cgialib 0.0.1
Sign up to get free protection for your applications and to get access to all the features.
- data/History.txt +4 -0
- data/Manifest.txt +35 -0
- data/PostInstall.txt +7 -0
- data/README.rdoc +49 -0
- data/Rakefile +28 -0
- data/bin/testgen +10 -0
- data/config/website.yml.sample +2 -0
- data/features/development.feature +13 -0
- data/features/steps/common.rb +174 -0
- data/features/steps/env.rb +6 -0
- data/lib/cgialib/lp/CLanguageScanner.rb +446 -0
- data/lib/cgialib/lp/CPPLanguageScanner.rb +326 -0
- data/lib/cgialib/lp/CTokenizer.rb +481 -0
- data/lib/cgialib/lp/Language.rb +293 -0
- data/lib/cgialib/lp/Tokenizer.rb +281 -0
- data/lib/cgialib/lp.rb +10 -0
- data/lib/cgialib/template/ut/c.rb +41 -0
- data/lib/cgialib/template/ut.rb +6 -0
- data/lib/cgialib/template.rb +6 -0
- data/lib/cgialib.rb +9 -0
- data/lib/testgen/cli.rb +210 -0
- data/script/console +10 -0
- data/script/destroy +14 -0
- data/script/generate +14 -0
- data/script/txt2html +82 -0
- data/spec/cgialib_spec.rb +11 -0
- data/spec/spec.opts +1 -0
- data/spec/spec_helper.rb +10 -0
- data/spec/testgen_cli_spec.rb +15 -0
- data/tasks/rspec.rake +21 -0
- data/website/index.html +86 -0
- data/website/index.txt +79 -0
- data/website/javascripts/rounded_corners_lite.inc.js +285 -0
- data/website/stylesheets/screen.css +159 -0
- data/website/template.html.erb +50 -0
- metadata +112 -0
@@ -0,0 +1,481 @@
|
|
1
|
+
# File: CTokenizer.rb
|
2
|
+
# Author: Jack Herrington
|
3
|
+
# Purpose: A tokenizer specialized to look for C style tokens
|
4
|
+
# Date: 12/21/02
|
5
|
+
|
6
|
+
#require "Tokenizer"
|
7
|
+
|
8
|
+
module LanguageParser
|
9
|
+
|
10
|
+
# class : CT_State
|
11
|
+
#
|
12
|
+
# The base class state object for the C-Tokenizer state machine.
|
13
|
+
|
14
|
+
class CT_State
|
15
|
+
# Special characters that are found as delineators in C
|
16
|
+
|
17
|
+
@@specials = { ";" => 1, "," => 1, ":" => 1, "{" => 1, "}" => 1,
|
18
|
+
"(" => 1, ")" => 1, "[" => 1, "]" => 1, "%" => 1,
|
19
|
+
"+" => 1, "-" => 1, "*" => 1, "." => 1 }
|
20
|
+
|
21
|
+
# initialize( newstate, addtoken )
|
22
|
+
#
|
23
|
+
# newstate - A method to be called to change state
|
24
|
+
# addtoken - The method to be called to add a token
|
25
|
+
#
|
26
|
+
# Intializes the state object
|
27
|
+
|
28
|
+
def initialize( newstate, addtoken )
|
29
|
+
|
30
|
+
@newstate = newstate
|
31
|
+
@addtoken = addtoken
|
32
|
+
|
33
|
+
end
|
34
|
+
|
35
|
+
# next( ch )
|
36
|
+
#
|
37
|
+
# ch - The character
|
38
|
+
#
|
39
|
+
# All states should override this method. This handles a
|
40
|
+
# character from the stream. Returning true means that the
|
41
|
+
# parsing should continue to the next character. Returning false
|
42
|
+
# means the parser should stay on the current character.
|
43
|
+
|
44
|
+
def next( ch )
|
45
|
+
|
46
|
+
true
|
47
|
+
|
48
|
+
end
|
49
|
+
|
50
|
+
end
|
51
|
+
|
52
|
+
# class : CT_OldComment
|
53
|
+
#
|
54
|
+
# Handles parsing an old-style C comment (e.g. /* ... */ )
|
55
|
+
|
56
|
+
class CT_OldComment < CT_State
|
57
|
+
|
58
|
+
# initialize( newstate, addtoken )
|
59
|
+
#
|
60
|
+
# newstate - A method to be called to change state
|
61
|
+
# addtoken - The method to be called to add a token
|
62
|
+
#
|
63
|
+
# Intializes the old-style comment state object
|
64
|
+
|
65
|
+
def initialize( newstate, addtoken )
|
66
|
+
|
67
|
+
super( newstate, addtoken )
|
68
|
+
|
69
|
+
# Initialize the text with the beginning /*
|
70
|
+
|
71
|
+
@text = "/*"
|
72
|
+
|
73
|
+
# True if the last character was a star
|
74
|
+
|
75
|
+
@last_was_star = false
|
76
|
+
|
77
|
+
end
|
78
|
+
|
79
|
+
# next( ch )
|
80
|
+
#
|
81
|
+
# ch - The character
|
82
|
+
#
|
83
|
+
# Handles the character in the parsing stream
|
84
|
+
|
85
|
+
def next( ch )
