6 |
|
* redistribute this software in source and binary code form, provided |
7 |
|
* that the following conditions are met: |
8 |
|
* |
9 |
< |
* 1. Acknowledgement of the program authors must be made in any |
10 |
< |
* publication of scientific results based in part on use of the |
11 |
< |
* program. An acceptable form of acknowledgement is citation of |
12 |
< |
* the article in which the program was described (Matthew |
13 |
< |
* A. Meineke, Charles F. Vardeman II, Teng Lin, Christopher |
14 |
< |
* J. Fennell and J. Daniel Gezelter, "OOPSE: An Object-Oriented |
15 |
< |
* Parallel Simulation Engine for Molecular Dynamics," |
16 |
< |
* J. Comput. Chem. 26, pp. 252-271 (2005)) |
17 |
< |
* |
18 |
< |
* 2. Redistributions of source code must retain the above copyright |
9 |
> |
* 1. Redistributions of source code must retain the above copyright |
10 |
|
* notice, this list of conditions and the following disclaimer. |
11 |
|
* |
12 |
< |
* 3. Redistributions in binary form must reproduce the above copyright |
12 |
> |
* 2. Redistributions in binary form must reproduce the above copyright |
13 |
|
* notice, this list of conditions and the following disclaimer in the |
14 |
|
* documentation and/or other materials provided with the |
15 |
|
* distribution. |
28 |
|
* arising out of the use of or inability to use software, even if the |
29 |
|
* University of Notre Dame has been advised of the possibility of |
30 |
|
* such damages. |
31 |
+ |
* |
32 |
+ |
* SUPPORT OPEN SCIENCE! If you use OpenMD or its source code in your |
33 |
+ |
* research, please cite the appropriate papers when you publish your |
34 |
+ |
* work. Good starting points are: |
35 |
+ |
* |
36 |
+ |
* [1] Meineke, et al., J. Comp. Chem. 26, 252-271 (2005). |
37 |
+ |
* [2] Fennell & Gezelter, J. Chem. Phys. 124, 234104 (2006). |
38 |
+ |
* [3] Sun, Lin & Gezelter, J. Chem. Phys. 128, 24107 (2008). |
39 |
+ |
* [4] Vardeman & Gezelter, in progress (2009). |
40 |
|
*/ |
41 |
|
|
42 |
|
#include "selection/SelectionCompiler.hpp" |
43 |
|
#include "utils/StringUtils.hpp" |
44 |
< |
namespace oopse { |
44 |
> |
namespace OpenMD { |
45 |
|
|
46 |
< |
bool SelectionCompiler::compile(const std::string& filename, const std::string& script) { |
46 |
> |
bool SelectionCompiler::compile(const std::string& filename, |
47 |
> |
const std::string& script) { |
48 |
|
|
49 |
|
this->filename = filename; |
50 |
|
this->script = script; |
53 |
|
aatokenCompiled.clear(); |
54 |
|
|
55 |
|
if (internalCompile()) { |
56 |
< |
return true; |
56 |
> |
return true; |
57 |
|
} |
58 |
|
|
59 |
|
int icharEnd; |
60 |
|
if ((icharEnd = script.find('\r', ichCurrentCommand)) == std::string::npos && |
61 |
|
(icharEnd = script.find('\n', ichCurrentCommand)) == std::string::npos) { |
62 |
< |
icharEnd = script.size(); |
62 |
> |
icharEnd = script.size(); |
63 |
|
} |
64 |
|
errorLine = script.substr(ichCurrentCommand, icharEnd); |
65 |
|
return false; |
66 |
< |
} |
66 |
> |
} |
67 |
|
|
68 |
< |
bool SelectionCompiler::internalCompile(){ |
68 |
> |
bool SelectionCompiler::internalCompile(){ |
69 |
|
|
70 |
|
cchScript = script.size(); |
71 |
|
ichToken = 0; |
81 |
|
int tokCommand = Token::nada; |
82 |
|
|
83 |
|
for ( ; true; ichToken += cchToken) { |
84 |
< |
if (lookingAtLeadingWhitespace()) |
85 |
< |
continue; |
86 |
< |
//if (lookingAtComment()) |
87 |
< |
// continue; |
88 |
< |
bool endOfLine = lookingAtEndOfLine(); |
89 |
< |
if (endOfLine || lookingAtEndOfStatement()) { |
90 |
< |
if (tokCommand != Token::nada) { |
91 |
< |
if (! compileCommand(ltoken)) { |
92 |
< |
return false; |
93 |
< |
} |
94 |
< |
aatokenCompiled.push_back(atokenCommand); |
95 |
< |
lineNumbers.push_back(lineCurrent); |
96 |
< |
lineIndices.push_back(ichCurrentCommand); |
97 |
< |
ltoken.clear(); |
98 |
< |
tokCommand = Token::nada; |
99 |
< |
} |
84 |
> |
if (lookingAtLeadingWhitespace()) |
85 |
> |
continue; |
86 |
> |
//if (lookingAtComment()) |
87 |
> |
// continue; |
88 |
> |
bool endOfLine = lookingAtEndOfLine(); |
89 |
> |
if (endOfLine || lookingAtEndOfStatement()) { |
90 |
> |
if (tokCommand != Token::nada) { |
91 |
> |
if (! compileCommand(ltoken)) { |
92 |
> |
return false; |
93 |
> |
} |
94 |
> |
aatokenCompiled.push_back(atokenCommand); |
95 |
> |
lineNumbers.push_back(lineCurrent); |
96 |
> |
