Debian Codegraph @

HISTORY

raw
1 This is the long-form history of the project. For the short form used
2 to generate release announcements, see NEWS.
3
4 == Repository head ==
5
6 === Changes by Eric S. Raymond ===
7
8 1. Typo and formatting fixes to THEORY.txt.
9
10 === Changes by Mark J. Reed, Alex Smith ===
11
12 2. Build system fixes required by recent versions of autoconf, which is
13 in randomness and ugliness and propensity to mutate a perfect match
14 for INTERCAL.
15
16 === Changes by Alex Smith ===
17
18 3. Some potential undefined behaviour (which happens to do the right
19 thing on every compiler tested) related to the -g option has been
20 fixed. (Thanks to Michael Tautschnig for reporting the bug.)
21
22 4. Document how the build system works. (That is, how everything fits
23 together, as opposed to the individual files.)
24
25 5. Fixes for two nasty dependency issues to do with testing. One issue
26 was that the tests could potentially be run against an old version
27 of the compiler (they now ensure the compiler is up-to-date); the
28 other was that tests on a non-installed compiler could potentially
29 use library files from an installed version, which might not
30 necessarily be the same version of the compiler and might not
31 contain the same bugfixes.
32
33 6. C-INTERCAL now understands C11 in the external calls system, the
34 same way that it understands C89 and C99. This gives us a gap of
35 (2011-1972 = 39) years between languages standards simultaneously
36 supported by the same compiler. Anyone know if this is a record?
37
38 7. Most of the tests in the pit are run by a common framework.
39 However, one of the tests is different; yapp's test involves
40 another call to ick to make sure that the output is correct
41 INTERCAL that produces the correct output. Guess what happened.
42 This change fixes it. (Hint if you're still guessing: what needs
43 to be done to make an INTERCAL program with no double-oh-sevens
44 run in a reproducible way, as might be required by an automated
45 testsuite?)
46
47 8. Fixed a typo in the definition of the "smudgeleft" and "smudgeright"
48 helper functions. This might theoretically have lead to code being
49 misoptimized, although I don't know of any specific situation in which
50 that would occur.
51
52 == Release 0.29 (by Alex Smith, Eric S. Raymond) 29 Nov 2010 ==
53
54 === Changes by Eric S. Raymond ===
55
56 1. There is now a compiler regression-test suite, invoked through the test
57 production in the pit makefile.
58
59 2. There are two new compiler error messages, E990 and E994.
60
61 3. A new program for the pit: Brendan Gregg's guess game.
62
63 4. The floating-point library is now automatically included if
64 a program NEXTs to labels in the range 5000-5999. Internally,
65 line-range to system library mappings are no longer hardcoded but
66 rather declared as an array of associations that can be easily
67 modified.
68
69 5. The source now uses explicit bool, rather than using int everywhere.
70
71 6. README, HISTORY, and BUGS are now mastered in asciidoc for easier
72 web-page generation.
73
74 7. The history of the hitherto-mysterious "Atari implementation" is now
75 documented.
76
77 === Bugs reported and changes patched by Arvid Norlander ===
78
79 8. Fixed a typo in the perpet.c version of the help information.
80
81 9. Fixed the CREATE statement to work on 64-bit architectures.
82
83 10. The availability of variable argument list mechanisms is now determined
84 automatically by the configure script, rather than relying on the old
85 method of checking which usually gets the wrong answer with modern
86 compilers, leading to build failures on many modern systems (previously
87 this was patched just for DJGPP, the patch now applies to all operating
88 systems and platforms). This bug is known to have affected the build on
89 DOS, FreeBSD and Mac OS X, and possibly affects other platforms. (also
90 reported independently by Elliott Hird)
91
92 11. Fixed some nonsense in the -@ help screen that was generated by an errant
93 search-and-replace.
94
95 12. Made several changes to bring the new Befunge-98 interface code (see
96 below) in line with changes in the Befunge-98 interpreter in use.
97
98 13. Added consts and statics in the code, hopefully to make bugs easier to
99 notice.
100
101 14. Refactored main(). It needed it, badly.
102
103 === Bugs reported by Elliott Hird ===
104
105 15. Fixed the configure script to work with the default Debian shell dash.
106 (also reported independently by Debian)
107
108 16. Actually fixed unlambda.i; the fix was meant to be in 0.28, and is
109 described in its changelog, but for some reason never made it to
110 the distributed tarball.
111
112 17. The -F option again works. (AIS: I accidentally broke it in the preceding
113 version, making it basically a no-op by mistake.)
114
115 18. The build now succeeds (with warnings) when lex or yacc are missing,
116 rather than failing with mysterious error messages that seem to have
117 nothing to do with the problems. (The success is made possible by using
118 the prebuilt versions; see change 11 of version 0.26, this is finishing
119 what that started.)
120
121 === Bugs reported, changes requested, and patches by Joris Huizer ===
122
123 19. The [xoritself] optimiser idiom has been generalised to handle some
124 situations where the redundant XOR operands are on different grouping
125 levels; this helps to optimise some heavily inlined greater-than tests.
126
127 20. Fixed unary operators applied to constants (AIS: it seems that my patch to
128 solve this issue earlier was itself buggy).
129
130 21. The -g option now causes optimisation of the output compiled C to be
131 turned off, and adds comments to the output file the same way -c does.
132
133 22. Added more optimiser idioms: [cxorand16], [cxorand32], [cfoldintoorinand],
134 [or16and], [xor16and], [lshift32half], [rshiftoutofneg], [andoutofneg];
135 also generalised the idiom-matching code to compare 16-bit and 32-bit
136 constants as equal if they have the same value, and fixed some typos in
137 the current 16-bit complement idioms.
138
139 23. ABSTAINING or REINSTATING are now a lot more efficient, taking an amount
140 of time proportional to the number of commands effected, rather than the
141 number of commands in the program. (AIS: I implemented this with the
142 algorithm in the patch Joris provided, but with different code, to make it
143 easier to maintain.)
144
145 === Changes by Stefan O'Rear ===
146
147 24. Added INTERCAL syntax highlighting information for vim. Why should Emacs
148 have all the fun?
149
150 === Changes reported by Phantom Hoover ===
151
152 25. Fixed a nasty documentation typo.
153
154 === Changes reported by the Debian autobuilders and Joey Hess ===
155
156 26. The build of the optimiser idioms is now split over many files, to reduce
157 the memory usage of gcc when it tries to compile the giant output C file.
