A blank pattern is a null-operation (neither includes nor
  # excludes files).
  elsif pattern.empty?
    @regex = nil
    @inclusive = nil

  # Patterns containing three or more consecutive stars are invalid and
  # will be ignored.
  elsif pattern =~ /\*\*\*+/
    @regex = nil
    @inclusive = nil

  # We have a valid pattern!
  else
    # A pattern starting with an exclamation mark ('!') negates the
    # pattern (exclude instead of include). Escape the exclamation
    # mark with a back-slash to match a literal exclamation mark
    # (i.e., '\!').
    if pattern.start_with?('!')
      @inclusive = false
      # Remove leading exclamation mark.
      pattern = pattern[1..-1]
    else
      @inclusive = true
    end

    # Remove leading back-slash escape for escaped hash ('#') or
    # exclamation mark ('!').
    if pattern.start_with?('\\')
      pattern = pattern[1..-1]
    end

    # Split pattern into segments. -1 to allow trailing slashes.
    pattern_segs = pattern.split('/', -1)

    # Normalize pattern to make processing easier.

    # A pattern beginning with a slash ('/') will only match paths
    # directly on the root directory instead of any descendant
    # paths. So, remove empty first segment to make pattern relative
    # to root.
    if pattern_segs[0].empty?
      pattern_segs.shift
    else
      # A pattern without a beginning slash ('/') will match any
      # descendant path. This is equivalent to "**/{pattern}". So,
      # prepend with double-asterisks to make pattern relative to
      # root.
      if pattern_segs.length == 1 && pattern_segs[0] != '**'
        pattern_segs.insert(0, '**')
      end
    end

    # A pattern ending with a slash ('/') will match all descendant
    # paths of if it is a directory but not if it is a regular file.
    # This is equivalent to "{pattern}/**". So, set last segment to
    # double asterisks to include all descendants.
    if pattern_segs[-1].empty?
      pattern_segs[-1] = '**'
    end

    # Handle platforms with backslash separated paths
    if File::SEPARATOR == '\\'
      path_sep = '\\\\'
    else
      path_sep = '/'
    end


    # Build regular expression from pattern.
    regex = '^'
    need_slash = false
    regex_end = pattern_segs.size - 1
    pattern_segs.each_index do |i|
      seg = pattern_segs[i]

      if seg == '**'
        # A pattern consisting solely of double-asterisks ('**')
        # will match every path.
        if i == 0 && i == regex_end
          regex.concat('.+')

        # A normalized pattern beginning with double-asterisks
        # ('**') will match any leading path segments.
        elsif i == 0
          regex.concat("(?:.+#{path_sep})?")
          need_slash = false

        # A normalized pattern ending with double-asterisks ('**')
        # will match any trailing path segments.
        elsif i == regex_end
          regex.concat("#{path_sep}.*")

        # A pattern with inner double-asterisks ('**') will match
        # multiple (or zero) inner path segments.
        else
          regex.concat("(?:#{path_sep}.+)?")
          need_slash = true
        end

      # Match single path segment.
      elsif seg == '*'
        if need_slash
          regex.concat(path_sep)
        end

        regex.concat("[^#{path_sep}]+")
        need_slash = true

      else
        # Match segment glob pattern.
        if need_slash
          regex.concat(path_sep)
        end

        regex.concat(translate_segment_glob(seg))
        need_slash = true
      end
    end

    regex.concat('$')
    super(regex)
  end
end

Instance Attribute Details

#regexObject (readonly)

Returns the value of attribute regex


6
7
8
# File 'lib/puppet/vendor/pathspec/lib/pathspec/gitignorespec.rb', line 6

def regex
  @regex
end

Instance Method Details

#inclusive?Boolean


272
273
274
# File 'lib/puppet/vendor/pathspec/lib/pathspec/gitignorespec.rb', line 272

def inclusive?
  @inclusive
end

#match(path) ⇒ Object


143
144
145
# File 'lib/puppet/vendor/pathspec/lib/pathspec/gitignorespec.rb', line 143

def match(path)
  super(path)
end

#translate_segment_glob(pattern) ⇒ Object


147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
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
202
203
204
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
# File 'lib/puppet/vendor/pathspec/lib/pathspec/gitignorespec.rb', line 147

def translate_segment_glob(pattern)
  """
  Translates the glob pattern to a regular expression. This is used in
  the constructor to translate a path segment glob pattern to its
  corresponding regular expression.

  *pattern* (``str``) is the glob pattern.

  Returns the regular expression (``str``).
  """
  # NOTE: This is derived from `fnmatch.translate()` and is similar to
  # the POSIX function `fnmatch()` with the `FNM_PATHNAME` flag set.

  escape = false
  regex = ''
  i = 0

  while i < pattern.size
    # Get next character.
    char = pattern[i].chr
    i += 1

    # Escape the character.
    if escape
      escape = false
      regex += Regexp.escape(char)

    # Escape character, escape next character.
    elsif char == '\\'
      escape = true

    # Multi-character wildcard. Match any string (except slashes),
    # including an empty string.
    elsif char == '*'
      regex += '[^/]*'

    # Single-character wildcard. Match any single character (except
    # a slash).
    elsif char == '?'
      regex += '[^/]'

    # Bracket expression wildcard. Except for the beginning
    # exclamation mark, the whole bracket expression can be used
    # directly as regex but we have to find where the expression
    # ends.
    # - "[][!]" matches ']', '[' and '!'.
    # - "[]-]" matches ']' and '-'.
    # - "[!]a-]" matches any character except ']', 'a' and '-'.
    elsif char == '['
      j = i
      # Pass brack expression negation.
      if j < pattern.size && pattern[j].chr == '!'
        j += 1
      end

      # Pass first closing bracket if it is at the beginning of the
      # expression.
      if j < pattern.size && pattern[j].chr == ']'
        j += 1
      end

      # Find closing bracket. Stop once we reach the end or find it.
      while j < pattern.size && pattern[j].chr != ']'
        j += 1
      end


      if j < pattern.size
        expr = '['

        # Bracket expression needs to be negated.
        if pattern[i].chr == '!'
          expr += '^'
          i += 1

        # POSIX declares that the regex bracket expression negation
        # "[^...]" is undefined in a glob pattern. Python's
        # `fnmatch.translate()` escapes the caret ('^') as a
        # literal. To maintain consistency with undefined behavior,
        # I am escaping the '^' as well.
        elsif pattern[i].chr == '^'
          expr += '\\^'
          i += 1
        end

        # Escape brackets contained within pattern
        if pattern[i].chr == ']' && i != j
          expr += '\]'
          i += 1
        end


        # Build regex bracket expression. Escape slashes so they are
        # treated as literal slashes by regex as defined by POSIX.
        expr += pattern[i..j].sub('\\', '\\\\')

        # Add regex bracket expression to regex result.
        regex += expr

        # Found end of bracket expression. Increment j to be one past
        # the closing bracket:
        #
        #  [...]
        #   ^   ^
        #   i   j
        #
        j += 1
        # Set i to one past the closing bracket.
        i = j

      # Failed to find closing bracket, treat opening bracket as a
      # bracket literal instead of as an expression.
      else
        regex += '\['
      end

    # Regular character, escape it for regex.
    else
      regex << Regexp.escape(char)
    end
  end

  regex
end