lineIndices.push_back(ichCurrentCommand); |
97 |
> |
ltoken.clear(); |
98 |
> |
tokCommand = Token::nada; |
99 |
> |
} |
100 |
|
|
101 |
< |
if (ichToken < cchScript) { |
102 |
< |
if (endOfLine) |
103 |
< |
++lineCurrent; |
104 |
< |
continue; |
105 |
< |
} |
106 |
< |
break; |
107 |
< |
} |
101 |
> |
if (ichToken < cchScript) { |
102 |
> |
if (endOfLine) |
103 |
> |
++lineCurrent; |
104 |
> |
continue; |
105 |
> |
} |
106 |
> |
break; |
107 |
> |
} |
108 |
|
|
109 |
< |
if (tokCommand != Token::nada) { |
110 |
< |
if (lookingAtString()) { |
111 |
< |
std::string str = getUnescapedStringLiteral(); |
112 |
< |
ltoken.push_back(Token(Token::string, str)); |
113 |
< |
continue; |
114 |
< |
} |
115 |
< |
//if ((tokCommand & Token::specialstring) != 0 && |
116 |
< |
// lookingAtSpecialString()) { |
117 |
< |
// std::string str = script.substr(ichToken, ichToken + cchToken); |
118 |
< |
// ltoken.push_back(Token(Token::string, str)); |
119 |
< |
// continue; |
120 |
< |
//} |
121 |
< |
if (lookingAtDecimal((tokCommand & Token::negnums) != 0)) { |
122 |
< |
float value = lexi_cast<float>(script.substr(ichToken, cchToken)); |
123 |
< |
std::cout << "encount an decimal: " << value << std::endl; |
124 |
< |
ltoken.push_back(Token(Token::decimal, boost::any(value))); |
125 |
< |
continue; |
126 |
< |
} |
127 |
< |
if (lookingAtInteger((tokCommand & Token::negnums) != 0)) { |
128 |
< |
|
129 |
< |
int val = lexi_cast<int>(script.substr(ichToken, cchToken)); |
130 |
< |
std::cout << "encount an integer: " << val << std::endl; |
131 |
< |
ltoken.push_back(Token(Token::integer, boost::any(val))); |
132 |
< |
continue; |
133 |
< |
} |
134 |
< |
} |
135 |
< |
|
136 |
< |
if (lookingAtLookupToken()) { |
137 |
< |
std::string ident = script.substr(ichToken, cchToken); |
138 |
< |
Token token; |
139 |
< |
Token* pToken = TokenMap::getInstance()->getToken(ident); |
140 |
< |
if (pToken != NULL) { |
141 |
< |
token = *pToken; |
142 |
< |
} else { |
142 |
< |
token = Token(Token::identifier, ident); |
143 |
< |
} |
144 |
< |
|
145 |
< |
int tok = token.tok; |
109 |
> |
if (tokCommand != Token::nada) { |
110 |
> |
if (lookingAtString()) { |
111 |
> |
std::string str = getUnescapedStringLiteral(); |
112 |
> |
ltoken.push_back(Token(Token::string, str)); |
113 |
> |
continue; |
114 |
> |
} |
115 |
> |
//if ((tokCommand & Token::specialstring) != 0 && |
116 |
> |
// lookingAtSpecialString()) { |
117 |
> |
// std::string str = script.substr(ichToken, ichToken + cchToken); |
118 |
> |
// ltoken.push_back(Token(Token::string, str)); |
119 |
> |
// continue; |
120 |
> |
//} |
121 |
> |
if (lookingAtDecimal((tokCommand & Token::negnums) != 0)) { |
122 |
> |
float value = lexi_cast<float>(script.substr(ichToken, cchToken)); |
123 |
> |
ltoken.push_back(Token(Token::decimal, boost::any(value))); |
124 |
> |
continue; |
125 |
> |
} |
126 |
> |
if (lookingAtInteger((tokCommand & Token::negnums) != 0)) { |
127 |
> |
|
128 |
> |
int val = lexi_cast<int>(script.substr(ichToken, cchToken)); |
129 |
> |
ltoken.push_back(Token(Token::integer, boost::any(val))); |
130 |
> |
continue; |
131 |
> |
} |
132 |
> |
} |
133 |
> |
|
134 |
> |
if (lookingAtLookupToken()) { |
135 |
> |
std::string ident = script.substr(ichToken, cchToken); |
136 |
> |
Token token; |
137 |
> |
Token* pToken = TokenMap::getInstance()->getToken(ident); |
138 |
> |
if (pToken != NULL) { |
139 |
> |
token = *pToken; |
140 |
> |
} else { |
141 |
> |
token = Token(Token::identifier, ident); |
142 |
> |
} |
143 |
|
|
144 |
< |
switch (tokCommand) { |
145 |
< |
case Token::nada: |
146 |
< |
ichCurrentCommand = ichToken; |
147 |
< |
//tokenCommand = token; |
148 |
< |
tokCommand = tok; |
149 |
< |
if ((tokCommand & Token::command) == 0) |
150 |
< |
return commandExpected(); |
151 |
< |
break; |
144 |
> |
int tok = token.tok; |
145 |
> |
|
146 |
> |
switch (tokCommand) { |
147 |
> |
case Token::nada: |
148 |
> |
ichCurrentCommand = ichToken; |
149 |
> |
//tokenCommand = token; |
150 |
> |
tokCommand = tok; |
151 |
> |
if ((tokCommand & Token::command) == 0) |
152 |
> |
return commandExpected(); |
153 |
> |
break; |
154 |
|
|
155 |
< |
case Token::define: |
156 |
< |
if (ltoken.size() == 1) { |
157 |
< |
// we are looking at the variable name |
158 |
< |
if (tok != Token::identifier && |
159 |
< |
(tok & Token::predefinedset) != Token::predefinedset) |
