aboutsummaryrefslogtreecommitdiff
path: root/spec/spec.go
blob: 3d462692924bcf456e06a02a7baa2bea477583ae (plain) (blame)
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
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
package spec

import (
	"fmt"
	"regexp"
	"sort"
	"strconv"
	"strings"
)

// LexKindID represents an ID of a lexical kind and is unique across all modes.
type LexKindID int

const (
	LexKindIDNil = LexKindID(0)
	LexKindIDMin = LexKindID(1)
)

func (id LexKindID) Int() int {
	return int(id)
}

// LexModeKindID represents an ID of a lexical kind and is unique within a mode.
// Use LexKindID to identify a kind across all modes uniquely.
type LexModeKindID int

const (
	LexModeKindIDNil = LexModeKindID(0)
	LexModeKindIDMin = LexModeKindID(1)
)

func (id LexModeKindID) Int() int {
	return int(id)
}

// LexKindName represents a name of a lexical kind.
type LexKindName string

const LexKindNameNil = LexKindName("")

func (k LexKindName) String() string {
	return string(k)
}

func (k LexKindName) validate() error {
	err := validateIdentifier(k.String())
	if err != nil {
		return fmt.Errorf("invalid kind name: %v", err)
	}
	return nil
}

// LexPattern represents a pattern of a lexeme.
// The pattern is written in regular expression.
type LexPattern string

func (p LexPattern) validate() error {
	if p == "" {
		return fmt.Errorf("pattern doesn't allow to be the empty string")
	}
	return nil
}

// LexModeID represents an ID of a lex mode.
type LexModeID int

const (
	LexModeIDNil     = LexModeID(0)
	LexModeIDDefault = LexModeID(1)
)

func (n LexModeID) String() string {
	return strconv.Itoa(int(n))
}

func (n LexModeID) Int() int {
	return int(n)
}

func (n LexModeID) IsNil() bool {
	return n == LexModeIDNil
}

// LexModeName represents a name of a lex mode.
type LexModeName string

const (
	LexModeNameNil     = LexModeName("")
	LexModeNameDefault = LexModeName("default")
)

func (m LexModeName) String() string {
	return string(m)
}

func (m LexModeName) validate() error {
	err := validateIdentifier(m.String())
	if err != nil {
		return fmt.Errorf("invalid mode name: %v", err)
	}
	return nil
}

const idPattern = `^[a-z](_?[0-9a-z]+)*$`

var idRE = regexp.MustCompile(idPattern)

func validateIdentifier(id string) error {
	if id == "" {
		return fmt.Errorf("identifier doesn't allow to be the empty string")
	}
	if !idRE.MatchString(id) {
		return fmt.Errorf("identifier must be %v", idPattern)
	}
	return nil
}

func SnakeCaseToUpperCamelCase(snake string) string {
	elems := strings.Split(snake, "_")
	for i, e := range elems {
		if len(e) == 0 {
			continue
		}
		elems[i] = strings.ToUpper(string(e[0])) + e[1:]
	}

	return strings.Join(elems, "")
}

type LexEntry struct {
	Kind     LexKindName   `json:"kind"`
	Pattern  LexPattern    `json:"pattern"`
	Modes    []LexModeName `json:"modes"`
	Push     LexModeName   `json:"push"`
	Pop      bool          `json:"pop"`
	Fragment bool          `json:"fragment"`
}

func (e *LexEntry) validate() error {
	err := e.Kind.validate()
	if err != nil {
		return err
	}
	err = e.Pattern.validate()
	if err != nil {
		return err
	}
	if len(e.Modes) > 0 {
		for _, mode := range e.Modes {
			err = mode.validate()
			if err != nil {
				return err
			}
		}
	}
	return nil
}

type LexSpec struct {
	Name    string      `json:"name"`
	Entries []*LexEntry `json:"entries"`
}

func (s *LexSpec) Validate() error {
	err := validateIdentifier(s.Name)
	if err != nil {
		return fmt.Errorf("invalid specification name: %v", err)
	}

	if len(s.Entries) <= 0 {
		return fmt.Errorf("the lexical specification must have at least one entry")
	}
	{
		var errs []error
		for i, e := range s.Entries {
			err := e.validate()
			if err != nil {
				errs = append(errs, fmt.Errorf("entry #%v: %w", i+1, err))
			}
		}
		if len(errs) > 0 {
			var b strings.Builder
			fmt.Fprintf(&b, "%v", errs[0])
			for _, err := range errs[1:] {
				fmt.Fprintf(&b, "\n%v", err)
			}
			return fmt.Errorf(b.String())
		}
	}
	{
		ks := map[string]struct{}{}
		fks := map[string]struct{}{}
		for _, e := range s.Entries {
			// Allow duplicate names between fragments and non-fragments.
			if e.Fragment {
				if _, exist := fks[e.Kind.String()]; exist {
					return fmt.Errorf("kinds `%v` are duplicates", e.Kind)
				}
				fks[e.Kind.String()] = struct{}{}
			} else {
				if _, exist := ks[e.Kind.String()]; exist {
					return fmt.Errorf("kinds `%v` are duplicates", e.Kind)
				}
				ks[e.Kind.String()] = struct{}{}
			}
		}
	}
	{
		kinds := []string{}
		modes := []string{
			LexModeNameDefault.String(), // This is a predefined mode.
		}
		for _, e := range s.Entries {
			if e.Fragment {
				continue
			}

			kinds = append(kinds, e.Kind.String())

