| 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@prep.ai.mit.edu. |
| 6 | \f |
| 7 | Changes since Guile 1.0 (Sun 5 Jan 1997): |
| 8 | |
| 9 | * Changes to the gh_ interface |
| 10 | |
| 11 | ** gh_eval_str() now returns an SCM object which is the result of the |
| 12 | evaluation |
| 13 | |
| 14 | * Changes to documentation |
| 15 | |
| 16 | ** the $(srcdir)/newdoc hierarchy now contains a new approach to the |
| 17 | manuals. The approach, recommended by Jim Blandy, is to have: (*) a |
| 18 | tutorial with the pedagogical style of guile-user, and a non-dry |
| 19 | reference manual in the style of the most excellent GNU libc reference |
| 20 | manual: the reference manual should be complete, but at the same time |
| 21 | it should have an introductory screen for each major topic, which can |
| 22 | be referenced if the user goes "up" a level in the info documentation. |
| 23 | |
| 24 | \f |
| 25 | Guile 1.0b3 |
| 26 | |
| 27 | User-visible changes from Thursday, September 5, 1996 until Guile 1.0: |
| 28 | |
| 29 | * Changes to the 'guile' program: |
| 30 | |
| 31 | ** Guile now loads some new files when it starts up. Guile first |
| 32 | searches the load path for init.scm, and loads it if found. Then, if |
| 33 | Guile is not being used to execute a script, and the user's home |
| 34 | directory contains a file named `.guile', Guile loads that. |
| 35 | |
| 36 | ** You can now use Guile as a shell script interpreter. |
| 37 | |
| 38 | To paraphrase the SCSH manual: |
| 39 | |
| 40 | When Unix tries to execute an executable file whose first two |
| 41 | characters are the `#!', it treats the file not as machine code to |
| 42 | be directly executed by the native processor, but as source code |
| 43 | to be executed by some interpreter. The interpreter to use is |
| 44 | specified immediately after the #! sequence on the first line of |
| 45 | the source file. The kernel reads in the name of the interpreter, |
| 46 | and executes that instead. It passes the interpreter the source |
| 47 | filename as its first argument, with the original arguments |
| 48 | following. Consult the Unix man page for the `exec' system call |
| 49 | for more information. |
| 50 | |
| 51 | Now you can use Guile as an interpreter, using a mechanism which is a |
| 52 | compatible subset of that provided by SCSH. |
| 53 | |
| 54 | Guile now recognizes a '-s' command line switch, whose argument is the |
| 55 | name of a file of Scheme code to load. It also treats the two |
| 56 | characters `#!' as the start of a comment, terminated by `!#'. Thus, |
| 57 | to make a file of Scheme code directly executable by Unix, insert the |
| 58 | following two lines at the top of the file: |
| 59 | |
| 60 | #!/usr/local/bin/guile -s |
| 61 | !# |
| 62 | |
| 63 | Guile treats the argument of the `-s' command-line switch as the name |
| 64 | of a file of Scheme code to load, and treats the sequence `#!' as the |
| 65 | start of a block comment, terminated by `!#'. |
| 66 | |
| 67 | For example, here's a version of 'echo' written in Scheme: |
| 68 | |
| 69 | #!/usr/local/bin/guile -s |
| 70 | !# |
| 71 | (let loop ((args (cdr (program-arguments)))) |
| 72 | (if (pair? args) |
| 73 | (begin |
| 74 | (display (car args)) |
| 75 | (if (pair? (cdr args)) |
| 76 | (display " ")) |
| 77 | (loop (cdr args))))) |
| 78 | (newline) |
| 79 | |
| 80 | Why does `#!' start a block comment terminated by `!#', instead of the |
| 81 | end of the line? That is the notation SCSH uses, and although we |
| 82 | don't yet support the other SCSH features that motivate that choice, |
| 83 | we would like to be backward-compatible with any existing Guile |
| 84 | scripts once we do. Furthermore, if the path to Guile on your system |
| 85 | is too long for your kernel, you can start the script with this |
| 86 | horrible hack: |
| 87 | |
| 88 | #!/bin/sh |
| 89 | exec /really/long/path/to/guile -s "$0" ${1+"$@"} |
| 90 | !# |
| 91 | |
| 92 | Note that some very old Unix systems don't support the `#!' syntax. |
| 93 | |
| 94 | |
| 95 | ** You can now run Guile without installing it. |
| 96 | |
| 97 | Previous versions of the interactive Guile interpreter (`guile') |
| 98 | couldn't start up unless Guile's Scheme library had been installed; |
| 99 | they used the value of the environment variable `SCHEME_LOAD_PATH' |
