Line data Source code
1 : // EnergyPlus, Copyright (c) 1996-2023, The Board of Trustees of the University of Illinois,
2 : // The Regents of the University of California, through Lawrence Berkeley National Laboratory
3 : // (subject to receipt of any required approvals from the U.S. Dept. of Energy), Oak Ridge
4 : // National Laboratory, managed by UT-Battelle, Alliance for Sustainable Energy, LLC, and other
5 : // contributors. All rights reserved.
6 : //
7 : // NOTICE: This Software was developed under funding from the U.S. Department of Energy and the
8 : // U.S. Government consequently retains certain rights. As such, the U.S. Government has been
9 : // granted for itself and others acting on its behalf a paid-up, nonexclusive, irrevocable,
10 : // worldwide license in the Software to reproduce, distribute copies to the public, prepare
11 : // derivative works, and perform publicly and display publicly, and to permit others to do so.
12 : //
13 : // Redistribution and use in source and binary forms, with or without modification, are permitted
14 : // provided that the following conditions are met:
15 : //
16 : // (1) Redistributions of source code must retain the above copyright notice, this list of
17 : // conditions and the following disclaimer.
18 : //
19 : // (2) Redistributions in binary form must reproduce the above copyright notice, this list of
20 : // conditions and the following disclaimer in the documentation and/or other materials
21 : // provided with the distribution.
22 : //
23 : // (3) Neither the name of the University of California, Lawrence Berkeley National Laboratory,
24 : // the University of Illinois, U.S. Dept. of Energy nor the names of its contributors may be
25 : // used to endorse or promote products derived from this software without specific prior
26 : // written permission.
27 : //
28 : // (4) Use of EnergyPlus(TM) Name. If Licensee (i) distributes the software in stand-alone form
29 : // without changes from the version obtained under this License, or (ii) Licensee makes a
30 : // reference solely to the software portion of its product, Licensee must refer to the
31 : // software as "EnergyPlus version X" software, where "X" is the version number Licensee
32 : // obtained under this License and may not use a different name for the software. Except as
33 : // specifically required in this Section (4), Licensee shall not use in a company name, a
34 : // product name, in advertising, publicity, or other promotional activities any name, trade
35 : // name, trademark, logo, or other designation of "EnergyPlus", "E+", "e+" or confusingly
36 : // similar designation, without the U.S. Department of Energy's prior written consent.
37 : //
38 : // THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR
39 : // IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY
40 : // AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
41 : // CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
42 : // CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
43 : // SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
44 : // THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
45 : // OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
46 : // POSSIBILITY OF SUCH DAMAGE.
47 :
48 : #include <EnergyPlus/FromChars.hh>
49 : #include <EnergyPlus/InputProcessing/IdfParser.hh>
50 : #include <cmath>
51 : #include <fast_float/fast_float.h>
52 : #include <fmt/format.h>
53 : #include <milo/dtoa.h>
54 : #include <milo/itoa.h>
55 :
56 : using json = nlohmann::json;
57 :
58 3356073 : auto const icompare = [](std::string_view a, std::string_view b) {
59 3356073 : return (a.length() == b.length()
60 8117654 : ? std::equal(a.begin(), a.end(), b.begin(), [](char const c, char const d) { return (::tolower(c) == ::tolower(d)); })
