| 1 | Guile NEWS --- history of user-visible changes. -*- text -*- |
| 2 | Copyright (C) 1996, 1997 Free Software Foundation, Inc. |
| 3 | See the end for copying conditions. |
| 4 | |
| 5 | Please send Guile bug reports to bug-guile@gnu.org. |
| 6 | \f |
| 7 | Changes since Guile 1.2: |
| 8 | |
| 9 | * Changes to the distribution |
| 10 | |
| 11 | ** We renamed the SCHEME_LOAD_PATH environment variable to GUILE_LOAD_PATH. |
| 12 | To avoid conflicts, programs should name environment variables after |
| 13 | themselves, except when there's a common practice establishing some |
| 14 | other convention. |
| 15 | |
| 16 | For now, Guile supports both GUILE_LOAD_PATH and SCHEME_LOAD_PATH, |
| 17 | giving the former precedence, and printing a warning message if the |
| 18 | latter is set. Guile 1.4 will not recognize SCHEME_LOAD_PATH at all. |
| 19 | |
| 20 | ** The header files related to multi-byte characters have been removed. |
| 21 | They were: libguile/extchrs.h and libguile/mbstrings.h. Any C code |
| 22 | which referred to these explicitly will probably need to be rewritten, |
| 23 | since the support for the variant string types has been removed; see |
| 24 | below. |
| 25 | |
| 26 | ** The header files append.h and sequences.h have been removed. These |
| 27 | files implemented non-R4RS operations which would encourage |
| 28 | non-portable programming style and less easy-to-read code. |
| 29 | |
| 30 | * Changes to the stand-alone interpreter |
| 31 | |
| 32 | ** New procedures have been added to implement a "batch mode": |
| 33 | |
| 34 | *** Function: batch-mode? |
| 35 | |
| 36 | Returns a boolean indicating whether the interpreter is in batch |
| 37 | mode. |
| 38 | |
| 39 | *** Function: set-batch-mode?! ARG |
| 40 | |
| 41 | If ARG is true, switches the interpreter to batch mode. The `#f' |
| 42 | case has not been implemented. |
| 43 | |
| 44 | ** Guile now provides full command-line editing, when run interactively. |
| 45 | To use this feature, you must have the readline library installed. |
| 46 | The Guile build process will notice it, and automatically include |
| 47 | support for it. |
| 48 | |
| 49 | The readline library is available via anonymous FTP from any GNU |
| 50 | mirror site; the canonical location is "ftp://prep.ai.mit.edu/pub/gnu". |
| 51 | |
| 52 | ** the-last-stack is now a fluid. |
| 53 | |
| 54 | * Changes to the procedure for linking libguile with your programs |
| 55 | |
| 56 | ** You can now use the `guile-config' utility to build programs that use Guile. |
| 57 | |
| 58 | Guile now includes a command-line utility called `guile-config', which |
| 59 | can provide information about how to compile and link programs that |
| 60 | use Guile. |
| 61 | |
| 62 | *** `guile-config compile' prints any C compiler flags needed to use Guile. |
| 63 | You should include this command's output on the command line you use |
| 64 | to compile C or C++ code that #includes the Guile header files. It's |
| 65 | usually just a `-I' flag to help the compiler find the Guile headers. |
| 66 | |
| 67 | |
| 68 | *** `guile-config link' prints any linker flags necessary to link with Guile. |
| 69 | |
| 70 | This command writes to its standard output a list of flags which you |
| 71 | must pass to the linker to link your code against the Guile library. |
| 72 | The flags include '-lguile' itself, any other libraries the Guile |
| 73 | library depends upon, and any `-L' flags needed to help the linker |
| 74 | find those libraries. |
| 75 | |
| 76 | For example, here is a Makefile rule that builds a program named 'foo' |
| 77 | from the object files ${FOO_OBJECTS}, and links them against Guile: |
| 78 | |
| 79 | foo: ${FOO_OBJECTS} |
| 80 | ${CC} ${CFLAGS} ${FOO_OBJECTS} `guile-config link` -o foo |
| 81 | |
| 82 | Previous Guile releases recommended that you use autoconf to detect |
| 83 | which of a predefined set of libraries were present on your system. |
| 84 | It is more robust to use `guile-config', since it records exactly which |
| 85 | libraries the installed Guile library requires. |
| 86 | |
| 87 | This was originally called `build-guile', but was renamed to |
| 88 | `guile-config' before Guile 1.3 was released, to be consistent with |
| 89 | the analogous script for the GTK+ GUI toolkit, which is called |
| 90 | `gtk-config'. |
| 91 | |
| 92 | |
| 93 | ** Use the GUILE_FLAGS macro in your configure.in file to find Guile. |
| 94 | |
| 95 | If you are using the GNU autoconf package to configure your program, |
| 96 | you can use the GUILE_FLAGS autoconf macro to call `guile-config' |
| 97 | (described above) and gather the necessary values for use in your |
| 98 | Makefiles. |
| 99 | |
| 100 | The GUILE_FLAGS macro expands to configure script code which runs the |
| 101 | `guile-config' script, to find out where Guile's header files and |
| 102 | libraries are installed. It sets two variables, marked for |
| 103 | substitution, as by AC_SUBST. |
| 104 | |
| 105 | GUILE_CFLAGS --- flags to pass to a C or C++ compiler to build |
| 106 | code that uses Guile header files. This is almost always just a |
| 107 | -I flag. |
| 108 | |
| 109 | GUILE_LDFLAGS --- flags to pass to the linker to link a |
| 110 | program against Guile. This includes `-lguile' for the Guile |
| 111 | library itself, any libraries that Guile itself requires (like |
| 112 | -lqthreads), and so on. It may also include a -L flag to tell the |
| 113 | compiler where to find the libraries. |
| 114 | |
| 115 | GUILE_FLAGS is defined in the file guile.m4, in the top-level |
| 116 | directory of the Guile distribution. You can copy it into your |
| 117 | package's aclocal.m4 file, and then use it in your configure.in file. |
| 118 | |
| 119 | If you are using the `aclocal' program, distributed with GNU automake, |
| 120 | to maintain your aclocal.m4 file, the Guile installation process |
| 121 | installs guile.m4 where aclocal will find it. All you need to do is |
| 122 | use GUILE_FLAGS in your configure.in file, and then run `aclocal'; |
| 123 | this will copy the definition of GUILE_FLAGS into your aclocal.m4 |
| 124 | file. |
| 125 | |
| 126 | |
| 127 | * Changes to Scheme functions and syntax |
| 128 | |
| 129 | ** Multi-byte strings have been removed, as have multi-byte and wide |
| 130 | ports. We felt that these were the wrong approach to |
| 131 | internationalization support. |
| 132 | |
| 133 | ** New function: readline [PROMPT] |
| 134 | Read a line from the terminal, and allow the user to edit it, |
| 135 | prompting with PROMPT. READLINE provides a large set of Emacs-like |
| 136 | editing commands, lets the user recall previously typed lines, and |
| 137 | works on almost every kind of terminal, including dumb terminals. |
| 138 | |
| 139 | READLINE assumes that the cursor is at the beginning of the line when |
| 140 | it is invoked. Thus, you can't print a prompt yourself, and then call |
| 141 | READLINE; you need to package up your prompt as a string, pass it to |
| 142 | the function, and let READLINE print the prompt itself. This is |
| 143 | because READLINE needs to know the prompt's screen width. |
| 144 | |
| 145 | For Guile to provide this function, you must have the readline library |
| 146 | installed on your system. |
| 147 | |
| 148 | See also ADD-HISTORY function. |
| 149 | |
| 150 | ** New function: add-history STRING |
| 151 | Add STRING as the most recent line in the history used by the READLINE |
| 152 | command. READLINE does not add lines to the history itself; you must |
| 153 | call ADD-HISTORY to make previous input available to the user. |
| 154 | |
| 155 | ** New module (ice-9 getopt-gnu-style): Parse command-line arguments. |
| 156 | |
| 157 | This module provides some simple argument parsing. It exports one |
| 158 | function: |
| 159 | |
| 160 | Function: getopt-gnu-style ARG-LS |
| 161 | Parse a list of program arguments into an alist of option |
| 162 | descriptions. |
| 163 | |
| 164 | Each item in the list of program arguments is examined to see if |
| 165 | it meets the syntax of a GNU long-named option. An argument like |
| 166 | `--MUMBLE' produces an element of the form (MUMBLE . #t) in the |
| 167 | returned alist, where MUMBLE is a keyword object with the same |
| 168 | name as the argument. An argument like `--MUMBLE=FROB' produces |
| 169 | an element of the form (MUMBLE . FROB), where FROB is a string. |
| 170 | |
| 171 | As a special case, the returned alist also contains a pair whose |
| 172 | car is the symbol `rest'. The cdr of this pair is a list |
| 173 | containing all the items in the argument list that are not options |
| 174 | of the form mentioned above. |
| 175 | |
| 176 | The argument `--' is treated specially: all items in the argument |
| 177 | list appearing after such an argument are not examined, and are |
| 178 | returned in the special `rest' list. |
| 179 | |
| 180 | This function does not parse normal single-character switches. |
| 181 | You will need to parse them out of the `rest' list yourself. |
| 182 | |
| 183 | ** macro-eval! is removed. Use local-eval instead. |
| 184 | |
| 185 | ** Some magic has been added to the printer to better handle user |
| 186 | written printing routines (like record printers, closure printers). |
| 187 | |
| 188 | The problem is that these user written routines must have access to |
| 189 | the current `print-state' to be able to handle fancy things like |
| 190 | detection of circular references. These print-states have to be |
| 191 | passed to the builtin printing routines (display, write, etc) to |
| 192 | properly continue the print chain. |
| 193 | |
| 194 | We didn't want to change all existing print code so that it |
| 195 | explicitely passes thru a print state in addition to a port. Instead, |
| 196 | we extented the possible values that the builtin printing routines |
| 197 | accept as a `port'. In addition to a normal port, they now also take |
| 198 | a pair of a normal port and a print-state. Printing will go to the |
| 199 | port and the print-state will be used to control the detection of |
| 200 | circular references, etc. If the builtin function does not care for a |
| 201 | print-state, it is simply ignored. |
| 202 | |
| 203 | User written callbacks are now called with such a pair as their |
| 204 | `port', but because every function now accepts this pair as a PORT |
| 205 | argument, you don't have to worry about that. In fact, it is probably |
| 206 | safest to not check for these pairs. |
| 207 | |
| 208 | However, it is sometimes necessary to continue a print chain on a |
| 209 | different port, for example to get a intermediate string |
| 210 | representation of the printed value, mangle that string somehow, and |
| 211 | then to finally print the mangled string. Use the new function |
| 212 | |
| 213 | inherit-print-state OLD-PORT NEW-PORT |
| 214 | |
| 215 | for this. It constructs a new `port' that prints to NEW-PORT but |
| 216 | inherits the print-state of OLD-PORT. |
| 217 | |
| 218 | ** struct-vtable-offset renamed to vtable-offset-user |
| 219 | |
| 220 | ** New constants: vtable-index-layout, vtable-index-vtable, vtable-index-printer |
| 221 | |
| 222 | ** There is now a fourth (optional) argument to make-vtable-vtable and |
| 223 | make-struct when constructing new types (vtables). This argument |
| 224 | initializes field vtable-index-printer of the vtable. |
| 225 | |
| 226 | ** The detection of circular references has been extended to structs. |
| 227 | That is, a structure that -- in the process of being printed -- prints |
| 228 | itself does not lead to infinite recursion. |
| 229 | |
| 230 | ** There is now some basic support for fluids. Please read |
| 231 | "libguile/fluid.h" to find out more. It is accessible from Scheme with |
| 232 | the following functions and macros: |
| 233 | |
| 234 | Function: make-fluid |
| 235 | |
| 236 | Create a new fluid object. Fluids are not special variables or |
| 237 | some other extension to the semantics of Scheme, but rather |
| 238 | ordinary Scheme objects. You can store them into variables (that |
| 239 | are still lexically scoped, of course) or into any other place you |
| 240 | like. Every fluid has a initial value of `#f'. |
| 241 | |
| 242 | Function: fluid? OBJ |
| 243 | |
| 244 | Test whether OBJ is a fluid. |
| 245 | |
| 246 | Function: fluid-ref FLUID |
| 247 | Function: fluid-set! FLUID VAL |
| 248 | |
| 249 | Access/modify the fluid FLUID. Modifications are only visible |
| 250 | within the current dynamic root (that includes threads). |
| 251 | |
| 252 | Function: with-fluids* FLUIDS VALUES THUNK |
| 253 | |
| 254 | FLUIDS is a list of fluids and VALUES a corresponding list of |
| 255 | values for these fluids. Before THUNK gets called the values are |
| 256 | installed in the fluids and the old values of the fluids are |
| 257 | saved in the VALUES list. When the flow of control leaves THUNK |
| 258 | or reenters it, the values get swapped again. You might think of |
| 259 | this as a `safe-fluid-excursion'. Note that the VALUES list is |
| 260 | modified by `with-fluids*'. |
| 261 | |
| 262 | Macro: with-fluids ((FLUID VALUE) ...) FORM ... |
| 263 | |
| 264 | The same as `with-fluids*' but with a different syntax. It looks |
| 265 | just like `let', but both FLUID and VALUE are evaluated. Remember, |
| 266 | fluids are not special variables but ordinary objects. FLUID |
| 267 | should evaluate to a fluid. |
| 268 | |
| 269 | ** Changes to system call interfaces: |
| 270 | |
| 271 | *** close-port, close-input-port and close-output-port now return a |
| 272 | boolean instead of an `unspecified' object. #t means that the port |
| 273 | was successfully closed, while #f means it was already closed. It is |
| 274 | also now possible for these procedures to raise an exception if an |
| 275 | error occurs (some errors from write can be delayed until close.) |
| 276 | |
| 277 | *** the first argument to chmod, fcntl, ftell and fseek can now be a |
| 278 | file descriptor. |
| 279 | |
| 280 | *** the third argument to fcntl is now optional. |
| 281 | |
| 282 | *** the first argument to chown can now be a file descriptor or a port. |
| 283 | |
| 284 | *** the argument to stat can now be a port. |
| 285 | |
