5 This is a production release that fixes one bug, changes two behaviors, and
6 removes one environment variable.
8 The bug is like the one in the last release except it applies if files are being
9 executed. I also made the fix more general.
11 The behavior that was changed is that `bc` now exits when given `-e`, `-f`,
12 `--expression` or `--file`. However, if the last one of those is `-f-` (using
13 `stdin` as the file), `bc` does not exit. If `-f-` exists and is not the last of
14 the `-e` and `-f` options (and equivalents), `bc` gives a fatal error and exits.
16 Next, I removed the `BC_EXPR_EXIT` and `DC_EXPR_EXIT` environment variables
17 since their use is not needed with the behavior change.
19 Finally, I made it so `bc` does not print the header, though the `-q` and
20 `--quiet` options were kept for compatibility with GNU `bc`.
24 This is a production release that fixes one minor bug: if `bc` was invoked like
25 the following, it would error:
28 echo "if (1 < 3) 1" | bc
31 Unless users run into this bug, they do not need to upgrade, but it is suggested
36 This is a production release that adds a way to install *all* locales. Users do
37 ***NOT*** need to upgrade.
39 For package maintainers wishing to make use of the change, just pass `-l` to
44 This is a production release that adds two Spanish locales. Users do ***NOT***
45 need to upgrade, unless they want those locales.
49 This is a production release that adjusts one behavior, fixes eight bugs, and
50 improves manpages for FreeBSD. Because this release fixes bugs, **users and
51 package maintainers should update to this version as soon as possible**.
53 The behavior that was adjusted was how code from the `-e` and `-f` arguments
54 (and equivalents) were executed. They used to be executed as one big chunk, but
55 in this release, they are now executed line-by-line.
57 The first bug fix in how output to `stdout` was handled in `SIGINT`. If a
58 `SIGINT` came in, the `stdout` buffer was not correctly flushed. In fact, a
59 clean-up function was not getting called. This release fixes that bug.
61 The second bug is in how `dc` handled input from `stdin`. This affected `bc` as
62 well since it was a mishandling of the `stdin` buffer.
64 The third fixed bug was that `bc` and `dc` could `abort()` (in debug mode) when
65 receiving a `SIGTERM`. This one was a race condition with pushing and popping
66 items onto and out of vectors.
68 The fourth bug fixed was that `bc` could leave extra items on the stack and
69 thus, not properly clean up some memory. (The memory would still get
70 `free()`'ed, but it would not be `free()`'ed when it could have been.)
72 The next two bugs were bugs in `bc`'s parser that caused crashes when executing
75 The last two bugs were crashes in `dc` that resulted from mishandling of
78 The manpage improvement was done by switching from [ronn][20] to [Pandoc][21] to
79 generate manpages. Pandoc generates much cleaner manpages and doesn't leave
80 blank lines where they shouldn't be.
84 This is a production release that adds one new feature: specific manpages.
86 Before this release, `bc` and `dc` only used one manpage each that referred to
87 various build options. This release changes it so there is one manpage set per
88 relevant build type. Each manual only has information about its particular
89 build, and `configure.sh` selects the correct set for install.
93 This is a production release that adds `utf8` locale symlinks and removes an
94 unused `auto` variable from the `ceil()` function in the [extended math
97 Users do ***NOT*** need to update unless they want the locales.
101 This is a production release with two small changes. Users do ***NOT*** need to
102 upgrade to this release; however, if they haven't upgraded to `3.0.0` yet, it
103 may be worthwhile to upgrade to this release.