| 100 | later on in the startup process, but not to find the startup code |
| 101 | itself. Now Guile uses `SCHEME_LOAD_PATH' in all searches for Scheme |
| 102 | code. |
| 103 | |
| 104 | To run Guile without installing it, build it in the normal way, and |
| 105 | then set the environment variable `SCHEME_LOAD_PATH' to a |
| 106 | colon-separated list of directories, including the top-level directory |
| 107 | of the Guile sources. For example, if you unpacked Guile so that the |
| 108 | full filename of this NEWS file is /home/jimb/guile-1.0b3/NEWS, then |
| 109 | you might say |
| 110 | |
| 111 | export SCHEME_LOAD_PATH=/home/jimb/my-scheme:/home/jimb/guile-1.0b3 |
| 112 | |
| 113 | |
| 114 | ** Guile's read-eval-print loop no longer prints #<unspecified> |
| 115 | results. If the user wants to see this, she can evaluate the |
| 116 | expression (assert-repl-print-unspecified #t), perhaps in her startup |
| 117 | file. |
| 118 | |
| 119 | ** Guile no longer shows backtraces by default when an error occurs; |
| 120 | however, it does display a message saying how to get one, and how to |
| 121 | request that they be displayed by default. After an error, evaluate |
| 122 | (backtrace) |
| 123 | to see a backtrace, and |
| 124 | (debug-enable 'backtrace) |
| 125 | to see them by default. |
| 126 | |
| 127 | |
| 128 | |
| 129 | * Changes to Guile Scheme: |
| 130 | |
| 131 | ** Guile now distinguishes between #f and the empty list. |
| 132 | |
| 133 | This is for compatibility with the IEEE standard, the (possibly) |
| 134 | upcoming Revised^5 Report on Scheme, and many extant Scheme |
| 135 | implementations. |
| 136 | |
| 137 | Guile used to have #f and '() denote the same object, to make Scheme's |
| 138 | type system more compatible with Emacs Lisp's. However, the change |
| 139 | caused too much trouble for Scheme programmers, and we found another |
| 140 | way to reconcile Emacs Lisp with Scheme that didn't require this. |
| 141 | |
| 142 | |
| 143 | ** Guile's delq, delv, delete functions, and their destructive |
| 144 | counterparts, delq!, delv!, and delete!, now remove all matching |
| 145 | elements from the list, not just the first. This matches the behavior |
| 146 | of the corresponding Emacs Lisp functions, and (I believe) the Maclisp |
| 147 | functions which inspired them. |
| 148 | |
| 149 | I recognize that this change may break code in subtle ways, but it |
| 150 | seems best to make the change before the FSF's first Guile release, |
| 151 | rather than after. |
| 152 | |
| 153 | |
| 154 | ** The compiled-library-path function has been deleted from libguile. |
| 155 | |
| 156 | ** The facilities for loading Scheme source files have changed. |
| 157 | |
| 158 | *** The variable %load-path now tells Guile which directories to search |
| 159 | for Scheme code. Its value is a list of strings, each of which names |
| 160 | a directory. |
| 161 | |
| 162 | *** The variable %load-extensions now tells Guile which extensions to |
| 163 | try appending to a filename when searching the load path. Its value |
| 164 | is a list of strings. Its default value is ("" ".scm"). |
| 165 | |
| 166 | *** (%search-load-path FILENAME) searches the directories listed in the |
| 167 | value of the %load-path variable for a Scheme file named FILENAME, |
| 168 | with all the extensions listed in %load-extensions. If it finds a |
| 169 | match, then it returns its full filename. If FILENAME is absolute, it |
| 170 | returns it unchanged. Otherwise, it returns #f. |
| 171 | |
| 172 | %search-load-path will not return matches that refer to directories. |
| 173 | |
| 174 | *** (primitive-load FILENAME :optional CASE-INSENSITIVE-P SHARP) |
| 175 | uses %seach-load-path to find a file named FILENAME, and loads it if |
| 176 | it finds it. If it can't read FILENAME for any reason, it throws an |
| 177 | error. |
| 178 | |
| 179 | The arguments CASE-INSENSITIVE-P and SHARP are interpreted as by the |
| 180 | `read' function. |
| 181 | |
| 182 | *** load uses the same searching semantics as primitive-load. |
| 183 | |
| 184 | *** The functions %try-load, try-load-with-path, %load, load-with-path, |
| 185 | basic-try-load-with-path, basic-load-with-path, try-load-module-with- |
| 186 | path, and load-module-with-path have been deleted. The functions |