| 286 | *** The following new procedures have been added (most use scsh |
| 287 | interfaces): |
| 288 | |
| 289 | *** procedure: close PORT/FD |
| 290 | Similar to close-port (*note close-port: Closing Ports.), but also |
| 291 | works on file descriptors. A side effect of closing a file |
| 292 | descriptor is that any ports using that file descriptor are moved |
| 293 | to a different file descriptor and have their revealed counts set |
| 294 | to zero. |
| 295 | |
| 296 | *** procedure: port->fdes PORT |
| 297 | Returns the integer file descriptor underlying PORT. As a side |
| 298 | effect the revealed count of PORT is incremented. |
| 299 | |
| 300 | *** procedure: fdes->ports FDES |
| 301 | Returns a list of existing ports which have FDES as an underlying |
| 302 | file descriptor, without changing their revealed counts. |
| 303 | |
| 304 | *** procedure: fdes->inport FDES |
| 305 | Returns an existing input port which has FDES as its underlying |
| 306 | file descriptor, if one exists, and increments its revealed count. |
| 307 | Otherwise, returns a new input port with a revealed count of 1. |
| 308 | |
| 309 | *** procedure: fdes->outport FDES |
| 310 | Returns an existing output port which has FDES as its underlying |
| 311 | file descriptor, if one exists, and increments its revealed count. |
| 312 | Otherwise, returns a new output port with a revealed count of 1. |
| 313 | |
| 314 | The next group of procedures perform a `dup2' system call, if NEWFD |
| 315 | (an integer) is supplied, otherwise a `dup'. The file descriptor to be |
| 316 | duplicated can be supplied as an integer or contained in a port. The |
| 317 | type of value returned varies depending on which procedure is used. |
| 318 | |
| 319 | All procedures also have the side effect when performing `dup2' that |
| 320 | any ports using NEWFD are moved to a different file descriptor and have |
| 321 | their revealed counts set to zero. |
| 322 | |
| 323 | *** procedure: dup->fdes PORT/FD [NEWFD] |
| 324 | Returns an integer file descriptor. |
| 325 | |
| 326 | *** procedure: dup->inport PORT/FD [NEWFD] |
| 327 | Returns a new input port using the new file descriptor. |
| 328 | |
| 329 | *** procedure: dup->outport PORT/FD [NEWFD] |
| 330 | Returns a new output port using the new file descriptor. |
| 331 | |
| 332 | *** procedure: dup PORT/FD [NEWFD] |
| 333 | Returns a new port if PORT/FD is a port, with the same mode as the |
| 334 | supplied port, otherwise returns an integer file descriptor. |
| 335 | |
| 336 | *** procedure: dup->port PORT/FD MODE [NEWFD] |
| 337 | Returns a new port using the new file descriptor. MODE supplies a |
| 338 | mode string for the port (*note open-file: File Ports.). |
| 339 | |
| 340 | *** procedure: setenv NAME VALUE |
| 341 | Modifies the environment of the current process, which is also the |
| 342 | default environment inherited by child processes. |
| 343 | |
| 344 | If VALUE is `#f', then NAME is removed from the environment. |
| 345 | Otherwise, the string NAME=VALUE is added to the environment, |
| 346 | replacing any existing string with name matching NAME. |
| 347 | |
| 348 | The return value is unspecified. |
| 349 | |
| 350 | *** procedure: truncate-file OBJ SIZE |
| 351 | Truncates the file referred to by OBJ to at most SIZE bytes. OBJ |
| 352 | can be a string containing a file name or an integer file |
| 353 | descriptor or port open for output on the file. The underlying |
| 354 | system calls are `truncate' and `ftruncate'. |
| 355 | |
| 356 | The return value is unspecified. |
| 357 | |
| 358 | *** procedure: setvbuf PORT MODE [SIZE] |
| 359 | Set the buffering mode for PORT. MODE can be: |
| 360 | `_IONBF' |
| 361 | non-buffered |
| 362 | |
| 363 | `_IOLBF' |
| 364 | line buffered |
| 365 | |
| 366 | `_IOFBF' |
| 367 | block buffered, using a newly allocated buffer of SIZE bytes. |
| 368 | However if SIZE is zero or unspecified, the port will be made |
| 369 | non-buffered. |
| 370 | |
| 371 | This procedure should not be used after I/O has been performed with |
| 372 | the port. |
| 373 | |
| 374 | Ports are usually block buffered by default, with a default buffer |
| 375 | size. Procedures e.g., *Note open-file: File Ports, which accept a |
| 376 | mode string allow `0' to be added to request an unbuffered port. |
| 377 | |
| 378 | *** procedure: fsync PORT/FD |
| 379 | Copies any unwritten data for the specified output file descriptor |
| 380 | to disk. If PORT/FD is a port, its buffer is flushed before the |
| 381 | underlying file descriptor is fsync'd. The return value is |
| 382 | unspecified. |
| 383 | |
| 384 | *** procedure: open-fdes PATH FLAGS [MODES] |
| 385 | Similar to `open' but returns a file descriptor instead of a port. |
| 386 | |
| 387 | *** procedure: execle PATH ENV [ARG] ... |
| 388 | Similar to `execl', but the environment of the new process is |
| 389 | specified by ENV, which must be a list of strings as returned by |
| 390 | the `environ' procedure. |
| 391 | |
| 392 | This procedure is currently implemented using the `execve' system |
| 393 | call, but we call it `execle' because of its Scheme calling |
| 394 | interface. |
| 395 | |
| 396 | *** procedure: strerror ERRNO |
| 397 | Returns the Unix error message corresponding to ERRNO, an integer. |
| 398 | |
| 399 | *** procedure: primitive-exit [STATUS] |
| 400 | Terminate the current process without unwinding the Scheme stack. |
| 401 | This is would typically be useful after a fork. The exit status |
| 402 | is STATUS if supplied, otherwise zero. |
| 403 | |
| 404 | *** procedure: times |
| 405 | Returns an object with information about real and processor time. |
| 406 | The following procedures accept such an object as an argument and |
| 407 | return a selected component: |
| 408 | |
| 409 | `tms:clock' |
| 410 | The current real time, expressed as time units relative to an |
| 411 | arbitrary base. |
| 412 | |
| 413 | `tms:utime' |
| 414 | The CPU time units used by the calling process. |
| 415 | |
| 416 | `tms:stime' |
| 417 | The CPU time units used by the system on behalf of the |
| 418 | calling process. |
| 419 | |
| 420 | `tms:cutime' |
| 421 | The CPU time units used by terminated child processes of the |
| 422 | calling process, whose status has been collected (e.g., using |
| 423 | `waitpid'). |
| 424 | |
| 425 | `tms:cstime' |
| 426 | Similarly, the CPU times units used by the system on behalf of |
| 427 | terminated child processes. |
| 428 | |
| 429 | ** Removed: list-length |
| 430 | ** Removed: list-append, list-append! |
| 431 | ** Removed: list-reverse, list-reverse! |
| 432 | |
| 433 | ** array-map renamed to array-map! |
| 434 | |
| 435 | ** serial-array-map renamed to serial-array-map! |
| 436 | |
| 437 | ** catch doesn't take #f as first argument any longer |
| 438 | |
| 439 | Previously, it was possible to pass #f instead of a key to `catch'. |
| 440 | That would cause `catch' to pass a jump buffer object to the procedure |
| 441 | passed as second argument. The procedure could then use this jump |
| 442 | buffer objekt as an argument to throw. |
| 443 | |
| 444 | This mechanism has been removed since its utility doesn't motivate the |
| 445 | extra complexity it introduces. |
| 446 | |
| 447 | * Changes to the gh_ interface |
| 448 | |
| 449 | ** The gh_enter function now takes care of loading the Guile startup files. |
| 450 | gh_enter works by calling scm_boot_guile; see the remarks below. |
| 451 | |
| 452 | ** Function: void gh_write (SCM x) |
| 453 | |
| 454 | Write the printed representation of the scheme object x to the current |
| 455 | output port. Corresponds to the scheme level `write'. |
| 456 | |
| 457 | ** gh_list_length renamed to gh_length. |
| 458 | |
| 459 | ** vector handling routines |
| 460 | |
| 461 | Several major changes. In particular, gh_vector() now resembles |
| 462 | (vector ...) (with a caveat -- see manual), and gh_make_vector() now |
| 463 | exists and behaves like (make-vector ...). gh_vset() and gh_vref() |
| 464 | have been renamed gh_vector_set_x() and gh_vector_ref(). Some missing |
| 465 | vector-related gh_ functions have been implemented. |
| 466 | |
| 467 | ** pair and list routines |
| 468 | |
| 469 | Implemented several of the R4RS pair and list functions that were |
| 470 | missing. |
| 471 | |
| 472 | ** gh_scm2doubles, gh_doubles2scm, gh_doubles2dvect |
| 473 | |
| 474 | New function. Converts double arrays back and forth between Scheme |
| 475 | and C. |
| 476 | |
| 477 | * Changes to the scm_ interface |
| 478 | |
| 479 | ** The function scm_boot_guile now takes care of loading the startup files. |
| 480 | |
| 481 | Guile's primary initialization function, scm_boot_guile, now takes |
| 482 | care of loading `boot-9.scm', in the `ice-9' module, to initialize |
| 483 | Guile, define the module system, and put together some standard |
| 484 | bindings. It also loads `init.scm', which is intended to hold |
| 485 | site-specific initialization code. |
| 486 | |
| 487 | Since Guile cannot operate properly until boot-9.scm is loaded, there |
| 488 | is no reason to separate loading boot-9.scm from Guile's other |
| 489 | initialization processes. |
| 490 | |
| 491 | This job used to be done by scm_compile_shell_switches, which didn't |
| 492 | make much sense; in particular, it meant that people using Guile for |
| 493 | non-shell-like applications had to jump through hoops to get Guile |
| 494 | initialized properly. |
| 495 | |
| 496 | ** The function scm_compile_shell_switches no longer loads the startup files. |
| 497 | Now, Guile always loads the startup files, whenever it is initialized; |
| 498 | see the notes above for scm_boot_guile and scm_load_startup_files. |
| 499 | |
| 500 | ** Function: scm_load_startup_files |
| 501 | This new function takes care of loading Guile's initialization file |
| 502 | (`boot-9.scm'), and the site initialization file, `init.scm'. Since |
| 503 | this is always called by the Guile initialization process, it's |
| 504 | probably not too useful to call this yourself, but it's there anyway. |
| 505 | |
| 506 | ** The semantics of smob marking have changed slightly. |
| 507 | |
| 508 | The smob marking function (the `mark' member of the scm_smobfuns |
| 509 | structure) is no longer responsible for setting the mark bit on the |
| 510 | smob. The generic smob handling code in the garbage collector will |
| 511 | set this bit. The mark function need only ensure that any other |
| 512 | objects the smob refers to get marked. |
| 513 | |
| 514 | Note that this change means that the smob's GC8MARK bit is typically |
| 515 | already set upon entry to the mark function. Thus, marking functions |
| 516 | which look like this: |
| 517 | |
| 518 | { |
| 519 | if (SCM_GC8MARKP (ptr)) |
| 520 | return SCM_BOOL_F; |
| 521 | SCM_SETGC8MARK (ptr); |
| 522 | ... mark objects to which the smob refers ... |
| 523 | } |
| 524 | |
| 525 | are now incorrect, since they will return early, and fail to mark any |
| 526 | other objects the smob refers to. Some code in the Guile library used |
| 527 | to work this way. |
| 528 | |
| 529 | ** The semantics of the I/O port functions in scm_ptobfuns have changed. |
| 530 | |
| 531 | If you have implemented your own I/O port type, by writing the |
| 532 | functions required by the scm_ptobfuns and then calling scm_newptob, |
| 533 | you will need to change your functions slightly. |
| 534 | |
| 535 | The functions in a scm_ptobfuns structure now expect the port itself |
| 536 | as their argument; they used to expect the `stream' member of the |
| 537 | port's scm_port_table structure. This allows functions in an |
| 538 | scm_ptobfuns structure to easily access the port's cell (and any flags |
| 539 | it its CAR), and the port's scm_port_table structure. |
| 540 | |
| 541 | Guile now passes the I/O port itself as the `port' argument in the |
| 542 | following scm_ptobfuns functions: |
| 543 | |
| 544 | int (*free) (SCM port); |
| 545 | int (*fputc) (int, SCM port); |
| 546 | int (*fputs) (char *, SCM port); |
| 547 | scm_sizet (*fwrite) SCM_P ((char *ptr, |
| 548 | scm_sizet size, |
| 549 | scm_sizet nitems, |
| 550 | SCM port)); |
| 551 | int (*fflush) (SCM port); |
| 552 | int (*fgetc) (SCM port); |
| 553 | int (*fclose) (SCM port); |
| 554 | |
| 555 | The interfaces to the `mark', `print', `equalp', and `fgets' methods |
| 556 | are unchanged. |
| 557 | |
| 558 | If you have existing code which defines its own port types, it is easy |
| 559 | to convert your code to the new interface; simply apply SCM_STREAM to |
| 560 | the port argument to yield the value you code used to expect. |
| 561 | |
| 562 | Note that since both the port and the stream have the same type in the |
| 563 | C code --- they are both SCM values --- the C compiler will not remind |
| 564 | you if you forget to update your scm_ptobfuns functions. |
| 565 | |
| 566 | |
| 567 | ** Function: int scm_internal_select (int fds, |
| 568 | SELECT_TYPE *rfds, |
| 569 | SELECT_TYPE *wfds, |
| 570 | SELECT_TYPE *efds, |
| 571 | struct timeval *timeout); |
| 572 | |
| 573 | This is a replacement for the `select' function provided by the OS. |
| 574 | It enables I/O blocking and sleeping to happen for one cooperative |
| 575 | thread without blocking other threads. It also avoids busy-loops in |
| 576 | these situations. It is intended that all I/O blocking and sleeping |
| 577 | will finally go through this function. Currently, this function is |
| 578 | only available on systems providing `gettimeofday' and `select'. |
| 579 | |
| 580 | ** Function: SCM scm_internal_stack_catch (SCM tag, |
| 581 | scm_catch_body_t body, |
| 582 | void *body_data, |
| 583 | scm_catch_handler_t handler, |
| 584 | void *handler_data) |
| 585 | |
| 586 | A new sibling to the other two C level `catch' functions |
| 587 | scm_internal_catch and scm_internal_lazy_catch. Use it if you want |
| 588 | the stack to be saved automatically into the variable `the-last-stack' |
| 589 | (scm_the_last_stack_var) on error. This is necessary if you want to |
| 590 | use advanced error reporting, such as calling scm_display_error and |
| 591 | scm_display_backtrace. (They both take a stack object as argument.) |
| 592 | |
