1 | #include <BufferedFileReader.h> |
---|
2 | #include "fun.h" |
---|
3 | #include "defs.h" |
---|
4 | #include "global.h" |
---|
5 | #include "reader.h" |
---|
6 | |
---|
7 | #include <cstdarg> |
---|
8 | #include <cerrno> |
---|
9 | |
---|
10 | bool scan_token(char *to, const char *from) { // __ATTR__USERESULT |
---|
11 | return sscanf(from, "%s", to) == 1; |
---|
12 | } |
---|
13 | |
---|
14 | void scan_token_or_die(char *to, const char *from) { |
---|
15 | if (!scan_token(to, from)) { |
---|
16 | throw_error(88, "expected to see a token here"); |
---|
17 | } |
---|
18 | } |
---|
19 | void scan_token_or_die(char *to, Reader& reader, int offset) { |
---|
20 | scan_token_or_die(to, reader.line()+offset); |
---|
21 | } |
---|
22 | |
---|
23 | void throw_error(int error_num, const char *error_message) { // __ATTR__NORETURN |
---|
24 | throw Convaln_exception(error_num, error_message); |
---|
25 | } |
---|
26 | |
---|
27 | char *strf(const char *format, ...) { // __ATTR__FORMAT(1) |
---|
28 | va_list parg; |
---|
29 | va_start(parg, format); |
---|
30 | |
---|
31 | const int BUFSIZE = 1000; |
---|
32 | char buffer[BUFSIZE]; |
---|
33 | int printed = vsprintf(buffer, format, parg); |
---|
34 | ca_assert(printed <= BUFSIZE); |
---|
35 | |
---|
36 | va_end(parg); |
---|
37 | |
---|
38 | return strndup(buffer, printed); |
---|
39 | } |
---|
40 | |
---|
41 | void throw_errorf(int error_num, const char *error_messagef, ...) { // __ATTR__FORMAT(2) __ATTR__NORETURN |
---|
42 | va_list parg; |
---|
43 | va_start(parg, error_messagef); |
---|
44 | |
---|
45 | const int BUFSIZE = 1000; |
---|
46 | static char buffer[BUFSIZE]; |
---|
47 | int printed = vsprintf(buffer, error_messagef, parg); |
---|
48 | |
---|
49 | va_end(parg); |
---|
50 | |
---|
51 | if (printed >= BUFSIZE) { |
---|
52 | throw_errorf(998, "Internal buffer overflow (while formatting error #%i '%s')", error_num, error_messagef); |
---|
53 | } |
---|
54 | throw_error(error_num, buffer); |
---|
55 | } |
---|
56 | |
---|
57 | // -------------------------------------------------------------------------------- |
---|
58 | |
---|
59 | bool Warnings::show_warnings = true; |
---|
60 | |
---|
61 | void warning(int warning_num, const char *warning_message) { |
---|
62 | // print out warning_message and continue execution. |
---|
63 | if (Warnings::shown()) |
---|
64 | fprintf(stderr, "WARNING(%d): %s\n", warning_num, warning_message); |
---|
65 | } |
---|
66 | void warningf(int warning_num, const char *warning_messagef, ...) { // __ATTR__FORMAT(2) |
---|
67 | if (Warnings::shown()) { |
---|
68 | va_list parg; |
---|
69 | va_start(parg, warning_messagef); |
---|
70 | |
---|
71 | const int BUFSIZE = 1000; |
---|
72 | static char buffer[BUFSIZE]; |
---|
73 | int printed = vsprintf(buffer, warning_messagef, parg); |
---|
74 | |
---|
75 | va_end(parg); |
---|
76 | |
---|
77 | if (printed >= BUFSIZE) { |
---|
78 | throw_errorf(997, "Internal buffer overflow (while formatting warning #%i '%s')", warning_num, warning_messagef); |
---|
79 | } |
---|
80 | warning(warning_num, buffer); |
---|
81 | } |
---|
82 | } |
---|
83 | |
---|
84 | int Skip_white_space(const char *line, int index) { |
---|
85 | // Skip white space from (index)th char of Str line. |
---|
86 | |
---|
87 | while (line[index] == ' ' || line[index] == '\t') |
---|
88 | ++index; |
---|
89 | return index; |
---|
90 | } |
---|
91 | |
---|
92 | void Getstr(char *line, int linenum) { |
---|
93 | // Get input Str from terminal. |
---|
94 | char c; |
---|
95 | int indi = 0; |
---|
96 | |
---|
97 | for (; (c = getchar()) != '\n' && indi < (linenum - 1); line[indi++] = c) {} |
---|
98 | |
---|
99 | line[indi] = '\0'; |
---|
100 | } |
---|
101 | |
---|
102 | inline void append_known_len(char*& string1, int len1, const char *string2, int len2) { |
---|
103 | ca_assert(len2); // else no need to call, string1 already correct |
---|
104 | int newlen = len1+len2; |
