summary refs log tree commit diff stats
path: root/rod/parsecfg.nim
blob: 0b9574a41fd1b2ec17f1fc330d352a8a9f9e5610 (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
#
#
#            Nimrod's Runtime Library
#        (c) Copyright 2009 Andreas Rumpf
#
#    See the file "copying.txt", included in this
#    distribution, for details about the copyright.
#

# A HIGH-PERFORMANCE configuration file parser;
# the Nimrod version of this file is part of the
# standard library.

import 
  llstream, nhashes, strutils, lexbase

type 
  TCfgEventKind* = enum 
    cfgEof,                   # end of file reached
    cfgSectionStart,          # a ``[section]`` has been parsed
    cfgKeyValuePair,          # a ``key=value`` pair has been detected
    cfgOption,                # a ``--key=value`` command line option
    cfgError # an error ocurred during parsing; msg contains the
             # error message
  TCfgEvent* = object of TObject
    case kind*: TCfgEventKind
    of cfgEof: 
        nil

    of cfgSectionStart: 
        section*: string

    of cfgKeyValuePair, cfgOption: 
        key*, value*: string

    of cfgError: 
        msg*: string

  
  TTokKind* = enum 
    tkInvalid, tkEof,         # order is important here!
    tkSymbol, tkEquals, tkColon, tkBracketLe, tkBracketRi, tkDashDash
  TToken*{.final.} = object   # a token
    kind*: TTokKind           # the type of the token
    literal*: string          # the parsed (string) literal
  
  TParserState* = enum 
    startState, commaState
  TCfgParser* = object of TBaseLexer
    tok*: TToken
    state*: TParserState
    filename*: string


proc Open*(c: var TCfgParser, filename: string, inputStream: PLLStream)
proc Close*(c: var TCfgParser)
proc next*(c: var TCfgParser): TCfgEvent
proc getColumn*(c: TCfgParser): int
proc getLine*(c: TCfgParser): int
proc getFilename*(c: TCfgParser): string
proc errorStr*(c: TCfgParser, msg: string): string
# implementation

const 
  SymChars: TCharSet = {'a'..'z', 'A'..'Z', '0'..'9', '_', '\x80'..'\xFF'} # 
                                                                           # ----------------------------------------------------------------------------

proc rawGetTok(c: var TCfgParser, tok: var TToken)
proc open(c: var TCfgParser, filename: string, inputStream: PLLStream) = 
  openBaseLexer(c, inputStream)
  c.filename = filename
  c.state = startState
  c.tok.kind = tkInvalid
  c.tok.literal = ""
  rawGetTok(c, c.tok)

proc close(c: var TCfgParser) = 
  closeBaseLexer(c)

proc getColumn(c: TCfgParser): int = 
  result = getColNumber(c, c.bufPos)

proc getLine(c: TCfgParser): int = 
  result = c.linenumber

proc getFilename(c: TCfgParser): string = 
  result = c.filename

proc handleHexChar(c: var TCfgParser, xi: var int) = 
  case c.buf[c.bufpos]
  of '0'..'9': 
    xi = (xi shl 4) or (ord(c.buf[c.bufpos]) - ord('0'))
    inc(c.bufpos)
  of 'a'..'f': 
    xi = (xi shl 4) or (ord(c.buf[c.bufpos]) - ord('a') + 10)
    inc(c.bufpos)
  of 'A'..'F': 
    xi = (xi shl 4) or (ord(c.buf[c.bufpos]) - ord('A') + 10)
    inc(c.bufpos)
  else: 
    nil

proc handleDecChars(c: var TCfgParser, xi: var int) = 
  while c.buf[c.bufpos] in {'0'..'9'}: 
    xi = (xi * 10) + (ord(c.buf[c.bufpos]) - ord('0'))
    inc(c.bufpos)