| 187 | above should serve their purposes. |
| 188 | |
| 189 | *** If the value of the variable %load-hook is a procedure, |
| 190 | `primitive-load' applies its value to the name of the file being |
| 191 | loaded (without the load path directory name prepended). If its value |
| 192 | is #f, it is ignored. Otherwise, an error occurs. |
| 193 | |
| 194 | This is mostly useful for printing load notification messages. |
| 195 | |
| 196 | |
| 197 | ** The function `eval!' is no longer accessible from the scheme level. |
| 198 | We can't allow operations which introduce glocs into the scheme level, |
| 199 | because Guile's type system can't handle these as data. Use `eval' or |
| 200 | `read-and-eval!' (see below) as replacement. |
| 201 | |
| 202 | ** The new function read-and-eval! reads an expression from PORT, |
| 203 | evaluates it, and returns the result. This is more efficient than |
| 204 | simply calling `read' and `eval', since it is not necessary to make a |
| 205 | copy of the expression for the evaluator to munge. |
| 206 | |
| 207 | Its optional arguments CASE_INSENSITIVE_P and SHARP are interpreted as |
| 208 | for the `read' function. |
| 209 | |
| 210 | |
| 211 | ** The function `int?' has been removed; its definition was identical |
| 212 | to that of `integer?'. |
| 213 | |
| 214 | ** The functions `<?', `<?', `<=?', `=?', `>?', and `>=?'. Code should |
| 215 | use the R4RS names for these functions. |
| 216 | |
| 217 | ** The function object-properties no longer returns the hash handle; |
| 218 | it simply returns the object's property list. |
| 219 | |
| 220 | ** Many functions have been changed to throw errors, instead of |
| 221 | returning #f on failure. The point of providing exception handling in |
| 222 | the language is to simplify the logic of user code, but this is less |
| 223 | useful if Guile's primitives don't throw exceptions. |
| 224 | |
| 225 | ** The function `fileno' has been renamed from `%fileno'. |
| 226 | |
| 227 | ** The function primitive-mode->fdes returns #t or #f now, not 1 or 0. |
| 228 | |
| 229 | |
| 230 | * Changes to Guile's C interface: |
| 231 | |
| 232 | ** The library's initialization procedure has been simplified. |
| 233 | scm_boot_guile now has the prototype: |
| 234 | |
| 235 | void scm_boot_guile (int ARGC, |
| 236 | char **ARGV, |
| 237 | void (*main_func) (), |
| 238 | void *closure); |
| 239 | |
| 240 | scm_boot_guile calls MAIN_FUNC, passing it CLOSURE, ARGC, and ARGV. |
| 241 | MAIN_FUNC should do all the work of the program (initializing other |
| 242 | packages, reading user input, etc.) before returning. When MAIN_FUNC |
| 243 | returns, call exit (0); this function never returns. If you want some |
| 244 | other exit value, MAIN_FUNC may call exit itself. |
| 245 | |
| 246 | scm_boot_guile arranges for program-arguments to return the strings |
| 247 | given by ARGC and ARGV. If MAIN_FUNC modifies ARGC/ARGV, should call |
| 248 | scm_set_program_arguments with the final list, so Scheme code will |
| 249 | know which arguments have been processed. |
| 250 | |
| 251 | scm_boot_guile establishes a catch-all catch handler which prints an |
| 252 | error message and exits the process. This means that Guile exits in a |
| 253 | coherent way when system errors occur and the user isn't prepared to |
| 254 | handle it. If the user doesn't like this behavior, they can establish |
| 255 | their own universal catcher in MAIN_FUNC to shadow this one. |
| 256 | |
| 257 | Why must the caller do all the real work from MAIN_FUNC? The garbage |
| 258 | collector assumes that all local variables of type SCM will be above |
| 259 | scm_boot_guile's stack frame on the stack. If you try to manipulate |
| 260 | SCM values after this function returns, it's the luck of the draw |
| 261 | whether the GC will be able to find the objects you allocate. So, |
| 262 | scm_boot_guile function exits, rather than returning, to discourage |
| 263 | people from making that mistake. |
| 264 | |
| 265 | The IN, OUT, and ERR arguments were removed; there are other |
| 266 | convenient ways to override these when desired. |
| 267 | |
| 268 | The RESULT argument was deleted; this function should never return. |
| 269 | |
| 270 | The BOOT_CMD argument was deleted; the MAIN_FUNC argument is more |
| 271 | general. |
| 272 | |
| 273 | |