61 : : false);
62 3356073 : };
63 :
64 0 : json IdfParser::decode(std::string_view idf, json const &schema)
65 : {
66 0 : bool success = true;
67 0 : return decode(idf, idf.size(), schema, success);
68 : }
69 :
70 770 : json IdfParser::decode(std::string_view idf, json const &schema, bool &success)
71 : {
72 770 : return decode(idf, idf.size(), schema, success);
73 : }
74 :
75 0 : json IdfParser::decode(std::string_view idf, size_t _idf_size, json const &schema)
76 : {
77 0 : bool success = true;
78 0 : return decode(idf, _idf_size, schema, success);
79 : }
80 :
81 770 : json IdfParser::decode(std::string_view idf, size_t _idf_size, json const &schema, bool &success)
82 : {
83 770 : success = true;
84 770 : cur_line_num = 1;
85 770 : index_into_cur_line = 0;
86 770 : beginning_of_line_index = 0;
87 770 : idf_size = _idf_size;
88 :
89 770 : if (idf.empty()) {
90 0 : success = false;
91 0 : return nullptr;
92 : }
93 :
94 770 : size_t index = 0;
95 770 : return parse_idf(idf, index, success, schema);
96 : }
97 :
98 0 : std::string IdfParser::encode(json const &root, json const &schema)
99 : {
100 : static constexpr std::string_view end_of_field(",\n ", 4);
101 : static constexpr std::string_view end_of_object(";\n\n", 3);
102 :
103 0 : std::string encoded, extension_key;
104 0 : if (idf_size > 0) {
105 0 : encoded.reserve(idf_size);
106 : } else {
107 0 : encoded.reserve(root.size() * 1024);
108 : }
109 :
110 0 : for (auto obj = root.begin(); obj != root.end(); ++obj) {
111 0 : const auto &legacy_idd = schema["properties"][obj.key()]["legacy_idd"];
112 0 : const auto &legacy_idd_field = legacy_idd["fields"];
113 0 : auto key = legacy_idd.find("extension");
114 0 : if (key != legacy_idd.end()) {
115 0 : extension_key = key.value().get<std::string>();
116 : }
117 0 : for (auto obj_in = obj.value().begin(); obj_in != obj.value().end(); ++obj_in) {
118 0 : encoded += obj.key();
119 0 : size_t skipped_fields = 0;
120 0 : for (size_t i = 0; i < legacy_idd_field.size(); i++) {
121 0 : std::string const &entry = legacy_idd_field[i].get<std::string>();
122 0 : if (obj_in.value().find(entry) == obj_in.value().end()) {
123 0 : if (entry == "name")
124 0 : encoded += std::string{end_of_field} + obj_in.key();
125 : else
126 0 : skipped_fields++;
127 0 : continue;
128 : }
129 0 : for (size_t j = 0; j < skipped_fields; j++)
130 0 : encoded += end_of_field;
131 0 : skipped_fields = 0;
132 0 : encoded += end_of_field;
133 0 : auto const &val = obj_in.value()[entry];
134 0 : if (val.is_string()) {
135 0 : encoded += val.get<std::string>();
136 0 : } else if (val.is_number_integer()) {
137 0 : encoded += std::to_string(val.get<int>());
138 : } else {
139 0 : dtoa(val.get<double>(), s);
140 0 : encoded += s;
141 : }
142 : }
143 :
144 0 : if (obj_in.value().find(extension_key) == obj_in.value().end()) {
145 0 : encoded += end_of_object;
146 0 : continue;
147 : }
148 :
149 0 : auto &extensions = obj_in.value()[extension_key];
150 0 : for (size_t extension_i = 0; extension_i < extensions.size(); extension_i++) {
151 0 : auto const &cur_extension_obj = extensions[extension_i];
152 0 : auto const &extensible = schema["properties"][obj.key()]["legacy_idd"]["extensibles"];
153 0 : for (size_t i = 0; i < extensible.size(); i++) {
154 0 : std::string const &tmp = extensible[i].get<std::string>();
155 0 : if (cur_extension_obj.find(tmp) == cur_extension_obj.end()) {
156 0 : skipped_fields++;
157 0 : continue;
158 : }
159 0 : for (size_t j = 0; j < skipped_fields; j++)
160 0 : encoded += end_of_field;
161 0 : skipped_fields = 0;
162 0 : encoded += end_of_field;
163 0 : if (cur_extension_obj[tmp].is_string()) {
164 0 : encoded += cur_extension_obj[tmp].get<std::string>();
165 : } else {
166 0 : dtoa(cur_extension_obj[tmp].get<double>(), s);
167 0 : encoded += s;
168 : }
169 : }