105 The first change is fixing a compiler warning on FreeBSD with strict warnings
108 The second change is to make the new implementation of `ceil()` in `lib2.bc`
113 *Notes for package maintainers:*
115 *First, the `2.7.0` release series saw a change in the option parsing. This made
116 me change one error message and add a few others. The error message that was
117 changed removed one format specifier. This means that `printf()` will seqfault
118 on old locale files. Unfortunately, `bc` cannot use any locale files except the
119 global ones that are already installed, so it will use the previous ones while
120 running tests during install. **If `bc` segfaults while running arg tests when
121 updating, it is because the global locale files have not been replaced. Make
122 sure to either prevent the test suite from running on update or remove the old
123 locale files before updating.** (Removing the locale files can be done with
124 `make uninstall` or by running the `locale_uninstall.sh` script.) Once this is
125 done, `bc` should install without problems.*
127 *Second, **the option to build without signal support has been removed**. See
128 below for the reasons why.*
130 This is a production release with some small bug fixes, a few improvements,
131 three major bug fixes, and a complete redesign of `bc`'s error and signal
132 handling. **Users and package maintainers should update to this version as soon
135 The first major bug fix was in how `bc` executed files. Previously, a whole file
136 was parsed before it was executed, but if a function is defined *after* code,
137 especially if the function definition was actually a redefinition, and the code
138 before the definition referred to the previous function, this `bc` would replace
139 the function before executing any code. The fix was to make sure that all code
140 that existed before a function definition was executed.
142 The second major bug fix was in `bc`'s `lib2.bc`. The `ceil()` function had a
143 bug where a `0` in the decimal place after the truncation position, caused it to
144 output the wrong numbers if there was any non-zero digit after.
146 The third major bug is that when passing parameters to functions, if an
147 expression included an array (not an array element) as a parameter, it was
148 accepted, when it should have been rejected. It is now correctly rejected.
150 Beyond that, this `bc` got several improvements that both sped it up, improved
151 the handling of signals, and improved the error handling.
153 First, the requirements for `bc` were pushed back to POSIX 2008. `bc` uses one
154 function, `strdup()`, which is not in POSIX 2001, and it is in the X/Open System
155 Interfaces group 2001. It is, however, in POSIX 2008, and since POSIX 2008 is
156 old enough to be supported anywhere that I care, that should be the requirement.
158 Second, the BcVm global variable was put into `bss`. This actually slightly
159 reduces the size of the executable from a massive code shrink, and it will stop
160 `bc` from allocating a large set of memory when `bc` starts.
162 Third, the default Karatsuba length was updated from 64 to 32 after making the
163 optimization changes below, since 32 is going to be better than 64 after the
166 Fourth, Spanish translations were added.
168 Fifth, the interpreter received a speedup to make performance on non-math-heavy
169 scripts more competitive with GNU `bc`. While improvements did, in fact, get it
170 much closer (see the [benchmarks][19]), it isn't quite there.
172 There were several things done to speed up the interpreter:
174 First, several small inefficiencies were removed. These inefficiencies included
175 calling the function `bc_vec_pop(v)` twice instead of calling
176 `bc_vec_npop(v, 2)`. They also included an extra function call for checking the
177 size of the stack and checking the size of the stack more than once on several
180 Second, since the current `bc` function is the one that stores constants and
181 strings, the program caches pointers to the current function's vectors of
182 constants and strings to prevent needing to grab the current function in order
183 to grab a constant or a string.
185 Third, `bc` tries to reuse `BcNum`'s (the internal representation of
186 arbitary-precision numbers). If a `BcNum` has the default capacity of
187 `BC_NUM_DEF_SIZE` (32 on 64-bit and 16 on 32-bit) when it is freed, it is added
188 to a list of available `BcNum`'s. And then, when a `BcNum` is allocated with a
189 capacity of `BC_NUM_DEF_SIZE` and any `BcNum`'s exist on the list of reusable
190 ones, one of those ones is grabbed instead.
192 In order to support these changes, the `BC_NUM_DEF_SIZE` was changed. It used to
193 be 16 bytes on all systems, but it was changed to more closely align with the
194 minimum allocation size on Linux, which is either 32 bytes (64-bit musl), 24
195 bytes (64-bit glibc), 16 bytes (32-bit musl), or 12 bytes (32-bit glibc). Since
196 these are the minimum allocation sizes, these are the sizes that would be
197 allocated anyway, making it worth it to just use the whole space, so the value
198 of `BC_NUM_DEF_SIZE` on 64-bit systems was changed to 32 bytes.