|
86
|
+
|
87
|
+
# Add this character to the comment
|
88
|
+
|
89
|
+
@text += ch
|
90
|
+
|
91
|
+
# See if we have a '/' if the last character was a star.
|
92
|
+
# If that is the case then return to normal parsing
|
93
|
+
# and add the comment token to the token array.
|
94
|
+
|
95
|
+
if ( ch == "/" && @last_was_star )
|
96
|
+
|
97
|
+
@addtoken.call( CommentToken.new( @text ) )
|
98
|
+
@newstate.call( CT_NormalState )
|
99
|
+
|
100
|
+
end
|
101
|
+
|
102
|
+
# Set the last_was_star to true if we see a star
|
103
|
+
|
104
|
+
@last_was_star = ( ch == "*" )
|
105
|
+
|
106
|
+
# Continue onto the next character
|
107
|
+
|
108
|
+
true
|
109
|
+
|
110
|
+
end
|
111
|
+
|
112
|
+
end
|
113
|
+
|
114
|
+
# class : CT_NewComment
|
115
|
+
#
|
116
|
+
# State object for new style C comments (e.g. //)
|
117
|
+
|
118
|
+
class CT_NewComment < CT_State
|
119
|
+
|
120
|
+
# initialize( newstate, addtoken )
|
121
|
+
#
|
122
|
+
# newstate - A method to be called to change state
|
123
|
+
# addtoken - The method to be called to add a token
|
124
|
+
#
|
125
|
+
|
126
|
+
def initialize( newstate, addtoken )
|
127
|
+
|
128
|
+
super( newstate, addtoken )
|
129
|
+
|
130
|
+
# Initialize the text buffer with the beginning //
|
131
|
+
|
132
|
+
@text = "//"
|
133
|
+
|
134
|
+
end
|
135
|
+
|
136
|
+
# next( ch )
|
137
|
+
#
|
138
|
+
# ch - The character
|
139
|
+
#
|
140
|
+
# Handles the character in the parsing stream
|
141
|
+
|
142
|
+
def next( ch )
|
143
|
+
|
144
|
+
# Add the character to the comment text
|
145
|
+
|
146
|
+
@text += ch
|
147
|
+
|
148
|
+
# Go back to the normal state if we find a return
|
149
|
+
|
150
|
+
if ( ch == "\n" )
|
151
|
+
|
152
|
+
@addtoken.call( CommentToken.new( @text ) )
|
153
|
+
@newstate.call( CT_NormalState )
|
154
|
+
|
155
|
+
end
|
156
|
+
|
157
|
+
# Proceed to the next character
|
158
|
+
|
159
|
+
true
|
160
|
+
|
161
|
+
end
|
162
|
+
|
163
|
+
end
|
164
|
+
|
165
|
+
# class : CT_DoubleQuote
|
166
|
+
#
|
167
|
+
# Handles parsing strings
|
168
|
+
|
169
|
+
class CT_DoubleQuote < CT_State
|
170
|
+
|
171
|
+
# initialize( newstate, addtoken )
|
172
|
+
#
|
173
|
+
# newstate - A method to be called to change state
|
174
|
+
# addtoken - The method to be called to add a token
|
175
|
+
#
|
176
|
+
|
177
|
+
def initialize( newstate, addtoken )
|
178
|
+
|
179
|
+
super( newstate, addtoken )
|
180
|
+
|
181
|
+
# Start the text buffer with the beginning double quote
|
182
|
+
|
183
|
+
@text = "\""
|
184
|
+
|
185
|
+
# Set the escaped flag to false. This will go true when
|
186
|
+
# we see a '\'
|
187
|
+
|
188
|
+
@escaped = false
|
189
|
+
|
190
|
+
end
|
191
|
+
|
192
|
+
# next( ch )
|
193
|
+
#
|
194
|
+
# ch - The character
|
195
|
+
#
|
196
|
+
# Handles the character in the parsing stream
|
197
|
+
|
198
|
+
def next( ch )
|
199
|
+
|
200
|
+
# Add this character to the text buffer
|
201
|
+
|
202
|
+
@text += ch
|
203
|
+
|
204
|
+
# If the character is a double qoute and we are not
|
205
|
+
# escape then go back to the normal state and add
|
206
|
+
# the string token to the array
|
207
|
+
|
208
|
+
if ( ch == "\"" && ! @escaped )