170 : }
171 0 : encoded += end_of_object;
172 : }
173 : }
174 0 : return encoded;
175 : }
176 :
177 282602 : std::string IdfParser::normalizeObjectType(std::string const &objectType)
178 : {
179 282602 : if (objectType.empty()) return std::string{};
180 565204 : auto key = convertToUpper(objectType);
181 565204 : auto tmp_umit = objectTypeMap.find(key);
182 282602 : if (tmp_umit != objectTypeMap.end()) {
183 282602 : return tmp_umit->second;
184 : }
185 0 : return std::string{};
186 : }
187 :
188 771 : std::vector<std::string> const &IdfParser::errors()
189 : {
190 771 : return errors_;
191 : }
192 :
193 771 : std::vector<std::string> const &IdfParser::warnings()
194 : {
195 771 : return warnings_;
196 : }
197 :
198 771 : bool IdfParser::hasErrors()
199 : {
200 771 : return !errors_.empty();
201 : }
202 :
203 770 : json IdfParser::parse_idf(std::string_view idf, size_t &index, bool &success, json const &schema)
204 : {
205 770 : json root;
206 : Token token;
207 770 : auto const &schema_properties = schema["properties"];
208 :
209 770 : objectTypeMap.reserve(schema_properties.size());
210 643720 : for (auto it = schema_properties.begin(); it != schema_properties.end(); ++it) {
211 1285900 : std::string key = convertToUpper(it.key());
212 642950 : objectTypeMap.emplace(std::move(key), it.key());
213 : }
214 :
215 770 : if (idf_size > 3) {
216 : // UTF-8 Byte Order Mark
217 770 : if (idf[0] == '\xEF' && idf[1] == '\xBB' && idf[2] == '\xBF') {
218 0 : index += 3;
219 0 : index_into_cur_line += 3;
220 : }
221 : }
222 :
223 770 : int idfObjectCount = 0;
224 : while (true) {
225 671916 : token = look_ahead(idf, index);
226 671916 : if (token == Token::END) {
227 770 : break;
228 671146 : } else if (token == Token::NONE) {
229 0 : success = false;
230 0 : return root;
231 671146 : } else if (token == Token::SEMICOLON) {
232 0 : next_token(idf, index);
233 0 : continue;
234 671146 : } else if (token == Token::COMMA) {
235 0 : errors_.emplace_back(fmt::format("Line: {} Index: {} - Extraneous comma found.", cur_line_num, index_into_cur_line));
236 0 : success = false;
237 0 : return root;
238 671146 : } else if (token == Token::EXCLAMATION) {
239 388544 : eat_comment(idf, index);
240 : } else {
241 282602 : ++idfObjectCount;
242 565204 : auto const parsed_obj_name = parse_string(idf, index);
243 565204 : auto const obj_name = normalizeObjectType(parsed_obj_name);
244 282602 : if (obj_name.empty()) {
245 0 : errors_.emplace_back(
246 0 : fmt::format("Line: {} Index: {} - \"{}\" is not a valid Object Type.", cur_line_num, index_into_cur_line, parsed_obj_name));
247 0 : while (token != Token::SEMICOLON && token != Token::END)
248 0 : token = next_token(idf, index);
249 0 : continue;
250 : }
251 :
252 282602 : bool object_success = true;
253 282602 : json const &obj_loc = schema_properties[obj_name];
254 282602 : json const &legacy_idd = obj_loc["legacy_idd"];
255 565204 : json obj = parse_object(idf, index, object_success, legacy_idd, obj_loc, idfObjectCount);
256 282602 : if (!object_success) {
257 0 : auto found_index = idf.find_first_of('\n', beginning_of_line_index);
258 0 : std::string line;
259 0 : if (found_index != std::string::npos) {
260 0 : line = idf.substr(beginning_of_line_index, found_index - beginning_of_line_index - 1);
261 : }
262 0 : errors_.emplace_back(
263 0 : fmt::format("Line: {} Index: {} - Error parsing \"{}\". Error in following line.", cur_line_num, index_into_cur_line, obj_name));
264 0 : errors_.emplace_back(fmt::format("~~~ {}", line));
265 0 : success = false;
266 0 : continue;
267 : }
268 282602 : u64toa(root[obj_name].size() + 1, s);
269 847806 : std::string name = fmt::format("{} {}", obj_name, s);
270 :
271 282602 : if (!obj.is_null()) {
272 565204 : auto const &name_iter = obj.find("name");
273 : // If you find a name field, use that
