| 1 | #!/usr/bin/env bash |
| 2 | # |
| 3 | # Test combination of var ops. |
| 4 | # |
| 5 | # NOTE: There are also slice tests in {array,arith-context}.test.sh. |
| 6 | |
| 7 | #### String length |
| 8 | v=foo |
| 9 | echo ${#v} |
| 10 | ## stdout: 3 |
| 11 | |
| 12 | #### Unicode string length (UTF-8) |
| 13 | v=$'_\u03bc_' |
| 14 | echo ${#v} |
| 15 | ## stdout: 3 |
| 16 | ## N-I dash stdout: 9 |
| 17 | ## N-I mksh stdout: 4 |
| 18 | |
| 19 | #### Unicode string length (spec/testdata/utf8-chars.txt) |
| 20 | v=$(cat spec/testdata/utf8-chars.txt) |
| 21 | echo ${#v} |
| 22 | ## stdout: 7 |
| 23 | ## N-I dash stdout: 13 |
| 24 | ## N-I mksh stdout: 13 |
| 25 | |
| 26 | #### String length with incomplete utf-8 |
| 27 | for num_bytes in 0 1 2 3 4 5 6 7 8 9 10 11 12 13; do |
| 28 | s=$(head -c $num_bytes spec/testdata/utf8-chars.txt) |
| 29 | echo ${#s} |
| 30 | done |
| 31 | ## STDOUT: |
| 32 | 0 |
| 33 | 1 |
| 34 | 2 |
| 35 | error: Incomplete utf-8 |
| 36 | 3 |
| 37 | 4 |
| 38 | error: Incomplete utf-8 |
| 39 | error: Incomplete utf-8 |
| 40 | 5 |
| 41 | 6 |
| 42 | error: Incomplete utf-8 |
| 43 | error: Incomplete utf-8 |
| 44 | error: Incomplete utf-8 |
| 45 | 7 |
| 46 | ## END |
| 47 | ## BUG bash STDOUT: |
| 48 | 0 |
| 49 | 1 |
| 50 | 2 |
| 51 | 3 |
| 52 | 3 |
| 53 | 4 |
| 54 | 5 |
| 55 | 6 |
| 56 | 5 |
| 57 | 6 |
| 58 | 7 |
| 59 | 8 |
| 60 | 9 |
| 61 | 7 |
| 62 | ## END |
| 63 | ## N-I dash/mksh STDOUT: |
| 64 | 0 |
| 65 | 1 |
| 66 | 2 |
| 67 | 3 |
| 68 | 4 |
| 69 | 5 |
| 70 | 6 |
| 71 | 7 |
| 72 | 8 |
| 73 | 9 |
| 74 | 10 |
| 75 | 11 |
| 76 | 12 |
| 77 | 13 |
| 78 | ## END |
| 79 | |
| 80 | #### String length with invalid utf-8 continuation bytes |
| 81 | for num_bytes in 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14; do |
| 82 | s=$(head -c $num_bytes spec/testdata/utf8-chars.txt)$(echo -e "\xFF") |
| 83 | echo ${#s} |
| 84 | done |
| 85 | ## STDOUT: |
| 86 | error: Invalid start of utf-8 char |
| 87 | error: Invalid start of utf-8 char |
| 88 | error: Invalid start of utf-8 char |
| 89 | error: Invalid utf-8 continuation byte |
| 90 | error: Invalid start of utf-8 char |
| 91 | error: Invalid start of utf-8 char |
| 92 | error: Invalid utf-8 continuation byte |
| 93 | error: Invalid utf-8 continuation byte |
| 94 | error: Invalid start of utf-8 char |
| 95 | error: Invalid start of utf-8 char |
| 96 | error: Invalid utf-8 continuation byte |
| 97 | error: Invalid utf-8 continuation byte |
| 98 | error: Invalid utf-8 continuation byte |
| 99 | error: Invalid start of utf-8 char |
| 100 | error: Invalid start of utf-8 char |
| 101 | ## END |
| 102 | ## BUG bash STDOUT: |
| 103 | 1 |
| 104 | 2 |
| 105 | 3 |
| 106 | 4 |
| 107 | 4 |
| 108 | 5 |
| 109 | 6 |
| 110 | 7 |