158 This was a problem on ia64. (The ulimit method of preventing thrashing has
159 been removed, as this method works better; likewise, the optimiser
160 optimisation now works again.)
161
162 === Changes by Alex Smith ===
163
164 27. The build system has been completely redone, again; it now uses automake
165 as well as a much newer version of autoconf. This means that there is now
166 one build system that works the same way on all systems, and that all
167 sorts of clever tricks like out-of-tree builds, a self-rebuilding build
168 system, and building the distribution directly from the Makefile, are
169 possible. As a side effect, the directory structure looks somewhat
170 different to how it looked in previous versions; src, doc, etc and pit are
171 still there, but now there's prebuilt instead of temp, an extra directory
172 buildaux, and bin, lib and include have disappeared.
173
174 28. The distribution now comes in compressed pax format. (Before anyone
175 complains that they have no idea what that is, or that I'm using something
176 strange and nonstandard, I'd like to point out that tar is no longer part
177 of POSIX, but that pax is. (See man 5 tar on a recent Linux system for
178 details of the whole mess.) Luckily, pax is forward-compatible with tar
179 so tar readers should be able to read it just fine, if you can't get your
180 hands on a pax reader.)
181
182 29. convickt can now produce its output as INTERCAL initialisation of a tail
183 array. (This doesn't auto-offset between the different characters; it is
184 therefore mostly useful for producing Baudot arrays for CLC-INTERCAL-style
185 IO.)
186
187 30. The function clock_gettime is now supported by the yuk profiler as an
188 alternative method of getting a high-resolution clock for profiling (it
189 is available on at least Linux).
190
191 31. The profiler now autodetects the same profiling mechanism at compile
192 time and at run time. (It was getting confused as to what mechanism it was
193 using, causing some really weird bugs.)
194
195 32. Added support for computed CREATE (the computation is done at create-time,
196 rather than when the unknown statement is encountered); partly because
197 this was rather easy, but mostly because it can be used to simulate a
198 computed NEXT, thus making the example expansion library compunex (which I
199 forgot to mention in the 0.28 release notes) redundant. Because it was
200 deprecated when I added it, compunex is now doubly deprecated, and anyone
201 tempted to use it should use computed CREATE instead. (I have always
202 maintained that if computed NEXT were ever added to C-INTERCAL, it would
203 be deprecated on arrival; the computed CREATE method implements all
204 features of NEXT, anyway, rather than the small subset COMPUNEX could
205 manage.)
206
207 33. Plugged a couple of small memory leaks; however, there are still several
208 much bigger memory leaks that could do with being fixed.
209
210 34. Added support for CREATING operators, with similar syntax to the CREATION
211 of commands. This requires the -ea options to be set to work properly.
212
213 35. More file extensions are now supported by the external calls system. The
214 .c99 extension indicates a C99 (rather than C89) file; all .c99 files are
215 preprocessed as C99, and if any C99 files are used the final compilation
216 which compiles and links everything will be done as C99 rather than C89
217 (this causes no problems for C-INTERCAL's output which is both legal C89
218 and legal C99). Also, libraries can be linked in; to link in the maths
219 library for instance (i.e. pass -lm to gcc), write libm.a as a filename at
220 the end of the command line. (The file's real name is libm.a; if a
221 filename is in lib*.a format, then ick will look in the normal library
222 locations instead of the current directory.) Other .a files can be linked
223 in from the current directory.
224
225 36. Fixed the handling of division by zero in the syslibc expansion library.
226 (The Revamped Manual states 'In all cases, division by 0 returns 0'.)
227
228 37. Made several changes to the documentation, including fixes for some rather
229 nasty typos (two of the worst were specifying variables as IGNORED by
230 default, and stating that select acts like OR in TriIntercal!).
231
232 38. Fixed a bug in the external calls system that made it impossible to write
233 a COME FROM in more than one file at a time whilst still having the
234 resulting program work reliably.
235
236 39. Added support for external calls to Befunge-98 programs; however, this
237 distribution does not ship with a Funge-98 interpreter, so in order to use
238 this feature you will first have to compile a Funge-98 interpreter into a
239 library for use with C-INTERCAL (see README.txt or the Revamped Manual for
240 information on this). Also added a test program for this to the pit.
241
242 40. Fixed some bugs in the go-away command in interfunge.i.
243
244 41. Fixed many bugs in the optimiser, and added a fuzz-tester ("make fuzz") for
245 the optimiser which has a chance at finding any new optimiser bugs that may
246 be introduced, or any that the maintainers don't know about.
247
248 == Release 0.28 (by Alex Smith) 1 Apr 2008 ==
249
250 === Patches, suggestions and bugs found by Joris Huizer ===
251
252 1. Fixed a typo in the [noopxor] optimiser idiom.
253
254 2. Added a [redundantdoublenot] optimiser idiom (a generalisation of Joris'
255 suggestion of and code for [triplenot]).
256
257 3. Fixed a typo that caused the flow optimiser to optimise some errors into
258 non-erroring code.
259
260 4. Fixed the [lshift16] optimiser idiom; previously it was giving incorrect
261 values in some cases where the bits being selected from the original
262 number weren't contiguous.
263
264 5. There is now no longer a segfault when the unusual option combination -f
265 (without specifying -O) is used.
266
267 6. It is again possible to recompile more than one source file at a time.
268
269 7. The flow optimiser now recognises that comments can be reinstated in
270 programs where they can be reinstated.
271
272 === Changes by Alex Smith ===
273
274 8. Fixed some typos in the documentation.
275
276 9. Applying an infix unary operator to a constant (as in &#26) now returns
277 the correct value; previously it returned the unary operator applied to a
278 sequence number that was 0 for the first unique constant that was the
279 argument of a unary operator, 1 for the second unique constant, and so
280 on. (This was a bug in the parser.)
281
282 10. Added doc/THEORY.txt back in. (It seems that it went missing some time
283 between 1.26 and 0.27.) Updated it to describe the current compiler.
284
285 11. Turned off optimisation of the optimiser, and also placed a 128MB limit on
286 the data segment and virtual memory it was allowed to use. This is to
287 prevent Linux thrashing and/or invoking the dreaded oom-killer when
288 recompiling idiotism.c. (This happened only rarely, but it happened to me
289 twice, each time making my computer unusable for about half an hour.