274 282602 : if (name_iter != obj.end()) {
275 227292 : name = name_iter.value().get<std::string>();
276 227292 : obj.erase(name_iter);
277 : } else {
278 : // Otherwise, see if it should have a name field
279 110620 : auto const &it = obj_loc.find("name");
280 55310 : if (it != obj_loc.end()) {
281 : // Let it slide, as a blank string, to be handled in the appropriate GetInput routine
282 1 : name = "";
283 : }
284 : }
285 : }
286 :
287 282602 : if (root[obj_name].find(name) != root[obj_name].end()) {
288 0 : errors_.emplace_back(
289 0 : fmt::format(R"(Duplicate name found for object of type "{}" named "{}". Overwriting existing object.)", obj_name, name));
290 : }
291 :
292 282602 : root[obj_name][name] = std::move(obj);
293 : }
294 671146 : }
295 :
296 770 : return root;
297 : }
298 :
299 282602 : json IdfParser::parse_object(
300 : std::string_view idf, size_t &index, bool &success, json const &legacy_idd, json const &schema_obj_loc, int idfObjectCount)
301 : {
302 282602 : json root = json::object();
303 565204 : json extensible = json::object();
304 565204 : json array_of_extensions = json::array();
305 : Token token;
306 565204 : std::string extension_key;
307 282602 : size_t legacy_idd_index = 0;
308 282602 : size_t extensible_index = 0;
309 282602 : success = true;
310 282602 : bool was_value_parsed = false;
311 282602 : auto const &legacy_idd_fields_array = legacy_idd["fields"];
312 565204 : auto const &legacy_idd_extensibles_iter = legacy_idd.find("extensibles");
313 :
314 282602 : auto const &schema_patternProperties = schema_obj_loc["patternProperties"];
315 565204 : std::string patternProperty;
316 282602 : int dot_star_present = schema_patternProperties.count(".*");
317 282602 : int no_whitespace_present = schema_patternProperties.count(R"(^.*\S.*$)");
318 282602 : if (dot_star_present) {
319 58820 : patternProperty = ".*";
320 223782 : } else if (no_whitespace_present) {
321 223782 : patternProperty = R"(^.*\S.*$)";
322 : } else {
323 0 : throw std::runtime_error(R"(The patternProperties value is not a valid choice (".*", "^.*\S.*$"))");
324 : }
325 282602 : auto const &schema_obj_props = schema_patternProperties[patternProperty]["properties"];
326 565204 : auto key = legacy_idd.find("extension");
327 :
328 282602 : json const *schema_obj_extensions = nullptr;
329 282602 : if (legacy_idd_extensibles_iter != legacy_idd.end()) {
330 95754 : if (key == legacy_idd.end()) {
331 0 : errors_.emplace_back("\"extension\" key not found in schema. Need to add to list in modify_schema.py.");
332 0 : success = false;
333 0 : return root;
334 : }
335 95754 : extension_key = key.value().get<std::string>();
336 95754 : schema_obj_extensions = &schema_obj_props[extension_key]["items"]["properties"];
337 : }
338 :
339 282602 : root["idf_order"] = idfObjectCount;
340 :
341 565204 : auto const &found_min_fields = schema_obj_loc.find("min_fields");
342 :
343 282602 : index += 1;
344 :
345 : while (true) {
346 10800470 : token = look_ahead(idf, index);
347 10800470 : root["idf_max_fields"] = legacy_idd_index;
348 10800470 : root["idf_max_extensible_fields"] = extensible_index;
349 10800470 : if (token == Token::NONE) {
350 0 : success = false;
351 0 : return root;
352 10800470 : } else if (token == Token::END) {
353 0 : return root;
354 10800470 : } else if (token == Token::COMMA || token == Token::SEMICOLON) {
355 4067701 : if (!was_value_parsed) {
356 298087 : int ext_size = 0;
357 298087 : if (legacy_idd_index < legacy_idd_fields_array.size()) {
358 : // std::string_view field_name = legacy_idd_fields_array[ legacy_idd_index ];
359 : // root[ field_name ] = "";
360 : } else {
361 9249 : auto const &legacy_idd_extensibles_array = legacy_idd_extensibles_iter.value();
362 9249 : ext_size = static_cast<int>(legacy_idd_extensibles_array.size());
