5 This is a production release that fixes the build on FreeBSD.
7 There was a syntax error in `configure.sh` that the Linux shell did not catch,
8 and FreeBSD depends on the existence of `tests/all.sh`.
10 All users that already upgraded to `3.2.5` should update to this release, with
11 my apologies for the poor release of `3.2.5`. Other users should skip `3.2.5` in
12 favor of this version.
16 This is a production release that fixes several bugs and adds a couple small
19 The two most important bugs were bugs that causes `dc` to access memory
20 out-of-bounds (crash in debug builds). This was found by upgrading to `afl++`
21 from `afl`. Both were caused by a failure to distinguish between the same two
24 Another bug was the failure to put all of the licenses in the `LICENSE.md` file.
26 Third, some warnings by `scan-build` were found and eliminated. This needed one
27 big change: `bc` and `dc` now bail out as fast as possible on fatal errors
28 instead of unwinding the stack.
30 Fourth, the pseudo-random number now attempts to seed itself with `/dev/random`
31 if `/dev/urandom` fails.
33 Finally, this release has a few quality-of-life changes to the build system. The
34 usage should not change at all; the only thing that changed was making sure the
35 `Makefile.in` was written to rebuild properly when headers changed and to not
36 rebuild when not necessary.
40 This is a production release that fixes a warning on `gcc` 6 or older, which
41 does not have an attribute that is used.
43 Users do ***NOT*** need to upgrade if they don't use `gcc` 6 or older.
47 This is a production release that fixes a bug in `gen/strgen.sh`. I recently
48 changed `gen/strgen.c`, but I did not change `gen/strgen.sh`.
50 Users that do not use `gen/strgen.sh` do not need to upgrade.
54 This is a production release that fixes a portability bug in `configure.sh`. The
55 bug was using the GNU `find` extension `-wholename`.
59 This is a production release that has one fix for `bcl(3)`. It is technically
60 not a bug fix since the behavior is undefined, but the `BclNumber`s that
61 `bcl_divmod()` returns will be set to `BCL_ERROR_INVALID_NUM` if there is an
62 error. Previously, they were not set.
66 This is a production release that has one bug fix and a major addition.
68 The bug fix was a missing `auto` variable in the bessel `j()` function in the
71 The major addition is a way to build a version of `bc`'s math code as a library.
72 This is done with the `-a` option to `configure.sh`. The API for the library can
73 be read in `./manuals/bcl.3.md` or `man bcl` once the library is installed with
76 This library was requested by developers before I even finished version 1.0, but
77 I could not figure out how to do it until now.
79 If the library has API breaking changes, the major version of `bc` will be
84 This is a production release that fixes a new warning from Clang 12 for FreeBSD
85 and also removes some possible undefined behavior found by UBSan that compilers
86 did not seem to take advantage of.
88 Users do ***NOT*** need to upgrade, if they do not want to.
92 This is a production release that fixes the Chinese locales (which caused `bc`
93 to crash) and a crash caused by `bc` executing code when it should not have been
96 ***ALL USERS SHOULD UPGRADE.***
100 This is a production release that fixes one bug, changes two behaviors, and
101 removes one environment variable.
103 The bug is like the one in the last release except it applies if files are being
104 executed. I also made the fix more general.
106 The behavior that was changed is that `bc` now exits when given `-e`, `-f`,
107 `--expression` or `--file`. However, if the last one of those is `-f-` (using
108 `stdin` as the file), `bc` does not exit. If `-f-` exists and is not the last of
109 the `-e` and `-f` options (and equivalents), `bc` gives a fatal error and exits.
111 Next, I removed the `BC_EXPR_EXIT` and `DC_EXPR_EXIT` environment variables
112 since their use is not needed with the behavior change.
114 Finally, I made it so `bc` does not print the header, though the `-q` and
115 `--quiet` options were kept for compatibility with GNU `bc`.
119 This is a production release that fixes one minor bug: if `bc` was invoked like
120 the following, it would error:
123 echo "if (1 < 3) 1" | bc
126 Unless users run into this bug, they do not need to upgrade, but it is suggested
131 This is a production release that adds a way to install *all* locales. Users do
132 ***NOT*** need to upgrade.
134 For package maintainers wishing to make use of the change, just pass `-l` to
139 This is a production release that adds two Spanish locales. Users do ***NOT***
140 need to upgrade, unless they want those locales.
144 This is a production release that adjusts one behavior, fixes eight bugs, and
145 improves manpages for FreeBSD. Because this release fixes bugs, **users and
146 package maintainers should update to this version as soon as possible**.
148 The behavior that was adjusted was how code from the `-e` and `-f` arguments
149 (and equivalents) were executed. They used to be executed as one big chunk, but
150 in this release, they are now executed line-by-line.