290 Therefore, a fix to prevent the situation happening again was added,
291 assuming your computer has at least 128MB free memory and disk space.) As
292 a bonus, the build process is now considerably faster, because the
293 optimisation of the optimiser was the slowest step of the process.
294
295 12. Internally, all symbols that are visible globally are now name-mangled (by
296 using the prefix ick_). This allows other files to be linked with files
297 compiled from INTERCAL without namespace collisions. This does not include
298 symbols that are only useful in multithreading (as INTERCAL's
299 multithreading model is incompatible with other programs), only for the
300 debugger (because linking yuk into a program causes all sorts of chaos if
301 non-INTERCAL programs are involved), or only used by the compiler (which
302 hopefully should never be linked to another program that isn't
303 specifically designed to be linked to it).
304
305 13. Thoroughly delinted the code, using Splint; the code is now annotated to
306 show Splint what is being done where (and all the non-obvious annotations
307 are explained so that humans can figure out what is going on as well);
308 over 800 warnings have been fixed, including some which identified bugs in
309 the original program (many of which were to do with edge cases involving
310 unusually long strings, and some of which were buffer overflows). All
311 comments from /*@ to @*/ are mine, and are annotations for Splint.
312
313 14. Implemented external calls, which allow multiple INTERCAL programs to be
314 linked to each other, and/or to C programs, via the -e option and NEXT /
315 COME FROM / NEXT FROM / FORGET / RESUME statements. See the manual for
316 more details. (Note that external calls require you to use gcc, GNU cpp,
317 and GNU ld, because they rely on features of those programs to work.)
318
319 15. Two other new compiler options have been added: -E (don't link system
320 library), and -w (enable +printflow debugging even for non-multithread
321 programs). +printflow has been enhanced to show what's going on in
322 externally-called programs to some extent, too.
323
324 16. Added a version of the INTERCAL system library written in C, as
325 pit/lib/syslibc.c. A new compiler feature allows the inclusion of such
326 expansion libraries via the command line; to include this library, you
327 would write the command line as ick -eE yourprogram.i syslibc (possibly
328 adding other options than -eE). The main advantage of the C-based system
329 library is speed; also, note that an implemented-in-a-faster-language-
330 than-INTERCAL system library existed in the Princeton compiler (which is
331 possibly how the INTERCAL version managed to stay buggy for decades
332 without anyone noticing), so this is another case of catching up with
333 historically available features. (The feature is also somewhat reminicent
334 of CLC-INTERCAL's 'preloads', but is implemented differently, and is
335 useful for different things, as it doesn't affect the compiler.)
336 Expansion libraries are included via -e, like non-INTERCAL programs are,
337 but are distinguished by not having an extension.
338
339 17. The handling of syntax errors has been improved; many lines that
340 previously caused E017 at compile time are now correctly interpreted as
341 run-time errors. The only way to cause E017 nowadays is to write a
342 constant out of the onespot range, and the manual description of the error
343 has been changed accordingly. (Note that I caused several regressions
344 while implementing this; I hope they're all fixed now, but there may be
345 some left. Please report any remaining parsing problems caused by this;
346 they mostly manifest as either the line after a comment being corrupted or
347 ignored, or as the line before a comment being taken to be part of the
348 comment.)
349
350 18. A new 'just-in-case' compilation method is used for syntax errors. If a
351 syntax error is encountered at compile-time, and contains just capital
352 letters with no other meaning (i.e. they don't form keywords and none of
353 them is the letter V), variables, and expressions, and there are no two
354 consecutive expressions or variables involved (i.e. they are separated by
355 keywords), the compiler will compile it anyway, just in case it happens to
356 gain a meaning later (see the next changelog entry). This is sort of the
357 converse of the way CLC-INTERCAL goes about compiling things that don't
358 have a meaning yet (it uses a 'just-too-late' compiler which waits for the
359 offending statement to error out, then recompiles it to see if it's gained
360 a meaning in the meantime), and this difference in handling typifies the
361 difference between the compilers quite well.
362
363 19. Implemented the CREATE statement, because I could (see point 18). This
364 allows new syntax to be created at runtime. CLC-INTERCAL also has a CREATE
365 statement, but its syntax is totally different (it specifies the bytecode
366 for the command, whereas C-INTERCAL's CREATE statement allows you to
367 specify what the new command does in INTERCAL itself.) A function call
368 ick_create is also available; it's usable from externally-called programs
369 and from expansion libraries, and means that INTERCAL commands can now be
370 defined to do things which were previously well out of INTERCAL's
371 reach. This makes expansion libraries act even more like CLC-INTERCAL
372 preloads.
373
374 20. Fixed a strange bug in yuk when the basename of the INTERCAL program was
375 the same as that of a shell builtin or command on the user's path.
376
377 21. Programs accept a new option +instapipe that causes output to be output
378 immediately, even if not followed by a newline, by flushing the output
379 after every character. This is most useful with binary output into pipes,
380 and occasionally useful for getting the output to very slow programs
381 faster.
382
383 22. Fixed an error in an optimiser idiom that could optimise erroring code
384 involving mingles into non-erroring code (which is of course incorrect);
385 also edited unlambda.i in the pit, which was taking advantage of this
386 error (and erroring out if optimisation was turned off).
387
388 23. Fixed a bug with the ON THE WAY TO line number displayed when COME FROMs
389 are involved that I accidentally introduced in 0.25.
390
391 === Changes from elsewhere ===
392
393 24. Added yet another nontrivial INTERCAL program to the pit; interfunge.i
394 (written by Jeffrey Lee) is a Befunge-93 interpreter written in INTERCAL.
395 (It appears to conform to the standard (I actually ran it against a
396 testsuite, and it passed); its only pecularities, apart from the language
397 in which it's written, appear to be the need for a blank line at the end
398 of the program, but no blank lines earlier (because INTERCAL can't sense
399 EOF, although there are plans to change that), the restriction to 80x25 is
400 enforced (this seems to be an implied requirement in the standard, but
401 most Befunge-93 interpreters don't enforce it), and numeric output is in
402 Roman numerals (what did you expect?).)
403
404 == Release 0.27 (by Alex Smith) 22 Dec 2007 ==
405
406 1. Joris Huizer pointed out that the manual nowhere says that a GIVE UP line
407 can't be abstained from by line number. This was probably a mistake in the
408 INTERCAL-72 manual, but the behaviour specified by the original manual is
409 now implemented; after all, it wouldn't do to violate the original de facto
410 standard. (Note that if a GIVE UP line is abstained by line number, there
411 is then no way to reinstate it; also, if the -X switch for CLC-INTERCAL
412 compatibility is used, ABSTAINING FROM a GIVE UP line by line number is
413 also silently ignored.)