363 : // std::string_view field_name = legacy_idd_extensibles_array[ extensible_index % ext_size ];
364 9249 : extensible_index++;
365 : // extensible[ field_name ] = "";
366 : }
367 298087 : if (ext_size && extensible_index % ext_size == 0) {
368 4598 : array_of_extensions.push_back(extensible);
369 4598 : extensible.clear();
370 : }
371 : }
372 4067701 : legacy_idd_index++;
373 4067701 : was_value_parsed = false;
374 4067701 : next_token(idf, index);
375 4067701 : if (token == Token::SEMICOLON) {
376 282602 : size_t min_fields = 0;
377 282602 : if (found_min_fields != schema_obj_loc.end()) {
378 155480 : min_fields = found_min_fields.value().get<size_t>();
379 : }
380 309750 : for (; legacy_idd_index < min_fields; legacy_idd_index++) {
381 : // std::string_view field_name = legacy_idd_fields_array[ legacy_idd_index ];
382 : // root[ field_name ] = "";
383 : }
384 282602 : if (!extensible.empty()) {
385 176 : array_of_extensions.push_back(extensible);
386 176 : extensible.clear();
387 : }
388 282602 : root["idf_max_fields"] = legacy_idd_index;
389 282602 : root["idf_max_extensible_fields"] = extensible_index;
390 282602 : break;
391 3785099 : }
392 6732769 : } else if (token == Token::EXCLAMATION) {
393 2963155 : eat_comment(idf, index);
394 3769614 : } else if (legacy_idd_index >= legacy_idd_fields_array.size()) {
395 1745170 : if (legacy_idd_extensibles_iter == legacy_idd.end()) {
396 0 : errors_.emplace_back(
397 0 : fmt::format("Line: {} Index: {} - Object contains more field values than maximum number of IDD fields and is not extensible.",
398 : cur_line_num,
399 0 : index_into_cur_line));
400 0 : success = false;
401 0 : return root;
402 : }
403 1745170 : if (schema_obj_extensions == nullptr) {
404 0 : errors_.emplace_back(fmt::format("Line: {} Index: {} - Object does not have extensible fields but should. Likely a parsing error.",
405 : cur_line_num,
406 0 : index_into_cur_line));
407 0 : success = false;
408 0 : return root;
409 : }
410 1745170 : auto const &legacy_idd_extensibles_array = legacy_idd_extensibles_iter.value();
411 1745170 : auto const size = legacy_idd_extensibles_array.size();
412 3490340 : std::string const &field_name = legacy_idd_extensibles_array[extensible_index % size].get<std::string>();
413 3490340 : auto val = parse_value(idf, index, success, schema_obj_extensions->at(field_name));
414 1745170 : if (!success) return root;
415 1745170 : extensible[field_name] = std::move(val);
416 1745170 : was_value_parsed = true;
417 1745170 : extensible_index++;
418 1745170 : if (extensible_index && extensible_index % size == 0) {
419 1305079 : array_of_extensions.push_back(extensible);
420 1305079 : extensible.clear();
421 : }
422 : } else {
423 2024444 : was_value_parsed = true;
424 4048888 : std::string const &field = legacy_idd_fields_array[legacy_idd_index].get<std::string>();
425 4048888 : auto const &find_field_iter = schema_obj_props.find(field);
426 2024444 : if (find_field_iter == schema_obj_props.end()) {
427 227292 : if (field == "name") {
428 227292 : root[field] = parse_string(idf, index);
429 : } else {
430 0 : u64toa(cur_line_num, s);
431 0 : errors_.emplace_back(fmt::format("Line: {} - Field \"{}\" was not found.", s, field));
432 : }
433 : } else {
434 3594304 : auto val = parse_value(idf, index, success, find_field_iter.value());
435 1797152 : if (!success) return root;
436 1797152 : root[field] = std::move(val);
437 : }
438 2024444 : if (!success) return root;
439 : }
440 10517868 : }
441 282602 : if (!array_of_extensions.empty()) {
442 95439 : root[extension_key] = std::move(array_of_extensions);
443 95439 : array_of_extensions = nullptr;
444 : }
445 282602 : return root;
446 : }
447 :
448 2288484 : json IdfParser::parse_number(std::string_view idf, size_t &index)
449 : {
450 2288484 : eat_whitespace(idf, index);