			for _, m := range e.Modes {
				modes = append(modes, m.String())
			}
		}

		kindErrs := findSpellingInconsistenciesErrors(kinds, nil)
		modeErrs := findSpellingInconsistenciesErrors(modes, func(ids []string) error {
			if SnakeCaseToUpperCamelCase(ids[0]) == SnakeCaseToUpperCamelCase(LexModeNameDefault.String()) {
				var b strings.Builder
				fmt.Fprintf(&b, "%+v", ids[0])
				for _, id := range ids[1:] {
					fmt.Fprintf(&b, ", %+v", id)
				}
				return fmt.Errorf("these identifiers are treated as the same. please use the same spelling as predefined '%v': %v", LexModeNameDefault, b.String())
			}
			return nil
		})
		errs := append(kindErrs, modeErrs...)
		if len(errs) > 0 {
			var b strings.Builder
			fmt.Fprintf(&b, "%v", errs[0])
			for _, err := range errs[1:] {
				fmt.Fprintf(&b, "\n%v", err)
			}
			return fmt.Errorf(b.String())
		}
	}

	return nil
}

func findSpellingInconsistenciesErrors(ids []string, hook func(ids []string) error) []error {
	duplicated := FindSpellingInconsistencies(ids)
	if len(duplicated) == 0 {
		return nil
	}

	var errs []error
	for _, dup := range duplicated {
		err := hook(dup)
		if err != nil {
			errs = append(errs, err)
			continue
		}

		var b strings.Builder
		fmt.Fprintf(&b, "%+v", dup[0])
		for _, id := range dup[1:] {
			fmt.Fprintf(&b, ", %+v", id)
		}
		err = fmt.Errorf("these identifiers are treated as the same. please use the same spelling: %v", b.String())
		errs = append(errs, err)
	}

	return errs
}

// FindSpellingInconsistencies finds spelling inconsistencies in identifiers. The identifiers are considered to be the same
// if they are spelled the same when expressed in UpperCamelCase. For example, `left_paren` and `LeftParen` are spelled the same
// in UpperCamelCase. Thus they are considere to be spelling inconsistency.
func FindSpellingInconsistencies(ids []string) [][]string {
	m := map[string][]string{}
	for _, id := range removeDuplicates(ids) {
		c := SnakeCaseToUpperCamelCase(id)
		m[c] = append(m[c], id)
	}

	var duplicated [][]string
	for _, camels := range m {
		if len(camels) == 1 {
			continue
		}
		duplicated = append(duplicated, camels)
	}

	for _, dup := range duplicated {
		sort.Slice(dup, func(i, j int) bool {
			return dup[i] < dup[j]
		})
	}
	sort.Slice(duplicated, func(i, j int) bool {
		return duplicated[i][0] < duplicated[j][0]
	})

	return duplicated
}

func removeDuplicates(s []string) []string {
	m := map[string]struct{}{}
	for _, v := range s {
		m[v] = struct{}{}
	}

	var unique []string
	for v := range m {
		unique = append(unique, v)
	}

	return unique
}

// StateID represents an ID of a state of a transition table.
type StateID int

const (
	// StateIDNil represents an empty entry of a transition table.
	// When the driver reads this value, it raises an error meaning lexical analysis failed.
	StateIDNil = StateID(0)

	// StateIDMin is the minimum value of the state ID. All valid state IDs are represented as
	// sequential numbers starting from this value.
	StateIDMin = StateID(1)
)

func (id StateID) Int() int {
	return int(id)
}

type RowDisplacementTable struct {
	OriginalRowCount int       `json:"original_row_count"`
	OriginalColCount int       `json:"original_col_count"`
	EmptyValue       StateID   `json:"empty_value"`
	Entries          []StateID `json:"entries"`
	Bounds           []int     `json:"bounds"`
	RowDisplacement  []int     `json:"row_displacement"`
}

type UniqueEntriesTable struct {
	UniqueEntries             *RowDisplacementTable `json:"unique_entries,omitempty"`
	UncompressedUniqueEntries []StateID             `json:"uncompressed_unique_entries,omitempty"`
	RowNums                   []int                 `json:"row_nums"`
	OriginalRowCount          int                   `json:"original_row_count"`
	OriginalColCount          int                   `json:"original_col_count"`
	EmptyValue                int                   `json:"empty_value"`
}

type TransitionTable struct {
	InitialStateID         StateID             `json:"initial_state_id"`
	AcceptingStates        []LexModeKindID     `json:"accepting_states"`
	RowCount               int                 `json:"row_count"`
	ColCount               int                 `json:"col_count"`
	Transition             *UniqueEntriesTable `json:"transition,omitempty"`
	UncompressedTransition []StateID           `json:"uncompressed_transition,omitempty"`
}

type CompiledLexModeSpec struct {
	KindNames []LexKindName    `json:"kind_names"`
	Push      []LexModeID      `json:"push"`
	Pop       []int            `json:"pop"`
	DFA       *TransitionTable `json:"dfa"`
}

type CompiledLexSpec struct {
	Name             string                 `json:"name"`
	InitialModeID    LexModeID              `json:"initial_mode_id"`
	ModeNames        []LexModeName          `json:"mode_names"`
	KindNames        []LexKindName          `json:"kind_names"`
	KindIDs          [][]LexKindID          `json:"kind_ids"`
	CompressionLevel int                    `json:"compression_level"`
	Specs            []*CompiledLexModeSpec `json:"specs"`
}