414
415 2. Rewrote the optimizer in a language (OIL) invented specifically for the
416 purpose; the set of optimizer idioms is now in idiotism.oil, and translated
417 into C and linked with the rest of the compiler at metacompile time (that
418 is, when the compiler itself is being compiled). It contains many of the
419 same idioms, but some have been generalised or reduced into smaller parts
420 and some new ones were added. Documentation for OIL has been added as part
421 of the documentation for the distribution as a whole.
422
423 3. Joris Huizer submitted a patch to fix a memory allocation bug which
424 occasionally caused memory corruption when a program had a number of
425 commands that was 1 less than a multiple of 256.
426
427 4. Completely redid the autoconfiguration. The install should now progress
428 with less user interaction than before. As a bonus, if you have a range of
429 GNU utilities including make, sed, tr, and bash (and possibly others), it's
430 now possible to compile using 'sh configdj.sh' followed by 'make' under
431 DJGPP. (To be specific, some settings that previously sometimes had to be
432 set by the user, such as whether lex provided yylineno and yyrestart, are
433 now either detected by config.sh or no longer relevant. Other settings that
434 were not previously detected, such as what extension executables have, are
435 now detected so that the config script works on DJGPP as well, now;
436 compiling the config script on a non-DJGPP system seems to result in a
437 script that fails on DJGPP, and likewise compiling the script on DJGPP
438 seems to result on a script that fails on other systems, so both scripts
439 are provided. The autoconfigurer now also adapts coopt.sh so that -F works
440 on DJGPP too, sort of (or else detects that it won't work and disables it).
441
442 5. Improved the action of 'make install', so that it actually installs
443 everything in the right place. It works under DJGPP, now, too, as long as
444 you have the programs installed that allow configdj.sh and make to work.
445
446 6. Fixed a bug involving COME FROM in multithreaded programs; now, if a line
447 COMEs FROM itself (either because this is written explicitly or because it
448 is a computed COME FROM computing to its own line number), the loop this
449 creates doesn't block all the other threads in the program.
450
451 7. Added support for statement WHILE statement, with similar semantics to CLC-
452 INTERCAL. (The second statement is repeated for as long as the first
453 statement is executing; NEXTing to the statement nexts to the first
454 statement and doesn't set off the loop, and COMING or NEXTING FROM the
455 first statement leaves the loop running until the NEXT FROM returns.) With
456 the current implementation, strange things happen if the same WHILE is
457 simultaneously run by multiple threads, although I decided to work out
458 exactly what the rules were anyway and document them, so those strange
459 things are now official.
460
461 8. Fixed some weird bugs in the multithreader that could cause segfaults in
462 several situations involving backtracking; it's kind of surprising that
463 they didn't turn up sooner.
464
465 9. Prevented a segfault in some situations where a syntax error was the
466 target of a noncomputed COME FROM or NEXT FROM.
467
468 10. CALCULATING now includes array dimensioning. (There previously was no
469 gerund for this situation; the original manual doesn't say what to do but
470 implies that this is a special case of calculate.)
471
472 11. Made operand overloading interact more sensibly with multithreading.
473 Previously, overloading information was shared between all threads, but the
474 stashes for operand overloading were separate in each thread. Now, the
475 overloading is not shared between threads, except when they are woven by a
476 WHILE statement.
477
478 12. There's no longer a need to set or modify copyloc in the makefile by hand;
479 COPYING.txt is now found the same way as syslib and the skeleton. It's also
480 now copied out of the distribution to the same place as the other data
481 files when installing, preventing the need to keep the original
482 installation directory around.
483
484 13. The compiler now actually looks for a temporary directory to place a file
485 in when creating a temporary file; this was meant to be fixed in 1.26, but
486 the fix didn't actually work.
487
488 14. Added a new program to the pit: an Unlambda interpreter written by Oerjan
489 Johansen in 2001, and also a commented version as documentation.
490
491 15. Section 23 of the release notes for version 0.26 is wrong; I was obviously
492 confused when I wrote them because they were speaking nonsense. For the
493 record: no, there is nothing special about NEXT in connection with
494 backtracking. It should have said "NEXT", not "choicepoint". Instead of
495 rewriting history, this changelog entry is added as a fix.
496
497 16. Added an entirely new set of documentation, the Revamped Manual, which is
498 available in a wide range of formats and documents recent language changes,
499 as well as many other things INTERCAL. The Revised Manual has been
500 preserved, but it seems unlikely to be updated in the near future (it is
501 still well worth a read anyway).
502
503 17. For compatibility, the COMMENT gerund can now be spelt COMMENTS or
504 COMMENTING.
505
506 18. Corrected the Emacs INTERCAL mode to work with more recent versions of
507 Emacs than it previously worked with, hopefully without sacrificing too
508 much backward compatibility.
509
510 19. Added a conversion program 'convickt'; it can convert between the Latin-1,
511 Baudot and EBCDIC character sets used by CLC-INTERCAL (Latin-1 is also
512 accepted by C-INTERCAL, so this gives a practical way to run CLC-INTERCAL
513 programs in unusual character sets on C-INTERCAL), and can also attempt to
514 convert into the 'Atari character set' (which uses $ instead of the cent
515 sign, ? instead of yen, and so on for the other characters that differ
516 between C-INTERCAL and CLC-INTERCAL).
517
518 20. Reorganised the Makefile and environment variable system for finding
519 locations to make more logical sense; various definitions and environment
520 variables now do what their name suggests rather than something completely
521 different, a situation which confused the programmers previously. The new
522 version should be better at finding files that it needs, and also cuts
523 down on the amount of useless copying that the Makefile did (such copying
524 is now only done if the destination file doesn't already exist or is out
525 of date).
526
527 21. Fixed some problems in the PIC-INTERCAL code: one problem with finding the
528 correct location for the skeleton, and some code duplication between the
529 two header files. Also clarified one of the comments.
