Mercurial > projects > qtd

comparison d1/qtd/ctfe/String.d @ 344:96a75b1e5b26

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
project structure changes
author Max Samukha <maxter@spambox.com>
date Fri, 14 May 2010 12:14:37 +0300
parents qt/qtd/ctfe/String.d@f9559a957be9
children
comparison
equal deleted inserted replaced
343:552647ec0f82 344:96a75b1e5b26
1 /**
2 * CTFE String routines.
3 *
4 * Authors: Daniel Keep <daniel.keep@gmail.com>
5 * Copyright: See LICENSE.
6 */
7 module qt.qtd.ctfe.String;
8
9 import Integer = qt.qtd.ctfe.Integer;
10
11 private
12 {
13 const HEX_CHARS = "0123456789abcdef";
14 }
15
16 /**
17 * Escapes a string into an equivalent string literal.
18 *
19 * Params:
20 * str = string to escape.
21 * aggressive = if set, the function will escape all non-printing
22 * characters, non-space whitespace and newlines. Defaults
23 * to true.
24 * Returns:
25 * Escaped string literal.
26 */
27 string escape_ctfe(string str, bool aggressive=true)
28 {
29 string head = "";
30
31 foreach( i,c ; str )
32 {
33 if( c == '"' || c == '\\' || c == '\0' )
34 {
35 head = "\""~str[0..i];
36 str = str[i..$];
37 break;
38 }
39
40 if( aggressive )
41 {
42 if( c < 0x20 || c == 0x7f )
43 {
44 head = "\""~str[0..i];
45 str = str[i..$];
46 break;
47 }
48 }
49 }
50
51 if( head.length == 0 )
52 return "\"" ~ str ~ "\"";
53
54 string tail = "";
55
56 foreach( c ; str )
57 {
58 if( c == '"' )
59 tail ~= `\"`;
60
61 else if( c == '\\' )
62 tail ~= "\\\\";
63
64 else if( c == '\0' )
65 tail ~= `\0`;
66
67 else if( aggressive )
68 {
69 switch( c )
70 {
71 case '\?':
72 tail ~= `\?`;
73 break;
74
75 case '\a':
76 tail ~= `\a`;
77 break;
78
79 case '\b':
80 tail ~= `\b`;
81 break;
82
83 case '\f':
84 tail ~= `\f`;
85 break;
86
87 case '\n':
88 tail ~= `\n`;
89 break;
90
91 case '\r':
92 tail ~= `\r`;
93 break;
94
95 case '\t':
96 tail ~= `\t`;
97 break;
98
99 case '\v':
100 tail ~= `\v`;
101 break;
102
103 default:
104 if( c < 0x20 || c == 0x75 )
105 {
106 tail ~= `\x`;
107 tail ~= HEX_CHARS[c/0xf];
108 tail ~= HEX_CHARS[c&0xf];
109 }
110 else
111 tail ~= c;
112 }
113 }
114 else
115 tail ~= c;
116 }
117
118 return head ~ tail ~ "\"";
119 }
120
121 version( Unittest )
122 {
123 static assert( escape_ctfe("abc") == "\"abc\"" );
124 static assert( escape_ctfe("a\"c") == "\"a\\\"c\"" );
125 }
126
127 /**
128 * Turns an array of bytes into a hexadecimal string.
129 *
130 * Params:
131 * arr = array to hexify.
132 * grouping = if non-zero, specifies after how many bytes to insert a
133 * space.
134 * Returns:
135 * String of hex bytes.