160 |
< |
return invalidExpressionToken(ident); |
161 |
< |
} else { |
162 |
< |
// we are looking at the expression |
163 |
< |
if (tok != Token::identifier && |
164 |
< |
(tok & (Token::expression | Token::predefinedset)) == 0) |
165 |
< |
return invalidExpressionToken(ident); |
166 |
< |
} |
155 |
> |
case Token::define: |
156 |
> |
if (ltoken.size() == 1) { |
157 |
> |
// we are looking at the variable name |
158 |
> |
if (tok != Token::identifier && |
159 |
> |
(tok & Token::predefinedset) != Token::predefinedset) |
160 |
> |
return invalidExpressionToken(ident); |
161 |
> |
} else { |
162 |
> |
// we are looking at the expression |
163 |
> |
if (tok != Token::identifier && |
164 |
> |
(tok & (Token::expression | Token::predefinedset)) == 0) |
165 |
> |
return invalidExpressionToken(ident); |
166 |
> |
} |
167 |
|
|
168 |
< |
break; |
168 |
> |
break; |
169 |
|
|
170 |
< |
case Token::select: |
171 |
< |
if (tok != Token::identifier && (tok & Token::expression) == 0) |
172 |
< |
return invalidExpressionToken(ident); |
173 |
< |
break; |
174 |
< |
} |
175 |
< |
ltoken.push_back(token); |
176 |
< |
continue; |
177 |
< |
} |
170 |
> |
case Token::select: |
171 |
> |
if (tok != Token::identifier && (tok & Token::expression) == 0) |
172 |
> |
return invalidExpressionToken(ident); |
173 |
> |
break; |
174 |
> |
} |
175 |
> |
ltoken.push_back(token); |
176 |
> |
continue; |
177 |
> |
} |
178 |
|
|
179 |
< |
if (ltoken.size() == 0) { |
180 |
< |
return commandExpected(); |
181 |
< |
} |
179 |
> |
if (ltoken.size() == 0) { |
180 |
> |
return commandExpected(); |
181 |
> |
} |
182 |
|
|
183 |
< |
return unrecognizedToken(); |
183 |
> |
return unrecognizedToken(); |
184 |
|
} |
185 |
|
|
186 |
|
return true; |
205 |
|
if (ch == '\r') { |
206 |
|
++ichT; |
207 |
|
if (ichT < cchScript && script[ichT] == '\n') |
208 |
< |
++ichT; |
208 |
> |
++ichT; |
209 |
|
} else if (ch == '\n') { |
210 |
|
++ichT; |
211 |
|
} else { |
244 |
|
} |
245 |
|
cchToken = ichT - ichToken; |
246 |
|
|
248 |
– |
|
249 |
– |
std::cout << "lookingAtString: encount " << script.substr(ichToken, cchToken) << std::endl; |
247 |
|
return true; |
248 |
|
} |
249 |
|
|
250 |
|
|
251 |
< |
std::string SelectionCompiler::getUnescapedStringLiteral() { |
251 |
> |
std::string SelectionCompiler::getUnescapedStringLiteral() { |
252 |
|
/** @todo */ |
253 |
|
std::string sb(cchToken - 2, ' '); |
254 |
|
|
256 |
|
int ich = ichToken + 1; |
257 |
|
|
258 |
|
while (ich < ichMax) { |
259 |
< |
char ch = script[ich++]; |
260 |
< |
if (ch == '\\' && ich < ichMax) { |
261 |
< |
ch = script[ich++]; |
262 |
< |
switch (ch) { |
263 |
< |
case 'b': |
264 |
< |
ch = '\b'; |
265 |
< |
break; |
266 |
< |
case 'n': |
267 |
< |
ch = '\n'; |
268 |
< |
break; |
269 |
< |
case 't': |
270 |
< |
ch = '\t'; |
271 |
< |
break; |
272 |
< |
case 'r': |
273 |
< |
ch = '\r'; |
274 |
< |
// fall into |
275 |
< |
case '"': |
276 |
< |
case '\\': |
277 |
< |
case '\'': |
278 |
< |
break; |
279 |
< |
case 'x': |
280 |
< |
case 'u': |
281 |
< |
int digitCount = ch == 'x' ? 2 : 4; |
282 |
< |
if (ich < ichMax) { |
283 |
< |
int unicode = 0; |
284 |
< |
for (int k = digitCount; --k >= 0 && ich < ichMax; ) { |
285 |
< |
char chT = script[ich]; |
286 |
< |
int hexit = getHexitValue(chT); |
287 |
< |
if (hexit < 0) |
288 |
< |
break; |
289 |
< |
unicode <<= 4; |
290 |
< |
unicode += hexit; |
291 |
< |
++ich; |
292 |
< |
} |
293 |
< |
ch = (char)unicode; |
294 |
< |
} |
295 |
< |
} |
296 |
< |
} |
297 |
< |
sb.append(1, ch); |
259 |
> |
char ch = script[ich++]; |
260 |
> |
if (ch == '\\' && ich < ichMax) { |
261 |
> |
ch = script[ich++]; |
262 |
> |
switch (ch) { |
263 |
> |
case 'b': |
264 |
> |
ch = '\b'; |
265 |
> |
break; |
266 |
> |
case 'n': |
267 |
> |
ch = '\n'; |
268 |
> |
break; |
269 |
> |
case 't': |
270 |
> |
ch = '\t'; |
271 |
> |
break; |
272 |
> |
case 'r': |
273 |
> |
ch = '\r'; |
274 |
> |
// fall into |
275 |
> |
case '"': |
276 |
> |
case '\\': |
277 |
> |
case '\'': |
278 |
> |
break; |
279 |
> |
case 'x': |
280 |
> |
case 'u': |
281 |
> |
int digitCount = ch == 'x' ? 2 : 4; |
282 |
> |
if (ich < ichMax) { |
283 |
> |
int unicode = 0; |
284 |
> |
for (int k = digitCount; --k >= 0 && ich < ichMax; ) { |
285 |
> |
char chT = script[ich]; |
286 |
> |
int hexit = getHexitValue(chT); |
287 |
> |
if (hexit < 0) |