|
209
|
+
|
210
|
+
@addtoken.call( CodeToken.new( @text ) )
|
211
|
+
@newstate.call( CT_NormalState )
|
212
|
+
|
213
|
+
end
|
214
|
+
|
215
|
+
# Set escaped to true if we see a \
|
216
|
+
|
217
|
+
@escaped = ( ch == "\\" )
|
218
|
+
|
219
|
+
# Proceed to the next character
|
220
|
+
|
221
|
+
true
|
222
|
+
|
223
|
+
end
|
224
|
+
|
225
|
+
end
|
226
|
+
|
227
|
+
# CT_WhitespaceTokenizer
|
228
|
+
#
|
229
|
+
# Handles whitespace in the character stream
|
230
|
+
|
231
|
+
class CT_WhitespaceTokenizer < CT_State
|
232
|
+
|
233
|
+
# initialize( newstate, addtoken )
|
234
|
+
#
|
235
|
+
# newstate - A method to be called to change state
|
236
|
+
# addtoken - The method to be called to add a token
|
237
|
+
#
|
238
|
+
|
239
|
+
def initialize( newstate, addtoken )
|
240
|
+
|
241
|
+
super( newstate, addtoken )
|
242
|
+
|
243
|
+
# Initialize the text buffer to blank
|
244
|
+
|
245
|
+
@text = ""
|
246
|
+
|
247
|
+
end
|
248
|
+
|
249
|
+
# next( ch )
|
250
|
+
#
|
251
|
+
# ch - The character
|
252
|
+
#
|
253
|
+
# Handles the character in the parsing stream
|
254
|
+
|
255
|
+
def next( ch )
|
256
|
+
|
257
|
+
if ( ch =~ /\s/ )
|
258
|
+
|
259
|
+
# If the character is whitespace add it to
|
260
|
+
# the buffer
|
261
|
+
|
262
|
+
@text += ch
|
263
|
+
return true
|
264
|
+
|
265
|
+
else
|
266
|
+
|
267
|
+
# Otherwise return to the normal state and
|
268
|
+
# add the token
|
269
|
+
|
270
|
+
@addtoken.call( WhitespaceToken.new( @text ) )
|
271
|
+
@newstate.call( CT_NormalState )
|
272
|
+
|
273
|
+
# Return false because we want the tokenizer
|
274
|
+
# to re-run on the current character
|
275
|
+
|
276
|
+
return false
|
277
|
+
|
278
|
+
end
|
279
|
+
|
280
|
+
end
|
281
|
+
|
282
|
+
end
|
283
|
+
|
284
|
+
# class : CT_WatingForComment
|
285
|
+
#
|
286
|
+
# Handles switching between old comments, new comments, and slashes.
|
287
|
+
|
288
|
+
class CT_WaitingForComment < CT_State
|
289
|
+
|
290
|
+
# next( ch )
|
291
|
+
#
|
292
|
+
# ch - The character
|
293
|
+
#
|
294
|
+
# Handles the character in the parsing stream
|
295
|
+
|
296
|
+
def next( ch )
|
297
|
+
|
298
|
+
# Check to see if we are looking at a new or old
|
299
|
+
# style comment
|
300
|
+
|
301
|
+
if ( ch == "*" )
|
302
|
+
|
303
|
+
@newstate.call( CT_OldComment )
|
304
|
+
|
305
|
+
elsif ( ch == "/" )
|
306
|
+
|
307
|
+
@newstate.call( CT_NewComment )
|
308
|
+
|
309
|
+
else
|
310
|
+
|
311
|
+
# Or if it was just a slash
|
312
|
+
|
313
|
+
@addtoken.call( CodeToken.new( "/" ) )
|
314
|
+
@newstate.call( CT_NormalState )
|
315
|
+
|
316
|
+
end
|
317
|
+
|
318
|
+
end
|
319
|
+
|
320
|
+
end
|
321
|
+
|
322
|
+
# class : CT_NormalState
|
323
|
+
#
|
324
|
+
# The default state machine to which all of the other states return.
|
325
|
+
|
326
|
+
class CT_NormalState < CT_State
|
327
|
+
|
328
|
+
# initialize( newstate, addtoken )
|
329
|
+
#
|
330
|
+
# newstate - A method to be called to change state
|
331
|
+
# addtoken - The method to be called to add a token
|
332
|
+
#
|
333
|
+
|
334
|
+
def initialize( newstate, addtoken )
|
335
|
+
|
336
|
+
super( newstate, addtoken )