152 The first bug fix in how output to `stdout` was handled in `SIGINT`. If a
153 `SIGINT` came in, the `stdout` buffer was not correctly flushed. In fact, a
154 clean-up function was not getting called. This release fixes that bug.
156 The second bug is in how `dc` handled input from `stdin`. This affected `bc` as
157 well since it was a mishandling of the `stdin` buffer.
159 The third fixed bug was that `bc` and `dc` could `abort()` (in debug mode) when
160 receiving a `SIGTERM`. This one was a race condition with pushing and popping
161 items onto and out of vectors.
163 The fourth bug fixed was that `bc` could leave extra items on the stack and
164 thus, not properly clean up some memory. (The memory would still get
165 `free()`'ed, but it would not be `free()`'ed when it could have been.)
167 The next two bugs were bugs in `bc`'s parser that caused crashes when executing
170 The last two bugs were crashes in `dc` that resulted from mishandling of
173 The manpage improvement was done by switching from [ronn][20] to [Pandoc][21] to
174 generate manpages. Pandoc generates much cleaner manpages and doesn't leave
175 blank lines where they shouldn't be.
179 This is a production release that adds one new feature: specific manpages.
181 Before this release, `bc` and `dc` only used one manpage each that referred to
182 various build options. This release changes it so there is one manpage set per
183 relevant build type. Each manual only has information about its particular
184 build, and `configure.sh` selects the correct set for install.
188 This is a production release that adds `utf8` locale symlinks and removes an
189 unused `auto` variable from the `ceil()` function in the [extended math
192 Users do ***NOT*** need to update unless they want the locales.
196 This is a production release with two small changes. Users do ***NOT*** need to
197 upgrade to this release; however, if they haven't upgraded to `3.0.0` yet, it
198 may be worthwhile to upgrade to this release.
200 The first change is fixing a compiler warning on FreeBSD with strict warnings
203 The second change is to make the new implementation of `ceil()` in `lib2.bc`
208 *Notes for package maintainers:*
210 *First, the `2.7.0` release series saw a change in the option parsing. This made
211 me change one error message and add a few others. The error message that was
212 changed removed one format specifier. This means that `printf()` will seqfault
213 on old locale files. Unfortunately, `bc` cannot use any locale files except the
214 global ones that are already installed, so it will use the previous ones while
215 running tests during install. **If `bc` segfaults while running arg tests when
216 updating, it is because the global locale files have not been replaced. Make
217 sure to either prevent the test suite from running on update or remove the old
218 locale files before updating.** (Removing the locale files can be done with
219 `make uninstall` or by running the `locale_uninstall.sh` script.) Once this is
220 done, `bc` should install without problems.*
222 *Second, **the option to build without signal support has been removed**. See
223 below for the reasons why.*
225 This is a production release with some small bug fixes, a few improvements,
226 three major bug fixes, and a complete redesign of `bc`'s error and signal
227 handling. **Users and package maintainers should update to this version as soon
230 The first major bug fix was in how `bc` executed files. Previously, a whole file
231 was parsed before it was executed, but if a function is defined *after* code,
232 especially if the function definition was actually a redefinition, and the code
233 before the definition referred to the previous function, this `bc` would replace
234 the function before executing any code. The fix was to make sure that all code
235 that existed before a function definition was executed.
237 The second major bug fix was in `bc`'s `lib2.bc`. The `ceil()` function had a
238 bug where a `0` in the decimal place after the truncation position, caused it to
239 output the wrong numbers if there was any non-zero digit after.
241 The third major bug is that when passing parameters to functions, if an
242 expression included an array (not an array element) as a parameter, it was
243 accepted, when it should have been rejected. It is now correctly rejected.
245 Beyond that, this `bc` got several improvements that both sped it up, improved
246 the handling of signals, and improved the error handling.
248 First, the requirements for `bc` were pushed back to POSIX 2008. `bc` uses one
249 function, `strdup()`, which is not in POSIX 2001, and it is in the X/Open System
250 Interfaces group 2001. It is, however, in POSIX 2008, and since POSIX 2008 is
251 old enough to be supported anywhere that I care, that should be the requirement.
253 Second, the BcVm global variable was put into `bss`. This actually slightly
254 reduces the size of the executable from a massive code shrink, and it will stop
255 `bc` from allocating a large set of memory when `bc` starts.
257 Third, the default Karatsuba length was updated from 64 to 32 after making the
258 optimization changes below, since 32 is going to be better than 64 after the
261 Fourth, Spanish translations were added.
263 Fifth, the interpreter received a speedup to make performance on non-math-heavy
264 scripts more competitive with GNU `bc`. While improvements did, in fact, get it
265 much closer (see the [benchmarks][19]), it isn't quite there.