288 |
> |
break; |
289 |
> |
unicode <<= 4; |
290 |
> |
unicode += hexit; |
291 |
> |
++ich; |
292 |
> |
} |
293 |
> |
ch = (char)unicode; |
294 |
> |
} |
295 |
> |
} |
296 |
> |
} |
297 |
> |
sb.append(1, ch); |
298 |
|
} |
299 |
|
|
300 |
|
return sb; |
301 |
< |
} |
301 |
> |
} |
302 |
|
|
303 |
< |
int SelectionCompiler::getHexitValue(char ch) { |
303 |
> |
int SelectionCompiler::getHexitValue(char ch) { |
304 |
|
if (ch >= '0' && ch <= '9') |
305 |
< |
return ch - '0'; |
305 |
> |
return ch - '0'; |
306 |
|
else if (ch >= 'a' && ch <= 'f') |
307 |
< |
return 10 + ch - 'a'; |
307 |
> |
return 10 + ch - 'a'; |
308 |
|
else if (ch >= 'A' && ch <= 'F') |
309 |
< |
return 10 + ch - 'A'; |
309 |
> |
return 10 + ch - 'A'; |
310 |
|
else |
311 |
< |
return -1; |
312 |
< |
} |
311 |
> |
return -1; |
312 |
> |
} |
313 |
|
|
314 |
< |
bool SelectionCompiler::lookingAtSpecialString() { |
314 |
> |
bool SelectionCompiler::lookingAtSpecialString() { |
315 |
|
int ichT = ichToken; |
316 |
|
char ch = script[ichT]; |
317 |
|
while (ichT < cchScript && ch != ';' && ch != '\r' && ch != '\n') { |
318 |
< |
++ichT; |
318 |
> |
++ichT; |
319 |
|
} |
320 |
|
cchToken = ichT - ichToken; |
321 |
|
return cchToken > 0; |
322 |
< |
} |
322 |
> |
} |
323 |
|
|
324 |
< |
bool SelectionCompiler::lookingAtDecimal(bool allowNegative) { |
324 |
> |
bool SelectionCompiler::lookingAtDecimal(bool allowNegative) { |
325 |
|
if (ichToken == cchScript) { |
326 |
< |
return false; |
326 |
> |
return false; |
327 |
|
} |
328 |
|
|
329 |
|
int ichT = ichToken; |
330 |
|
if (script[ichT] == '-') { |
331 |
< |
++ichT; |
331 |
> |
++ichT; |
332 |
|
} |
333 |
|
bool digitSeen = false; |
334 |
|
char ch = 'X'; |
335 |
|
while (ichT < cchScript && std::isdigit(ch = script[ichT])) { |
336 |
< |
++ichT; |
337 |
< |
digitSeen = true; |
336 |
> |
++ichT; |
337 |
> |
digitSeen = true; |
338 |
|
} |
339 |
|
|
340 |
|
if (ichT == cchScript || ch != '.') { |
341 |
< |
return false; |
341 |
> |
return false; |
342 |
|
} |
343 |
|
|
344 |
< |
// to support 1.ca, let's check the character after the dot |
345 |
< |
// to determine if it is an alpha |
346 |
< |
if (ch == '.' && (ichT + 1 < cchScript) && std::isalpha(script[ichT + 1])) { |
350 |
< |
return false; |
344 |
> |
// to support DMPC.1, let's check the character before the dot |
345 |
> |
if (ch == '.' && (ichT > 0) && std::isalpha(script[ichT - 1])) { |
346 |
> |
return false; |
347 |
|
} |
348 |
|
|
349 |
|
++ichT; |
350 |
|
while (ichT < cchScript && std::isdigit(script[ichT])) { |
351 |
< |
++ichT; |
352 |
< |
digitSeen = true; |
351 |
> |
++ichT; |
352 |
> |
digitSeen = true; |
353 |
|
} |
354 |
|
cchToken = ichT - ichToken; |
355 |
|
return digitSeen; |
356 |
< |
} |
356 |
> |
} |
357 |
|
|
358 |
< |
bool SelectionCompiler::lookingAtInteger(bool allowNegative) { |
358 |
> |
bool SelectionCompiler::lookingAtInteger(bool allowNegative) { |
359 |
|
if (ichToken == cchScript) { |
360 |
< |
return false; |
360 |
> |
return false; |
361 |
|
} |
362 |
|
int ichT = ichToken; |
363 |
|
if (allowNegative && script[ichToken] == '-') { |
364 |
< |
++ichT; |
364 |
> |
++ichT; |
365 |
|
} |
366 |
|
int ichBeginDigits = ichT; |
367 |
|
while (ichT < cchScript && std::isdigit(script[ichT])) { |
368 |
< |
++ichT; |
368 |
> |
++ichT; |
369 |
|
} |
370 |
|
if (ichBeginDigits == ichT) { |
371 |
< |
return false; |
371 |
> |
return false; |
372 |
|
} |
373 |
|
cchToken = ichT - ichToken; |
374 |
|
return true; |
375 |
< |
} |
375 |
> |
} |
376 |
|
|
377 |
< |
bool SelectionCompiler::lookingAtLookupToken() { |
377 |
> |
bool SelectionCompiler::lookingAtLookupToken() { |
378 |
|
if (ichToken == cchScript) { |
379 |
< |
return false; |
379 |
> |
return false; |
380 |
|
} |
381 |
|
|
382 |
|
int ichT = ichToken; |
383 |
|
char ch; |
384 |
|
switch (ch = script[ichT++]) { |
385 |
< |
case '(': |
386 |
< |
case ')': |
387 |
< |
case ',': |
388 |
< |
case '*': |
389 |
< |
case '-': |
390 |
< |
case '[': |
391 |
< |
case ']': |
392 |
< |
case '+': |
393 |
< |
case ':': |
394 |
< |
case '@': |
395 |
< |
case '.': |
396 |
< |
case '%': |
397 |
< |
break; |
398 |
< |
case '&': |
399 |
< |
case '|': |
400 |
< |
if (ichT < cchScript && script[ichT] == ch) { |
401 |
< |
++ichT; |
402 |
< |
} |
403 |
< |
break; |
404 |
< |
case '<': |
405 |
< |
case '=': |
406 |
< |