200 On top of that, at least on 64-bit, `BC_NUM_DEF_SIZE` supports numbers with
201 either 72 integer digits or 45 integer digits and 27 fractional digits. This
202 should be more than enough for most cases since `bc`'s default `scale` values
203 are 0 or 20, meaning that, by default, it has at most 20 fractional digits. And
204 45 integer digits are *a lot*; it's enough to calculate the amount of mass in
205 the Milky Way galaxy in kilograms. Also, 72 digits is enough to calculate the
206 diameter of the universe in Planck lengths.
208 (For 32-bit, these numbers are either 32 integer digits or 12 integer digits and
209 20 fractional digits. These are also quite big, and going much bigger on a
210 32-bit system seems a little pointless since 12 digits in just under a trillion
211 and 20 fractional digits is still enough for about any use since `10^-20` light
212 years is just under a millimeter.)
214 All of this together means that for ordinary uses, and even uses in scientific
215 work, the default number size will be all that is needed, which means that
216 nearly all, if not all, numbers will be reused, relieving pressure on the system
219 I did several experiments to find the changes that had the most impact,
220 especially with regard to reusing `BcNum`'s. One was putting `BcNum`'s into
221 buckets according to their capacity in powers of 2 up to 512. That performed
222 worse than `bc` did in `2.7.2`. Another was putting any `BcNum` on the reuse
223 list that had a capacity of `BC_NUM_DEF_SIZE * 2` and reusing them for `BcNum`'s
224 that requested `BC_NUM_DEF_SIZE`. This did reduce the amount of time spent, but
225 it also spent a lot of time in the system allocator for an unknown reason. (When
226 using `strace`, a bunch more `brk` calls showed up.) Just reusing `BcNum`'s that
227 had exactly `BC_NUM_DEF_SIZE` capacity spent the smallest amount of time in both
228 user and system time. This makes sense, especially with the changes to make
229 `BC_NUM_DEF_SIZE` bigger on 64-bit systems, since the vast majority of numbers
230 will only ever use numbers with a size less than or equal to `BC_NUM_DEF_SIZE`.
232 Last of all, `bc`'s signal handling underwent a complete redesign. (This is the
233 reason that this version is `3.0.0` and not `2.8.0`.) The change was to move
234 from a polling approach to signal handling to an interrupt-based approach.
236 Previously, every single loop condition had a check for signals. I suspect that
237 this could be expensive when in tight loops.
239 Now, the signal handler just uses `longjmp()` (actually `siglongjmp()`) to start
240 an unwinding of the stack until it is stopped or the stack is unwound to
241 `main()`, which just returns. If `bc` is currently executing code that cannot be
242 safely interrupted (according to POSIX), then signals are "locked." The signal
243 handler checks if the lock is taken, and if it is, it just sets the status to
244 indicate that a signal arrived. Later, when the signal lock is released, the
245 status is checked to see if a signal came in. If so, the stack unwinding starts.
247 This design eliminates polling in favor of maintaining a stack of `jmp_buf`'s.
248 This has its own performance implications, but it gives better interaction. And
249 the cost of pushing and popping a `jmp_buf` in a function is paid at most twice.
250 Most functions do not pay that price, and most of the rest only pay it once.
251 (There are only some 3 functions in `bc` that push and pop a `jmp_buf` twice.)
253 As a side effect of this change, I had to eliminate the use of `stdio.h` in `bc`
254 because `stdio` does not play nice with signals and `longjmp()`. I implemented
255 custom I/O buffer code that takes a fraction of the size. This means that static
256 builds will be smaller, but non-static builds will be bigger, though they will
257 have less linking time.
259 This change is also good because my history implementation was already bypassing
260 `stdio` for good reasons, and unifying the architecture was a win.