267 There were several things done to speed up the interpreter:
269 First, several small inefficiencies were removed. These inefficiencies included
270 calling the function `bc_vec_pop(v)` twice instead of calling
271 `bc_vec_npop(v, 2)`. They also included an extra function call for checking the
272 size of the stack and checking the size of the stack more than once on several
275 Second, since the current `bc` function is the one that stores constants and
276 strings, the program caches pointers to the current function's vectors of
277 constants and strings to prevent needing to grab the current function in order
278 to grab a constant or a string.
280 Third, `bc` tries to reuse `BcNum`'s (the internal representation of
281 arbitary-precision numbers). If a `BcNum` has the default capacity of
282 `BC_NUM_DEF_SIZE` (32 on 64-bit and 16 on 32-bit) when it is freed, it is added
283 to a list of available `BcNum`'s. And then, when a `BcNum` is allocated with a
284 capacity of `BC_NUM_DEF_SIZE` and any `BcNum`'s exist on the list of reusable
285 ones, one of those ones is grabbed instead.
287 In order to support these changes, the `BC_NUM_DEF_SIZE` was changed. It used to
288 be 16 bytes on all systems, but it was changed to more closely align with the
289 minimum allocation size on Linux, which is either 32 bytes (64-bit musl), 24
290 bytes (64-bit glibc), 16 bytes (32-bit musl), or 12 bytes (32-bit glibc). Since
291 these are the minimum allocation sizes, these are the sizes that would be
292 allocated anyway, making it worth it to just use the whole space, so the value
293 of `BC_NUM_DEF_SIZE` on 64-bit systems was changed to 32 bytes.
295 On top of that, at least on 64-bit, `BC_NUM_DEF_SIZE` supports numbers with
296 either 72 integer digits or 45 integer digits and 27 fractional digits. This
297 should be more than enough for most cases since `bc`'s default `scale` values
298 are 0 or 20, meaning that, by default, it has at most 20 fractional digits. And
299 45 integer digits are *a lot*; it's enough to calculate the amount of mass in
300 the Milky Way galaxy in kilograms. Also, 72 digits is enough to calculate the
301 diameter of the universe in Planck lengths.
303 (For 32-bit, these numbers are either 32 integer digits or 12 integer digits and
304 20 fractional digits. These are also quite big, and going much bigger on a
305 32-bit system seems a little pointless since 12 digits in just under a trillion
306 and 20 fractional digits is still enough for about any use since `10^-20` light
307 years is just under a millimeter.)
309 All of this together means that for ordinary uses, and even uses in scientific
310 work, the default number size will be all that is needed, which means that
311 nearly all, if not all, numbers will be reused, relieving pressure on the system
314 I did several experiments to find the changes that had the most impact,
315 especially with regard to reusing `BcNum`'s. One was putting `BcNum`'s into
316 buckets according to their capacity in powers of 2 up to 512. That performed
317 worse than `bc` did in `2.7.2`. Another was putting any `BcNum` on the reuse
318 list that had a capacity of `BC_NUM_DEF_SIZE * 2` and reusing them for `BcNum`'s
319 that requested `BC_NUM_DEF_SIZE`. This did reduce the amount of time spent, but
320 it also spent a lot of time in the system allocator for an unknown reason. (When
321 using `strace`, a bunch more `brk` calls showed up.) Just reusing `BcNum`'s that
322 had exactly `BC_NUM_DEF_SIZE` capacity spent the smallest amount of time in both
323 user and system time. This makes sense, especially with the changes to make
324 `BC_NUM_DEF_SIZE` bigger on 64-bit systems, since the vast majority of numbers
325 will only ever use numbers with a size less than or equal to `BC_NUM_DEF_SIZE`.
327 Last of all, `bc`'s signal handling underwent a complete redesign. (This is the
328 reason that this version is `3.0.0` and not `2.8.0`.) The change was to move
329 from a polling approach to signal handling to an interrupt-based approach.
331 Previously, every single loop condition had a check for signals. I suspect that
332 this could be expensive when in tight loops.
334 Now, the signal handler just uses `longjmp()` (actually `siglongjmp()`) to start
335 an unwinding of the stack until it is stopped or the stack is unwound to
336 `main()`, which just returns. If `bc` is currently executing code that cannot be
337 safely interrupted (according to POSIX), then signals are "locked." The signal
338 handler checks if the lock is taken, and if it is, it just sets the status to
339 indicate that a signal arrived. Later, when the signal lock is released, the
340 status is checked to see if a signal came in. If so, the stack unwinding starts.
342 This design eliminates polling in favor of maintaining a stack of `jmp_buf`'s.
343 This has its own performance implications, but it gives better interaction. And
344 the cost of pushing and popping a `jmp_buf` in a function is paid at most twice.
345 Most functions do not pay that price, and most of the rest only pay it once.
346 (There are only some 3 functions in `bc` that push and pop a `jmp_buf` twice.)