case '>': |
407 |
< |
if (ichT < cchScript && ((ch = script[ichT]) == '<' || ch == '=' || ch == '>')) { |
408 |
< |
++ichT; |
409 |
< |
} |
410 |
< |
break; |
411 |
< |
case '/': |
412 |
< |
case '!': |
413 |
< |
if (ichT < cchScript && script[ichT] == '=') { |
414 |
< |
++ichT; |
415 |
< |
} |
416 |
< |
break; |
417 |
< |
default: |
422 |
< |
if ((ch < 'a' || ch > 'z') && (ch < 'A' && ch > 'Z') && ch != '_') { |
423 |
< |
return false; |
424 |
< |
} |
425 |
< |
case '?': // include question marks in identifier for atom expressions |
426 |
< |
while (ichT < cchScript && !std::isspace(ch = script[ichT]) && (std::isalpha(ch) ||std::isdigit(ch) || |
427 |
< |
ch == '_' || ch == '?') ) { |
385 |
> |
case '(': |
386 |
> |
case ')': |
387 |
> |
case ',': |
388 |
> |
case '[': |
389 |
> |
case ']': |
390 |
> |
break; |
391 |
> |
case '&': |
392 |
> |
case '|': |
393 |
> |
if (ichT < cchScript && script[ichT] == ch) { |
394 |
> |
++ichT; |
395 |
> |
} |
396 |
> |
break; |
397 |
> |
case '<': |
398 |
> |
case '=': |
399 |
> |
case '>': |
400 |
> |
if (ichT < cchScript && ((ch = script[ichT]) == '<' || ch == '=' || ch == '>')) { |
401 |
> |
++ichT; |
402 |
> |
} |
403 |
> |
break; |
404 |
> |
case '/': |
405 |
> |
case '!': |
406 |
> |
if (ichT < cchScript && script[ichT] == '=') { |
407 |
> |
++ichT; |
408 |
> |
} |
409 |
> |
break; |
410 |
> |
default: |
411 |
> |
if ((ch < 'a' || ch > 'z') && (ch < 'A' && ch > 'Z') && ch != '_') { |
412 |
> |
return false; |
413 |
> |
} |
414 |
> |
case '*': |
415 |
> |
case '?': // include question marks in identifier for atom expressions |
416 |
> |
while (ichT < cchScript && !std::isspace(ch = script[ichT]) && |
417 |
> |
(std::isalpha(ch) ||std::isdigit(ch) || ch == '_' || ch == '.' || ch == '*' || ch == '?' || ch == '+' || ch == '-' || ch == '[' || ch == ']') ){ |
418 |
|
|
419 |
< |
++ichT; |
420 |
< |
} |
421 |
< |
break; |
419 |
> |
++ichT; |
420 |
> |
} |
421 |
> |
break; |
422 |
|
} |
423 |
|
|
424 |
|
cchToken = ichT - ichToken; |
425 |
|
|
436 |
– |
std::cout << "lookingAtLookupToken: encount " << script.substr(ichToken, cchToken) << std::endl; |
426 |
|
return true; |
427 |
< |
} |
427 |
> |
} |
428 |
|
|
429 |
< |
bool SelectionCompiler::compileCommand(const std::vector<Token>& ltoken) { |
429 |
> |
bool SelectionCompiler::compileCommand(const std::vector<Token>& ltoken) { |
430 |
|
const Token& tokenCommand = ltoken[0]; |
431 |
|
int tokCommand = tokenCommand.tok; |
432 |
|
|
433 |
|
atokenCommand = ltoken; |
434 |
|
if ((tokCommand & Token::expressionCommand) != 0 && !compileExpression()) { |
435 |
< |
return false; |
435 |
> |
return false; |
436 |
|
} |
437 |
|
|
438 |
|
return true; |
439 |
< |
} |
439 |
> |
} |
440 |
|
|
441 |
< |
bool SelectionCompiler::compileExpression() { |
441 |
> |
bool SelectionCompiler::compileExpression() { |
442 |
|
/** todo */ |
443 |
|
int i = 1; |
444 |
|
int tokCommand = atokenCommand[0].tok; |
445 |
|
if (tokCommand == Token::define) { |
446 |
< |
i = 2; |
446 |
> |
i = 2; |
447 |
|
} else if ((tokCommand & Token::embeddedExpression) != 0) { |
448 |
< |
// look for the open parenthesis |
449 |
< |
while (i < atokenCommand.size() && |
450 |
< |
atokenCommand[i].tok != Token::leftparen) |
448 |
> |
// look for the open parenthesis |
449 |
> |
while (i < atokenCommand.size() && |
450 |
> |
atokenCommand[i].tok != Token::leftparen) |
451 |
|
++i; |
452 |
|
} |
453 |
|
|
454 |
|
if (i >= atokenCommand.size()) { |
455 |
< |
return true; |
455 |
> |
return true; |
456 |
|
} |
457 |
|
return compileExpression(i); |
458 |
|
} |
459 |
|
|
460 |
|
|
461 |
< |
bool SelectionCompiler::addTokenToPostfix(const Token& token) { |
461 |
> |
bool SelectionCompiler::addTokenToPostfix(const Token& token) { |
462 |
|
ltokenPostfix.push_back(token); |
463 |
|
return true; |
464 |
< |
} |
464 |
> |
} |
465 |
|
|
466 |
< |
bool SelectionCompiler::compileExpression(int itoken) { |
466 |
> |
bool SelectionCompiler::compileExpression(int itoken) { |
467 |
|
ltokenPostfix.clear(); |
468 |
|
for (int i = 0; i < itoken; ++i) { |
469 |
< |
addTokenToPostfix(atokenCommand[i]); |
469 |
> |
addTokenToPostfix(atokenCommand[i]); |
470 |
|
} |
471 |
|
|
472 |
|
atokenInfix = atokenCommand; |
474 |
|
|
475 |
|
addTokenToPostfix(Token::tokenExpressionBegin); |
476 |
|
if (!clauseOr()) { |
477 |
< |
return false; |