proc getEscapedChar(c: var TCfgParser, tok: var TToken) = 
  var xi: int
  inc(c.bufpos)               # skip '\'
  case c.buf[c.bufpos]
  of 'n', 'N': 
    tok.literal = tok.literal & "\n"
    Inc(c.bufpos)
  of 'r', 'R', 'c', 'C': 
    add(tok.literal, CR)
    Inc(c.bufpos)
  of 'l', 'L': 
    add(tok.literal, LF)
    Inc(c.bufpos)
  of 'f', 'F': 
    add(tok.literal, FF)
    inc(c.bufpos)
  of 'e', 'E': 
    add(tok.literal, ESC)
    Inc(c.bufpos)
  of 'a', 'A': 
    add(tok.literal, BEL)
    Inc(c.bufpos)
  of 'b', 'B': 
    add(tok.literal, BACKSPACE)
    Inc(c.bufpos)
  of 'v', 'V': 
    add(tok.literal, VT)
    Inc(c.bufpos)
  of 't', 'T': 
    add(tok.literal, Tabulator)
    Inc(c.bufpos)
  of '\'', '\"': 
    add(tok.literal, c.buf[c.bufpos])
    Inc(c.bufpos)
  of '\\': 
    add(tok.literal, '\\')
    Inc(c.bufpos)
  of 'x', 'X': 
    inc(c.bufpos)
    xi = 0
    handleHexChar(c, xi)
    handleHexChar(c, xi)
    add(tok.literal, Chr(xi))
  of '0'..'9': 
    xi = 0
    handleDecChars(c, xi)
    if (xi <= 255): add(tok.literal, Chr(xi))
    else: tok.kind = tkInvalid
  else: tok.kind = tkInvalid
  
proc HandleCRLF(c: var TCfgParser, pos: int): int = 
  case c.buf[pos]
  of CR: result = lexbase.HandleCR(c, pos)
  of LF: result = lexbase.HandleLF(c, pos)
  else: result = pos
  
proc getString(c: var TCfgParser, tok: var TToken, rawMode: bool) = 
  var 
    pos: int
    ch: Char
    buf: cstring
  pos = c.bufPos + 1          # skip "
  buf = c.buf                 # put `buf` in a register
  tok.kind = tkSymbol
  if (buf[pos] == '\"') and (buf[pos + 1] == '\"'): 
    # long string literal:
    inc(pos, 2)               # skip ""
                              # skip leading newline:
    pos = HandleCRLF(c, pos)
    buf = c.buf
    while true: 
      case buf[pos]
      of '\"': 
        if (buf[pos + 1] == '\"') and (buf[pos + 2] == '\"'): break 
        add(tok.literal, '\"')
        Inc(pos)
      of CR, LF: 
        pos = HandleCRLF(c, pos)
        buf = c.buf
        tok.literal = tok.literal & "\n"
      of lexbase.EndOfFile: 
        tok.kind = tkInvalid
        break 
      else: 
        add(tok.literal, buf[pos])
        Inc(pos)
    c.bufpos = pos +
        3                     # skip the three """
  else: 
    # ordinary string literal
    while true: 
      ch = buf[pos]
      if ch == '\"': 
        inc(pos)              # skip '"'
        break 
      if ch in {CR, LF, lexbase.EndOfFile}: 
        tok.kind = tkInvalid
        break 
      if (ch == '\\') and not rawMode: 
        c.bufPos = pos
        getEscapedChar(c, tok)
        pos = c.bufPos
      else: 
        add(tok.literal, ch)
        Inc(pos)
    c.bufpos = pos

proc getSymbol(c: var TCfgParser, tok: var TToken) = 
  var 
    pos: int
    buf: cstring
  pos = c.bufpos
  buf = c.buf
  while true: 
    add(tok.literal, buf[pos])
    Inc(pos)
    if not (buf[pos] in SymChars): break 
  c.bufpos = pos
  tok.kind = tkSymbol