348 As a side effect of this change, I had to eliminate the use of `stdio.h` in `bc`
349 because `stdio` does not play nice with signals and `longjmp()`. I implemented
350 custom I/O buffer code that takes a fraction of the size. This means that static
351 builds will be smaller, but non-static builds will be bigger, though they will
352 have less linking time.
354 This change is also good because my history implementation was already bypassing
355 `stdio` for good reasons, and unifying the architecture was a win.
357 Another reason for this change is that my `bc` should *always* behave correctly
358 in the presence of signals like `SIGINT`, `SIGTERM`, and `SIGQUIT`. With the
359 addition of my own I/O buffering, I needed to also make sure that the buffers
360 were correctly flushed even when such signals happened.
362 For this reason, I **removed the option to build without signal support**.
364 As a nice side effect of this change, the error handling code could be changed
365 to take advantage of the stack unwinding that signals used. This means that
366 signals and error handling use the same code paths, which means that the stack
367 unwinding is well-tested. (Errors are tested heavily in the test suite.)
369 It also means that functions do not need to return a status code that
370 ***every*** caller needs to check. This eliminated over 100 branches that simply
371 checked return codes and then passed that return code up the stack if necessary.
372 The code bloat savings from this is at least 1700 bytes on `x86_64`, *before*
373 taking into account the extra code from removing `stdio.h`.
377 This is a production release with one major bug fix.
379 The `length()` built-in function can take either a number or an array. If it
380 takes an array, it returns the length of the array. Arrays can be passed by
381 reference. The bug is that the `length()` function would not properly
382 dereference arrays that were references. This is a bug that affects all users.
384 **ALL USERS SHOULD UPDATE `bc`**.
388 This is a production release with fixes for new locales and fixes for compiler
393 This is a production release with a bug fix for Linux, new translations, and new
398 * Option parsing in `BC_ENV_ARGS` was broken on Linux in 2.6.1 because `glibc`'s
399 `getopt_long()` is broken. To get around that, and to support long options on
400 every platform, an adapted version of [`optparse`][17] was added. Now, `bc`
401 does not even use `getopt()`.
402 * Parsing `BC_ENV_ARGS` with quotes now works. It isn't the smartest, but it
403 does the job if there are spaces in file names.
405 The following new languages are supported:
413 All of these translations were generated using [DeepL][18], so improvements are
416 There is only one new feature: **`bc` now has a built-in pseudo-random number
419 The PRNG is seeded, making it useful for applications where
420 `/dev/urandom` does not work because output needs to be reproducible. However,
421 it also uses `/dev/urandom` to seed itself by default, so it will start with a
422 good seed by default.
424 It also outputs 32 bits on 32-bit platforms and 64 bits on 64-bit platforms, far
425 better than the 15 bits of C's `rand()` and `bash`'s `$RANDOM`.
427 In addition, the PRNG can take a bound, and when it gets a bound, it
428 automatically adjusts to remove bias. It can also generate numbers of arbitrary
429 size. (As of the time of release, the largest pseudo-random number generated by
430 this `bc` was generated with a bound of `2^(2^20)`.)
432 ***IMPORTANT: read the [`bc` manual][9] and the [`dc` manual][10] to find out
433 exactly what guarantees the PRNG provides. The underlying implementation is not
434 guaranteed to stay the same, but the guarantees that it provides are guaranteed
435 to stay the same regardless of the implementation.***
437 On top of that, four functions were added to `bc`'s [extended math library][16]
438 to make using the PRNG easier:
440 * `frand(p)`: Generates a number between `[0,1)` to `p` decimal places.
441 * `ifrand(i, p)`: Generates an integer with bound `i` and adds it to `frand(p)`.
442 * `srand(x)`: Randomizes the sign of `x`. In other words, it flips the sign of
443 `x` with probability `0.5`.
444 * `brand()`: Returns a random boolean value (either `0` or `1`).
448 This is a production release with a bug fix for FreeBSD.
450 The bug was that when `bc` was built without long options, it would give a fatal
451 error on every run. This was caused by a mishandling of `optind`.
455 This release is a production release ***with no bugfixes***. If you do not want
456 to upgrade, you don't have to.
458 No source code changed; the only thing that changed was `lib2.bc`.
460 This release adds one function to the [extended math library][16]: `p(x, y)`,
461 which calculates `x` to the power of `y`, whether or not `y` is an integer. (The
462 `^` operator can only accept integer powers.)
464 This release also includes a couple of small tweaks to the [extended math
465 library][16], mostly to fix returning numbers with too high of `scale`.