---|
105 | ARB_realloc(string1, newlen+1); |
---|
106 | memcpy(string1+len1, string2, len2); |
---|
107 | string1[newlen] = 0; |
---|
108 | } |
---|
109 | |
---|
110 | void terminate_with(char*& str, char ch) { |
---|
111 | // append 'ch' to end of 'str' (before \n) |
---|
112 | // - if it's not already there and |
---|
113 | // - 'str' contains more than just '\n' |
---|
114 | |
---|
115 | int len = str0len(str); |
---|
116 | if (!len) return; |
---|
117 | |
---|
118 | ca_assert(str[len-1] == '\n'); |
---|
119 | if (len == 1) return; |
---|
120 | |
---|
121 | if (str[len-2] == ch) return; |
---|
122 | |
---|
123 | char temp[] = { ch, '\n' }; |
---|
124 | append_known_len(str, len-1, temp, 2); |
---|
125 | } |
---|
126 | |
---|
127 | void skip_eolnl_and_append(char*& string1, const char *string2) { |
---|
128 | int len1 = str0len(string1); |
---|
129 | if (len1 && string1[len1-1] == '\n') len1--; |
---|
130 | int len2 = str0len(string2); |
---|
131 | if (len2) append_known_len(string1, len1, string2, len2); |
---|
132 | else { string1[len1] = 0; } |
---|
133 | } |
---|
134 | |
---|
135 | void skip_eolnl_and_append_spaced(char*& string1, const char *string2) { |
---|
136 | int len1 = str0len(string1); |
---|
137 | if (len1 && string1[len1-1] == '\n') string1[len1-1] = ' '; |
---|
138 | append_known_len(string1, len1, string2, str0len(string2)); |
---|
139 | } |
---|
140 | |
---|
141 | void Append(char*& string1, const char *string2) { |
---|
142 | int len2 = str0len(string2); |
---|
143 | if (len2) append_known_len(string1, str0len(string1), string2, len2); |
---|
144 | } |
---|
145 | void Append(char*& string1, char ch) { |
---|
146 | append_known_len(string1, str0len(string1), &ch, 1); |
---|
147 | } |
---|
148 | |
---|
149 | void upcase(char *str) { |
---|
150 | // Capitalize all char in the str. |
---|
151 | if (!str) return; |
---|
152 | for (int i = 0; str[i]; ++i) str[i] = toupper(str[i]); |
---|
153 | } |
---|
154 | |
---|
155 | int fputs_len(const char *str, int len, Writer& write) { |
---|
156 | // like fputs(), but does not print more than 'len' characters |
---|
157 | // returns number of chars written or throws write error |
---|
158 | for (int i = 0; i<len; ++i) { |
---|
159 | if (!str[i]) return i; |
---|
160 | write.out(str[i]); |
---|
161 | } |
---|
162 | return len; |
---|
163 | } |
---|
164 | |
---|
165 | // ------------------------- |
---|
166 | // pattern matching |
---|
167 | |
---|
168 | enum FindMode { FIND_START, FIND_SKIP_OVER }; |
---|
169 | |
---|
170 | static int findPattern(const char *text, const char *pattern, FindMode mode) { |
---|
171 | // Return offset of 'pattern' in 'text' or -1 |
---|
172 | // (compares case insensitive) |
---|
173 | // return position after found 'pattern' in 'endPtr' |
---|
174 | |
---|
175 | if (text && pattern) { |
---|
176 | for (int t = 0; text[t]; ++t) { |
---|
177 | bool mismatch = false; |
---|
178 | |
---|
179 | int p = 0; |
---|
180 | for (; !mismatch && pattern[p]; ++p) { |
---|
181 | mismatch = tolower(text[t+p]) != tolower(pattern[p]); |
---|
182 | } |
---|
183 | if (!mismatch) { |
---|
184 | switch (mode) { |
---|
185 | case FIND_START: return t; |
---|
186 | case FIND_SKIP_OVER: return t+p; |
---|
187 | } |
---|
188 | ca_assert(0); |
---|
189 | } |
---|
190 | } |
---|
191 | } |
---|
192 | return -1; |
---|
193 | } |
---|
194 | |
---|
195 | static int findMultipattern(const char *str, const char** const& pattern, char expect_behind, FindMode mode) { |
---|
196 | // search 'str' for the occurrence of any 'pattern' |
---|
197 | // if 'expect_behind' != 0 -> expect that char behind the found pattern |
---|
198 | |
---|
199 | int offset = -1; |
---|
200 | |
---|
201 | if (str) { |
---|
202 | FindMode use = expect_behind ? FIND_SKIP_OVER : mode; |
---|
203 | |
---|
204 | int p; |
---|
205 | for (p = 0; offset == -1 && pattern[p]; ++p) { |
---|