| 593 | ** Function: SCM scm_spawn_thread (scm_catch_body_t body, |
| 594 | void *body_data, |
| 595 | scm_catch_handler_t handler, |
| 596 | void *handler_data) |
| 597 | |
| 598 | Spawns a new thread. It does a job similar to |
| 599 | scm_call_with_new_thread but takes arguments more suitable when |
| 600 | spawning threads from application C code. |
| 601 | |
| 602 | ** The hook scm_error_callback has been removed. It was originally |
| 603 | intended as a way for the user to install his own error handler. But |
| 604 | that method works badly since it intervenes between throw and catch, |
| 605 | thereby changing the semantics of expressions like (catch #t ...). |
| 606 | The correct way to do it is to use one of the C level catch functions |
| 607 | in throw.c: scm_internal_catch/lazy_catch/stack_catch. |
| 608 | |
| 609 | ** Removed functions: |
| 610 | |
| 611 | scm_obj_length, scm_list_length, scm_list_append, scm_list_append_x, |
| 612 | scm_list_reverse, scm_list_reverse_x |
| 613 | |
| 614 | ** New macros: SCM_LISTn where n is one of the integers 0-9. |
| 615 | |
| 616 | These can be used for pretty list creation from C. The idea is taken |
| 617 | from Erick Gallesio's STk. |
| 618 | |
| 619 | ** scm_array_map renamed to scm_array_map_x |
| 620 | |
| 621 | ** mbstrings are now removed |
| 622 | |
| 623 | This means that the type codes scm_tc7_mb_string and |
| 624 | scm_tc7_mb_substring has been removed. |
| 625 | |
| 626 | ** The macros SCM_TYP7D and SCM_TYP7SD has been removed. |
| 627 | |
| 628 | ** The macro SCM_TYP7S has taken the role of the old SCM_TYP7D |
| 629 | |
| 630 | SCM_TYP7S now masks away the bit which distinguishes substrings from |
| 631 | strings. |
| 632 | |
| 633 | ** All genio functions changed names and interfaces; new functions are |
| 634 | scm_putc, scm_puts, scm_lfwrite, scm_getc, scm_ungetc, and |
| 635 | scm_do_read_line. |
| 636 | |
| 637 | ** scm_catch_body_t: Backward incompatible change! |
| 638 | |
| 639 | Body functions to scm_internal_catch and friends do not any longer |
| 640 | take a second argument. This is because it is no longer possible to |
| 641 | pass a #f arg to catch. |
| 642 | |
| 643 | ** Calls to scm_protect_object and scm_unprotect now nest properly. |
| 644 | |
| 645 | The function scm_protect_object protects its argument from being freed |
| 646 | by the garbage collector. scm_unprotect_object removes that |
| 647 | protection. |
| 648 | |
| 649 | These functions now nest properly. That is, for every object O, there |
| 650 | is a counter which scm_protect_object(O) increments and |
| 651 | scm_unprotect_object(O) decrements, if the counter is greater than |
| 652 | zero. Every object's counter is zero when it is first created. If an |
| 653 | object's counter is greater than zero, the garbage collector will not |
| 654 | reclaim its storage. |
| 655 | |
| 656 | This allows you to use scm_protect_object in your code without |
| 657 | worrying that some other function you call will call |
| 658 | scm_unprotect_object, and allow it to be freed. Assuming that the |
| 659 | functions you call are well-behaved, and unprotect only those objects |
| 660 | they protect, you can follow the same rule and have confidence that |
| 661 | objects will be freed only at appropriate times. |
| 662 | |
| 663 | \f |
| 664 | Changes in Guile 1.2 (released Tuesday, June 24 1997): |
| 665 | |
| 666 | * Changes to the distribution |
| 667 | |
| 668 | ** Nightly snapshots are now available from ftp.red-bean.com. |
| 669 | The old server, ftp.cyclic.com, has been relinquished to its rightful |
| 670 | owner. |
| 671 | |
| 672 | Nightly snapshots of the Guile development sources are now available via |
| 673 | anonymous FTP from ftp.red-bean.com, as /pub/guile/guile-snap.tar.gz. |
| 674 | |
| 675 | Via the web, that's: ftp://ftp.red-bean.com/pub/guile/guile-snap.tar.gz |
| 676 | For getit, that's: ftp.red-bean.com:/pub/guile/guile-snap.tar.gz |
| 677 | |
| 678 | ** To run Guile without installing it, the procedure has changed a bit. |
| 679 | |
| 680 | If you used a separate build directory to compile Guile, you'll need |
| 681 | to include the build directory in SCHEME_LOAD_PATH, as well as the |
| 682 | source directory. See the `INSTALL' file for examples. |
| 683 | |
| 684 | * Changes to the procedure for linking libguile with your programs |
| 685 | |
| 686 | ** The standard Guile load path for Scheme code now includes |
| 687 | $(datadir)/guile (usually /usr/local/share/guile). This means that |
| 688 | you can install your own Scheme files there, and Guile will find them. |
| 689 | (Previous versions of Guile only checked a directory whose name |
| 690 | contained the Guile version number, so you had to re-install or move |
| 691 | your Scheme sources each time you installed a fresh version of Guile.) |
| 692 | |
| 693 | The load path also includes $(datadir)/guile/site; we recommend |
| 694 | putting individual Scheme files there. If you want to install a |
| 695 | package with multiple source files, create a directory for them under |
| 696 | $(datadir)/guile. |
| 697 | |
| 698 | ** Guile 1.2 will now use the Rx regular expression library, if it is |
| 699 | installed on your system. When you are linking libguile into your own |
| 700 | programs, this means you will have to link against -lguile, -lqt (if |
| 701 | you configured Guile with thread support), and -lrx. |
| 702 | |
| 703 | If you are using autoconf to generate configuration scripts for your |
| 704 | application, the following lines should suffice to add the appropriate |
| 705 | libraries to your link command: |
| 706 | |
| 707 | ### Find Rx, quickthreads and libguile. |
| 708 | AC_CHECK_LIB(rx, main) |
| 709 | AC_CHECK_LIB(qt, main) |
| 710 | AC_CHECK_LIB(guile, scm_shell) |
| 711 | |
| 712 | The Guile 1.2 distribution does not contain sources for the Rx |
| 713 | library, as Guile 1.0 did. If you want to use Rx, you'll need to |
| 714 | retrieve it from a GNU FTP site and install it separately. |
| 715 | |
| 716 | * Changes to Scheme functions and syntax |
| 717 | |
| 718 | ** The dynamic linking features of Guile are now enabled by default. |
| 719 | You can disable them by giving the `--disable-dynamic-linking' option |
| 720 | to configure. |
| 721 | |
| 722 | (dynamic-link FILENAME) |
| 723 | |
| 724 | Find the object file denoted by FILENAME (a string) and link it |
| 725 | into the running Guile application. When everything works out, |
| 726 | return a Scheme object suitable for representing the linked object |
| 727 | file. Otherwise an error is thrown. How object files are |
| 728 | searched is system dependent. |
| 729 | |
| 730 | (dynamic-object? VAL) |
| 731 | |
| 732 | Determine whether VAL represents a dynamically linked object file. |
| 733 | |
| 734 | (dynamic-unlink DYNOBJ) |
| 735 | |
| 736 | Unlink the indicated object file from the application. DYNOBJ |
| 737 | should be one of the values returned by `dynamic-link'. |
| 738 | |
| 739 | (dynamic-func FUNCTION DYNOBJ) |
| 740 | |
| 741 | Search the C function indicated by FUNCTION (a string or symbol) |
| 742 | in DYNOBJ and return some Scheme object that can later be used |
| 743 | with `dynamic-call' to actually call this function. Right now, |
| 744 | these Scheme objects are formed by casting the address of the |
| 745 | function to `long' and converting this number to its Scheme |
| 746 | representation. |
| 747 | |
| 748 | (dynamic-call FUNCTION DYNOBJ) |
| 749 | |
| 750 | Call the C function indicated by FUNCTION and DYNOBJ. The |
| 751 | function is passed no arguments and its return value is ignored. |
| 752 | When FUNCTION is something returned by `dynamic-func', call that |
| 753 | function and ignore DYNOBJ. When FUNCTION is a string (or symbol, |
| 754 | etc.), look it up in DYNOBJ; this is equivalent to |
| 755 | |
| 756 | (dynamic-call (dynamic-func FUNCTION DYNOBJ) #f) |
| 757 | |
| 758 | Interrupts are deferred while the C function is executing (with |
| 759 | SCM_DEFER_INTS/SCM_ALLOW_INTS). |
| 760 | |
| 761 | (dynamic-args-call FUNCTION DYNOBJ ARGS) |
| 762 | |
| 763 | Call the C function indicated by FUNCTION and DYNOBJ, but pass it |
| 764 | some arguments and return its return value. The C function is |
| 765 | expected to take two arguments and return an `int', just like |
| 766 | `main': |
| 767 | |
| 768 | int c_func (int argc, char **argv); |
| 769 | |
| 770 | ARGS must be a list of strings and is converted into an array of |
| 771 | `char *'. The array is passed in ARGV and its size in ARGC. The |
| 772 | return value is converted to a Scheme number and returned from the |
| 773 | call to `dynamic-args-call'. |
| 774 | |
| 775 | When dynamic linking is disabled or not supported on your system, |
| 776 | the above functions throw errors, but they are still available. |
| 777 | |
| 778 | Here is a small example that works on GNU/Linux: |
| 779 | |
| 780 | (define libc-obj (dynamic-link "libc.so")) |
| 781 | (dynamic-args-call 'rand libc-obj '()) |
| 782 | |
| 783 | See the file `libguile/DYNAMIC-LINKING' for additional comments. |
| 784 | |
| 785 | ** The #/ syntax for module names is depreciated, and will be removed |
| 786 | in a future version of Guile. Instead of |
| 787 | |
| 788 | #/foo/bar/baz |
| 789 | |
| 790 | instead write |
| 791 | |
| 792 | (foo bar baz) |
| 793 | |
| 794 | The latter syntax is more consistent with existing Lisp practice. |
| 795 | |
| 796 | ** Guile now does fancier printing of structures. Structures are the |
| 797 | underlying implementation for records, which in turn are used to |
| 798 | implement modules, so all of these object now print differently and in |
| 799 | a more informative way. |
| 800 | |
| 801 | The Scheme printer will examine the builtin variable *struct-printer* |
| 802 | whenever it needs to print a structure object. When this variable is |
| 803 | not `#f' it is deemed to be a procedure and will be applied to the |
| 804 | structure object and the output port. When *struct-printer* is `#f' |
| 805 | or the procedure return `#f' the structure object will be printed in |
| 806 | the boring #<struct 80458270> form. |
| 807 | |
| 808 | This hook is used by some routines in ice-9/boot-9.scm to implement |
| 809 | type specific printing routines. Please read the comments there about |
| 810 | "printing structs". |
| 811 | |
| 812 | One of the more specific uses of structs are records. The printing |
| 813 | procedure that could be passed to MAKE-RECORD-TYPE is now actually |
| 814 | called. It should behave like a *struct-printer* procedure (described |
| 815 | above). |
| 816 | |
| 817 | ** Guile now supports a new R4RS-compliant syntax for keywords. A |
| 818 | token of the form #:NAME, where NAME has the same syntax as a Scheme |
| 819 | symbol, is the external representation of the keyword named NAME. |
| 820 | Keyword objects print using this syntax as well, so values containing |
| 821 | keyword objects can be read back into Guile. When used in an |
| 822 | expression, keywords are self-quoting objects. |
| 823 | |
| 824 | Guile suports this read syntax, and uses this print syntax, regardless |
| 825 | of the current setting of the `keyword' read option. The `keyword' |
| 826 | read option only controls whether Guile recognizes the `:NAME' syntax, |
| 827 | which is incompatible with R4RS. (R4RS says such token represent |
| 828 | symbols.) |
| 829 | |
| 830 | ** Guile has regular expression support again. Guile 1.0 included |
| 831 | functions for matching regular expressions, based on the Rx library. |
| 832 | In Guile 1.1, the Guile/Rx interface was removed to simplify the |
| 833 | distribution, and thus Guile had no regular expression support. Guile |
| 834 | 1.2 again supports the most commonly used functions, and supports all |
| 835 | of SCSH's regular expression functions. |
| 836 | |
| 837 | If your system does not include a POSIX regular expression library, |
| 838 | and you have not linked Guile with a third-party regexp library such as |
| 839 | Rx, these functions will not be available. You can tell whether your |
| 840 | Guile installation includes regular expression support by checking |
| 841 | whether the `*features*' list includes the `regex' symbol. |
| 842 | |
| 843 | *** regexp functions |
| 844 | |
| 845 | By default, Guile supports POSIX extended regular expressions. That |
| 846 | means that the characters `(', `)', `+' and `?' are special, and must |
| 847 | be escaped if you wish to match the literal characters. |
| 848 | |
| 849 | This regular expression interface was modeled after that implemented |
| 850 | by SCSH, the Scheme Shell. It is intended to be upwardly compatible |
| 851 | with SCSH regular expressions. |
| 852 | |
| 853 | **** Function: string-match PATTERN STR [START] |
| 854 | Compile the string PATTERN into a regular expression and compare |
| 855 | it with STR. The optional numeric argument START specifies the |
| 856 | position of STR at which to begin matching. |
| 857 | |
| 858 | `string-match' returns a "match structure" which describes what, |
| 859 | if anything, was matched by the regular expression. *Note Match |
| 860 | Structures::. If STR does not match PATTERN at all, |
| 861 | `string-match' returns `#f'. |
| 862 | |
| 863 | Each time `string-match' is called, it must compile its PATTERN |
| 864 | argument into a regular expression structure. This operation is |
| 865 | expensive, which makes `string-match' inefficient if the same regular |
| 866 | expression is used several times (for example, in a loop). For better |
| 867 | performance, you can compile a regular expression in advance and then |
| 868 | match strings against the compiled regexp. |
| 869 | |
| 870 | **** Function: make-regexp STR [FLAGS] |
| 871 | Compile the regular expression described by STR, and return the |
| 872 | compiled regexp structure. If STR does not describe a legal |
| 873 | regular expression, `make-regexp' throws a |
| 874 | `regular-expression-syntax' error. |
| 875 | |
| 876 | FLAGS may be the bitwise-or of one or more of the following: |
| 877 | |
| 878 | **** Constant: regexp/extended |
| 879 | Use POSIX Extended Regular Expression syntax when interpreting |