530
531 22. Added support for CLC-INTERCAL-style Baudot I/O (generalised array I/O is
532 currently unimplemented, though). This has a two minor differences to the
533 CLC-INTERCAL version: unrecognised characters, instead of causing an
534 error, become 0 when WRITING IN (except that tabs are translated to single
535 spaces) and ASCII code 26 when READING OUT. (Baudot characters not in
536 ASCII will be translated to Latin-1; there are more of these than might be
537 expected, because the CLC-INTERCAL extended Baudot is being used).
538
539 23. The man pages and some fixes for the documentation Makefile from the
540 Debian package were copied into the main distribution. The man page was
541 written originally by Mark W. Eichin and later modified by Joey Hess (and
542 was further modified by me prior to placing it in the distribution); the
543 Makefile fixes are due to Joey Hess. Strangely, many of the changes I made
544 to the Makefile for this version have much the same effect as fixes that
545 the Debian package maintainers had come up with earlier; a new Makefile
546 variable ick_specific_subdir has been added to hopefully make the Debian
547 maintainer's job easier (it allows elimination of versioned
548 subdirectories, replacing them with versionless subdirectories).
549
550 24. Added a new test program to the pit, tests/ignorret.i, that demonstrates
551 how the interaction between IGNORE and RETRIEVE is different between
552 C-INTERCAL, CLC-INTERCAL, and J-INTERCAL, and documentation showing the
553 results on each version.
554
555 Release 1.26 (by Alex Smith) [a bugfix of 0.26] 3 Sep 2007
556
557 1. Fixed an optimizer bug (thanks to Joris Huizer for pointing it out),
558 relating to active-bits deduction in some expressions involving select.
559
560 2. Fixed a bug in the flow optimizer that caused it to occasionally confuse
561 ABSTAIN with REINSTATE when gerunds were used.
562
563 3. Simplified the compilation of INTERCAL files to .exe under DJGPP, to avoid
564 some temporary files (hopefully solving some problems users may be having
565 with finding which directory the temporary files are in under DJGPP);
566 when under DJGPP a temporary file other than .c or .exe is needed, the
567 compiler now looks for a directory reserved for temporary files to place it
568 in rather than just using the current directory.
569
570 4. The original INTERCAL manual specifies that GIVE UP cannot be REINSTATED
571 or ABSTAINED FROM; this is now implemented, by taking only the original
572 DO/DO NOT status into account when executing a GIVE UP statement.
573
574 5. Dealt with problems with what happens with the 'original' stdin (this
575 information was previously lost, making it hard for the debugger to accept
576 input because it had no obvious place to accept it from), and removed the
577 previous hacky workarounds for this problem (which were my fault).
578
579 6. Made several improvements to the documentation, especially with respect to
580 installing and running under DJGPP.
581
582 7. Modified the version number scheme, so that minor revisions and bugfixes
583 now increment the major version number rather than the minor version number
584 (it was getting a little bored stuck at 0). The next major release will be
585 version 0.27.
586
587 == Release 0.26 (by Alex Smith) 1 Apr 2007 ==
588
589 1. Fixed a few bugs in the previous release (some bugfixed by Joris Huizer).
590
591 2. Ported to DJGPP. The compiler now works on both POSIX and DJGPP systems.
592
593 3. Reorganized the directory structure.
594
595 4. Added support for cross-compiling to the PIC microcontroller.
596
597 5. Added a simple test program for PIC cross-compilation.
598
599 6. Improved further searching for skeletons and executables.
600
601 7. Delinted some of the code, and turned up the default gcc warning level.
602
603 8. Reformatted some of the documentation.
604
605 9. Renamed and split files; all files used in the build now have 8.3
606 filenames, to prevent any problems using DJGPP on 8.3 systems. The
607 optimizer was split from the code degeneration into dekludge.c (optimizer)
608 and feh2.c (degeneration), because feh.c was getting too large.
609
610 10. The text documentation (README, COPYING, NEWS, and BUGS) was renamed to
611 end .txt (again for DJGPP compatibility); formatting changes were made to
612 README in addition to the addition of extra readme information.
613
614 11. Included precompiled-to-C versions of the parser and the lexer so that
615 DJGPP (or I suppose, POSIX if you hacked the makefile) implementations
616 without bison or flex could still compile the rest of C-INTERCAL and give
617 a running program.
618
619 12. Fixed a bug involving optimization of Tri-INTERCAL programs.
620
621 13. Added to the lexer, so that it supports CLC-INTERCAL syntax for operators
622 that were already present in C-INTERCAL, as long as they don't conflict
623 with C-INTERCAL syntax. (Using the -X switch will support CLC-INTERCAL
624 syntax for operators that were already present in C-INTERCAL even if they
625 conflict with C-INTERCAL syntax.) For the symbols present in the
626 CLC-INTERCAL character set but not in ASCII, ISO-8859-1 is supported as an
627 encoding as well as UTF-8 (to increase the chance that CLC-INTERCAL
628 programs will run without modification); it was meant to be supported to
629 some extent before, but seemed not to be working.
630
631 14. Added the COMMENT gerund, with the same semantics as in CLC-INTERCAL.
632
633 15. Added COME FROM gerund. This suffers from the same restrictions as
634 computed COME FROM, in that it can't be used in PIC programs and causes
635 the program to end up substantially less efficient.
636
637 16. Added support for prefix as well as infix unary operators. The rule is
638 that any number of prefix unary operators, but only one infix unary
639 operator, is allowed on each group, variable, and/or mesh. (Previously,
640 only infix operators were supported, and then only one per mesh.)
641
642 17. Added positional precedences to the parser. INTERCAL still doesn't have
643 operator precedences (that would be completely contrary to the spirit of
644 INTERCAL) but it now has rules for deciding on the meaning of previously
645 ambiguous expressions. Chains of select and mingle operations right-
646 associate, and when a unary operator could legitimately be in either a
647 prefix or infix position, the infix position is assumed (although remember
648 that only one infix operator is allowed for each group, variable, and/or
649 mesh).
650
651 18. Added the not character as a synonym for NOT.
652
653 19. Added operand overloading, for onespot and twospot variables. This uses
654 the slat operator in almost, but not quite, the same way as
655 CLC-INTERCAL. Note that the width of unary operations will mostly depend
656 on the type a variable appears to have in an expression, rather than on
657 its actual bitwidth (if a onespot is overloaded to a twospot or vice
658 versa). Note that meshes won't change as the result of operand overloading
659 unless the -v option is used (an error will be generated instead).