| 111 | 6 |
| 112 | 7 |
| 113 | 8 |
| 114 | 9 |
| 115 | 10 |
| 116 | 8 |
| 117 | 8 |
| 118 | ## N-I dash STDOUT: |
| 119 | 7 |
| 120 | 8 |
| 121 | 9 |
| 122 | 10 |
| 123 | 11 |
| 124 | 12 |
| 125 | 13 |
| 126 | 14 |
| 127 | 15 |
| 128 | 16 |
| 129 | 17 |
| 130 | 18 |
| 131 | 19 |
| 132 | 20 |
| 133 | 20 |
| 134 | ## END |
| 135 | ## N-I mksh STDOUT: |
| 136 | 1 |
| 137 | 2 |
| 138 | 3 |
| 139 | 4 |
| 140 | 5 |
| 141 | 6 |
| 142 | 7 |
| 143 | 8 |
| 144 | 9 |
| 145 | 10 |
| 146 | 11 |
| 147 | 12 |
| 148 | 13 |
| 149 | 14 |
| 150 | 14 |
| 151 | ## END |
| 152 | |
| 153 | #### Length of undefined variable |
| 154 | echo ${#undef} |
| 155 | ## stdout: 0 |
| 156 | |
| 157 | #### Length of undefined variable with nounset |
| 158 | set -o nounset |
| 159 | echo ${#undef} |
| 160 | ## status: 1 |
| 161 | ## OK dash status: 2 |
| 162 | |
| 163 | #### Cannot take length of substring slice |
| 164 | # These are runtime errors, but we could make them parse time errors. |
| 165 | v=abcde |
| 166 | echo ${#v:1:3} |
| 167 | ## status: 1 |
| 168 | ## OK osh status: 2 |
| 169 | ## N-I dash status: 0 |
| 170 | ## N-I dash stdout: 5 |
| 171 | |
| 172 | #### Pattern replacement |
| 173 | v=abcde |
| 174 | echo ${v/c*/XX} |
| 175 | ## stdout: abXX |
| 176 | ## N-I dash status: 2 |
| 177 | ## N-I dash stdout-json: "" |
| 178 | |
| 179 | #### Pattern replacement on unset variable |
| 180 | echo [${v/x/y}] |
| 181 | echo status=$? |
| 182 | set -o nounset # make sure this fails |
| 183 | echo [${v/x/y}] |
| 184 | ## STDOUT: |
| 185 | [] |
| 186 | status=0 |
| 187 | ## BUG mksh STDOUT: |
| 188 | # patsub disrespects nounset! |
| 189 | [] |
| 190 | status=0 |
| 191 | [] |
| 192 | ## status: 1 |
| 193 | ## BUG mksh status: 0 |
| 194 | ## N-I dash status: 2 |
| 195 | ## N-I dash stdout-json: "" |
| 196 | |
| 197 | #### Global Pattern replacement with / |
| 198 | s=xx_xx_xx |
| 199 | echo ${s/xx?/yy_} ${s//xx?/yy_} |
| 200 | ## stdout: yy_xx_xx yy_yy_xx |
| 201 | ## N-I dash status: 2 |
| 202 | ## N-I dash stdout-json: "" |
| 203 | |
| 204 | #### Left Anchored Pattern replacement with # |
| 205 | s=xx_xx_xx |
| 206 | echo ${s/?xx/_yy} ${s/#?xx/_yy} |
| 207 | ## stdout: xx_yy_xx xx_xx_xx |
| 208 | ## N-I dash status: 2 |
| 209 | ## N-I dash stdout-json: "" |
| 210 | |
| 211 | #### Right Anchored Pattern replacement with % |
| 212 | s=xx_xx_xx |
| 213 | echo ${s/?xx/_yy} ${s/%?xx/_yy} |
| 214 | ## stdout: xx_yy_xx xx_xx_yy |
| 215 | ## N-I dash status: 2 |
| 216 | ## N-I dash stdout-json: "" |
| 217 | |
| 218 | #### Replace fixed strings |
| 219 | s=xx_xx |
| 220 | echo ${s/xx/yy} ${s//xx/yy} ${s/#xx/yy} ${s/%xx/yy} |
| 221 | ## stdout: yy_xx yy_yy yy_xx xx_yy |
| 222 | ## N-I dash status: 2 |