|
337
|
+
|
338
|
+
# This normal state handles adding CodeTokens in the
|
339
|
+
# basic stream (e.g. not in a string). So we have a
|
340
|
+
# text buffer.
|
341
|
+
|
342
|
+
@text = ""
|
343
|
+
|
344
|
+
end
|
345
|
+
|
346
|
+
# next( ch )
|
347
|
+
#
|
348
|
+
# ch - The character
|
349
|
+
#
|
350
|
+
# Handles the character in the parsing stream
|
351
|
+
|
352
|
+
def next( ch )
|
353
|
+
|
354
|
+
if @@specials[ch]
|
355
|
+
|
356
|
+
# If this is a special character (e.g. ;,*,+, etc.)
|
357
|
+
# then dump the current token and add the special
|
358
|
+
# characer token
|
359
|
+
|
360
|
+
@addtoken.call( CodeToken.new( @text ) )
|
361
|
+
@text = ""
|
362
|
+
|
363
|
+
@addtoken.call( CodeToken.new( ch ) )
|
364
|
+
|
365
|
+
elsif ch == "\""
|
366
|
+
|
367
|
+
# Start the double quote state if we see a
|
368
|
+
# double quote
|
369
|
+
|
370
|
+
@addtoken.call( CodeToken.new( @text ) )
|
371
|
+
@newstate.call( CT_DoubleQuote )
|
372
|
+
|
373
|
+
elsif ch == "/"
|
374
|
+
|
375
|
+
# Start the comment switcher state if we
|
376
|
+
# see a slash
|
377
|
+
|
378
|
+
@addtoken.call( CodeToken.new( @text ) )
|
379
|
+
@newstate.call( CT_WaitingForComment )
|
380
|
+
|
381
|
+
elsif ch =~ /\s/
|
382
|
+
|
383
|
+
# Move into the whitespace state if we
|
384
|
+
# see whitespace. Return true to re-run
|
385
|
+
# the parser on this character.
|
386
|
+
|
387
|
+
@addtoken.call( CodeToken.new( @text ) )
|
388
|
+
@newstate.call( CT_WhitespaceTokenizer )
|
389
|
+
return false
|
390
|
+
|
391
|
+
else
|
392
|
+
|
393
|
+
# Otherwise add this character to the buffer
|
394
|
+
|
395
|
+
@text += ch
|
396
|
+
|
397
|
+
end
|
398
|
+
|
399
|
+
# Continue onto the next character
|
400
|
+
|
401
|
+
true
|
402
|
+
|
403
|
+
end
|
404
|
+
|
405
|
+
end
|
406
|
+
|
407
|
+
# class : CTokenizer
|
408
|
+
#
|
409
|
+
# The main entry class that parses C text into a set of tokens
|
410
|
+
|
411
|
+
class CTokenizer < Tokenizer
|
412
|
+
|
413
|
+
# parse( text )
|
414
|
+
#
|
415
|
+
# text - The C text
|
416
|
+
#
|
417
|
+
# Parses the C text string into tokens
|
418
|
+
|
419
|
+
def parse( text )
|
420
|
+
|
421
|
+
# Set the current state to the normal state
|
422
|
+
|
423
|
+
@state = CT_NormalState.new( method( :newstate ), method( :addtoken ) )
|
424
|
+
|
425
|
+
# Iterate through the text
|
426
|
+
|
427
|
+
index = 0
|
428
|
+
|
429
|
+
while index < text.length
|
430
|
+
|
431
|
+
# Dispatch the character to the current state
|
432
|
+
|
433
|
+
if ( @state.next( text[ index ].chr() ) )
|
434
|
+
|
435
|
+
index += 1
|
436
|
+
|
437
|
+
end
|
438
|
+
|
439
|
+
end
|
440
|
+
|
441
|
+
end
|
442
|
+
|
443
|
+
protected
|
444
|
+
|
445
|
+
# newstate( classref )
|
446
|
+
#
|
447
|
+
# classref - The new static class type
|
448
|
+
#
|
449
|
+
# Called when we are requesting a change of state. This method creates the
|
450
|
+
# new state from the class reference that is passed in.
|
451
|
+
|
452
|
+
def newstate( classref )
|
453
|
+
|
454
|
+
# Sets the state to a new state based on the class
|
455
|
+
# given
|
456
|
+
|
457
|
+
@state = classref.new( method( :newstate ), method( :addtoken ) )
|
458
|
+
|
459
|
+
end
|
460
|
+
|
461
|
+
# addtoken( token )
|
462
|
+
#
|
463
|
+
# token - The new token
|
464
|
+
#
|
465
|
+
# This adds a token to the token list.
|
466
|
+
|
467
|
+
def addtoken( token )
|
468
|
+
|
469
|
+
# Adds a token to the stack. If the token text is empty
|
470
|
+
# then ignore it
|
471
|
+
|
472
|
+
return if ( token.to_s().length < 1 )
|
473
|
+
|
474
|
+
# Add the token to the array
|
475
|
+
|
476
|
+
@tokens.push( token )
|
477
|
+
|
478
|
+
end
|
479
|
+
|
480
|
+
end
|
481
|
+
end
|