| 880 | STR. If not set, POSIX Basic Regular Expression syntax is used. |
| 881 | If the FLAGS argument is omitted, we assume regexp/extended. |
| 882 | |
| 883 | **** Constant: regexp/icase |
| 884 | Do not differentiate case. Subsequent searches using the |
| 885 | returned regular expression will be case insensitive. |
| 886 | |
| 887 | **** Constant: regexp/newline |
| 888 | Match-any-character operators don't match a newline. |
| 889 | |
| 890 | A non-matching list ([^...]) not containing a newline matches a |
| 891 | newline. |
| 892 | |
| 893 | Match-beginning-of-line operator (^) matches the empty string |
| 894 | immediately after a newline, regardless of whether the FLAGS |
| 895 | passed to regexp-exec contain regexp/notbol. |
| 896 | |
| 897 | Match-end-of-line operator ($) matches the empty string |
| 898 | immediately before a newline, regardless of whether the FLAGS |
| 899 | passed to regexp-exec contain regexp/noteol. |
| 900 | |
| 901 | **** Function: regexp-exec REGEXP STR [START [FLAGS]] |
| 902 | Match the compiled regular expression REGEXP against `str'. If |
| 903 | the optional integer START argument is provided, begin matching |
| 904 | from that position in the string. Return a match structure |
| 905 | describing the results of the match, or `#f' if no match could be |
| 906 | found. |
| 907 | |
| 908 | FLAGS may be the bitwise-or of one or more of the following: |
| 909 | |
| 910 | **** Constant: regexp/notbol |
| 911 | The match-beginning-of-line operator always fails to match (but |
| 912 | see the compilation flag regexp/newline above) This flag may be |
| 913 | used when different portions of a string are passed to |
| 914 | regexp-exec and the beginning of the string should not be |
| 915 | interpreted as the beginning of the line. |
| 916 | |
| 917 | **** Constant: regexp/noteol |
| 918 | The match-end-of-line operator always fails to match (but see the |
| 919 | compilation flag regexp/newline above) |
| 920 | |
| 921 | **** Function: regexp? OBJ |
| 922 | Return `#t' if OBJ is a compiled regular expression, or `#f' |
| 923 | otherwise. |
| 924 | |
| 925 | Regular expressions are commonly used to find patterns in one string |
| 926 | and replace them with the contents of another string. |
| 927 | |
| 928 | **** Function: regexp-substitute PORT MATCH [ITEM...] |
| 929 | Write to the output port PORT selected contents of the match |
| 930 | structure MATCH. Each ITEM specifies what should be written, and |
| 931 | may be one of the following arguments: |
| 932 | |
| 933 | * A string. String arguments are written out verbatim. |
| 934 | |
| 935 | * An integer. The submatch with that number is written. |
| 936 | |
| 937 | * The symbol `pre'. The portion of the matched string preceding |
| 938 | the regexp match is written. |
| 939 | |
| 940 | * The symbol `post'. The portion of the matched string |
| 941 | following the regexp match is written. |
| 942 | |
| 943 | PORT may be `#f', in which case nothing is written; instead, |
| 944 | `regexp-substitute' constructs a string from the specified ITEMs |
| 945 | and returns that. |
| 946 | |
| 947 | **** Function: regexp-substitute/global PORT REGEXP TARGET [ITEM...] |
| 948 | Similar to `regexp-substitute', but can be used to perform global |
| 949 | substitutions on STR. Instead of taking a match structure as an |
| 950 | argument, `regexp-substitute/global' takes two string arguments: a |
| 951 | REGEXP string describing a regular expression, and a TARGET string |
| 952 | which should be matched against this regular expression. |
| 953 | |
| 954 | Each ITEM behaves as in REGEXP-SUBSTITUTE, with the following |
| 955 | exceptions: |
| 956 | |
| 957 | * A function may be supplied. When this function is called, it |
| 958 | will be passed one argument: a match structure for a given |
| 959 | regular expression match. It should return a string to be |
| 960 | written out to PORT. |
| 961 | |
| 962 | * The `post' symbol causes `regexp-substitute/global' to recurse |
| 963 | on the unmatched portion of STR. This *must* be supplied in |
| 964 | order to perform global search-and-replace on STR; if it is |
| 965 | not present among the ITEMs, then `regexp-substitute/global' |
| 966 | will return after processing a single match. |
| 967 | |
| 968 | *** Match Structures |
| 969 | |
| 970 | A "match structure" is the object returned by `string-match' and |
| 971 | `regexp-exec'. It describes which portion of a string, if any, matched |
| 972 | the given regular expression. Match structures include: a reference to |
| 973 | the string that was checked for matches; the starting and ending |
| 974 | positions of the regexp match; and, if the regexp included any |
| 975 | parenthesized subexpressions, the starting and ending positions of each |
| 976 | submatch. |
| 977 | |
| 978 | In each of the regexp match functions described below, the `match' |
| 979 | argument must be a match structure returned by a previous call to |
| 980 | `string-match' or `regexp-exec'. Most of these functions return some |
| 981 | information about the original target string that was matched against a |
| 982 | regular expression; we will call that string TARGET for easy reference. |
| 983 | |
| 984 | **** Function: regexp-match? OBJ |
| 985 | Return `#t' if OBJ is a match structure returned by a previous |
| 986 | call to `regexp-exec', or `#f' otherwise. |
| 987 | |
| 988 | **** Function: match:substring MATCH [N] |
| 989 | Return the portion of TARGET matched by subexpression number N. |
| 990 | Submatch 0 (the default) represents the entire regexp match. If |
| 991 | the regular expression as a whole matched, but the subexpression |
| 992 | number N did not match, return `#f'. |
| 993 | |
| 994 | **** Function: match:start MATCH [N] |
| 995 | Return the starting position of submatch number N. |
| 996 | |
| 997 | **** Function: match:end MATCH [N] |
| 998 | Return the ending position of submatch number N. |
| 999 | |
| 1000 | **** Function: match:prefix MATCH |
| 1001 | Return the unmatched portion of TARGET preceding the regexp match. |
| 1002 | |
| 1003 | **** Function: match:suffix MATCH |
| 1004 | Return the unmatched portion of TARGET following the regexp match. |
| 1005 | |
| 1006 | **** Function: match:count MATCH |
| 1007 | Return the number of parenthesized subexpressions from MATCH. |
| 1008 | Note that the entire regular expression match itself counts as a |
| 1009 | subexpression, and failed submatches are included in the count. |
| 1010 | |
| 1011 | **** Function: match:string MATCH |
| 1012 | Return the original TARGET string. |
| 1013 | |
| 1014 | *** Backslash Escapes |
| 1015 | |
| 1016 | Sometimes you will want a regexp to match characters like `*' or `$' |
| 1017 | exactly. For example, to check whether a particular string represents |
| 1018 | a menu entry from an Info node, it would be useful to match it against |
| 1019 | a regexp like `^* [^:]*::'. However, this won't work; because the |
| 1020 | asterisk is a metacharacter, it won't match the `*' at the beginning of |
| 1021 | the string. In this case, we want to make the first asterisk un-magic. |
| 1022 | |
| 1023 | You can do this by preceding the metacharacter with a backslash |
| 1024 | character `\'. (This is also called "quoting" the metacharacter, and |
| 1025 | is known as a "backslash escape".) When Guile sees a backslash in a |
| 1026 | regular expression, it considers the following glyph to be an ordinary |
| 1027 | character, no matter what special meaning it would ordinarily have. |
| 1028 | Therefore, we can make the above example work by changing the regexp to |
| 1029 | `^\* [^:]*::'. The `\*' sequence tells the regular expression engine |
| 1030 | to match only a single asterisk in the target string. |
| 1031 | |
| 1032 | Since the backslash is itself a metacharacter, you may force a |
| 1033 | regexp to match a backslash in the target string by preceding the |
| 1034 | backslash with itself. For example, to find variable references in a |
| 1035 | TeX program, you might want to find occurrences of the string `\let\' |
| 1036 | followed by any number of alphabetic characters. The regular expression |
| 1037 | `\\let\\[A-Za-z]*' would do this: the double backslashes in the regexp |
| 1038 | each match a single backslash in the target string. |
| 1039 | |
| 1040 | **** Function: regexp-quote STR |
| 1041 | Quote each special character found in STR with a backslash, and |
| 1042 | return the resulting string. |
| 1043 | |
| 1044 | *Very important:* Using backslash escapes in Guile source code (as |
| 1045 | in Emacs Lisp or C) can be tricky, because the backslash character has |
| 1046 | special meaning for the Guile reader. For example, if Guile encounters |
| 1047 | the character sequence `\n' in the middle of a string while processing |
| 1048 | Scheme code, it replaces those characters with a newline character. |
| 1049 | Similarly, the character sequence `\t' is replaced by a horizontal tab. |
| 1050 | Several of these "escape sequences" are processed by the Guile reader |
| 1051 | before your code is executed. Unrecognized escape sequences are |
| 1052 | ignored: if the characters `\*' appear in a string, they will be |
| 1053 | translated to the single character `*'. |
| 1054 | |
| 1055 | This translation is obviously undesirable for regular expressions, |
| 1056 | since we want to be able to include backslashes in a string in order to |
| 1057 | escape regexp metacharacters. Therefore, to make sure that a backslash |
| 1058 | is preserved in a string in your Guile program, you must use *two* |
| 1059 | consecutive backslashes: |
| 1060 | |
| 1061 | (define Info-menu-entry-pattern (make-regexp "^\\* [^:]*")) |
| 1062 | |
| 1063 | The string in this example is preprocessed by the Guile reader before |
| 1064 | any code is executed. The resulting argument to `make-regexp' is the |
| 1065 | string `^\* [^:]*', which is what we really want. |
| 1066 | |
| 1067 | This also means that in order to write a regular expression that |
| 1068 | matches a single backslash character, the regular expression string in |
| 1069 | the source code must include *four* backslashes. Each consecutive pair |
| 1070 | of backslashes gets translated by the Guile reader to a single |
| 1071 | backslash, and the resulting double-backslash is interpreted by the |
| 1072 | regexp engine as matching a single backslash character. Hence: |
| 1073 | |
| 1074 | (define tex-variable-pattern (make-regexp "\\\\let\\\\=[A-Za-z]*")) |
| 1075 | |
| 1076 | The reason for the unwieldiness of this syntax is historical. Both |
| 1077 | regular expression pattern matchers and Unix string processing systems |
| 1078 | have traditionally used backslashes with the special meanings described |
| 1079 | above. The POSIX regular expression specification and ANSI C standard |
| 1080 | both require these semantics. Attempting to abandon either convention |
| 1081 | would cause other kinds of compatibility problems, possibly more severe |
| 1082 | ones. Therefore, without extending the Scheme reader to support |
| 1083 | strings with different quoting conventions (an ungainly and confusing |
| 1084 | extension when implemented in other languages), we must adhere to this |
| 1085 | cumbersome escape syntax. |
| 1086 | |
| 1087 | * Changes to the gh_ interface |
| 1088 | |
| 1089 | * Changes to the scm_ interface |
| 1090 | |
| 1091 | * Changes to system call interfaces: |
| 1092 | |
| 1093 | ** The value returned by `raise' is now unspecified. It throws an exception |
| 1094 | if an error occurs. |
| 1095 | |
| 1096 | *** A new procedure `sigaction' can be used to install signal handlers |
| 1097 | |
| 1098 | (sigaction signum [action] [flags]) |
| 1099 | |
| 1100 | signum is the signal number, which can be specified using the value |
| 1101 | of SIGINT etc. |
| 1102 | |
| 1103 | If action is omitted, sigaction returns a pair: the CAR is the current |
| 1104 | signal hander, which will be either an integer with the value SIG_DFL |
| 1105 | (default action) or SIG_IGN (ignore), or the Scheme procedure which |
| 1106 | handles the signal, or #f if a non-Scheme procedure handles the |
| 1107 | signal. The CDR contains the current sigaction flags for the handler. |
| 1108 | |
| 1109 | If action is provided, it is installed as the new handler for signum. |
| 1110 | action can be a Scheme procedure taking one argument, or the value of |
| 1111 | SIG_DFL (default action) or SIG_IGN (ignore), or #f to restore |
| 1112 | whatever signal handler was installed before sigaction was first used. |
| 1113 | Flags can optionally be specified for the new handler (SA_RESTART is |
| 1114 | always used if the system provides it, so need not be specified.) The |
| 1115 | return value is a pair with information about the old handler as |
| 1116 | described above. |
| 1117 | |
| 1118 | This interface does not provide access to the "signal blocking" |
| 1119 | facility. Maybe this is not needed, since the thread support may |
| 1120 | provide solutions to the problem of consistent access to data |
| 1121 | structures. |
| 1122 | |
| 1123 | *** A new procedure `flush-all-ports' is equivalent to running |
| 1124 | `force-output' on every port open for output. |
| 1125 | |
| 1126 | ** Guile now provides information on how it was built, via the new |
| 1127 | global variable, %guile-build-info. This variable records the values |
| 1128 | of the standard GNU makefile directory variables as an assocation |
| 1129 | list, mapping variable names (symbols) onto directory paths (strings). |
| 1130 | For example, to find out where the Guile link libraries were |
| 1131 | installed, you can say: |
| 1132 | |
| 1133 | guile -c "(display (assq-ref %guile-build-info 'libdir)) (newline)" |
| 1134 | |
| 1135 | |
| 1136 | * Changes to the scm_ interface |
| 1137 | |
| 1138 | ** The new function scm_handle_by_message_noexit is just like the |
| 1139 | existing scm_handle_by_message function, except that it doesn't call |
| 1140 | exit to terminate the process. Instead, it prints a message and just |
| 1141 | returns #f. This might be a more appropriate catch-all handler for |
| 1142 | new dynamic roots and threads. |
| 1143 | |
| 1144 | \f |
