File: split.go

package info (click to toggle)
golang-github-go-openapi-swag 1%3A0.25.1-1
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid
  • size: 3,200 kB
  • sloc: sh: 30; makefile: 3
file content (312 lines) | stat: -rw-r--r-- 8,719 bytes parent folder | download
1
2
3
4
5
6
7
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
142
143
144
145
146
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
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
// Copyright 2015 go-swagger maintainers
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
//    http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.

package mangling

import (
	"unicode"
)

type splitterOption func(*splitter)

// withPostSplitInitialismCheck allows to catch initialisms after main split process
func withPostSplitInitialismCheck(s *splitter) {
	s.postSplitInitialismCheck = true
}

func withReplaceFunc(fn ReplaceFunc) func(*splitter) {
	return func(s *splitter) {
		s.replaceFunc = fn
	}
}

func withInitialismsCache(c *initialismsCache) splitterOption {
	return func(s *splitter) {
		s.initialismsCache = c
	}
}

type (
	initialismMatch struct {
		body       []rune
		start, end int
		complete   bool
		hasPlural  pluralForm
	}
	initialismMatches []initialismMatch
)

func (m initialismMatch) isZero() bool {
	return m.start == 0 && m.end == 0
}

type splitter struct {
	*initialismsCache

	postSplitInitialismCheck bool
	replaceFunc              ReplaceFunc
}

func newSplitter(options ...splitterOption) splitter {
	var s splitter

	for _, option := range options {
		option(&s)
	}

	if s.replaceFunc == nil {
		s.replaceFunc = defaultReplaceTable
	}

	return s
}

func (s splitter) split(name string) *[]nameLexem {
	nameRunes := []rune(name)
	matches := s.gatherInitialismMatches(nameRunes)
	if matches == nil {
		return poolOfLexems.BorrowLexems()
	}

	return s.mapMatchesToNameLexems(nameRunes, matches)
}

func (s splitter) gatherInitialismMatches(nameRunes []rune) *initialismMatches {
	var matches *initialismMatches

	for currentRunePosition, currentRune := range nameRunes {
		// recycle these allocations as we loop over runes
		// with such recycling, only 2 slices should be allocated per call
		// instead of o(n).
		newMatches := poolOfMatches.BorrowMatches()

		// check current initialism matches
		if matches != nil { // skip first iteration
			for _, match := range *matches {
				if keepCompleteMatch := match.complete; keepCompleteMatch {
					*newMatches = append(*newMatches, match)

					// the match is complete: keep it then move on to next rune
					continue
				}

				currentMatchRune := match.body[currentRunePosition-match.start]
				if currentMatchRune != currentRune {
					// failed match, move on to next rune
					continue
				}

				// try to complete ongoing match
				if currentRunePosition-match.start == len(match.body)-1 {
					// we are close; the next step is to check the symbol ahead
					// if it is a lowercase letter, then it is not the end of match
					// but the beginning of the next word.
					//
					// NOTE(fredbi): this heuristic sometimes leads to counterintuitive splits and
					// perhaps (not sure yet) we should check against case _alternance_.
					//
					// Example:
					//
					// In the current version, in the sentence "IDS initialism", "ID" is recognized as an initialism,
					// leading to a split like "id_s_initialism" (or IDSInitialism),
					// whereas in the sentence "IDx initialism", it is not and produces something like
					// "i_d_x_initialism" (or IDxInitialism). The generated file name is not great.
					//
					// Both go identifiers are tolerated by linters.
					//
					// Notice that the slightly different input "IDs initialism" is correctly detected
					// as a pluralized initialism and produces something like "ids_initialism" (or IDsInitialism).

					if currentRunePosition < len(nameRunes)-1 {
						nextRune := nameRunes[currentRunePosition+1]

						// recognize a plural form for this initialism (only simple pluralization is supported)
						if nextRune == 's' && match.hasPlural == simplePlural {
							// detected a pluralized initialism
							match.body = append(match.body, nextRune)
							currentRunePosition++
							if currentRunePosition < len(nameRunes)-1 {
								nextRune = nameRunes[currentRunePosition+1]
								if newWord := unicode.IsLower(nextRune); newWord {
									// it is the start of a new word.
									// Match is only partial and the initialism is not recognized : move on
									continue
								}
							}

							// this is a pluralized match: keep it
							match.complete = true
							match.hasPlural = simplePlural
							match.end = currentRunePosition
							*newMatches = append(*newMatches, match)