451 :
452 2288484 : size_t save_i = index;
453 :
454 2288484 : bool running = true;
455 36756650 : while (running) {
456 17234083 : if (save_i == idf_size) {
457 0 : break;
458 : }
459 :
460 17234083 : char const c = idf[save_i];
461 17234083 : switch (c) {
462 2288484 : case '!':
463 : case ',':
464 : case ';':
465 : case '\r':
466 : case '\n':
467 2288484 : running = false;
468 2288484 : break;
469 14945599 : default:
470 14945599 : ++save_i;
471 : }
472 : }
473 :
474 2288484 : auto diff = save_i - index;
475 2288484 : auto value = idf.substr(index, diff);
476 2288484 : index_into_cur_line += diff;
477 2288484 : index = save_i;
478 :
479 2125946 : auto const convert_double = [&index, this](std::string_view str) -> json {
480 2125942 : size_t plus_sign = 0;
481 2125942 : if (str.front() == '+') {
482 1 : plus_sign = 1;
483 : }
484 2125942 : auto const str_end = str.data() + str.size(); // have to do this for MSVC
485 : double val;
486 2125942 : auto result = fast_float::from_chars(str.data() + plus_sign, str.data() + str.size(), val);
487 2125942 : if (result.ec == std::errc::invalid_argument || result.ec == std::errc::result_out_of_range) {
488 0 : return rtrim(str);
489 2125942 : } else if (result.ptr != str_end) {
490 1 : auto const initial_ptr = result.ptr;
491 3 : while (result.ptr != str_end) {
492 1 : if (*result.ptr != ' ') {
493 0 : break;
494 : }
495 1 : ++result.ptr;
496 : }
497 1 : if (result.ptr == str_end) {
498 2 : index -= (str_end - initial_ptr);
499 2 : this->index_into_cur_line -= (str_end - initial_ptr);
500 1 : return val;
501 : }
502 0 : return rtrim(str);
503 : }
504 2125941 : return val;
505 2288484 : };
506 :
507 4414426 : auto const convert_int = [&convert_double, &index, this](std::string_view str) -> json {
508 2288484 : auto const str_end = str.data() + str.size(); // have to do this for MSVC
509 : int val;
510 2288484 : auto result = FromChars::from_chars(str.data(), str.data() + str.size(), val);
511 2288484 : if (result.ec == std::errc::result_out_of_range || result.ec == std::errc::invalid_argument) {
512 3101 : return convert_double(str);
513 2285383 : } else if (result.ptr != str_end) {
514 2122841 : if (*result.ptr == '.' || *result.ptr == 'e' || *result.ptr == 'E') {
515 2122841 : return convert_double(str);
516 : } else {
517 0 : auto const initial_ptr = result.ptr;
518 0 : while (result.ptr != str_end) {
519 0 : if (*result.ptr != ' ') {
520 0 : break;
521 : }
522 0 : ++result.ptr;
523 : }
524 0 : if (result.ptr == str_end) {
525 0 : index -= (str_end - initial_ptr);
526 0 : this->index_into_cur_line -= (str_end - initial_ptr);
527 0 : return val;
528 : }
529 0 : return rtrim(str);
530 : }
531 : }
532 162542 : return val;
533 2288484 : };
534 :
535 2288484 : return convert_int(value);
536 : }
537 :
538 34531 : json IdfParser::parse_integer(std::string_view idf, size_t &index)
539 : {
540 34531 : eat_whitespace(idf, index);
541 :
542 34531 : size_t save_i = index;
543 :
544 34531 : bool running = true;
545 191007 : while (running) {
546 78238 : if (save_i == idf_size) {
547 0 : break;
548 : }
549 :
550 78238 : char const c = idf[save_i];
551 78238 : switch (c) {
552 34531 : case '!':
553 : case ',':
554 : case ';':
555 : case '\r':
556 : case '\n':
557 34531 : running = false;
558 34531 : break;
559 43707 : default:
560 43707 : ++save_i;
561 : }
562 : }
563 :
564 34531 : auto diff = save_i - index;
565 34531 : auto string_value = idf.substr(index, diff);
566 34531 : index_into_cur_line += diff;
567 34531 : index = save_i;
568 :
569 34531 : auto const string_end = string_value.data() + string_value.size(); // have to do this for MSVC
570 : int int_value;
571 : // Try using from_chars
572 34531 : auto result = FromChars::from_chars(string_value.data(), string_value.data() + string_value.size(), int_value);