660
661 20. Added an option for generalised assignments. With the -v option,
662 expressions (rather than just variables) can appear on the left of an
663 assignment. Any meshes on the left of an assignment (even if in apparently
664 inoccuous positions) may change value as a result. (Note that such
665 assignments aren't always possible.) If the left-hand-side of an
666 expression is anything other than simply a variable or a constant, the
667 compiler will try to change the values of meshes and non-ignored variables
668 in the expression to make it fit. (In bases other than 2, it will nearly
669 always fail.) Unlike CLC-INTERCAL, assigning to a mesh only changes that
670 mesh, not line labels or other uses of the number.
671
672 21. Restricted the compiler to just one input file at a time (it used to break
673 if this wasn't done anyway).
674
675 22. Fixed a nonportability involving stdin and stdout in cesspool.c.
676
677 23. Added support for the NEXT FROM command, with the same semantics as COME
678 FROM (including computed NEXT FROM and NEXT FROM gerund), except that it
679 saves the line that was NEXTed FROM on the choicepoint stack.
680
681 24. Fixed a bug where the yuk debugger accessed unallocated memory.
682
683 == Release 0.25 (by Alex Smith) 21 Jun 2006 ==
684
685 1. Added support for computed ABSTAIN and TRY AGAIN.
686
687 2. Added support for computed COME FROM.
688
689 3. Added support for multithread programs. This includes ONCE and AGAIN,
690 which also work in singlethread programs.
691
692 4. Expanded intercal.el to support Font Lock and some utilities for
693 generating INTERCAL-style constants. It also works with the
694 compilation routines of Emacs to compile INTERCAL programs without
695 the need for make. Degenerated C code now has a mode line to help
696 with Emacs compatibility.
697
698 5. Added a rudimentary debugger and profiler for singlethread programs. The
699 debugger's features include useful things like 'write out INTERCAL expression
700 in simpler terms' (which is only possible if -O is on). The profiler works
701 much like gcov, counting the length of time spent in and number of executions
702 of each command in the code.
703
704 6. Changed searching for includes, libraries, syslib, skeleton. This
705 is so that non-root users, or people who want to evaluate ick, can
706 do so more simply. The compiler will search for the above files in
707 the directory they should have been installed in, the current
708 directory, and the directory in which ick's executable resides (in
709 that order). Skeleton problems, in particular, seem to have been a
710 major problem for unfamiliar users trying out ick.
711
712 7. Added several new test programs to the Pit.
713
714 8. Greatly expanded the original optimizer (which optimizes
715 expressions). It can now recognize most of the idioms in syslib,
716 and many idioms in other Pit programs. Also added a new optimizer,
717 to try to optimize program flow, which is important because
718 otherwise INTERCAL wouldn't have any rudimentary optimizers. There
719 is also the -F optimizer, which I hope will become infamous. It
720 creates object code which is close to maximally fast in some
721 specialized situations, at the expense of compile time (compilation
722 can take a very long time when -F is active) and object file
723 size. I suggest checking its effect on primes.i.
724
725 9. Fixed the grammar (again) in a spark/ears situation involving array
726 subscripts, after proving that the previous fixes weren't adequate.
727
728 10. Added support for backtracking (MAYBE, GO AHEAD, GO BACK). Note
729 that it is possible for multithreading, backtracking, computed
730 ABSTAIN, and computed COME FROM to all be used simultaneously in
731 the same program, to help maximize obfuscation.
732
733 11. Added an option (-l) for generating warnings. To prevent this
734 actually being useful, it is designed to generate far too many
735 warnings, which all have to be trawled through to find the
736 important ones, and has no way to turn the unimportant ones off.
737
738 12. Added the +printflow and +mystery options for executing INTERCAL programs.
739 Note that it is not entirely obvious what either of them do (+printflow is a
740 debugging aid, but the output is somewhat obfuscated, and I am not going to
741 explain what +mystery does in this document).
742
743 == Release 0.24 (by Eric S. Raymond) 31 Dec 2003 ==
744
745 1. RPMs can now be built by non-root users.
746
747 == Release 0.23 (by Eric S. Raymond) 9 Mar 2003 ==
748
749 1. More fixes by Don Knuth. He found a bug in the type-propagation logic.
750
751 == Release 0.22 (by Eric S. Raymond) 4 Mar 2003 ==
752
753 1. Incorporated a minor fix patch from Donald Knuth. Yes, *the*
754 Donald Knuth. Seems he found a weakness in the code for handling
755 comments, and fixed it to do multiline comments properly. He also made
756 some improvements to intercal.el, and contributed a test program.
757
758 2. Updated build machinery. This package now uses GNU autoconf.
759
760 == Release 0.21 (by Eric S. Raymond) 30 July 2002 ==
761
762 1. Jonathan Nieder sent a grammar fix for BLR's old bug #5 (a bad spark-ears
763 interaction.)
764
765 2. Latin-1 support and makefile fixes from Magnus Bodin.
766
767 == Release 0.20 (by Eric S. Raymond) 16 September 1999 ==
768
769 (The following changes are due to John Cowan <[email protected]>)
770
771 1. In addition to Latin-1, UTF-8 is now acceptable in source code as well.
772 All the numerous currency symbols of Unicode (ISO 10646) can be used as
773 big-money alternatives in their UTF-8 format. Indeed, a random mixture of
774 Latin-1 and UTF-8 also works fine; this seems very much in the spirit of
775 Intercal.
776
777 2. Intercal has now been localized for the Ancient Roman as well as the
778 Posix/C locale. You may input numbers in Latin and write Intercal programs
779 using Latin keywords. See lexer.l for a list of equivalences.
780
781 3. The Makefile in the pit directory now works like a real Makefile, and
782 won't compile anything that's already been compiled.
783
784 == Release 0.19 (by Eric S. Raymond) 16 August 1998 ==
785
786 1. Fixed a code-generation error in assignments from tail and hybrid
787 variables. (Error pointed out by Malcom Ryan.)
788
789 == Release 0.18 (by Eric S. Raymond) 29 April 1998 ==
790
791 (The following changes are due to John Cowan <[email protected]>)
792
793 1. The source code is now POSIXly correct. By making the POSIX
794 variable in the Makefile undefined, you can restore the use of
795 <varargs.h>, but otherwise <stdarg.h> is now used.