469 This release is a production release which addresses inconsistencies in the
470 Portuguese locales. No `bc` code was changed.
472 The issues were that the ISO files used different naming, and also that the
473 files that should have been symlinks were not. I did not catch that because
474 GitHub rendered them the exact same way.
478 This release is a production release.
480 No code was changed, but the build system was changed to allow `CFLAGS` to be
481 given to `CC`, like this:
484 CC="gcc -O3 -march=native" ./configure.sh
487 If this happens, the flags are automatically put into `CFLAGS`, and the compiler
488 is set appropriately. In the example above this means that `CC` will be "gcc"
489 and `CFLAGS` will be "-O3 -march=native".
491 This behavior was added to conform to GNU autotools practices.
495 This is a production release which addresses portability concerns discovered
496 in the `bc` build system. No `bc` code was changed.
498 * Support for Solaris SPARC and AIX were added.
499 * Minor documentations edits were performed.
500 * An option for `configure.sh` was added to disable long options if
501 `getopt_long()` is missing.
505 This is a production release with new translations. No code changed.
507 The translations were contributed by [bugcrazy][15], and they are for
508 Portuguese, both Portugal and Brazil locales.
512 This is a production release primarily aimed at improving `dc`.
514 * A couple of copy and paste errors in the [`dc` manual][10] were fixed.
515 * `dc` startup was optimized by making sure it didn't have to set up `bc`-only
517 * The `bc` `&&` and `||` operators were made available to `dc` through the `M`
518 and `m` commands, respectively.
519 * `dc` macros were changed to be tail call-optimized.
521 The last item, tail call optimization, means that if the last thing in a macro
522 is a call to another macro, then the old macro is popped before executing the
523 new macro. This change was made to stop `dc` from consuming more and more memory
524 as macros are executed in a loop.
526 The `q` and `Q` commands still respect the "hidden" macros by way of recording
527 how many macros were removed by tail call optimization.
531 This is a production release meant to fix warnings in the Gentoo `ebuild` by
532 making it possible to disable binary stripping. Other users do *not* need to
537 This is a production release. It fixes a bug that caused `-1000000000 < -1` to
538 return `0`. This only happened with negative numbers and only if the value on
539 the left was more negative by a certain amount. That said, this bug *is* a bad
540 bug, and needs to be fixed.
542 **ALL USERS SHOULD UPDATE `bc`**.
546 This is a production release with changes to the build system.
550 This release is a production release. It only has new features and performance
553 1. The performance of `sqrt(x)` was improved.
554 2. The new function `root(x, n)` was added to the extended math library to
555 calculate `n`th roots.
556 3. The new function `cbrt(x)` was added to the extended math library to
557 calculate cube roots.
561 This is a non-critical release; it just changes the build system, and in
564 1. Linked locale files were changed to link to their sources with a relative
566 2. A bug in `configure.sh` that caused long option parsing to fail under `bash`
571 This release is not a critical release.
573 1. A few codes were added to history.
574 2. Multiplication was optimized a bit more.
575 3. Addition and subtraction were both optimized a bit more.
579 This release contains a fix for the test suite made for Linux from Scratch: now
580 the test suite prints `pass` when a test is passed.
582 Other than that, there is no change in this release, so distros and other users
583 do not need to upgrade.
587 This release is a production release.
589 The following bugs were fixed:
591 1. A `dc` bug that caused stack mishandling was fixed.
592 2. A warning on OpenBSD was fixed.
593 3. Bugs in `ctrl+arrow` operations in history were fixed.
594 4. The ability to paste multiple lines in history was added.
595 5. A `bc` bug, mishandling of array arguments to functions, was fixed.
596 6. A crash caused by freeing the wrong pointer was fixed.
597 7. A `dc` bug where strings, in a rare case, were mishandled in parsing was
600 In addition, the following changes were made:
602 1. Division was slightly optimized.
603 2. An option was added to the build to disable printing of prompts.
604 3. The special case of empty arguments is now handled. This is to prevent
605 errors in scripts that end up passing empty arguments.
606 4. A harmless bug was fixed. This bug was that, with the pop instructions
607 (mostly) removed (see below), `bc` would leave extra values on its stack for
608 `void` functions and in a few other cases. These extra items would not
609 affect anything put on the stack and would not cause any sort of crash or
610 even buggy behavior, but they would cause `bc` to take more memory than it
613 On top of the above changes, the following optimizations were added:
615 1. The need for pop instructions in `bc` was removed.
616 2. Extra tests on every iteration of the interpreter loop were removed.
617 3. Updating function and code pointers on every iteration of the interpreter
618 loop was changed to only updating them when necessary.
619 4. Extra assignments to pointers were removed.
621 Altogether, these changes sped up the interpreter by around 2x.
623 ***NOTE***: This is the last release with new features because this `bc` is now
624 considered complete. From now on, only bug fixes and new translations will be
629 This is a production, bug-fix release.
631 Two bugs were fixed in this release:
633 1. A rare and subtle signal handling bug was fixed.