| 1145 | Changes in Guile 1.1 (released Friday, May 16 1997): |
| 1146 | |
| 1147 | * Changes to the distribution. |
| 1148 | |
| 1149 | The Guile 1.0 distribution has been split up into several smaller |
| 1150 | pieces: |
| 1151 | guile-core --- the Guile interpreter itself. |
| 1152 | guile-tcltk --- the interface between the Guile interpreter and |
| 1153 | Tcl/Tk; Tcl is an interpreter for a stringy language, and Tk |
| 1154 | is a toolkit for building graphical user interfaces. |
| 1155 | guile-rgx-ctax --- the interface between Guile and the Rx regular |
| 1156 | expression matcher, and the translator for the Ctax |
| 1157 | programming language. These are packaged together because the |
| 1158 | Ctax translator uses Rx to parse Ctax source code. |
| 1159 | |
| 1160 | This NEWS file describes the changes made to guile-core since the 1.0 |
| 1161 | release. |
| 1162 | |
| 1163 | We no longer distribute the documentation, since it was either out of |
| 1164 | date, or incomplete. As soon as we have current documentation, we |
| 1165 | will distribute it. |
| 1166 | |
| 1167 | |
| 1168 | |
| 1169 | * Changes to the stand-alone interpreter |
| 1170 | |
| 1171 | ** guile now accepts command-line arguments compatible with SCSH, Olin |
| 1172 | Shivers' Scheme Shell. |
| 1173 | |
| 1174 | In general, arguments are evaluated from left to right, but there are |
| 1175 | exceptions. The following switches stop argument processing, and |
| 1176 | stash all remaining command-line arguments as the value returned by |
| 1177 | the (command-line) function. |
| 1178 | -s SCRIPT load Scheme source code from FILE, and exit |
| 1179 | -c EXPR evalute Scheme expression EXPR, and exit |
| 1180 | -- stop scanning arguments; run interactively |
| 1181 | |
| 1182 | The switches below are processed as they are encountered. |
| 1183 | -l FILE load Scheme source code from FILE |
| 1184 | -e FUNCTION after reading script, apply FUNCTION to |
| 1185 | command line arguments |
| 1186 | -ds do -s script at this point |
| 1187 | --emacs enable Emacs protocol (experimental) |
| 1188 | -h, --help display this help and exit |
| 1189 | -v, --version display version information and exit |
| 1190 | \ read arguments from following script lines |
| 1191 | |
| 1192 | So, for example, here is a Guile script named `ekko' (thanks, Olin) |
| 1193 | which re-implements the traditional "echo" command: |
| 1194 | |
| 1195 | #!/usr/local/bin/guile -s |
| 1196 | !# |
| 1197 | (define (main args) |
| 1198 | (map (lambda (arg) (display arg) (display " ")) |
| 1199 | (cdr args)) |
| 1200 | (newline)) |
| 1201 | |
| 1202 | (main (command-line)) |
| 1203 | |
| 1204 | Suppose we invoke this script as follows: |
| 1205 | |
| 1206 | ekko a speckled gecko |
| 1207 | |
| 1208 | Through the magic of Unix script processing (triggered by the `#!' |
| 1209 | token at the top of the file), /usr/local/bin/guile receives the |
| 1210 | following list of command-line arguments: |
| 1211 | |
| 1212 | ("-s" "./ekko" "a" "speckled" "gecko") |
| 1213 | |
| 1214 | Unix inserts the name of the script after the argument specified on |
| 1215 | the first line of the file (in this case, "-s"), and then follows that |
| 1216 | with the arguments given to the script. Guile loads the script, which |
| 1217 | defines the `main' function, and then applies it to the list of |
| 1218 | remaining command-line arguments, ("a" "speckled" "gecko"). |
| 1219 | |
| 1220 | In Unix, the first line of a script file must take the following form: |
| 1221 | |
| 1222 | #!INTERPRETER ARGUMENT |
| 1223 | |
| 1224 | where INTERPRETER is the absolute filename of the interpreter |
| 1225 | executable, and ARGUMENT is a single command-line argument to pass to |
| 1226 | the interpreter. |
| 1227 | |
| 1228 | You may only pass one argument to the interpreter, and its length is |
| 1229 | limited. These restrictions can be annoying to work around, so Guile |
| 1230 | provides a general mechanism (borrowed from, and compatible with, |
| 1231 | SCSH) for circumventing them. |
| 1232 | |
| 1233 | If the ARGUMENT in a Guile script is a single backslash character, |
| 1234 | `\', Guile will open the script file, parse arguments from its second |
| 1235 | and subsequent lines, and replace the `\' with them. So, for example, |
| 1236 | here is another implementation of the `ekko' script: |
| 1237 | |
| 1238 | #!/usr/local/bin/guile \ |
| 1239 | -e main -s |
| 1240 | !# |
| 1241 | (define (main args) |
| 1242 | (for-each (lambda (arg) (display arg) (display " ")) |
| 1243 | (cdr args)) |
| 1244 | (newline)) |
| 1245 | |
| 1246 | If the user invokes this script as follows: |
| 1247 | |
| 1248 | ekko a speckled gecko |
| 1249 | |
| 1250 | Unix expands this into |
| 1251 | |
| 1252 | /usr/local/bin/guile \ ekko a speckled gecko |
| 1253 | |
| 1254 | When Guile sees the `\' argument, it replaces it with the arguments |
| 1255 | read from the second line of the script, producing: |
| 1256 | |
| 1257 | /usr/local/bin/guile -e main -s ekko a speckled gecko |
| 1258 | |
| 1259 | This tells Guile to load the `ekko' script, and apply the function |
| 1260 | `main' to the argument list ("a" "speckled" "gecko"). |
| 1261 | |
| 1262 | Here is how Guile parses the command-line arguments: |
| 1263 | - Each space character terminates an argument. This means that two |
| 1264 | spaces in a row introduce an empty-string argument. |
| 1265 | - The tab character is not permitted (unless you quote it with the |
| 1266 | backslash character, as described below), to avoid confusion. |
| 1267 | - The newline character terminates the sequence of arguments, and will |
| 1268 | also terminate a final non-empty argument. (However, a newline |
| 1269 | following a space will not introduce a final empty-string argument; |
| 1270 | it only terminates the argument list.) |
| 1271 | - The backslash character is the escape character. It escapes |
| 1272 | backslash, space, tab, and newline. The ANSI C escape sequences |
| 1273 | like \n and \t are also supported. These produce argument |
| 1274 | constituents; the two-character combination \n doesn't act like a |
| 1275 | terminating newline. The escape sequence \NNN for exactly three |
| 1276 | octal digits reads as the character whose ASCII code is NNN. As |
| 1277 | above, characters produced this way are argument constituents. |
| 1278 | Backslash followed by other characters is not allowed. |
| 1279 | |
| 1280 | * Changes to the procedure for linking libguile with your programs |
| 1281 | |
| 1282 | ** Guile now builds and installs a shared guile library, if your |
| 1283 | system support shared libraries. (It still builds a static library on |
| 1284 | all systems.) Guile automatically detects whether your system |
| 1285 | supports shared libraries. To prevent Guile from buildisg shared |
| 1286 | libraries, pass the `--disable-shared' flag to the configure script. |
| 1287 | |
| 1288 | Guile takes longer to compile when it builds shared libraries, because |
| 1289 | it must compile every file twice --- once to produce position- |
| 1290 | independent object code, and once to produce normal object code. |
| 1291 | |
| 1292 | ** The libthreads library has been merged into libguile. |
| 1293 | |
| 1294 | To link a program against Guile, you now need only link against |
| 1295 | -lguile and -lqt; -lthreads is no longer needed. If you are using |
| 1296 | autoconf to generate configuration scripts for your application, the |
| 1297 | following lines should suffice to add the appropriate libraries to |
| 1298 | your link command: |
| 1299 | |
| 1300 | ### Find quickthreads and libguile. |
| 1301 | AC_CHECK_LIB(qt, main) |
| 1302 | AC_CHECK_LIB(guile, scm_shell) |
| 1303 | |
| 1304 | * Changes to Scheme functions |
| 1305 | |
| 1306 | ** Guile Scheme's special syntax for keyword objects is now optional, |
| 1307 | and disabled by default. |
| 1308 | |
| 1309 | The syntax variation from R4RS made it difficult to port some |
| 1310 | interesting packages to Guile. The routines which accepted keyword |
| 1311 | arguments (mostly in the module system) have been modified to also |
| 1312 | accept symbols whose names begin with `:'. |
| 1313 | |
| 1314 | To change the keyword syntax, you must first import the (ice-9 debug) |
| 1315 | module: |
| 1316 | (use-modules (ice-9 debug)) |
| 1317 | |
| 1318 | Then you can enable the keyword syntax as follows: |
| 1319 | (read-set! keywords 'prefix) |
| 1320 | |
| 1321 | To disable keyword syntax, do this: |
| 1322 | (read-set! keywords #f) |
| 1323 | |
| 1324 | ** Many more primitive functions accept shared substrings as |
| 1325 | arguments. In the past, these functions required normal, mutable |
| 1326 | strings as arguments, although they never made use of this |
| 1327 | restriction. |
| 1328 | |
| 1329 | ** The uniform array functions now operate on byte vectors. These |
| 1330 | functions are `array-fill!', `serial-array-copy!', `array-copy!', |
| 1331 | `serial-array-map', `array-map', `array-for-each', and |
| 1332 | `array-index-map!'. |
| 1333 | |
| 1334 | ** The new functions `trace' and `untrace' implement simple debugging |
| 1335 | support for Scheme functions. |
| 1336 | |
| 1337 | The `trace' function accepts any number of procedures as arguments, |
| 1338 | and tells the Guile interpreter to display each procedure's name and |
| 1339 | arguments each time the procedure is invoked. When invoked with no |
| 1340 | arguments, `trace' returns the list of procedures currently being |
| 1341 | traced. |
| 1342 | |
| 1343 | The `untrace' function accepts any number of procedures as arguments, |
| 1344 | and tells the Guile interpreter not to trace them any more. When |
| 1345 | invoked with no arguments, `untrace' untraces all curretly traced |
| 1346 | procedures. |
| 1347 | |
| 1348 | The tracing in Guile has an advantage over most other systems: we |
| 1349 | don't create new procedure objects, but mark the procedure objects |
| 1350 | themselves. This means that anonymous and internal procedures can be |
| 1351 | traced. |
| 1352 | |
| 1353 | ** The function `assert-repl-prompt' has been renamed to |
| 1354 | `set-repl-prompt!'. It takes one argument, PROMPT. |
| 1355 | - If PROMPT is #f, the Guile read-eval-print loop will not prompt. |
| 1356 | - If PROMPT is a string, we use it as a prompt. |
| 1357 | - If PROMPT is a procedure accepting no arguments, we call it, and |
| 1358 | display the result as a prompt. |
| 1359 | - Otherwise, we display "> ". |
| 1360 | |
| 1361 | ** The new function `eval-string' reads Scheme expressions from a |
| 1362 | string and evaluates them, returning the value of the last expression |
| 1363 | in the string. If the string contains no expressions, it returns an |
| 1364 | unspecified value. |
| 1365 | |
| 1366 | ** The new function `thunk?' returns true iff its argument is a |
| 1367 | procedure of zero arguments. |
| 1368 | |
| 1369 | ** `defined?' is now a builtin function, instead of syntax. This |
| 1370 | means that its argument should be quoted. It returns #t iff its |
| 1371 | argument is bound in the current module. |
| 1372 | |
| 1373 | ** The new syntax `use-modules' allows you to add new modules to your |
| 1374 | environment without re-typing a complete `define-module' form. It |
| 1375 | accepts any number of module names as arguments, and imports their |
| 1376 | public bindings into the current module. |
| 1377 | |
| 1378 | ** The new function (module-defined? NAME MODULE) returns true iff |
| 1379 | NAME, a symbol, is defined in MODULE, a module object. |
| 1380 | |
| 1381 | ** The new function `builtin-bindings' creates and returns a hash |
| 1382 | table containing copies of all the root module's bindings. |
| 1383 | |
| 1384 | ** The new function `builtin-weak-bindings' does the same as |
| 1385 | `builtin-bindings', but creates a doubly-weak hash table. |
| 1386 | |
| 1387 | ** The `equal?' function now considers variable objects to be |
| 1388 | equivalent if they have the same name and the same value. |
| 1389 | |
| 1390 | ** The new function `command-line' returns the command-line arguments |
| 1391 | given to Guile, as a list of strings. |
| 1392 | |
| 1393 | When using guile as a script interpreter, `command-line' returns the |
| 1394 | script's arguments; those processed by the interpreter (like `-s' or |
| 1395 | `-c') are omitted. (In other words, you get the normal, expected |
| 1396 | behavior.) Any application that uses scm_shell to process its |
| 1397 | command-line arguments gets this behavior as well. |
| 1398 | |
| 1399 | ** The new function `load-user-init' looks for a file called `.guile' |
| 1400 | in the user's home directory, and loads it if it exists. This is |
| 1401 | mostly for use by the code generated by scm_compile_shell_switches, |
| 1402 | but we thought it might also be useful in other circumstances. |
| 1403 | |
| 1404 | ** The new function `log10' returns the base-10 logarithm of its |
| 1405 | argument. |
| 1406 | |
| 1407 | ** Changes to I/O functions |
| 1408 | |
| 1409 | *** The functions `read', `primitive-load', `read-and-eval!', and |
| 1410 | `primitive-load-path' no longer take optional arguments controlling |
| 1411 | case insensitivity and a `#' parser. |
| 1412 | |
| 1413 | Case sensitivity is now controlled by a read option called |
| 1414 | `case-insensitive'. The user can add new `#' syntaxes with the |
| 1415 | `read-hash-extend' function (see below). |
| 1416 | |
| 1417 | *** The new function `read-hash-extend' allows the user to change the |
| 1418 | syntax of Guile Scheme in a somewhat controlled way. |
| 1419 | |
| 1420 | (read-hash-extend CHAR PROC) |
| 1421 | When parsing S-expressions, if we read a `#' character followed by |
| 1422 | the character CHAR, use PROC to parse an object from the stream. |
| 1423 | If PROC is #f, remove any parsing procedure registered for CHAR. |
| 1424 | |
| 1425 | The reader applies PROC to two arguments: CHAR and an input port. |
| 1426 | |
| 1427 | *** The new functions read-delimited and read-delimited! provide a |
| 1428 | general mechanism for doing delimited input on streams. |
| 1429 | |
| 1430 | (read-delimited DELIMS [PORT HANDLE-DELIM]) |
| 1431 | Read until we encounter one of the characters in DELIMS (a string), |