477 |
> |
return false; |
478 |
|
} |
479 |
|
|
480 |
|
addTokenToPostfix(Token::tokenExpressionEnd); |
481 |
|
if (itokenInfix != atokenInfix.size()) { |
482 |
< |
return endOfExpressionExpected(); |
482 |
> |
return endOfExpressionExpected(); |
483 |
|
} |
484 |
|
|
485 |
|
atokenCommand = ltokenPostfix; |
486 |
|
return true; |
487 |
< |
} |
487 |
> |
} |
488 |
|
|
489 |
< |
Token SelectionCompiler::tokenNext() { |
489 |
> |
Token SelectionCompiler::tokenNext() { |
490 |
|
if (itokenInfix == atokenInfix.size()) { |
491 |
< |
return Token(); |
491 |
> |
return Token(); |
492 |
|
} |
493 |
|
return atokenInfix[itokenInfix++]; |
494 |
< |
} |
494 |
> |
} |
495 |
|
|
496 |
< |
boost::any SelectionCompiler::valuePeek() { |
496 |
> |
boost::any SelectionCompiler::valuePeek() { |
497 |
|
if (itokenInfix == atokenInfix.size()) { |
498 |
< |
return boost::any(); |
498 |
> |
return boost::any(); |
499 |
|
} else { |
500 |
< |
return atokenInfix[itokenInfix].value; |
500 |
> |
return atokenInfix[itokenInfix].value; |
501 |
|
} |
502 |
< |
} |
502 |
> |
} |
503 |
|
|
504 |
< |
int SelectionCompiler::tokPeek() { |
504 |
> |
int SelectionCompiler::tokPeek() { |
505 |
|
if (itokenInfix == atokenInfix.size()) { |
506 |
< |
return 0; |
506 |
> |
return 0; |
507 |
|
}else { |
508 |
< |
return atokenInfix[itokenInfix].tok; |
508 |
> |
return atokenInfix[itokenInfix].tok; |
509 |
|
} |
510 |
< |
} |
510 |
> |
} |
511 |
|
|
512 |
< |
bool SelectionCompiler::clauseOr() { |
512 |
> |
bool SelectionCompiler::clauseOr() { |
513 |
|
if (!clauseAnd()) { |
514 |
< |
return false; |
514 |
> |
return false; |
515 |
|
} |
516 |
|
|
517 |
|
while (tokPeek() == Token::opOr) { |
518 |
< |
Token tokenOr = tokenNext(); |
519 |
< |
if (!clauseAnd()) { |
520 |
< |
return false; |
521 |
< |
} |
522 |
< |
addTokenToPostfix(tokenOr); |
518 |
> |
Token tokenOr = tokenNext(); |
519 |
> |
if (!clauseAnd()) { |
520 |
> |
return false; |
521 |
> |
} |
522 |
> |
addTokenToPostfix(tokenOr); |
523 |
|
} |
524 |
|
return true; |
525 |
< |
} |
525 |
> |
} |
526 |
|
|
527 |
< |
bool SelectionCompiler::clauseAnd() { |
527 |
> |
bool SelectionCompiler::clauseAnd() { |
528 |
|
if (!clauseNot()) { |
529 |
< |
return false; |
529 |
> |
return false; |
530 |
|
} |
531 |
|
|
532 |
|
while (tokPeek() == Token::opAnd) { |
533 |
< |
Token tokenAnd = tokenNext(); |
534 |
< |
if (!clauseNot()) { |
535 |
< |
return false; |
536 |
< |
} |
537 |
< |
addTokenToPostfix(tokenAnd); |
533 |
> |
Token tokenAnd = tokenNext(); |
534 |
> |
if (!clauseNot()) { |
535 |
> |
return false; |
536 |
> |
} |
537 |
> |
addTokenToPostfix(tokenAnd); |
538 |
|
} |
539 |
|
return true; |
540 |
< |
} |
540 |
> |
} |
541 |
|
|
542 |
< |
bool SelectionCompiler::clauseNot() { |
542 |
> |
bool SelectionCompiler::clauseNot() { |
543 |
|
if (tokPeek() == Token::opNot) { |
544 |
< |
Token tokenNot = tokenNext(); |
545 |
< |
if (!clauseNot()) { |
546 |
< |
return false; |
547 |
< |
} |
548 |
< |
return addTokenToPostfix(tokenNot); |
544 |
> |
Token tokenNot = tokenNext(); |
545 |
> |
if (!clauseNot()) { |
546 |
> |
return false; |
547 |
> |
} |
548 |
> |
return addTokenToPostfix(tokenNot); |
549 |
|
} |
550 |
|
return clausePrimitive(); |
551 |
< |
} |
551 |
> |
} |
552 |
|
|
553 |
< |
bool SelectionCompiler::clausePrimitive() { |
553 |
> |
bool SelectionCompiler::clausePrimitive() { |
554 |
|
int tok = tokPeek(); |
555 |
|
switch (tok) { |
556 |
< |
case Token::within: |
557 |
< |
return clauseWithin(); |
556 |
> |
case Token::within: |
557 |
> |
return clauseWithin(); |
558 |
|
|
559 |
< |
case Token::asterisk: |
560 |
< |
case Token::identifier: |
561 |
< |
return clauseChemObjName(); |
559 |
> |
case Token::asterisk: |
560 |
> |
case Token::identifier: |
561 |
> |
return clauseChemObjName(); |
562 |
|
|
563 |
< |
case Token::integer : |
564 |
< |
return clauseIndex(); |
565 |
< |
default: |
566 |
< |
if ((tok & Token::atomproperty) == Token::atomproperty) { |
567 |
< |
return clauseComparator(); |
568 |
< |
} |
569 |
< |
if ((tok & Token::predefinedset) != Token::predefinedset) { |
570 |
< |
break; |
571 |
< |
} |
572 |
< |
// fall into the code and below and just add the token |
573 |
< |
case Token::all: |
574 |
< |
case Token::none: |
575 |
< |
return addTokenToPostfix(tokenNext()); |
576 |
< |
case Token::leftparen: |
577 |
< |
tokenNext(); |