136 */
137
138 string hexify_ctfe(ubyte[] arr, int grouping = 0)
139 {
140 string r = "";
141 int bytes = grouping;
142 foreach( b ; arr )
143 {
144 if( bytes == 0 && grouping > 0 )
145 {
146 r ~= ' ';
147 bytes = grouping;
148 }
149
150 auto bh = b/16;
151 auto bl = b&15;
152
153 assert( bh < 16 );
154 assert( bl < 16 );
155
156 r ~= HEX_CHARS[bh];
157 r ~= HEX_CHARS[bl];
158
159 if( grouping > 0 )
160 -- bytes;
161 }
162 return r;
163 }
164
165 /// ditto
166
167 string hexify_ctfe(string arr, int grouping = 0)
168 {
169 string r = "";
170 int bytes = grouping;
171 foreach( b ; arr )
172 {
173 if( bytes == 0 && grouping > 0 )
174 {
175 r ~= ' ';
176 bytes = grouping;
177 }
178
179 auto bh = b/16;
180 auto bl = b&15;
181
182 assert( bh < 16 );
183 assert( bl < 16 );
184
185 r ~= HEX_CHARS[bh];
186 r ~= HEX_CHARS[bl];
187
188 if( grouping > 0 )
189 -- bytes;
190 }
191 return r;
192 }
193
194 version( Unittest )
195 {
196 static const ubyte[] DATA_1 = [0x00,0x01,0x02,0x03];
197 static const ubyte[] DATA_2 = [0x0f,0x10,0xef,0xf0];
198
199 static assert( hexify_ctfe(DATA_1) == "00010203" );
200 static assert( hexify_ctfe(DATA_2) == "0f10eff0" );
201
202 static assert( hexify_ctfe(DATA_1, 1) == "00 01 02 03" );
203 static assert( hexify_ctfe(DATA_2, 1) == "0f 10 ef f0" );
204
205 static assert( hexify_ctfe(DATA_1, 2) == "0001 0203" );
206 static assert( hexify_ctfe(DATA_2, 2) == "0f10 eff0" );
207
208 static assert( hexify_ctfe(DATA_1, 4) == "00010203" );
209 static assert( hexify_ctfe(DATA_2, 4) == "0f10eff0" );
210 }
211
212 /**
213 * Pads a string. padl adds padding to the left, padr adds it to the right.
214 * Params:
215 * str = string to pad.
216 * len = length to pad to.
217 * padding = character to use for padding. Defaults to space.
218 * Returns:
219 * padded string.
220 */
221
222 string padl_ctfe(string str, int len, char padding = ' ')
223 {
224 while( str.length < len )
225 str = padding ~ str;
226 return str;
227 }
228
229 /// ditto
230
231 string padr_ctfe(string str, int len, char padding = ' ')
232 {
233 while( str.length < len )
234 str ~= padding;
235 return str;
236 }
237
238 version( Unittest )
239 {
240 static assert( padl_ctfe("abc", 2) == "abc" );
241 static assert( padl_ctfe("abc", 3) == "abc" );
242 static assert( padl_ctfe("abc", 4) == " abc" );
243 static assert( padl_ctfe("abc", 4, 'x') == "xabc" );
244
245 static assert( padr_ctfe("abc", 2) == "abc" );
246 static assert( padr_ctfe("abc", 3) == "abc" );
247 static assert( padr_ctfe("abc", 4) == "abc " );
248 static assert( padr_ctfe("abc", 4, 'x') == "abcx" );
249 }
250
251 /**
252 * Returns the tail of a string after a given splitting character. The Rev
253 * variant returns the tail after the last instance of the splitting
254 * character.
255 */
256
257 string tail_ctfe(string str, char split)
258 {
259 foreach( i,c ; str )
260 {
261 if( c == split )
262 return str[i+1..$];
263 }
264 return str;
265 }
266
267 /// ditto
268
269 string tailRev_ctfe(string str, char split)
270 {
271 foreach_reverse( i,c ; str )
272 {
273 if( c == split )
274 return str[i+1..$];
275 }
276 return str;
277 }
278
279 /**
280 * Determines whether a character is valid in an identifier in a
281 * non-initial position.
282 *
283 * Does not support the full range of valid D identifier characters.
284 */
285
286 bool isIdentChar_ctfe(char c)
287 {
288 return ('a' <= c && c <= 'z')
289 || ('A' <= c && c <= 'Z')
290 || ('0' <= c && c <= '9')
291 || (c == '_');
292 }
293
294 /**
295 * Determines whether a character is valid in an identifier in an
296 * initial position.
297 *
298 * Does not support the full range of valid D identifier characters.
299 */
300
301 bool isIdentStartChar_ctfe(char c)
302 {
303 return ('a' <= c && c <= 'z')
304 || ('A' <= c && c <= 'Z')
305 || (c == '_');
306 }
307
308 /**
309 * Returns a line spec suitable for mixing in. This can be used with string
310 * mixins to ensure compile errors appear on the "correct" line in the source.
311 */
312
313 string linespec_ctfe(string file, long line)
314 {
315 return "#line "~Integer.format_ctfe(line)~" \"" ~ file ~ "\"\n";
316 }
317