| 1432 | or end-of-file. PORT is the input port to read from; it defaults to |
| 1433 | the current input port. The HANDLE-DELIM parameter determines how |
| 1434 | the terminating character is handled; it should be one of the |
| 1435 | following symbols: |
| 1436 | |
| 1437 | 'trim omit delimiter from result |
| 1438 | 'peek leave delimiter character in input stream |
| 1439 | 'concat append delimiter character to returned value |
| 1440 | 'split return a pair: (RESULT . TERMINATOR) |
| 1441 | |
| 1442 | HANDLE-DELIM defaults to 'peek. |
| 1443 | |
| 1444 | (read-delimited! DELIMS BUF [PORT HANDLE-DELIM START END]) |
| 1445 | A side-effecting variant of `read-delimited'. |
| 1446 | |
| 1447 | The data is written into the string BUF at the indices in the |
| 1448 | half-open interval [START, END); the default interval is the whole |
| 1449 | string: START = 0 and END = (string-length BUF). The values of |
| 1450 | START and END must specify a well-defined interval in BUF, i.e. |
| 1451 | 0 <= START <= END <= (string-length BUF). |
| 1452 | |
| 1453 | It returns NBYTES, the number of bytes read. If the buffer filled |
| 1454 | up without a delimiter character being found, it returns #f. If the |
| 1455 | port is at EOF when the read starts, it returns the EOF object. |
| 1456 | |
| 1457 | If an integer is returned (i.e., the read is successfully terminated |
| 1458 | by reading a delimiter character), then the HANDLE-DELIM parameter |
| 1459 | determines how to handle the terminating character. It is described |
| 1460 | above, and defaults to 'peek. |
| 1461 | |
| 1462 | (The descriptions of these functions were borrowed from the SCSH |
| 1463 | manual, by Olin Shivers and Brian Carlstrom.) |
| 1464 | |
| 1465 | *** The `%read-delimited!' function is the primitive used to implement |
| 1466 | `read-delimited' and `read-delimited!'. |
| 1467 | |
| 1468 | (%read-delimited! DELIMS BUF GOBBLE? [PORT START END]) |
| 1469 | |
| 1470 | This returns a pair of values: (TERMINATOR . NUM-READ). |
| 1471 | - TERMINATOR describes why the read was terminated. If it is a |
| 1472 | character or the eof object, then that is the value that terminated |
| 1473 | the read. If it is #f, the function filled the buffer without finding |
| 1474 | a delimiting character. |
| 1475 | - NUM-READ is the number of characters read into BUF. |
| 1476 | |
| 1477 | If the read is successfully terminated by reading a delimiter |
| 1478 | character, then the gobble? parameter determines what to do with the |
| 1479 | terminating character. If true, the character is removed from the |
| 1480 | input stream; if false, the character is left in the input stream |
| 1481 | where a subsequent read operation will retrieve it. In either case, |
| 1482 | the character is also the first value returned by the procedure call. |
| 1483 | |
| 1484 | (The descriptions of this function was borrowed from the SCSH manual, |
| 1485 | by Olin Shivers and Brian Carlstrom.) |
| 1486 | |
| 1487 | *** The `read-line' and `read-line!' functions have changed; they now |
| 1488 | trim the terminator by default; previously they appended it to the |
| 1489 | returned string. For the old behavior, use (read-line PORT 'concat). |
| 1490 | |
| 1491 | *** The functions `uniform-array-read!' and `uniform-array-write!' now |
| 1492 | take new optional START and END arguments, specifying the region of |
| 1493 | the array to read and write. |
| 1494 | |
| 1495 | *** The `ungetc-char-ready?' function has been removed. We feel it's |
| 1496 | inappropriate for an interface to expose implementation details this |
| 1497 | way. |
| 1498 | |
| 1499 | ** Changes to the Unix library and system call interface |
| 1500 | |
| 1501 | *** The new fcntl function provides access to the Unix `fcntl' system |
| 1502 | call. |
| 1503 | |
| 1504 | (fcntl PORT COMMAND VALUE) |
| 1505 | Apply COMMAND to PORT's file descriptor, with VALUE as an argument. |
| 1506 | Values for COMMAND are: |
| 1507 | |
| 1508 | F_DUPFD duplicate a file descriptor |
| 1509 | F_GETFD read the descriptor's close-on-exec flag |
| 1510 | F_SETFD set the descriptor's close-on-exec flag to VALUE |
| 1511 | F_GETFL read the descriptor's flags, as set on open |
| 1512 | F_SETFL set the descriptor's flags, as set on open to VALUE |
| 1513 | F_GETOWN return the process ID of a socket's owner, for SIGIO |
| 1514 | F_SETOWN set the process that owns a socket to VALUE, for SIGIO |
| 1515 | FD_CLOEXEC not sure what this is |
| 1516 | |
| 1517 | For details, see the documentation for the fcntl system call. |
| 1518 | |
| 1519 | *** The arguments to `select' have changed, for compatibility with |
| 1520 | SCSH. The TIMEOUT parameter may now be non-integral, yielding the |
| 1521 | expected behavior. The MILLISECONDS parameter has been changed to |
| 1522 | MICROSECONDS, to more closely resemble the underlying system call. |
| 1523 | The RVEC, WVEC, and EVEC arguments can now be vectors; the type of the |
| 1524 | corresponding return set will be the same. |
| 1525 | |
| 1526 | *** The arguments to the `mknod' system call have changed. They are |
| 1527 | now: |
| 1528 | |
| 1529 | (mknod PATH TYPE PERMS DEV) |
| 1530 | Create a new file (`node') in the file system. PATH is the name of |
| 1531 | the file to create. TYPE is the kind of file to create; it should |
| 1532 | be 'fifo, 'block-special, or 'char-special. PERMS specifies the |
| 1533 | permission bits to give the newly created file. If TYPE is |
| 1534 | 'block-special or 'char-special, DEV specifies which device the |
| 1535 | special file refers to; its interpretation depends on the kind of |
| 1536 | special file being created. |
| 1537 | |
| 1538 | *** The `fork' function has been renamed to `primitive-fork', to avoid |
| 1539 | clashing with various SCSH forks. |
| 1540 | |
| 1541 | *** The `recv' and `recvfrom' functions have been renamed to `recv!' |
| 1542 | and `recvfrom!'. They no longer accept a size for a second argument; |
| 1543 | you must pass a string to hold the received value. They no longer |
| 1544 | return the buffer. Instead, `recv' returns the length of the message |
| 1545 | received, and `recvfrom' returns a pair containing the packet's length |
| 1546 | and originating address. |
| 1547 | |
| 1548 | *** The file descriptor datatype has been removed, as have the |
| 1549 | `read-fd', `write-fd', `close', `lseek', and `dup' functions. |
| 1550 | We plan to replace these functions with a SCSH-compatible interface. |
| 1551 | |
| 1552 | *** The `create' function has been removed; it's just a special case |
| 1553 | of `open'. |
| 1554 | |
| 1555 | *** There are new functions to break down process termination status |
| 1556 | values. In the descriptions below, STATUS is a value returned by |
| 1557 | `waitpid'. |
| 1558 | |
| 1559 | (status:exit-val STATUS) |
| 1560 | If the child process exited normally, this function returns the exit |
| 1561 | code for the child process (i.e., the value passed to exit, or |
| 1562 | returned from main). If the child process did not exit normally, |
| 1563 | this function returns #f. |
| 1564 | |
| 1565 | (status:stop-sig STATUS) |
| 1566 | If the child process was suspended by a signal, this function |
| 1567 | returns the signal that suspended the child. Otherwise, it returns |
| 1568 | #f. |
| 1569 | |
| 1570 | (status:term-sig STATUS) |
| 1571 | If the child process terminated abnormally, this function returns |
| 1572 | the signal that terminated the child. Otherwise, this function |
| 1573 | returns false. |
| 1574 | |
| 1575 | POSIX promises that exactly one of these functions will return true on |
| 1576 | a valid STATUS value. |
| 1577 | |
| 1578 | These functions are compatible with SCSH. |
| 1579 | |
| 1580 | *** There are new accessors and setters for the broken-out time vectors |
| 1581 | returned by `localtime', `gmtime', and that ilk. They are: |
| 1582 | |
| 1583 | Component Accessor Setter |
| 1584 | ========================= ============ ============ |
| 1585 | seconds tm:sec set-tm:sec |
| 1586 | minutes tm:min set-tm:min |
| 1587 | hours tm:hour set-tm:hour |
| 1588 | day of the month tm:mday set-tm:mday |
| 1589 | month tm:mon set-tm:mon |
| 1590 | year tm:year set-tm:year |
| 1591 | day of the week tm:wday set-tm:wday |
| 1592 | day in the year tm:yday set-tm:yday |
| 1593 | daylight saving time tm:isdst set-tm:isdst |
| 1594 | GMT offset, seconds tm:gmtoff set-tm:gmtoff |
| 1595 | name of time zone tm:zone set-tm:zone |
| 1596 | |
| 1597 | *** There are new accessors for the vectors returned by `uname', |
| 1598 | describing the host system: |
| 1599 | |
| 1600 | Component Accessor |
| 1601 | ============================================== ================ |
| 1602 | name of the operating system implementation utsname:sysname |
| 1603 | network name of this machine utsname:nodename |
| 1604 | release level of the operating system utsname:release |
| 1605 | version level of the operating system utsname:version |
| 1606 | machine hardware platform utsname:machine |
| 1607 | |
| 1608 | *** There are new accessors for the vectors returned by `getpw', |
| 1609 | `getpwnam', `getpwuid', and `getpwent', describing entries from the |
| 1610 | system's user database: |
| 1611 | |
| 1612 | Component Accessor |
| 1613 | ====================== ================= |
| 1614 | user name passwd:name |
| 1615 | user password passwd:passwd |
| 1616 | user id passwd:uid |
| 1617 | group id passwd:gid |
| 1618 | real name passwd:gecos |
| 1619 | home directory passwd:dir |
| 1620 | shell program passwd:shell |
| 1621 | |
| 1622 | *** There are new accessors for the vectors returned by `getgr', |
| 1623 | `getgrnam', `getgrgid', and `getgrent', describing entries from the |
| 1624 | system's group database: |
| 1625 | |
| 1626 | Component Accessor |
| 1627 | ======================= ============ |
| 1628 | group name group:name |
| 1629 | group password group:passwd |
| 1630 | group id group:gid |
| 1631 | group members group:mem |
| 1632 | |
| 1633 | *** There are new accessors for the vectors returned by `gethost', |
| 1634 | `gethostbyaddr', `gethostbyname', and `gethostent', describing |
| 1635 | internet hosts: |
| 1636 | |
| 1637 | Component Accessor |
| 1638 | ========================= =============== |
| 1639 | official name of host hostent:name |
| 1640 | alias list hostent:aliases |
| 1641 | host address type hostent:addrtype |
| 1642 | length of address hostent:length |
| 1643 | list of addresses hostent:addr-list |
| 1644 | |
| 1645 | *** There are new accessors for the vectors returned by `getnet', |
| 1646 | `getnetbyaddr', `getnetbyname', and `getnetent', describing internet |
| 1647 | networks: |
| 1648 | |
| 1649 | Component Accessor |
| 1650 | ========================= =============== |
| 1651 | official name of net netent:name |
| 1652 | alias list netent:aliases |
| 1653 | net number type netent:addrtype |
| 1654 | net number netent:net |
| 1655 | |
| 1656 | *** There are new accessors for the vectors returned by `getproto', |
| 1657 | `getprotobyname', `getprotobynumber', and `getprotoent', describing |
| 1658 | internet protocols: |
| 1659 | |
| 1660 | Component Accessor |
| 1661 | ========================= =============== |
| 1662 | official protocol name protoent:name |
| 1663 | alias list protoent:aliases |
| 1664 | protocol number protoent:proto |
| 1665 | |
| 1666 | *** There are new accessors for the vectors returned by `getserv', |
| 1667 | `getservbyname', `getservbyport', and `getservent', describing |
| 1668 | internet protocols: |
| 1669 | |
| 1670 | Component Accessor |
| 1671 | ========================= =============== |
| 1672 | official service name servent:name |
| 1673 | alias list servent:aliases |
| 1674 | port number servent:port |
| 1675 | protocol to use servent:proto |
| 1676 | |
| 1677 | *** There are new accessors for the sockaddr structures returned by |
| 1678 | `accept', `getsockname', `getpeername', `recvfrom!': |
| 1679 | |
| 1680 | Component Accessor |
| 1681 | ======================================== =============== |
| 1682 | address format (`family') sockaddr:fam |
| 1683 | path, for file domain addresses sockaddr:path |
| 1684 | address, for internet domain addresses sockaddr:addr |
| 1685 | TCP or UDP port, for internet sockaddr:port |
| 1686 | |
| 1687 | *** The `getpwent', `getgrent', `gethostent', `getnetent', |
| 1688 | `getprotoent', and `getservent' functions now return #f at the end of |
| 1689 | the user database. (They used to throw an exception.) |
| 1690 | |
| 1691 | Note that calling MUMBLEent function is equivalent to calling the |
| 1692 | corresponding MUMBLE function with no arguments. |
| 1693 | |
| 1694 | *** The `setpwent', `setgrent', `sethostent', `setnetent', |
| 1695 | `setprotoent', and `setservent' routines now take no arguments. |
| 1696 | |
| 1697 | *** The `gethost', `getproto', `getnet', and `getserv' functions now |
| 1698 | provide more useful information when they throw an exception. |
| 1699 | |
| 1700 | *** The `lnaof' function has been renamed to `inet-lnaof'. |
| 1701 | |
| 1702 | *** Guile now claims to have the `current-time' feature. |
| 1703 | |
| 1704 | *** The `mktime' function now takes an optional second argument ZONE, |
| 1705 | giving the time zone to use for the conversion. ZONE should be a |
| 1706 | string, in the same format as expected for the "TZ" environment variable. |
| 1707 | |
| 1708 | *** The `strptime' function now returns a pair (TIME . COUNT), where |
| 1709 | TIME is the parsed time as a vector, and COUNT is the number of |
| 1710 | characters from the string left unparsed. This function used to |
| 1711 | return the remaining characters as a string. |
| 1712 | |
| 1713 | *** The `gettimeofday' function has replaced the old `time+ticks' function. |
| 1714 | The return value is now (SECONDS . MICROSECONDS); the fractional |
| 1715 | component is no longer expressed in "ticks". |
| 1716 | |
| 1717 | *** The `ticks/sec' constant has been removed, in light of the above change. |
| 1718 | |
| 1719 | * Changes to the gh_ interface |
| 1720 | |
| 1721 | ** gh_eval_str() now returns an SCM object which is the result of the |
| 1722 | evaluation |
| 1723 | |