578 |
< |
if (!clauseOr()) { |
579 |
< |
return false; |
580 |
< |
} |
581 |
< |
if (tokenNext().tok != Token::rightparen) { |
582 |
< |
return rightParenthesisExpected(); |
583 |
< |
} |
584 |
< |
return true; |
563 |
> |
case Token::integer : |
564 |
> |
return clauseIndex(); |
565 |
> |
default: |
566 |
> |
if ((tok & Token::atomproperty) == Token::atomproperty) { |
567 |
> |
return clauseComparator(); |
568 |
> |
} |
569 |
> |
if ((tok & Token::predefinedset) != Token::predefinedset) { |
570 |
> |
break; |
571 |
> |
} |
572 |
> |
// fall into the code and below and just add the token |
573 |
> |
case Token::all: |
574 |
> |
case Token::none: |
575 |
> |
case Token::hull: |
576 |
> |
return addTokenToPostfix(tokenNext()); |
577 |
> |
case Token::leftparen: |
578 |
> |
tokenNext(); |
579 |
> |
if (!clauseOr()) { |
580 |
> |
return false; |
581 |
> |
} |
582 |
> |
if (tokenNext().tok != Token::rightparen) { |
583 |
> |
return rightParenthesisExpected(); |
584 |
> |
} |
585 |
> |
return true; |
586 |
|
} |
587 |
|
return unrecognizedExpressionToken(); |
588 |
< |
} |
588 |
> |
} |
589 |
|
|
590 |
< |
bool SelectionCompiler::clauseComparator() { |
590 |
> |
bool SelectionCompiler::clauseComparator() { |
591 |
|
Token tokenAtomProperty = tokenNext(); |
592 |
|
Token tokenComparator = tokenNext(); |
593 |
|
if ((tokenComparator.tok & Token::comparator) == 0) { |
594 |
< |
return comparisonOperatorExpected(); |
594 |
> |
return comparisonOperatorExpected(); |
595 |
|
} |
596 |
|
|
597 |
|
Token tokenValue = tokenNext(); |
598 |
|
if (tokenValue.tok != Token::integer && tokenValue.tok != Token::decimal) { |
599 |
< |
return numberExpected(); |
599 |
> |
return numberExpected(); |
600 |
|
} |
601 |
|
|
602 |
|
float val; |
603 |
|
if (tokenValue.value.type() == typeid(int)) { |
604 |
< |
val = boost::any_cast<int>(tokenValue.value); |
604 |
> |
val = boost::any_cast<int>(tokenValue.value); |
605 |
|
} else if (tokenValue.value.type() == typeid(float)) { |
606 |
< |
val = boost::any_cast<float>(tokenValue.value); |
606 |
> |
val = boost::any_cast<float>(tokenValue.value); |
607 |
|
} else { |
608 |
< |
return false; |
608 |
> |
return false; |
609 |
|
} |
610 |
|
|
611 |
|
boost::any floatVal; |
612 |
|
floatVal = val; |
613 |
|
return addTokenToPostfix(Token(tokenComparator.tok, |
614 |
< |
tokenAtomProperty.tok, floatVal)); |
615 |
< |
} |
614 |
> |
tokenAtomProperty.tok, floatVal)); |
615 |
> |
} |
616 |
|
|
617 |
< |
bool SelectionCompiler::clauseWithin() { |
617 |
> |
bool SelectionCompiler::clauseWithin() { |
618 |
|
tokenNext(); // WITHIN |
619 |
|
if (tokenNext().tok != Token::leftparen) { // ( |
620 |
< |
return leftParenthesisExpected(); |
620 |
> |
return leftParenthesisExpected(); |
621 |
|
} |
622 |
|
|
623 |
|
boost::any distance; |
624 |
|
Token tokenDistance = tokenNext(); // distance |
625 |
|
switch(tokenDistance.tok) { |
626 |
< |
case Token::integer: |
627 |
< |
case Token::decimal: |
628 |
< |
distance = tokenDistance.value; |
629 |
< |
break; |
630 |
< |
default: |
631 |
< |
return numberOrKeywordExpected(); |
626 |
> |
case Token::integer: |
627 |
> |
case Token::decimal: |
628 |
> |
distance = tokenDistance.value; |
629 |
> |
break; |
630 |
> |
default: |
631 |
> |
return numberOrKeywordExpected(); |
632 |
|
} |
633 |
|
|
634 |
|
if (tokenNext().tok != Token::opOr) { // , |
635 |
< |
return commaExpected(); |
635 |
> |
return commaExpected(); |
636 |
|
} |
637 |
|
|
638 |
|
if (! clauseOr()) { // *expression* |
639 |
< |
return false; |
639 |
> |
return false; |
640 |
|
} |
641 |
|
|
642 |
|
if (tokenNext().tok != Token::rightparen) { // )T |
643 |
< |
return rightParenthesisExpected(); |
643 |
> |
return rightParenthesisExpected(); |
644 |
|
} |
645 |
|
|
646 |
|
return addTokenToPostfix(Token(Token::within, distance)); |
647 |
< |
} |
647 |
> |
} |
648 |
|
|
649 |
< |
bool SelectionCompiler::clauseChemObjName() { |
650 |
< |
std::string chemObjName; |
651 |
< |
int tok = tokPeek(); |
662 |
< |
if (!clauseName(chemObjName)){ |
663 |
< |
return false; |
664 |
< |
} |
649 |
> |
bool SelectionCompiler::clauseChemObjName() { |
650 |
> |
Token token = tokenNext(); |
651 |
> |
if (token.tok == Token::identifier && token.value.type() == typeid(std::string)) { |
652 |
|
|
653 |
+ |
std::string name = boost::any_cast<std::string>(token.value); |