796
797 2. Latin-1 (ISO 8859-1) is now the standard input charset of C-INTERCAL.
798 This means that the change operator (0xA2) has returned as an alternative to
799 big money ($). Furthermore, the quid (0xA3) and zlotnik (0xA4) are also
800 acceptable. For those using Latin-9 (ISO 8859-15), the euro (0xA4) will also
801 work. ("Euro", like "ampersand", cannot be sillified further.)
802 New names have been added to intercal.mm.
803
804 3. A fortiori, input of Volap\"uk digits may now be in Latin-1 or TeX format.
805
806 4. Yacc is now deprecated in favor of bison; some old yaccs don't do proper
807 error recovery, and lines like "DO ~ ERROR" failed at compile time.
808
809 5. Some minor portability bugs on systems where ptr_diff is not the same
810 as int were removed by casting expressions of the form (ptr - ptr) to int.
811
812 == Release 0.17 (by Eric S. Raymond) 27 Feb 1998 ==
813
814 Incorporates a minor bug fix by Jeff Uphoff <[email protected]>.
815
816 == Release 0.16 (by Brian Raiter and Eric S. Raymond) 26 May 1997 ==
817
818 This is the 25th-anniversary release of INTERCAL. This language, conceived
819 in irony a quarter century ago today, has outlasted many more serious and
820 high-minded essays in computer language design. This fact should give us
821 pause to reflect, or possibly frighten the crap out of us. Or both.
822
823 1. Corrected manual bugs.
824
825 2. Fixed over-enthusiastic install production in the Makefile.
826
827 3. DO PLEASE is no longer accepted (Brian Raiter pointed out that the
828 manual does not sanction it).
829
830 4. Fixed various bugs in the optimizer.
831
832 5. Incorporated fixes made by William Walter Patterson for bugs that
833 appeared when multiple files are given on the command line.
834
835 (The following changes are due to Brian Raiter)
836
837 6. The system library is now available for TriINTERCAL (all bases, 3 to 7).
838 The compiler has been updated to use the right library automagically.
839
840 7. The error 774 feature has been brought into conformance with the
841 Princeton compiler. The odds are now a flat 10% (but still 0% with -b).
842
843 8. The compiler has been modified to allocate memory dynamically, and
844 the hard-coded limits on program size and number of variables has
845 been removed. Programs are now only limited by the amount of
846 available memory.
847
848 9. Array output now fflushes after outputting newlines.
849
850 10. Two minor bug fixes: one in wimpmode input, which was clamping
851 input to the signed integer range, and one in politesse checking.
852 The latter was causing programs to be recognized as impolite only
853 if their ratio was less that one-sixth, not one-fifth. (Fixing this
854 also necessitated adding a few PLEASEs to pit/rot13.i.)
855
856 11. Several new programs have been tossed into the pit.
857
858 == Release 0.15 (by Eric S. Raymond) 20 Jun 1996 ==
859
860 1. Switched the INTERCAL manual master to mgm in order to be able to handle
861 section crossreferences properly. The manual has now been carefully
862 proofread against the paper version by ESR and seems to match. It's
863 also been independently proofread for typos by Brian Raiter.
864
865 2. The optimizer now does complete analysis and evaluation of constant
866 expressions at compile time.
867
868 3. Added optimization templates for ~, |, &, ^ expressions. Also added
869 |, &, ^ idioms to the system library, and documented them.
870
871 (The following changes are due to Brian Raiter)
872
873 4. Every one of the INTERCAL-72 messages is now emitted when appropriate.
874 including the previously un-emitted E000, E197, and E200 (and see below).
875
876 5. A close reading (while proofing) of the sections of the manual on
877 error messages seems to indicate that splatted lines were only
878 splatted in the default listing printed after a successful
879 compile. (This seems to be a standard feature of IBM compilers; I
880 remember that RPG did this as well.) My understanding of the manual is
881 when the syntactically offensive line is encountered at runtime, it is
882 printed as error E000. I have modified the code to do this.
883
884 6. I have fixed INTERCAL in order to add another bug. Namely, I have
885 implemented the BUG/NOBUG option in the Princeton compiler. By
886 default, there is a chance that INTERCAL will randomly insert error
887 E774 into your executable. (The odds are 1/128 per statement, not
888 counting COME FROMs. Is there any way to find out what the
889 Princeton compiler's odds were?) The new option -b turns this
890 feature off.
891
892 7. The new file pit/lib/numio.i contains routines for wimpmode I/O.
893 These routines are documented in the file pit/lib/numio.doc.
894 It is my earnest hope that C-INTERCAL programmers everywhere will
895 flock to these handy routines, and C-INTERCAL users will thus no
896 longer need to fear that they might succumb to the foul and awful
897 temptation to use the dastardly +wimpmode option.
898
899 == Release 0.14 (by Eric S. Raymond) 9 Jun 1996 ==
900
901 1. The documentation has been reorganized and cleaned up (you can now
902 format either a C-INTERCAL or INTERCAL-72 manual from the same master).
903 The diagrams in the paper original have been transcribed as pic figures.
904 See doc/READ.ME for details.
905
906 2. Yet more code. The pit now includes a second random-number generator,
907 a ROT13 filter, and even two programs in TriINTERCAL!
908
909 3. Louis Howell's corrected system library is now included.
910
911 4. The distribution is now formally GPLed.
912
913 == Release 0.13 (by Eric S. Raymond) 5 Jun 1996 ==
914
915 1. I folded in patches by Brian Raiter <[email protected]> that port
916 this code to Linux, ANSIfies it, and enables it to use flex. Note: This
917 version has been tested on Linux and SunOS but not yet elsewhere.
918
919 2. Another piece of nontrivial INTERCAL code has been found! (Doubtless
920 hell is freezing over even as I write.) We've added to the pit the
921 `beer' program by Matt DiMeo <[email protected]>, courtesy of the
922 infamous Beer List (URL:http://www.ionet.net/~timtroyr/funhouse/beer.html).
923
924 == Release 0.12 (by Eric S. Raymond) 22 Mar 1995 ==
925
926 1. The code now compiles and runs under Linux, using bison and lex. It does
927 *not* yet work with flex; C.P. Osborne's patches for 0.11 turned out not
928 to be sufficient (at least for flex 2.4.6 under Linux using the -l option).