							// match is complete: keep it then move on to next rune
							continue
						}

						if newWord := unicode.IsLower(nextRune); newWord {
							// it is the start of a new word
							// Match is only partial and the initialism is not recognized : move on
							continue
						}
					}

					match.complete = true
					match.end = currentRunePosition
				}

				// append the ongoing matching attempt (not necessarily complete)
				*newMatches = append(*newMatches, match)
			}
		}

		// check for new initialism matches
		for i, r := range s.initialismsRunes {
			if r[0] == currentRune {
				*newMatches = append(*newMatches, initialismMatch{
					start:     currentRunePosition,
					body:      r,
					complete:  false,
					hasPlural: s.initialismsPluralForm[i],
				})
			}
		}

		if matches != nil {
			poolOfMatches.RedeemMatches(matches)
		}
		matches = newMatches
	}

	// up to the caller to redeem this last slice
	return matches
}

func (s splitter) mapMatchesToNameLexems(nameRunes []rune, matches *initialismMatches) *[]nameLexem {
	nameLexems := poolOfLexems.BorrowLexems()

	var lastAcceptedMatch initialismMatch
	for _, match := range *matches {
		if !match.complete {
			continue
		}

		if firstMatch := lastAcceptedMatch.isZero(); firstMatch {
			s.appendBrokenDownCasualString(nameLexems, nameRunes[:match.start])
			*nameLexems = append(*nameLexems, s.breakInitialism(string(match.body)))

			lastAcceptedMatch = match

			continue
		}

		if overlappedMatch := match.start <= lastAcceptedMatch.end; overlappedMatch {
			continue
		}

		middle := nameRunes[lastAcceptedMatch.end+1 : match.start]
		s.appendBrokenDownCasualString(nameLexems, middle)
		*nameLexems = append(*nameLexems, s.breakInitialism(string(match.body)))

		lastAcceptedMatch = match
	}

	// we have not found any accepted matches
	if lastAcceptedMatch.isZero() {
		*nameLexems = (*nameLexems)[:0]
		s.appendBrokenDownCasualString(nameLexems, nameRunes)
	} else if lastAcceptedMatch.end+1 != len(nameRunes) {
		rest := nameRunes[lastAcceptedMatch.end+1:]
		s.appendBrokenDownCasualString(nameLexems, rest)
	}

	poolOfMatches.RedeemMatches(matches)

	return nameLexems
}

func (s splitter) breakInitialism(original string) nameLexem {
	return newInitialismNameLexem(original, original)
}

func (s splitter) appendBrokenDownCasualString(segments *[]nameLexem, str []rune) {
	currentSegment := poolOfBuffers.BorrowBuffer(len(str)) // unlike strings.Builder, bytes.Buffer initial storage can reused
	defer func() {
		poolOfBuffers.RedeemBuffer(currentSegment)
	}()

	addCasualNameLexem := func(original string) {
		*segments = append(*segments, newCasualNameLexem(original))
	}

	addInitialismNameLexem := func(original, match string) {
		*segments = append(*segments, newInitialismNameLexem(original, match))
	}

	var addNameLexem func(string)
	if s.postSplitInitialismCheck {
		addNameLexem = func(original string) {
			for i := range s.initialisms {
				if isEqualFoldIgnoreSpace(s.initialismsUpperCased[i], original) {
					addInitialismNameLexem(original, s.initialisms[i])

					return
				}
			}

			addCasualNameLexem(original)
		}
	} else {
		addNameLexem = addCasualNameLexem
	}

	// NOTE: (performance). The few remaining non-amortized allocations
	// lay in the code below: using String() forces
	for _, rn := range str {
		if replace, found := s.replaceFunc(rn); found {
			if currentSegment.Len() > 0 {
				addNameLexem(currentSegment.String())
				currentSegment.Reset()
			}

			if replace != "" {
				addNameLexem(replace)
			}

			continue
		}

		if !unicode.In(rn, unicode.L, unicode.M, unicode.N, unicode.Pc) {
			if currentSegment.Len() > 0 {
				addNameLexem(currentSegment.String())
				currentSegment.Reset()
			}

			continue
		}

		if unicode.IsUpper(rn) {
			if currentSegment.Len() > 0 {
				addNameLexem(currentSegment.String())
			}
			currentSegment.Reset()
		}

		currentSegment.WriteRune(rn)
	}

	if currentSegment.Len() > 0 {
		addNameLexem(currentSegment.String())
	}
}