654 |
+ |
if (isNameValid(name)) { |
655 |
+ |
return addTokenToPostfix(Token(Token::name, name)); |
656 |
+ |
} else { |
657 |
+ |
return compileError("invalid name: " + name); |
658 |
+ |
} |
659 |
+ |
} |
660 |
|
|
661 |
< |
tok = tokPeek(); |
662 |
< |
//allow two dot at most |
663 |
< |
if (tok == Token::dot) { |
670 |
< |
tokenNext(); |
671 |
< |
chemObjName += "."; |
672 |
< |
if (!clauseName(chemObjName)) { |
673 |
< |
return false; |
674 |
< |
} |
675 |
< |
tok = tokPeek(); |
676 |
< |
if (tok == Token::dot) { |
677 |
< |
tokenNext(); |
678 |
< |
chemObjName += "."; |
661 |
> |
return false; |
662 |
> |
|
663 |
> |
} |
664 |
|
|
665 |
< |
if (!clauseName(chemObjName)) { |
666 |
< |
return false; |
667 |
< |
} |
668 |
< |
} |
669 |
< |
} |
665 |
> |
bool SelectionCompiler::isNameValid(const std::string& name) { |
666 |
> |
int nbracket = 0; |
667 |
> |
int ndot = 0; |
668 |
> |
for (int i =0 ; i < name.size(); ++i) { |
669 |
> |
switch(name[i]) { |
670 |
|
|
671 |
< |
return addTokenToPostfix(Token(Token::name, chemObjName)); |
672 |
< |
} |
673 |
< |
|
674 |
< |
bool SelectionCompiler:: clauseName(std::string& name) { |
675 |
< |
|
676 |
< |
int tok = tokPeek(); |
677 |
< |
|
678 |
< |
if (tok == Token::asterisk || tok == Token::identifier) { |
679 |
< |
name += boost::any_cast<std::string>(tokenNext().value); |
680 |
< |
|
696 |
< |
while(true){ |
697 |
< |
tok = tokPeek(); |
698 |
< |
switch (tok) { |
699 |
< |
case Token::asterisk : |
700 |
< |
name += "*"; |
701 |
< |
tokenNext(); |
702 |
< |
break; |
703 |
< |
case Token::identifier : |
704 |
< |
name += boost::any_cast<std::string>(tokenNext().value); |
705 |
< |
break; |
706 |
< |
case Token::integer : |
707 |
< |
name += toString(boost::any_cast<int>(tokenNext().value)); |
708 |
< |
break; |
709 |
< |
case Token::dot : |
710 |
< |
return true; |
711 |
< |
default : |
712 |
< |
return true; |
713 |
< |
} |
714 |
< |
} |
715 |
< |
|
716 |
< |
}else { |
717 |
< |
return false; |
671 |
> |
case '[' : |
672 |
> |
++nbracket; |
673 |
> |
break; |
674 |
> |
case ']' : |
675 |
> |
--nbracket; |
676 |
> |
break; |
677 |
> |
case '.' : |
678 |
> |
++ndot; |
679 |
> |
break; |
680 |
> |
} |
681 |
|
} |
682 |
|
|
683 |
< |
} |
683 |
> |
//only allow 3 dots at most |
684 |
> |
return (ndot <=3 && nbracket == 0) ? true : false; |
685 |
> |
} |
686 |
|
|
687 |
< |
bool SelectionCompiler::clauseIndex(){ |
687 |
> |
bool SelectionCompiler::clauseIndex(){ |
688 |
|
Token token = tokenNext(); |
689 |
|
if (token.tok == Token::integer) { |
690 |
< |
int index = boost::any_cast<int>(token.value); |
691 |
< |
int tok = tokPeek(); |
692 |
< |
std::cout << "Token::to is " << Token::to << ", tok = " << tok << std::endl; |
693 |
< |
if (tok == Token::to) { |
694 |
< |
tokenNext(); |
695 |
< |
tok = tokPeek(); |
696 |
< |
if (tok != Token::integer) { |
697 |
< |
return numberExpected(); |
698 |
< |
} |
690 |
> |
int index = boost::any_cast<int>(token.value); |
691 |
> |
int tok = tokPeek(); |
692 |
> |
std::cout << "Token::to is " << Token::to << ", tok = " << tok << std::endl; |
693 |
> |
if (tok == Token::to) { |
694 |
> |
tokenNext(); |
695 |
> |
tok = tokPeek(); |
696 |
> |
if (tok != Token::integer) { |
697 |
> |
return numberExpected(); |
698 |
> |
} |
699 |
|
|
700 |
< |
boost::any intVal = tokenNext().value; |
701 |
< |
int first = index; |
702 |
< |
if (intVal.type() != typeid(int)){ |
703 |
< |
return false; |
704 |
< |
} |
705 |
< |
int second = boost::any_cast<int>(intVal); |
700 |
> |
boost::any intVal = tokenNext().value; |
701 |
> |
int first = index; |
702 |
> |
if (intVal.type() != typeid(int)){ |
703 |
> |
return false; |
704 |
> |
} |
705 |
> |
int second = boost::any_cast<int>(intVal); |
706 |
|
|
707 |
< |
return addTokenToPostfix(Token(Token::index, boost::any(std::make_pair(first, second)))); |
707 |
> |
return addTokenToPostfix(Token(Token::index, boost::any(std::make_pair(first, second)))); |
708 |
|
|
709 |
< |
}else { |
710 |
< |
return addTokenToPostfix(Token(Token::index, boost::any(index))); |
711 |
< |
} |
709 |
> |
}else { |
710 |
> |
return addTokenToPostfix(Token(Token::index, boost::any(index))); |
711 |
> |
} |
712 |
|
} else { |
713 |
< |
return numberExpected(); |
713 |
> |
return numberExpected(); |
714 |
|
} |
715 |
< |
} |
715 |
> |
} |
716 |
|
|
717 |
|
} |