206 | offset = findPattern(str, pattern[p], use); |
---|
207 | } |
---|
208 | |
---|
209 | if (offset != -1) { |
---|
210 | if (expect_behind) { |
---|
211 | if (str[offset] != expect_behind) { // mismatch |
---|
212 | offset = findMultipattern(str+offset, pattern, expect_behind, mode); |
---|
213 | } |
---|
214 | else { |
---|
215 | switch (mode) { |
---|
216 | case FIND_START: |
---|
217 | offset -= str0len(pattern[p]); |
---|
218 | ca_assert(offset >= 0); |
---|
219 | break; |
---|
220 | case FIND_SKIP_OVER: |
---|
221 | offset++; |
---|
222 | break; |
---|
223 | } |
---|
224 | } |
---|
225 | } |
---|
226 | } |
---|
227 | } |
---|
228 | return offset; |
---|
229 | } |
---|
230 | |
---|
231 | static int findSubspecies(const char *str, char expect_behind, FindMode mode) { |
---|
232 | const char *subspecies_pattern[] = { "subspecies", "sub-species", "subsp.", NULp }; |
---|
233 | return findMultipattern(str, subspecies_pattern, expect_behind, mode); |
---|
234 | } |
---|
235 | |
---|
236 | static int findStrain(const char *str, char expect_behind, FindMode mode) { |
---|
237 | const char *strain_pattern[] = { "strain", "str.", NULp }; |
---|
238 | return findMultipattern(str, strain_pattern, expect_behind, mode); |
---|
239 | } |
---|
240 | |
---|
241 | int find_pattern(const char *text, const char *pattern) { return findPattern(text, pattern, FIND_START); } |
---|
242 | int skip_pattern(const char *text, const char *pattern) { return findPattern(text, pattern, FIND_SKIP_OVER); } |
---|
243 | |
---|
244 | int find_subspecies(const char *str, char expect_behind) { return findSubspecies(str, expect_behind, FIND_START); } |
---|
245 | int skip_subspecies(const char *str, char expect_behind) { return findSubspecies(str, expect_behind, FIND_SKIP_OVER); } |
---|
246 | |
---|
247 | int find_strain(const char *str, char expect_behind) { return findStrain(str, expect_behind, FIND_START); } |
---|
248 | int skip_strain(const char *str, char expect_behind) { return findStrain(str, expect_behind, FIND_SKIP_OVER); } |
---|
249 | |
---|
250 | const char *stristr(const char *str, const char *substring) { |
---|
251 | int offset = find_pattern(str, substring); |
---|
252 | return offset >= 0 ? str+offset : NULp; |
---|
253 | } |
---|
254 | |
---|
255 | int ___lookup_keyword(const char *keyword, const char * const *lookup_table, int lookup_table_size) { |
---|
256 | // returns the index [0..n-1] of 'keyword' in lookup_table |
---|
257 | // or -1 if not found. |
---|
258 | |
---|
259 | for (int i = 0; i<lookup_table_size; ++i) { |
---|
260 | if (str_equal(keyword, lookup_table[i])) return i; |
---|
261 | } |
---|
262 | return -1; |
---|
263 | } |
---|
264 | |
---|
265 | int parse_key_word(const char *line, char *key, const char *separator) { |
---|
266 | // Copy keyword starting at position 'index' of 'line' delimited by 'separator' into 'key'. |
---|
267 | // Do not copy more than 'TOKENSIZE-1' characters. |
---|
268 | // Returns length of keyword. |
---|
269 | |
---|
270 | int k = 0; |
---|
271 | if (line) { |
---|
272 | while (k<(TOKENSIZE-1) && line[k] && !occurs_in(line[k], separator)) { |
---|
273 | key[k] = line[k]; |
---|
274 | ++k; |
---|
275 | } |
---|
276 | } |
---|
277 | key[k] = 0; |
---|
278 | return k; |
---|
279 | } |
---|
280 | |
---|
281 | // ---------------------- |
---|
282 | // FormattedFile |
---|
283 | |
---|
284 | FormattedFile::FormattedFile(const char *Name, Format Type) |
---|
285 | : name_(ARB_strdup(Name)), |
---|
286 | type_(Type) |
---|
287 | {} |
---|
288 | FormattedFile::~FormattedFile() { |
---|
289 | free(name_); |
---|
290 | } |
---|
291 | |
---|
292 | void FormattedFile::init(const char *Name, Format Type) { |
---|
293 | ca_assert(!name_); // do not init twice |
---|
294 | ca_assert(Name); |
---|
295 | ca_assert(Type != UNKNOWN); |
---|
296 | |
---|
297 | name_ = nulldup(Name); |
---|
298 | type_ = Type; |
---|
299 | } |
---|
300 | |
---|