| 223 | ## N-I dash stdout-json: "" |
| 224 | |
| 225 | #### Replace is longest match |
| 226 | # If it were shortest, then you would just replace the first <html> |
| 227 | s='begin <html></html> end' |
| 228 | echo ${s/<*>/[]} |
| 229 | ## stdout: begin [] end |
| 230 | ## N-I dash status: 2 |
| 231 | ## N-I dash stdout-json: "" |
| 232 | |
| 233 | #### Replace char class |
| 234 | s=xx_xx_xx |
| 235 | echo ${s//[[:alpha:]]/y} ${s//[^[:alpha:]]/-} |
| 236 | ## stdout: yy_yy_yy xx-xx-xx |
| 237 | ## N-I mksh stdout: xx_xx_xx xx_xx_xx |
| 238 | ## N-I dash status: 2 |
| 239 | ## N-I dash stdout-json: "" |
| 240 | |
| 241 | #### Replace hard glob |
| 242 | s='aa*bb+cc' |
| 243 | echo ${s//\**+/__} # Literal *, then any sequence of characters, then literal + |
| 244 | ## stdout: aa__cc |
| 245 | ## N-I dash status: 2 |
| 246 | ## N-I dash stdout-json: "" |
| 247 | |
| 248 | #### Pattern replacement ${v/} is not valid |
| 249 | v=abcde |
| 250 | echo -${v/}- |
| 251 | echo status=$? |
| 252 | ## status: 2 |
| 253 | ## stdout-json: "" |
| 254 | ## N-I dash status: 2 |
| 255 | ## N-I dash stdout-json: "" |
| 256 | ## BUG bash/mksh status: 0 |
| 257 | ## BUG bash/mksh stdout-json: "-abcde-\nstatus=0\n" |
| 258 | |
| 259 | #### Pattern replacement ${v//} is not valid |
| 260 | v='a/b/c' |
| 261 | echo -${v//}- |
| 262 | echo status=$? |
| 263 | ## status: 2 |
| 264 | ## stdout-json: "" |
| 265 | ## N-I dash status: 2 |
| 266 | ## N-I dash stdout-json: "" |
| 267 | ## BUG bash/mksh status: 0 |
| 268 | ## BUG bash/mksh stdout-json: "-a/b/c-\nstatus=0\n" |
| 269 | |
| 270 | #### ${v/a} is the same as ${v/a/} -- no replacement string |
| 271 | v='aabb' |
| 272 | echo ${v/a} |
| 273 | echo status=$? |
| 274 | ## stdout-json: "abb\nstatus=0\n" |
| 275 | ## N-I dash stdout-json: "" |
| 276 | ## N-I dash status: 2 |
| 277 | |
| 278 | #### String slice |
| 279 | foo=abcdefg |
| 280 | echo ${foo:1:3} |
| 281 | ## stdout: bcd |
| 282 | ## N-I dash status: 2 |
| 283 | ## N-I dash stdout-json: "" |
| 284 | |
| 285 | #### Negative string slice |
| 286 | foo=abcdefg |
| 287 | echo ${foo: -4:3} |
| 288 | ## stdout: def |
| 289 | ## N-I dash status: 2 |
| 290 | ## N-I dash stdout-json: "" |
| 291 | |
| 292 | #### String slice with math |
| 293 | # I think this is the $(()) language inside? |
| 294 | i=1 |
| 295 | foo=abcdefg |
| 296 | echo ${foo: i-3-2 : i + 2} |
| 297 | ## stdout: def |
| 298 | ## N-I dash status: 2 |
| 299 | ## N-I dash stdout-json: "" |
| 300 | |
| 301 | #### Slice String with Unicode |
| 302 | # mksh slices by bytes. |
| 303 | foo='--μ--' |
| 304 | echo ${foo:1:3} |
| 305 | ## stdout: -μ- |
| 306 | ## BUG mksh stdout: -μ |
| 307 | ## N-I dash status: 2 |
| 308 | ## N-I dash stdout-json: "" |