634 2. A misbehavior on `0` to a negative power was fixed.
636 The last bug bears some mentioning.
638 When I originally wrote power, I did not thoroughly check its error cases;
639 instead, I had it check if the first number was `0` and then if so, just return
640 `0`. However, `0` to a negative power means that `1` will be divided by `0`,
643 I caught this, but only after I stopped being cocky. You see, sometime later, I
644 had noticed that GNU `bc` returned an error, correctly, but I thought it was
645 wrong simply because that's not what my `bc` did. I saw it again later and had a
646 double take. I checked for real, finally, and found out that my `bc` was wrong
649 That was bad on me. But the bug was easy to fix, so it is fixed now.
651 There are two other things in this release:
653 1. Subtraction was optimized by [Stefan Eßer][14].
654 2. Division was also optimized, also by Stefan Eßer.
658 This release contains a fix for a possible overflow in the signal handling. I
659 would be surprised if any users ran into it because it would only happen after 2
660 billion (`2^31-1`) `SIGINT`'s, but I saw it and had to fix it.
664 This release contains very few things that will apply to any users.
666 1. A slight bug in `dc`'s interactive mode was fixed.
667 2. A bug in the test suite that was only triggered on NetBSD was fixed.
668 3. **The `-P`/`--no-prompt` option** was added for users that do not want a
670 4. A `make check` target was added as an alias for `make test`.
671 5. `dc` got its own read prompt: `?> `.
675 This release is a production release.
677 This release is also a little different from previous releases. From here on
678 out, I do not plan on adding any more features to this `bc`; I believe that it
679 is complete. However, there may be bug fix releases in the future, if I or any
680 others manage to find bugs.
682 This release has only a few new features:
684 1. `atan2(y, x)` was added to the extended math library as both `a2(y, x)` and
686 2. Locales were fixed.
687 3. A **POSIX shell-compatible script was added as an alternative to compiling
688 `gen/strgen.c`** on a host machine. More details about making the choice
689 between the two can be found by running `./configure.sh --help` or reading
690 the [build manual][13].
691 4. Multiplication was optimized by using **diagonal multiplication**, rather
692 than straight brute force.
693 5. The `locale_install.sh` script was fixed.
694 6. `dc` was given the ability to **use the environment variable
696 7. `dc` was also given the ability to **use the `-i` or `--interactive`**
698 8. Printing the prompt was fixed so that it did not print when it shouldn't.
699 9. Signal handling was fixed.
700 10. **Handling of `SIGTERM` and `SIGQUIT`** was fixed.
701 11. The **built-in functions `maxibase()`, `maxobase()`, and `maxscale()`** (the
702 commands `T`, `U`, `V` in `dc`, respectively) were added to allow scripts to
703 query for the max allowable values of those globals.
704 12. Some incompatibilities with POSIX were fixed.
706 In addition, this release is `2.0.0` for a big reason: the internal format for
707 numbers changed. They used to be a `char` array. Now, they are an array of
708 larger integers, packing more decimal digits into each integer. This has
709 delivered ***HUGE*** performance improvements, especially for multiplication,
712 This `bc` should now be the fastest `bc` available, but I may be wrong.
716 This release contains a fix for a harmless bug (it is harmless in that it still
717 works, but it just copies extra data) in the [`locale_install.sh`][12] script.
721 This version contains fixes for the build on Arch Linux.
725 This release removes the use of `local` in shell scripts because it's not POSIX
726 shell-compatible, and also updates a man page that should have been updated a
727 long time ago but was missed.
731 This release contains some missing locale `*.msg` files.
735 This release contains a few bug fixes and new French translations.
739 This release contains a fix for a bug: use of uninitialized data. Such data was
740 only used when outputting an error message, but I am striving for perfection. As
741 Michelangelo said, "Trifles make perfection, and perfection is no trifle."
745 This release contains fixes for OpenBSD.
749 This release contains bug fixes for some rare bugs.
753 This is a production release.
755 There have been several changes since `1.1.0`:
757 1. The build system had some changes.
758 2. Locale support has been added. (Patches welcome for translations.)
759 3. **The ability to turn `ibase`, `obase`, and `scale` into stacks** was added
760 with the `-g` command-line option. (See the [`bc` manual][9] for more
762 4. Support for compiling on Mac OSX out of the box was added.
763 5. The extended math library got `t(x)`, `ceil(x)`, and some aliases.
764 6. The extended math library also got `r2d(x)` (for converting from radians to
765 degrees) and `d2r(x)` (for converting from degrees to radians). This is to
766 allow using degrees with the standard library.
767 7. Both calculators now accept numbers in **scientific notation**. See the
768 [`bc` manual][9] and the [`dc` manual][10] for details.
769 8. Both calculators can **output in either scientific or engineering
770 notation**. See the [`bc` manual][9] and the [`dc` manual][10] for details.