| 1724 | ** gh_scm2str() now copies the Scheme data to a caller-provided C |
| 1725 | array |
| 1726 | |
| 1727 | ** gh_scm2newstr() now makes a C array, copies the Scheme data to it, |
| 1728 | and returns the array |
| 1729 | |
| 1730 | ** gh_scm2str0() is gone: there is no need to distinguish |
| 1731 | null-terminated from non-null-terminated, since gh_scm2newstr() allows |
| 1732 | the user to interpret the data both ways. |
| 1733 | |
| 1734 | * Changes to the scm_ interface |
| 1735 | |
| 1736 | ** The new function scm_symbol_value0 provides an easy way to get a |
| 1737 | symbol's value from C code: |
| 1738 | |
| 1739 | SCM scm_symbol_value0 (char *NAME) |
| 1740 | Return the value of the symbol named by the null-terminated string |
| 1741 | NAME in the current module. If the symbol named NAME is unbound in |
| 1742 | the current module, return SCM_UNDEFINED. |
| 1743 | |
| 1744 | ** The new function scm_sysintern0 creates new top-level variables, |
| 1745 | without assigning them a value. |
| 1746 | |
| 1747 | SCM scm_sysintern0 (char *NAME) |
| 1748 | Create a new Scheme top-level variable named NAME. NAME is a |
| 1749 | null-terminated string. Return the variable's value cell. |
| 1750 | |
| 1751 | ** The function scm_internal_catch is the guts of catch. It handles |
| 1752 | all the mechanics of setting up a catch target, invoking the catch |
| 1753 | body, and perhaps invoking the handler if the body does a throw. |
| 1754 | |
| 1755 | The function is designed to be usable from C code, but is general |
| 1756 | enough to implement all the semantics Guile Scheme expects from throw. |
| 1757 | |
| 1758 | TAG is the catch tag. Typically, this is a symbol, but this function |
| 1759 | doesn't actually care about that. |
| 1760 | |
| 1761 | BODY is a pointer to a C function which runs the body of the catch; |
| 1762 | this is the code you can throw from. We call it like this: |
| 1763 | BODY (BODY_DATA, JMPBUF) |
| 1764 | where: |
| 1765 | BODY_DATA is just the BODY_DATA argument we received; we pass it |
| 1766 | through to BODY as its first argument. The caller can make |
| 1767 | BODY_DATA point to anything useful that BODY might need. |
| 1768 | JMPBUF is the Scheme jmpbuf object corresponding to this catch, |
| 1769 | which we have just created and initialized. |
| 1770 | |
| 1771 | HANDLER is a pointer to a C function to deal with a throw to TAG, |
| 1772 | should one occur. We call it like this: |
| 1773 | HANDLER (HANDLER_DATA, THROWN_TAG, THROW_ARGS) |
| 1774 | where |
| 1775 | HANDLER_DATA is the HANDLER_DATA argument we recevied; it's the |
| 1776 | same idea as BODY_DATA above. |
| 1777 | THROWN_TAG is the tag that the user threw to; usually this is |
| 1778 | TAG, but it could be something else if TAG was #t (i.e., a |
| 1779 | catch-all), or the user threw to a jmpbuf. |
| 1780 | THROW_ARGS is the list of arguments the user passed to the THROW |
| 1781 | function. |
| 1782 | |
| 1783 | BODY_DATA is just a pointer we pass through to BODY. HANDLER_DATA |
| 1784 | is just a pointer we pass through to HANDLER. We don't actually |
| 1785 | use either of those pointers otherwise ourselves. The idea is |
| 1786 | that, if our caller wants to communicate something to BODY or |
| 1787 | HANDLER, it can pass a pointer to it as MUMBLE_DATA, which BODY and |
| 1788 | HANDLER can then use. Think of it as a way to make BODY and |
| 1789 | HANDLER closures, not just functions; MUMBLE_DATA points to the |
| 1790 | enclosed variables. |
| 1791 | |
| 1792 | Of course, it's up to the caller to make sure that any data a |
| 1793 | MUMBLE_DATA needs is protected from GC. A common way to do this is |
| 1794 | to make MUMBLE_DATA a pointer to data stored in an automatic |
| 1795 | structure variable; since the collector must scan the stack for |
| 1796 | references anyway, this assures that any references in MUMBLE_DATA |
| 1797 | will be found. |
| 1798 | |
| 1799 | ** The new function scm_internal_lazy_catch is exactly like |
| 1800 | scm_internal_catch, except: |
| 1801 | |
| 1802 | - It does not unwind the stack (this is the major difference). |
| 1803 | - If handler returns, its value is returned from the throw. |
| 1804 | - BODY always receives #f as its JMPBUF argument (since there's no |
| 1805 | jmpbuf associated with a lazy catch, because we don't unwind the |
| 1806 | stack.) |
| 1807 | |
| 1808 | ** scm_body_thunk is a new body function you can pass to |
| 1809 | scm_internal_catch if you want the body to be like Scheme's `catch' |
| 1810 | --- a thunk, or a function of one argument if the tag is #f. |
| 1811 | |
| 1812 | BODY_DATA is a pointer to a scm_body_thunk_data structure, which |
| 1813 | contains the Scheme procedure to invoke as the body, and the tag |
| 1814 | we're catching. If the tag is #f, then we pass JMPBUF (created by |
| 1815 | scm_internal_catch) to the body procedure; otherwise, the body gets |
| 1816 | no arguments. |
| 1817 | |
| 1818 | ** scm_handle_by_proc is a new handler function you can pass to |
| 1819 | scm_internal_catch if you want the handler to act like Scheme's catch |
| 1820 | --- call a procedure with the tag and the throw arguments. |
| 1821 | |
| 1822 | If the user does a throw to this catch, this function runs a handler |
| 1823 | procedure written in Scheme. HANDLER_DATA is a pointer to an SCM |
| 1824 | variable holding the Scheme procedure object to invoke. It ought to |
| 1825 | be a pointer to an automatic variable (i.e., one living on the stack), |
| 1826 | or the procedure object should be otherwise protected from GC. |
| 1827 | |
| 1828 | ** scm_handle_by_message is a new handler function to use with |
| 1829 | `scm_internal_catch' if you want Guile to print a message and die. |
| 1830 | It's useful for dealing with throws to uncaught keys at the top level. |
| 1831 | |
| 1832 | HANDLER_DATA, if non-zero, is assumed to be a char * pointing to a |
| 1833 | message header to print; if zero, we use "guile" instead. That |
| 1834 | text is followed by a colon, then the message described by ARGS. |
| 1835 | |
| 1836 | ** The return type of scm_boot_guile is now void; the function does |
| 1837 | not return a value, and indeed, never returns at all. |
| 1838 | |
| 1839 | ** The new function scm_shell makes it easy for user applications to |
| 1840 | process command-line arguments in a way that is compatible with the |
| 1841 | stand-alone guile interpreter (which is in turn compatible with SCSH, |
| 1842 | the Scheme shell). |
| 1843 | |
| 1844 | To use the scm_shell function, first initialize any guile modules |
| 1845 | linked into your application, and then call scm_shell with the values |
| 1846 | of ARGC and ARGV your `main' function received. scm_shell will add |
| 1847 | any SCSH-style meta-arguments from the top of the script file to the |
| 1848 | argument vector, and then process the command-line arguments. This |
| 1849 | generally means loading a script file or starting up an interactive |
| 1850 | command interpreter. For details, see "Changes to the stand-alone |
| 1851 | interpreter" above. |
| 1852 | |
| 1853 | ** The new functions scm_get_meta_args and scm_count_argv help you |
| 1854 | implement the SCSH-style meta-argument, `\'. |
| 1855 | |
| 1856 | char **scm_get_meta_args (int ARGC, char **ARGV) |
| 1857 | If the second element of ARGV is a string consisting of a single |
| 1858 | backslash character (i.e. "\\" in Scheme notation), open the file |
| 1859 | named by the following argument, parse arguments from it, and return |
| 1860 | the spliced command line. The returned array is terminated by a |
| 1861 | null pointer. |
| 1862 | |
| 1863 | For details of argument parsing, see above, under "guile now accepts |
| 1864 | command-line arguments compatible with SCSH..." |
| 1865 | |
| 1866 | int scm_count_argv (char **ARGV) |
| 1867 | Count the arguments in ARGV, assuming it is terminated by a null |
| 1868 | pointer. |
| 1869 | |
| 1870 | For an example of how these functions might be used, see the source |
| 1871 | code for the function scm_shell in libguile/script.c. |
| 1872 | |
| 1873 | You will usually want to use scm_shell instead of calling this |
| 1874 | function yourself. |
| 1875 | |
| 1876 | ** The new function scm_compile_shell_switches turns an array of |
| 1877 | command-line arguments into Scheme code to carry out the actions they |
| 1878 | describe. Given ARGC and ARGV, it returns a Scheme expression to |
| 1879 | evaluate, and calls scm_set_program_arguments to make any remaining |
| 1880 | command-line arguments available to the Scheme code. For example, |
| 1881 | given the following arguments: |
| 1882 | |
| 1883 | -e main -s ekko a speckled gecko |
| 1884 | |
| 1885 | scm_set_program_arguments will return the following expression: |
| 1886 | |
| 1887 | (begin (load "ekko") (main (command-line)) (quit)) |
| 1888 | |
| 1889 | You will usually want to use scm_shell instead of calling this |
| 1890 | function yourself. |
| 1891 | |
| 1892 | ** The function scm_shell_usage prints a usage message appropriate for |
| 1893 | an interpreter that uses scm_compile_shell_switches to handle its |
| 1894 | command-line arguments. |
| 1895 | |
| 1896 | void scm_shell_usage (int FATAL, char *MESSAGE) |
| 1897 | Print a usage message to the standard error output. If MESSAGE is |
| 1898 | non-zero, write it before the usage message, followed by a newline. |
| 1899 | If FATAL is non-zero, exit the process, using FATAL as the |
| 1900 | termination status. (If you want to be compatible with Guile, |
| 1901 | always use 1 as the exit status when terminating due to command-line |
| 1902 | usage problems.) |
| 1903 | |
| 1904 | You will usually want to use scm_shell instead of calling this |
| 1905 | function yourself. |
| 1906 | |
| 1907 | ** scm_eval_0str now returns SCM_UNSPECIFIED if the string contains no |
| 1908 | expressions. It used to return SCM_EOL. Earth-shattering. |
| 1909 | |
| 1910 | ** The macros for declaring scheme objects in C code have been |
| 1911 | rearranged slightly. They are now: |
| 1912 | |
| 1913 | SCM_SYMBOL (C_NAME, SCHEME_NAME) |
| 1914 | Declare a static SCM variable named C_NAME, and initialize it to |
| 1915 | point to the Scheme symbol whose name is SCHEME_NAME. C_NAME should |
| 1916 | be a C identifier, and SCHEME_NAME should be a C string. |
| 1917 | |
| 1918 | SCM_GLOBAL_SYMBOL (C_NAME, SCHEME_NAME) |
| 1919 | Just like SCM_SYMBOL, but make C_NAME globally visible. |
| 1920 | |
| 1921 | SCM_VCELL (C_NAME, SCHEME_NAME) |
| 1922 | Create a global variable at the Scheme level named SCHEME_NAME. |
| 1923 | Declare a static SCM variable named C_NAME, and initialize it to |
| 1924 | point to the Scheme variable's value cell. |
| 1925 | |
| 1926 | SCM_GLOBAL_VCELL (C_NAME, SCHEME_NAME) |
| 1927 | Just like SCM_VCELL, but make C_NAME globally visible. |
| 1928 | |
| 1929 | The `guile-snarf' script writes initialization code for these macros |
| 1930 | to its standard output, given C source code as input. |
| 1931 | |
| 1932 | The SCM_GLOBAL macro is gone. |
| 1933 | |
| 1934 | ** The scm_read_line and scm_read_line_x functions have been replaced |
| 1935 | by Scheme code based on the %read-delimited! procedure (known to C |
| 1936 | code as scm_read_delimited_x). See its description above for more |
| 1937 | information. |
| 1938 | |
| 1939 | ** The function scm_sys_open has been renamed to scm_open. It now |
| 1940 | returns a port instead of an FD object. |
| 1941 | |
| 1942 | * The dynamic linking support has changed. For more information, see |
| 1943 | libguile/DYNAMIC-LINKING. |
| 1944 | |
| 1945 | \f |
| 1946 | Guile 1.0b3 |
| 1947 | |
| 1948 | User-visible changes from Thursday, September 5, 1996 until Guile 1.0 |
| 1949 | (Sun 5 Jan 1997): |
| 1950 | |
| 1951 | * Changes to the 'guile' program: |
| 1952 | |
| 1953 | ** Guile now loads some new files when it starts up. Guile first |
| 1954 | searches the load path for init.scm, and loads it if found. Then, if |
| 1955 | Guile is not being used to execute a script, and the user's home |
| 1956 | directory contains a file named `.guile', Guile loads that. |
| 1957 | |
| 1958 | ** You can now use Guile as a shell script interpreter. |
| 1959 | |
| 1960 | To paraphrase the SCSH manual: |
| 1961 | |
| 1962 | When Unix tries to execute an executable file whose first two |
| 1963 | characters are the `#!', it treats the file not as machine code to |
| 1964 | be directly executed by the native processor, but as source code |
| 1965 | to be executed by some interpreter. The interpreter to use is |
| 1966 | specified immediately after the #! sequence on the first line of |
| 1967 | the source file. The kernel reads in the name of the interpreter, |
| 1968 | and executes that instead. It passes the interpreter the source |
| 1969 | filename as its first argument, with the original arguments |
| 1970 | following. Consult the Unix man page for the `exec' system call |
| 1971 | for more information. |
| 1972 | |
| 1973 | Now you can use Guile as an interpreter, using a mechanism which is a |
| 1974 | compatible subset of that provided by SCSH. |
| 1975 | |
| 1976 | Guile now recognizes a '-s' command line switch, whose argument is the |
| 1977 | name of a file of Scheme code to load. It also treats the two |
| 1978 | characters `#!' as the start of a comment, terminated by `!#'. Thus, |
| 1979 | to make a file of Scheme code directly executable by Unix, insert the |
| 1980 | following two lines at the top of the file: |
| 1981 | |
| 1982 | #!/usr/local/bin/guile -s |
| 1983 | !# |
| 1984 | |
| 1985 | Guile treats the argument of the `-s' command-line switch as the name |
| 1986 | of a file of Scheme code to load, and treats the sequence `#!' as the |
| 1987 | start of a block comment, terminated by `!#'. |
| 1988 | |
| 1989 | For example, here's a version of 'echo' written in Scheme: |
| 1990 | |
| 1991 | #!/usr/local/bin/guile -s |
| 1992 | !# |
| 1993 | (let loop ((args (cdr (program-arguments)))) |
| 1994 | (if (pair? args) |
| 1995 | (begin |
| 1996 | (display (car args)) |
| 1997 | (if (pair? (cdr args)) |
| 1998 | (display " ")) |
| 1999 | (loop (cdr args))))) |
| 2000 | (newline) |
| 2001 | |