| 274 | ** Guile's header files should no longer conflict with your system's |
| 275 | header files. |
| 276 | |
| 277 | In order to compile code which #included <libguile.h>, previous |
| 278 | versions of Guile required you to add a directory containing all the |
| 279 | Guile header files to your #include path. This was a problem, since |
| 280 | Guile's header files have names which conflict with many systems' |
| 281 | header files. |
| 282 | |
| 283 | Now only <libguile.h> need appear in your #include path; you must |
| 284 | refer to all Guile's other header files as <libguile/mumble.h>. |
| 285 | Guile's installation procedure puts libguile.h in $(includedir), and |
| 286 | the rest in $(includedir)/libguile. |
| 287 | |
| 288 | |
| 289 | ** Two new C functions, scm_protect_object and scm_unprotect_object, |
| 290 | have been added to the Guile library. |
| 291 | |
| 292 | scm_protect_object (OBJ) protects OBJ from the garbage collector. |
| 293 | OBJ will not be freed, even if all other references are dropped, |
| 294 | until someone does scm_unprotect_object (OBJ). Both functions |
| 295 | return OBJ. |
| 296 | |
| 297 | Note that calls to scm_protect_object do not nest. You can call |
| 298 | scm_protect_object any number of times on a given object, and the |
| 299 | next call to scm_unprotect_object will unprotect it completely. |
| 300 | |
| 301 | Basically, scm_protect_object and scm_unprotect_object just |
| 302 | maintain a list of references to things. Since the GC knows about |
| 303 | this list, all objects it mentions stay alive. scm_protect_object |
| 304 | adds its argument to the list; scm_unprotect_object remove its |
| 305 | argument from the list. |
| 306 | |
| 307 | |
| 308 | ** scm_eval_0str now returns the value of the last expression |
| 309 | evaluated. |
| 310 | |
| 311 | ** The new function scm_read_0str reads an s-expression from a |
| 312 | null-terminated string, and returns it. |
| 313 | |
| 314 | ** The new function `scm_stdio_to_port' converts a STDIO file pointer |
| 315 | to a Scheme port object. |
| 316 | |
| 317 | ** The new function `scm_set_program_arguments' allows C code to set |
| 318 | the value teruturned by the Scheme `program-arguments' function. |
| 319 | |
| 320 | \f |
| 321 | Older changes: |
| 322 | |
| 323 | * Guile no longer includes sophisticated Tcl/Tk support. |
| 324 | |
| 325 | The old Tcl/Tk support was unsatisfying to us, because it required the |
| 326 | user to link against the Tcl library, as well as Tk and Guile. The |
| 327 | interface was also un-lispy, in that it preserved Tcl/Tk's practice of |
| 328 | referring to widgets by names, rather than exporting widgets to Scheme |
| 329 | code as a special datatype. |
| 330 | |
| 331 | In the Usenix Tk Developer's Workshop held in July 1996, the Tcl/Tk |
| 332 | maintainers described some very interesting changes in progress to the |
| 333 | Tcl/Tk internals, which would facilitate clean interfaces between lone |
| 334 | Tk and other interpreters --- even for garbage-collected languages |
| 335 | like Scheme. They expected the new Tk to be publicly available in the |
| 336 | fall of 1996. |
| 337 | |
| 338 | Since it seems that Guile might soon have a new, cleaner interface to |
| 339 | lone Tk, and that the old Guile/Tk glue code would probably need to be |
| 340 | completely rewritten, we (Jim Blandy and Richard Stallman) have |
| 341 | decided not to support the old code. We'll spend the time instead on |
| 342 | a good interface to the newer Tk, as soon as it is available. |
| 343 | |
| 344 | Until then, gtcltk-lib provides trivial, low-maintenance functionality. |
| 345 | |
| 346 | \f |
| 347 | Copyright information: |
| 348 | |
| 349 | Copyright (C) 1996,1997 Free Software Foundation, Inc. |
| 350 | |
| 351 | Permission is granted to anyone to make or distribute verbatim copies |
| 352 | of this document as received, in any medium, provided that the |
| 353 | copyright notice and this permission notice are preserved, |
| 354 | thus giving the recipient permission to redistribute in turn. |
| 355 | |
| 356 | Permission is granted to distribute modified versions |
| 357 | of this document, or of portions of it, |
| 358 | under the above conditions, provided also that they |
| 359 | carry prominent notices stating who last changed them. |
| 360 | |