proc skip(c: var TCfgParser) = 
  var 
    buf: cstring
    pos: int
  pos = c.bufpos
  buf = c.buf
  while true: 
    case buf[pos]
    of ' ': 
      Inc(pos)
    of Tabulator: 
      inc(pos)
    of '#', ';': 
      while not (buf[pos] in {CR, LF, lexbase.EndOfFile}): inc(pos)
    of CR, LF: 
      pos = HandleCRLF(c, pos)
      buf = c.buf
    else: 
      break                   # EndOfFile also leaves the loop
  c.bufpos = pos

proc rawGetTok(c: var TCfgParser, tok: var TToken) = 
  tok.kind = tkInvalid
  setlen(tok.literal, 0)
  skip(c)
  case c.buf[c.bufpos]
  of '=': 
    tok.kind = tkEquals
    inc(c.bufpos)
    tok.literal = "="
  of '-': 
    inc(c.bufPos)
    if c.buf[c.bufPos] == '-': inc(c.bufPos)
    tok.kind = tkDashDash
    tok.literal = "--"
  of ':': 
    tok.kind = tkColon
    inc(c.bufpos)
    tok.literal = ":"
  of 'r', 'R': 
    if c.buf[c.bufPos + 1] == '\"': 
      Inc(c.bufPos)
      getString(c, tok, true)
    else: 
      getSymbol(c, tok)
  of '[': 
    tok.kind = tkBracketLe
    inc(c.bufpos)
    tok.literal = "["
  of ']': 
    tok.kind = tkBracketRi
    Inc(c.bufpos)
    tok.literal = "]"
  of '\"': 
    getString(c, tok, false)
  of lexbase.EndOfFile: 
    tok.kind = tkEof
  else: getSymbol(c, tok)
  
proc errorStr(c: TCfgParser, msg: string): string = 
  result = `%`("$1($2, $3) Error: $4", 
               [c.filename, $(getLine(c)), $(getColumn(c)), msg])

proc getKeyValPair(c: var TCfgParser, kind: TCfgEventKind): TCfgEvent = 
  if c.tok.kind == tkSymbol: 
    result.kind = kind
    result.key = c.tok.literal
    result.value = ""
    rawGetTok(c, c.tok)
    while c.tok.literal == ".": 
      add(result.key, '.')
      rawGetTok(c, c.tok)
      if c.tok.kind == tkSymbol: 
        add(result.key, c.tok.literal)
        rawGetTok(c, c.tok)
      else: 
        result.kind = cfgError
        result.msg = errorStr(c, "symbol expected, but found: " & c.tok.literal)
        break 
    if c.tok.kind in {tkEquals, tkColon}: 
      rawGetTok(c, c.tok)
      if c.tok.kind == tkSymbol: 
        result.value = c.tok.literal
      else: 
        result.kind = cfgError
        result.msg = errorStr(c, "symbol expected, but found: " & c.tok.literal)
      rawGetTok(c, c.tok)
  else: 
    result.kind = cfgError
    result.msg = errorStr(c, "symbol expected, but found: " & c.tok.literal)
    rawGetTok(c, c.tok)

proc next(c: var TCfgParser): TCfgEvent = 
  case c.tok.kind
  of tkEof: 
    result.kind = cfgEof
  of tkDashDash: 
    rawGetTok(c, c.tok)
    result = getKeyValPair(c, cfgOption)
  of tkSymbol: 
    result = getKeyValPair(c, cfgKeyValuePair)
  of tkBracketLe: 
    rawGetTok(c, c.tok)
    if c.tok.kind == tkSymbol: 
      result.kind = cfgSectionStart
      result.section = c.tok.literal
    else: 
      result.kind = cfgError
      result.msg = errorStr(c, "symbol expected, but found: " & c.tok.literal)
    rawGetTok(c, c.tok)
    if c.tok.kind == tkBracketRi: 
      rawGetTok(c, c.tok)
    else: 
      result.kind = cfgError
      result.msg = errorStr(c, "\']\' expected, but found: " & c.tok.literal)
  of tkInvalid, tkBracketRi, tkEquals, tkColon: 
    result.kind = cfgError
    result.msg = errorStr(c, "invalid token: " & c.tok.literal)
    rawGetTok(c, c.tok)