573 34531 : if (result.ec == std::errc::result_out_of_range || result.ec == std::errc::invalid_argument) {
574 : // Failure, return the string
575 0 : return rtrim(string_value);
576 34531 : } else if (result.ptr != string_end) {
577 : // Didn't use the entire string, try again via double conversion + rounding
578 673 : size_t plus_sign = 0;
579 673 : if (string_value.front() == '+') {
580 0 : plus_sign = 1;
581 : }
582 : double double_value;
583 673 : auto fresult = fast_float::from_chars(string_value.data() + plus_sign, string_value.data() + string_value.size(), double_value);
584 673 : if (fresult.ec == std::errc::invalid_argument || fresult.ec == std::errc::result_out_of_range) {
585 : // Failure, return the string
586 0 : return rtrim(string_value);
587 : }
588 673 : int_value = static_cast<int>(std::round(double_value));
589 : }
590 34531 : return int_value;
591 : }
592 :
593 3542322 : json IdfParser::parse_value(std::string_view idf, size_t &index, bool &success, json const &field_loc)
594 : {
595 : Token token;
596 7084644 : auto const &field_type = field_loc.find("type");
597 3542322 : if (field_type != field_loc.end()) {
598 3180395 : if (field_type.value() == "number") {
599 2127607 : token = Token::NUMBER;
600 1052788 : } else if (field_type.value() == "integer") {
601 34531 : token = Token::INTEGER;
602 : } else {
603 1018257 : token = Token::STRING;
604 : }
605 : } else {
606 361927 : token = look_ahead(idf, index);
607 : }
608 :
609 3542322 : switch (token) {
610 1219307 : case Token::STRING: {
611 2438614 : auto const parsed_string = parse_string(idf, index);
612 2438614 : auto const &enum_it = field_loc.find("enum");
613 1219307 : if (enum_it != field_loc.end()) {
614 1708624 : for (auto const &enum_str : enum_it.value()) {
615 3042372 : auto const &str = enum_str.get<std::string>();
616 1708624 : if (icompare(str, parsed_string)) {
617 374876 : return str;
618 : }
619 : }
620 844431 : } else if (icompare(parsed_string, "Autosize") || icompare(parsed_string, "Autocalculate")) {
621 141948 : auto const &default_it = field_loc.find("default");
622 141948 : auto const &anyOf_it = field_loc.find("anyOf");
623 :
624 70974 : if (anyOf_it == field_loc.end()) {
625 0 : errors_.emplace_back(
626 0 : fmt::format("Line: {} Index: {} - Field cannot be Autosize or Autocalculate", cur_line_num, index_into_cur_line));
627 0 : return parsed_string;
628 : }
629 : // The following is hacky because it abuses knowing the consistent generated structure
630 : // in the future this might not hold true for the array indexes.
631 70974 : if (default_it != field_loc.end()) {
632 51423 : return field_loc.at("anyOf")[1]["enum"][1];
633 : } else {
634 19551 : return field_loc.at("anyOf")[1]["enum"][0];
635 : }
636 : }
637 773457 : return parsed_string;
638 : }
639 2288484 : case Token::NUMBER: {
640 2288484 : return parse_number(idf, index);
641 : }
642 34531 : case Token::INTEGER: {
643 34531 : return parse_integer(idf, index);
644 : }
645 0 : case Token::NONE:
646 : case Token::END:
647 : case Token::EXCLAMATION:
648 : case Token::COMMA:
649 : case Token::SEMICOLON:
650 : default:
651 0 : break;
652 : }
653 0 : success = false;
654 0 : return nullptr;
655 : }
656 :
657 1729201 : std::string IdfParser::parse_string(std::string_view idf, size_t &index)
658 : {
659 1729201 : eat_whitespace(idf, index);
660 :
661 3458402 : std::string str;
662 : char c;
663 :
664 : while (true) {
665 62939445 : if (index == idf_size) {
666 0 : break;
667 : }
668 :
669 32334323 : c = idf[index];
670 32334323 : increment_both_index(index, index_into_cur_line);
671 32334323 : if (c == ',' || c == ';' || c == '!') {
672 1729201 : decrement_both_index(index, index_into_cur_line);
673 1729201 : break;
674 : } else {
675 30605122 : str += c;
676 : }
677 : }
678 :