771 9. Some inefficiencies were removed.
772 10. Some bugs were fixed.
773 11. Some bugs in the extended library were fixed.
774 12. Some defects from [Coverity Scan][11] were fixed.
778 This release contains a fix to the build system that allows it to build on older
783 This release contains a fix for a bug in the test suite where `bc` tests and
784 `dc` tests could not be run in parallel.
788 This release has a fix for a history bug; the down arrow did not work.
792 This release fixes a bug in the `1.1.0` build system. The source is exactly the
795 The bug that was fixed was a failure to install if no `EXECSUFFIX` was used.
799 This is a production release. However, many new features were added since `1.0`.
801 1. **The build system has been changed** to use a custom, POSIX
802 shell-compatible configure script ([`configure.sh`][6]) to generate a POSIX
803 make-compatible `Makefile`, which means that `bc` and `dc` now build out of
804 the box on any POSIX-compatible system.
805 2. Out-of-memory and output errors now cause the `bc` to report the error,
806 clean up, and die, rather than just reporting and trying to continue.
807 3. **Strings and constants are now garbage collected** when possible.
808 4. Signal handling and checking has been made more simple and more thorough.
809 5. `BcGlobals` was refactored into `BcVm` and `BcVm` was made global. Some
810 procedure names were changed to reflect its difference to everything else.
811 6. Addition got a speed improvement.
812 7. Some common code for addition and multiplication was refactored into its own
814 8. A bug was removed where `dc` could have been selected, but the internal
815 `#define` that returned `true` for a query about `dc` would not have
817 9. Useless calls to `bc_num_zero()` were removed.
818 10. **History support was added.** The history support is based off of a
819 [UTF-8 aware fork][7] of [`linenoise`][8], which has been customized with
820 `bc`'s own data structures and signal handling.
821 11. Generating C source from the math library now removes tabs from the library,
822 shrinking the size of the executable.
823 12. The math library was shrunk.
824 13. Error handling and reporting was improved.
825 14. Reallocations were reduced by giving access to the request size for each
827 15. **`abs()` (`b` command for `dc`) was added as a builtin.**
828 16. Both calculators were tested on FreeBSD.
829 17. Many obscure parse bugs were fixed.
830 18. Markdown and man page manuals were added, and the man pages are installed by
832 19. Executable size was reduced, though the added features probably made the
833 executable end up bigger.
834 20. **GNU-style array references were added as a supported feature.**
835 21. Allocations were reduced.
836 22. **New operators were added**: `$` (`$` for `dc`), `@` (`@` for `dc`), `@=`,
837 `<<` (`H` for `dc`), `<<=`, `>>` (`h` for `dc`), and `>>=`. See the
838 [`bc` manual][9] and the [`dc` manual][10] for more details.
839 23. **An extended math library was added.** This library contains code that
840 makes it so I can replace my desktop calculator with this `bc`. See the
841 [`bc` manual][3] for more details.
842 24. Support for all capital letters as numbers was added.
843 25. **Support for GNU-style void functions was added.**
844 26. A bug fix for improper handling of function parameters was added.
845 27. Precedence for the or (`||`) operator was changed to match GNU `bc`.
846 28. `dc` was given an explicit negation command.
847 29. `dc` was changed to be able to handle strings in arrays.
849 ## 1.1 Release Candidate 3
851 This release is the eighth release candidate for 1.1, though it is the third
852 release candidate meant as a general release candidate. The new code has not
853 been tested as thoroughly as it should for release.
855 ## 1.1 Release Candidate 2
857 This release is the seventh release candidate for 1.1, though it is the second
858 release candidate meant as a general release candidate. The new code has not
859 been tested as thoroughly as it should for release.
861 ## 1.1 FreeBSD Beta 5
863 This release is the sixth release candidate for 1.1, though it is the fifth
864 release candidate meant specifically to test if `bc` works on FreeBSD. The new
865 code has not been tested as thoroughly as it should for release.
867 ## 1.1 FreeBSD Beta 4
869 This release is the fifth release candidate for 1.1, though it is the fourth
870 release candidate meant specifically to test if `bc` works on FreeBSD. The new
871 code has not been tested as thoroughly as it should for release.
873 ## 1.1 FreeBSD Beta 3
875 This release is the fourth release candidate for 1.1, though it is the third
876 release candidate meant specifically to test if `bc` works on FreeBSD. The new
877 code has not been tested as thoroughly as it should for release.
879 ## 1.1 FreeBSD Beta 2
881 This release is the third release candidate for 1.1, though it is the second
882 release candidate meant specifically to test if `bc` works on FreeBSD. The new
883 code has not been tested as thoroughly as it should for release.