| 2002 | Why does `#!' start a block comment terminated by `!#', instead of the |
| 2003 | end of the line? That is the notation SCSH uses, and although we |
| 2004 | don't yet support the other SCSH features that motivate that choice, |
| 2005 | we would like to be backward-compatible with any existing Guile |
| 2006 | scripts once we do. Furthermore, if the path to Guile on your system |
| 2007 | is too long for your kernel, you can start the script with this |
| 2008 | horrible hack: |
| 2009 | |
| 2010 | #!/bin/sh |
| 2011 | exec /really/long/path/to/guile -s "$0" ${1+"$@"} |
| 2012 | !# |
| 2013 | |
| 2014 | Note that some very old Unix systems don't support the `#!' syntax. |
| 2015 | |
| 2016 | |
| 2017 | ** You can now run Guile without installing it. |
| 2018 | |
| 2019 | Previous versions of the interactive Guile interpreter (`guile') |
| 2020 | couldn't start up unless Guile's Scheme library had been installed; |
| 2021 | they used the value of the environment variable `SCHEME_LOAD_PATH' |
| 2022 | later on in the startup process, but not to find the startup code |
| 2023 | itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme |
| 2024 | code. |
| 2025 | |
| 2026 | To run Guile without installing it, build it in the normal way, and |
| 2027 | then set the environment variable `SCHEME_LOAD_PATH' to a |
| 2028 | colon-separated list of directories, including the top-level directory |
| 2029 | of the Guile sources. For example, if you unpacked Guile so that the |
| 2030 | full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then |
| 2031 | you might say |
| 2032 | |
| 2033 | export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3 |
| 2034 | |
| 2035 | |
| 2036 | ** Guile's read-eval-print loop no longer prints #<unspecified> |
| 2037 | results. If the user wants to see this, she can evaluate the |
| 2038 | expression (assert-repl-print-unspecified #t), perhaps in her startup |
| 2039 | file. |
| 2040 | |
| 2041 | ** Guile no longer shows backtraces by default when an error occurs; |
| 2042 | however, it does display a message saying how to get one, and how to |
| 2043 | request that they be displayed by default. After an error, evaluate |
| 2044 | (backtrace) |
| 2045 | to see a backtrace, and |
| 2046 | (debug-enable 'backtrace) |
| 2047 | to see them by default. |
| 2048 | |
| 2049 | |
| 2050 | |
| 2051 | * Changes to Guile Scheme: |
| 2052 | |
| 2053 | ** Guile now distinguishes between #f and the empty list. |
| 2054 | |
| 2055 | This is for compatibility with the IEEE standard, the (possibly) |
| 2056 | upcoming Revised^5 Report on Scheme, and many extant Scheme |
| 2057 | implementations. |
| 2058 | |
| 2059 | Guile used to have #f and '() denote the same object, to make Scheme's |
| 2060 | type system more compatible with Emacs Lisp's. However, the change |
| 2061 | caused too much trouble for Scheme programmers, and we found another |
| 2062 | way to reconcile Emacs Lisp with Scheme that didn't require this. |
| 2063 | |
| 2064 | |
| 2065 | ** Guile's delq, delv, delete functions, and their destructive |
| 2066 | counterparts, delq!, delv!, and delete!, now remove all matching |
| 2067 | elements from the list, not just the first. This matches the behavior |
| 2068 | of the corresponding Emacs Lisp functions, and (I believe) the Maclisp |
| 2069 | functions which inspired them. |
| 2070 | |
| 2071 | I recognize that this change may break code in subtle ways, but it |
| 2072 | seems best to make the change before the FSF's first Guile release, |
| 2073 | rather than after. |
| 2074 | |
| 2075 | |
| 2076 | ** The compiled-library-path function has been deleted from libguile. |
| 2077 | |
| 2078 | ** The facilities for loading Scheme source files have changed. |
| 2079 | |
| 2080 | *** The variable %load-path now tells Guile which directories to search |
| 2081 | for Scheme code. Its value is a list of strings, each of which names |
| 2082 | a directory. |
| 2083 | |
| 2084 | *** The variable %load-extensions now tells Guile which extensions to |
| 2085 | try appending to a filename when searching the load path. Its value |
| 2086 | is a list of strings. Its default value is ("" ".scm"). |
| 2087 | |
| 2088 | *** (%search-load-path FILENAME) searches the directories listed in the |
| 2089 | value of the %load-path variable for a Scheme file named FILENAME, |
| 2090 | with all the extensions listed in %load-extensions. If it finds a |
| 2091 | match, then it returns its full filename. If FILENAME is absolute, it |
| 2092 | returns it unchanged. Otherwise, it returns #f. |
| 2093 | |
| 2094 | %search-load-path will not return matches that refer to directories. |
| 2095 | |
| 2096 | *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP) |
| 2097 | uses %seach-load-path to find a file named FILENAME, and loads it if |
| 2098 | it finds it. If it can't read FILENAME for any reason, it throws an |
| 2099 | error. |
| 2100 | |
| 2101 | The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the |
| 2102 | `read' function. |
| 2103 | |
| 2104 | *** load uses the same searching semantics as primitive-load. |
| 2105 | |
| 2106 | *** The functions %try-load, try-load-with-path, %load, load-with-path, |
| 2107 | basic-try-load-with-path, basic-load-with-path, try-load-module-with- |
| 2108 | path, and load-module-with-path have been deleted. The functions |
| 2109 | above should serve their purposes. |
| 2110 | |
| 2111 | *** If the value of the variable %load-hook is a procedure, |
| 2112 | `primitive-load' applies its value to the name of the file being |
| 2113 | loaded (without the load path directory name prepended). If its value |
| 2114 | is #f, it is ignored. Otherwise, an error occurs. |
| 2115 | |
| 2116 | This is mostly useful for printing load notification messages. |
| 2117 | |
| 2118 | |
| 2119 | ** The function `eval!' is no longer accessible from the scheme level. |
| 2120 | We can't allow operations which introduce glocs into the scheme level, |
| 2121 | because Guile's type system can't handle these as data. Use `eval' or |
| 2122 | `read-and-eval!' (see below) as replacement. |
| 2123 | |
| 2124 | ** The new function read-and-eval! reads an expression from PORT, |
| 2125 | evaluates it, and returns the result. This is more efficient than |
| 2126 | simply calling `read' and `eval', since it is not necessary to make a |
| 2127 | copy of the expression for the evaluator to munge. |
| 2128 | |
| 2129 | Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as |
| 2130 | for the `read' function. |
| 2131 | |
| 2132 | |
| 2133 | ** The function `int?' has been removed; its definition was identical |
| 2134 | to that of `integer?'. |
| 2135 | |
| 2136 | ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should |
| 2137 | use the R4RS names for these functions. |
| 2138 | |
| 2139 | ** The function object-properties no longer returns the hash handle; |
| 2140 | it simply returns the object's property list. |
| 2141 | |
| 2142 | ** Many functions have been changed to throw errors, instead of |
| 2143 | returning #f on failure. The point of providing exception handling in |
| 2144 | the language is to simplify the logic of user code, but this is less |
| 2145 | useful if Guile's primitives don't throw exceptions. |
| 2146 | |
| 2147 | ** The function `fileno' has been renamed from `%fileno'. |
| 2148 | |
| 2149 | ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0. |
| 2150 | |
| 2151 | |
| 2152 | * Changes to Guile's C interface: |
| 2153 | |
| 2154 | ** The library's initialization procedure has been simplified. |
| 2155 | scm_boot_guile now has the prototype: |
| 2156 | |
| 2157 | void scm_boot_guile (int ARGC, |
| 2158 | char **ARGV, |
| 2159 | void (*main_func) (), |
| 2160 | void *closure); |
| 2161 | |
| 2162 | scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV. |
| 2163 | MAIN_FUNC should do all the work of the program (initializing other |
| 2164 | packages, reading user input, etc.) before returning. When MAIN_FUNC |
| 2165 | returns, call exit (0); this function never returns. If you want some |
| 2166 | other exit value, MAIN_FUNC may call exit itself. |
| 2167 | |
| 2168 | scm_boot_guile arranges for program-arguments to return the strings |
| 2169 | given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call |
| 2170 | scm_set_program_arguments with the final list, so Scheme code will |
| 2171 | know which arguments have been processed. |
| 2172 | |
| 2173 | scm_boot_guile establishes a catch-all catch handler which prints an |
| 2174 | error message and exits the process. This means that Guile exits in a |
| 2175 | coherent way when system errors occur and the user isn't prepared to |
| 2176 | handle it. If the user doesn't like this behavior, they can establish |
| 2177 | their own universal catcher in MAIN_FUNC to shadow this one. |
| 2178 | |
| 2179 | Why must the caller do all the real work from MAIN_FUNC? The garbage |
| 2180 | collector assumes that all local variables of type SCM will be above |
| 2181 | scm_boot_guile's stack frame on the stack. If you try to manipulate |
| 2182 | SCM values after this function returns, it's the luck of the draw |
| 2183 | whether the GC will be able to find the objects you allocate. So, |
| 2184 | scm_boot_guile function exits, rather than returning, to discourage |
| 2185 | people from making that mistake. |
| 2186 | |
| 2187 | The IN, OUT, and ERR arguments were removed; there are other |
| 2188 | convenient ways to override these when desired. |
| 2189 | |
| 2190 | The RESULT argument was deleted; this function should never return. |
| 2191 | |
| 2192 | The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more |
| 2193 | general. |
| 2194 | |
| 2195 | |
| 2196 | ** Guile's header files should no longer conflict with your system's |
| 2197 | header files. |
| 2198 | |
| 2199 | In order to compile code which #included <libguile.h>, previous |
| 2200 | versions of Guile required you to add a directory containing all the |
| 2201 | Guile header files to your #include path. This was a problem, since |
| 2202 | Guile's header files have names which conflict with many systems' |
| 2203 | header files. |
| 2204 | |
| 2205 | Now only <libguile.h> need appear in your #include path; you must |
| 2206 | refer to all Guile's other header files as <libguile/mumble.h>. |
| 2207 | Guile's installation procedure puts libguile.h in $(includedir), and |
| 2208 | the rest in $(includedir)/libguile. |
| 2209 | |
| 2210 | |
| 2211 | ** Two new C functions, scm_protect_object and scm_unprotect_object, |
| 2212 | have been added to the Guile library. |
| 2213 | |
| 2214 | scm_protect_object (OBJ) protects OBJ from the garbage collector. |
| 2215 | OBJ will not be freed, even if all other references are dropped, |
| 2216 | until someone does scm_unprotect_object (OBJ). Both functions |
| 2217 | return OBJ. |
| 2218 | |
| 2219 | Note that calls to scm_protect_object do not nest. You can call |
| 2220 | scm_protect_object any number of times on a given object, and the |
| 2221 | next call to scm_unprotect_object will unprotect it completely. |
| 2222 | |
| 2223 | Basically, scm_protect_object and scm_unprotect_object just |
| 2224 | maintain a list of references to things. Since the GC knows about |
| 2225 | this list, all objects it mentions stay alive. scm_protect_object |
| 2226 | adds its argument to the list; scm_unprotect_object remove its |
| 2227 | argument from the list. |
| 2228 | |
| 2229 | |
| 2230 | ** scm_eval_0str now returns the value of the last expression |
| 2231 | evaluated. |
| 2232 | |
| 2233 | ** The new function scm_read_0str reads an s-expression from a |
| 2234 | null-terminated string, and returns it. |
| 2235 | |
| 2236 | ** The new function `scm_stdio_to_port' converts a STDIO file pointer |
| 2237 | to a Scheme port object. |
| 2238 | |
| 2239 | ** The new function `scm_set_program_arguments' allows C code to set |
| 2240 | the value teruturned by the Scheme `program-arguments' function. |
| 2241 | |
| 2242 | \f |
| 2243 | Older changes: |
| 2244 | |
| 2245 | * Guile no longer includes sophisticated Tcl/Tk support. |
| 2246 | |
| 2247 | The old Tcl/Tk support was unsatisfying to us, because it required the |
| 2248 | user to link against the Tcl library, as well as Tk and Guile. The |
| 2249 | interface was also un-lispy, in that it preserved Tcl/Tk's practice of |
| 2250 | referring to widgets by names, rather than exporting widgets to Scheme |
| 2251 | code as a special datatype. |
| 2252 | |
| 2253 | In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk |
| 2254 | maintainers described some very interesting changes in progress to the |
| 2255 | Tcl/Tk internals, which would facilitate clean interfaces between lone |
| 2256 | Tk and other interpreters --- even for garbage-collected languages |
| 2257 | like Scheme. They expected the new Tk to be publicly available in the |
| 2258 | fall of 1996. |
| 2259 | |
| 2260 | Since it seems that Guile might soon have a new, cleaner interface to |
| 2261 | lone Tk, and that the old Guile/Tk glue code would probably need to be |
| 2262 | completely rewritten, we (Jim Blandy and Richard Stallman) have |
| 2263 | decided not to support the old code. We'll spend the time instead on |
| 2264 | a good interface to the newer Tk, as soon as it is available. |
| 2265 | |
| 2266 | Until then, gtcltk-lib provides trivial, low-maintenance functionality. |
| 2267 | |
| 2268 | \f |
| 2269 | Copyright information: |
| 2270 | |
| 2271 | Copyright (C) 1996,1997 Free Software Foundation, Inc. |
| 2272 | |
| 2273 | Permission is granted to anyone to make or distribute verbatim copies |
| 2274 | of this document as received, in any medium, provided that the |
| 2275 | copyright notice and this permission notice are preserved, |
| 2276 | thus giving the recipient permission to redistribute in turn. |
| 2277 | |
| 2278 | Permission is granted to distribute modified versions |
| 2279 | of this document, or of portions of it, |
| 2280 | under the above conditions, provided also that they |
| 2281 | carry prominent notices stating who last changed them. |
| 2282 | |
| 2283 | \f |
| 2284 | Local variables: |
| 2285 | mode: outline |
| 2286 | paragraph-separate: "[ \f]*$" |
| 2287 | end: |
| 2288 | |