679 3458402 : return rtrim(str);
680 : }
681 :
682 246292819 : void IdfParser::increment_both_index(size_t &index, size_t &line_index)
683 : {
684 246292819 : index++;
685 246292819 : line_index++;
686 246292819 : }
687 :
688 1729201 : void IdfParser::decrement_both_index(size_t &index, size_t &line_index)
689 : {
690 1729201 : index--;
691 1729201 : line_index--;
692 1729201 : }
693 :
694 95710393 : void IdfParser::eat_whitespace(std::string_view idf, size_t &index)
695 : {
696 171466556 : while (index < idf_size) {
697 171465786 : switch (idf[index]) {
698 74471271 : case ' ':
699 : case '\r':
700 : case '\t':
701 74471271 : increment_both_index(index, index_into_cur_line);
702 74471271 : continue;
703 1284892 : case '\n':
704 1284892 : increment_both_index(index, cur_line_num);
705 1284892 : beginning_of_line_index = index;
706 1284892 : index_into_cur_line = 0;
707 1284892 : continue;
708 19953460 : default:
709 19953460 : return;
710 : }
711 : }
712 : }
713 :
714 122301090 : void IdfParser::eat_comment(std::string_view idf, size_t &index)
715 : {
716 : while (true) {
717 241250481 : if (index == idf_size) break;
718 122301089 : if (idf[index] == '\n') {
719 3351698 : increment_both_index(index, cur_line_num);
720 3351698 : index_into_cur_line = 0;
721 3351698 : beginning_of_line_index = index;
722 3351698 : break;
723 : }
724 118949391 : increment_both_index(index, index_into_cur_line);
725 : }
726 3351699 : }
727 :
728 11834313 : IdfParser::Token IdfParser::look_ahead(std::string_view idf, size_t index)
729 : {
730 11834313 : size_t save_index = index;
731 11834313 : size_t save_line_num = cur_line_num;
732 11834313 : size_t save_line_index = index_into_cur_line;
733 11834313 : Token token = next_token(idf, save_index);
734 11834313 : cur_line_num = save_line_num;
735 11834313 : index_into_cur_line = save_line_index;
736 11834313 : return token;
737 : }
738 :
739 15902014 : IdfParser::Token IdfParser::next_token(std::string_view idf, size_t &index)
740 : {
741 15902014 : eat_whitespace(idf, index);
742 :
743 15902014 : if (index == idf_size) {
744 770 : return Token::END;
745 : }
746 :
747 15901244 : char const c = idf[index];
748 15901244 : increment_both_index(index, index_into_cur_line);
749 15901244 : switch (c) {
750 3351699 : case '!':
751 3351699 : return Token::EXCLAMATION;
752 7570198 : case ',':
753 7570198 : return Token::COMMA;
754 565204 : case ';':
755 565204 : return Token::SEMICOLON;
756 4414143 : default:
757 : static constexpr std::string_view numeric(".-+0123456789");
758 4414143 : if (numeric.find_first_of(c) != std::string::npos) {
759 2494300 : return Token::NUMBER;
760 : }
761 1919843 : return Token::STRING;
762 : }
763 : decrement_both_index(index, index_into_cur_line);
764 : return Token::NONE;
765 : }
766 :
767 0 : IdfParser::Token IdfParser::next_limited_token(std::string_view idf, size_t &index)
768 : {
769 0 : if (index == idf_size) {
770 0 : return Token::END;
771 : }
772 :
773 0 : char const c = idf[index];
774 0 : increment_both_index(index, index_into_cur_line);
775 0 : switch (c) {
776 0 : case '!':
777 0 : return Token::EXCLAMATION;
778 0 : case ',':
779 0 : return Token::COMMA;
780 0 : case ';':
781 0 : return Token::SEMICOLON;
782 0 : default:
783 0 : return Token::NONE;
784 : }
785 : }
786 :
787 1729201 : std::string IdfParser::rtrim(std::string_view str)
788 : {
789 : static constexpr std::string_view whitespace(" \t\0", 3);
790 1729201 : if (str.empty()) {
791 0 : return std::string{};
792 : }
793 1729201 : auto const index = str.find_last_not_of(whitespace);
794 1729201 : if (index == std::string::npos) {
795 0 : return std::string{};
796 1729201 : } else if (index + 1 < str.length()) {
797 29 : return std::string{str.substr(0, index + 1)};
798 : }
799 1729172 : return std::string{str};
800 : }
|