885 ## 1.1 FreeBSD Beta 1
887 This release is the second release candidate for 1.1, though it is meant
888 specifically to test if `bc` works on FreeBSD. The new code has not been tested as
889 thoroughly as it should for release.
891 ## 1.1 Release Candidate 1
893 This is the first release candidate for 1.1. The new code has not been tested as
894 thoroughly as it should for release.
898 This is the first non-beta release. `bc` is ready for production use.
900 As such, a lot has changed since 0.5.
902 1. `dc` has been added. It has been tested even more thoroughly than `bc` was
903 for `0.5`. It does not have the `!` command, and for security reasons, it
904 never will, so it is complete.
905 2. `bc` has been more thoroughly tested. An entire section of the test suite
906 (for both programs) has been added to test for errors.
907 3. A prompt (`>>> `) has been added for interactive mode, making it easier to
908 see inputs and outputs.
909 4. Interrupt handling has been improved, including elimination of race
910 conditions (as much as possible).
911 5. MinGW and [Windows Subsystem for Linux][1] support has been added (see
912 [xstatic][2] for binaries).
913 6. Memory leaks and errors have been eliminated (as far as ASan and Valgrind
915 7. Crashes have been eliminated (as far as [afl][3] can tell).
916 8. Karatsuba multiplication was added (and thoroughly) tested, speeding up
917 multiplication and power by orders of magnitude.
918 9. Performance was further enhanced by using a "divmod" function to reduce
919 redundant divisions and by removing superfluous `memset()` calls.
920 10. To switch between Karatsuba and `O(n^2)` multiplication, the config variable
921 `BC_NUM_KARATSUBA_LEN` was added. It is set to a sane default, but the
922 optimal number can be found with [`karatsuba.py`][4] (requires Python 3)
923 and then configured through `make`.
924 11. The random math test generator script was changed to Python 3 and improved.
925 `bc` and `dc` have together been run through 30+ million random tests.
926 12. All known math bugs have been fixed, including out of control memory
927 allocations in `sine` and `cosine` (that was actually a parse bug), certain
928 cases of infinite loop on square root, and slight inaccuracies (as much as
929 possible; see the [README][5]) in transcendental functions.
930 13. Parsing has been fixed as much as possible.
931 14. Test coverage was improved to 94.8%. The only paths not covered are ones
932 that happen when `malloc()` or `realloc()` fails.
933 15. An extension to get the length of an array was added.
934 16. The boolean not (`!`) had its precedence change to match negation.
935 17. Data input was hardened.
936 18. `bc` was made fully compliant with POSIX when the `-s` flag is used or
937 `POSIXLY_CORRECT` is defined.
938 19. Error handling was improved.
939 20. `bc` now checks that files it is given are not directories.
941 ## 1.0 Release Candidate 7
943 This is the seventh release candidate for 1.0. It fixes a few bugs in 1.0
946 ## 1.0 Release Candidate 6
948 This is the sixth release candidate for 1.0. It fixes a few bugs in 1.0 Release
951 ## 1.0 Release Candidate 5
953 This is the fifth release candidate for 1.0. It fixes a few bugs in 1.0 Release
956 ## 1.0 Release Candidate 4
958 This is the fourth release candidate for 1.0. It fixes a few bugs in 1.0 Release
961 ## 1.0 Release Candidate 3
963 This is the third release candidate for 1.0. It fixes a few bugs in 1.0 Release
966 ## 1.0 Release Candidate 2
968 This is the second release candidate for 1.0. It fixes a few bugs in 1.0 Release
971 ## 1.0 Release Candidate 1
973 This is the first Release Candidate for 1.0. `bc` is complete, with `dc`, but it
978 This beta release completes more features, but it is still not complete nor
979 tested as thoroughly as necessary.
983 This beta release fixes a few bugs in 0.4.
987 This is a beta release. It does not have the complete set of features, and it is
988 not thoroughly tested.
990 [1]: https://docs.microsoft.com/en-us/windows/wsl/install-win10
991 [2]: https://pkg.musl.cc/bc/
992 [3]: http://lcamtuf.coredump.cx/afl/
996 [7]: https://github.com/rain-1/linenoise-mob
997 [8]: https://github.com/antirez/linenoise
998 [9]: ./manuals/bc/A.1.md
999 [10]: ./manuals/dc/A.1.md
1000 [11]: https://scan.coverity.com/projects/gavinhoward-bc
1001 [12]: ./locale_install.sh
1002 [13]: ./manuals/build.md
1003 [14]: https://github.com/stesser
1004 [15]: https://github.com/bugcrazy
1005 [16]: ./manuals/bc/A.1.md#extended-library
1006 [17]: https://github.com/skeeto/optparse
1007 [18]: https://www.deepl.com/translator
1008 [19]: ./manuals/benchmarks.md
1009 [20]: https://github.com/apjanke/ronn-ng
1010 [21]: https://pandoc.org/