929 See the BUGS list.
930
931 2. Note 3 under 0.11 was incorrect. We're in conformance after all.
932
933 3. The politesse check is a little smarter about programs too small to
934 conform to the 1/5-to-1/3 criterion. All the pit programs have been
935 checked for conformance. Also, the politesse error message now tells
936 you the offending program's line count.
937
938 == Release 0.11 (by Eric S. Raymond) 24 Sep 1994 ==
939
940 1. Incorporated changes by Jan-Pieter Cornet <[email protected]> for
941 magical inclusion of the system library. These address an outstanding bug
942
943 2. (SS) The `system library' should be precompiled and `linked' to user
944 written code, at least optionally. Doing this with the current compiler
945 presents a few problems.
946
947 which has accordingly been removed from the BUGS file. They also fix a
948 previously unknown bug; due to a minor oversight in the lexer, it was not
949 possible to ABSTAIN FROM WRITING IN or READING OUT. Jan also contributed
950 a test program that demonstrates gerund abstention.
951
952 2. Despite the obvious appeal of the idea, and much discussion on
953 alt.lang.intercal, I decided *not* to add Klingon; maintaining two parallel
954 grammars just so the compiler could support SVO and OSV languages would be
955 too painful. There was also a minor mess near case-sensitivity in the
956 digits. The workaround kluges I was offered instead failed to satisfy.
957
958 3. We fess up to a violation of the INTERCAL-72 standard. That implementation,
959 instead of accepting an entire multiple-digit number per line of input,
960 expected one digit per *card image* (input line). It is unclear how
961 INTERCAL-72 programs ever managed to input more than one number.
962
963 4. Added patches by C.P.Osborne <[email protected]> to make the lexer work
964 using flex, for Linux and BSD/386 support. I didn't use his error-reporting
965 patch, though; instead, I changed the parser to catch errors at logical-line
966 level. I also kept the N in `RESUBNIT'; this is a carefully preserved
967 feature of INTERCAL-72.
968
969 5. Volapuk digits are now recognized on input. TO DO: Recognize Volapuk
970 keywords, including gerunds.
971
972 == Release 0.10 (by Eric S. Raymond) 9 Sep 1994 ==
973
974 1. It is now possible to undo a line-number abstention with a gerund
975 reinstatement, or undo a gerund abstention with a line-number
976 reinstatement. In previous versions, gerund abstention/reinstatements and
977 line-number abstention/reinstatements were independent of each other.
978
979 2. After release 0.5, I wrote in the TO DO list as follows:
980
981 2. (ESR) Input format internationalization -- allow WRITE IN input digits in
982 major languages such as Nahuatl, Tagalog, Sanskrit, and Basque.
983
984 The twisted loons in the alt.folklore.computers crowd loved this idea,
985 and I actually got sent digit lists for Nahuatl, Tagalog, Sanskrit, and
986 Basque -- also, Kwakiutl, Georgian, Ojibwe, and Albanian. I've left
987 out Albanian (didn't want to keep track of the dipthong diacritical)
988 and Ojibwe (no zero digit). So: Nahuatl, Tagalog, Sanskrit, Basque,
989 Georgian, and Kwakiutl are now supported in addition to English.
990
991 3. The THEORY document has been significantly revised.
992
993 == Release 0.9 (by Eric S. Raymond) 5 Mar 1993 ==
994
995 1. An embarrassing port bug in the INTERCAL skeleton has been fixed
996
997 2. Jamie Zawinski's suggestion for an improved E333 error message has
998 been accepted.
999
1000 3. Added and improved documentation for pit programs.
1001
1002 4. The install production in the Makefile now puts lose.h in place
1003 as well as cesspool.h.
1004
1005 5. You can now set the C compiler to be used as a back end by INTERCAL
1006 by setting the environment variable CC. The default is cc.
1007
1008 == Release 0.8 (many hacks on 0.7 by Eric Raymond) 21 Jan 1992 ==
1009
1010 1. A previously undocumented `feature' of the Princeton compiler,
1011 revealed by Don Woods, is now emulated. If fewer than 1/5 or more
1012 than 1/3 of a program's statements contain the qualifier `PLEASE', the
1013 program is rejected with an entertaining error message. Programs in pit
1014 have been hacked to pass the politesse test.
1015
1016 2. The `sample' program now works correctly, thanks to Louis Howell. Also,
1017 he has contributed new programs to calculate pi and to search for prime
1018 numbers.
1019
1020 3. An intercal mode for GNU Emacs is included. It helps programmers deal with
1021 the new politesse check by randomly expanding the abbrev "do" to DO PLEASE
1022 or PLEASE DO about 1/4 of the time.
1023
1024 4. Improvements in code generation for NEXT/RESUME.
1025
1026 5. Fixes to the optimizer; it handles the results of optimized MINGLEs
1027 correctly now. Also, added a rewrite rule that checks for the
1028 equality-test idiom using XOR.
1029
1030 6. The original error messages from the Princeton compiler have been restored.
1031
1032 The documentation has been updated to reflect these changes.
1033
1034 == Release 0.7 (by Louis Howell) 21 Dec 1991 ==
1035
1036 In addition to bug fixes, new features in this version include arrays,
1037 character I/O, modified (hopefully improved!) semantics for the COME
1038 FROM statement, extensions to bases other than 2, new INTERCAL programs,
1039 and new documentation.
1040
1041 == Release 0.6 (by Steve Swales) 5 Jan 1991 ==
1042
1043 1. Restored Princeton documentation (courtesy of Michael Ernst).
1044
1045 2. Added -@ usage option to ick.
1046
1047 3. Added [+/-]help, [+/-]traditional, and [+/-]wimpmode options for
1048 generated programs; [+/-]traditional is a no-op at this point.
1049
1050 4. Various bug fixes, esp. in lexical analysis.
1051
1052 == Release 0.5 (ESR's second pre-release) 5 Jun 1990 ==
1053
1054 1. As a convenience to all you junior birdmen out there, `NINER' is accepted as
1055 a synonym for `NINE' in INTERCAL input.
1056
1057 2. The COME FROM statement is now compiled.
1058
1059 == Release 0.3 (ESR's original pre-release) ==
1060
1061 The resurrection of INTERCAL from its grave took place on 12 May 1990
1062 when ESR released 0.3 on an unsuspecting USENET.