262 Another reason for this change is that my `bc` should *always* behave correctly
263 in the presence of signals like `SIGINT`, `SIGTERM`, and `SIGQUIT`. With the
264 addition of my own I/O buffering, I needed to also make sure that the buffers
265 were correctly flushed even when such signals happened.
267 For this reason, I **removed the option to build without signal support**.
269 As a nice side effect of this change, the error handling code could be changed
270 to take advantage of the stack unwinding that signals used. This means that
271 signals and error handling use the same code paths, which means that the stack
272 unwinding is well-tested. (Errors are tested heavily in the test suite.)
274 It also means that functions do not need to return a status code that
275 ***every*** caller needs to check. This eliminated over 100 branches that simply
276 checked return codes and then passed that return code up the stack if necessary.
277 The code bloat savings from this is at least 1700 bytes on `x86_64`, *before*
278 taking into account the extra code from removing `stdio.h`.
282 This is a production release with one major bug fix.
284 The `length()` built-in function can take either a number or an array. If it
285 takes an array, it returns the length of the array. Arrays can be passed by
286 reference. The bug is that the `length()` function would not properly
287 dereference arrays that were references. This is a bug that affects all users.
289 **ALL USERS SHOULD UPDATE `bc`**.
293 This is a production release with fixes for new locales and fixes for compiler
298 This is a production release with a bug fix for Linux, new translations, and new
303 * Option parsing in `BC_ENV_ARGS` was broken on Linux in 2.6.1 because `glibc`'s
304 `getopt_long()` is broken. To get around that, and to support long options on
305 every platform, an adapted version of [`optparse`][17] was added. Now, `bc`
306 does not even use `getopt()`.
307 * Parsing `BC_ENV_ARGS` with quotes now works. It isn't the smartest, but it
308 does the job if there are spaces in file names.
310 The following new languages are supported:
318 All of these translations were generated using [DeepL][18], so improvements are
321 There is only one new feature: **`bc` now has a built-in pseudo-random number
324 The PRNG is seeded, making it useful for applications where
325 `/dev/urandom` does not work because output needs to be reproducible. However,
326 it also uses `/dev/urandom` to seed itself by default, so it will start with a
327 good seed by default.
329 It also outputs 32 bits on 32-bit platforms and 64 bits on 64-bit platforms, far
330 better than the 15 bits of C's `rand()` and `bash`'s `$RANDOM`.
332 In addition, the PRNG can take a bound, and when it gets a bound, it
333 automatically adjusts to remove bias. It can also generate numbers of arbitrary
334 size. (As of the time of release, the largest pseudo-random number generated by
335 this `bc` was generated with a bound of `2^(2^20)`.)
337 ***IMPORTANT: read the [`bc` manual][9] and the [`dc` manual][10] to find out
338 exactly what guarantees the PRNG provides. The underlying implementation is not
339 guaranteed to stay the same, but the guarantees that it provides are guaranteed
340 to stay the same regardless of the implementation.***
342 On top of that, four functions were added to `bc`'s [extended math library][16]
343 to make using the PRNG easier:
345 * `frand(p)`: Generates a number between `[0,1)` to `p` decimal places.
346 * `ifrand(i, p)`: Generates an integer with bound `i` and adds it to `frand(p)`.
347 * `srand(x)`: Randomizes the sign of `x`. In other words, it flips the sign of
348 `x` with probability `0.5`.
349 * `brand()`: Returns a random boolean value (either `0` or `1`).
353 This is a production release with a bug fix for FreeBSD.
355 The bug was that when `bc` was built without long options, it would give a fatal
356 error on every run. This was caused by a mishandling of `optind`.
360 This release is a production release ***with no bugfixes***. If you do not want
361 to upgrade, you don't have to.
363 No source code changed; the only thing that changed was `lib2.bc`.
365 This release adds one function to the [extended math library][16]: `p(x, y)`,
366 which calculates `x` to the power of `y`, whether or not `y` is an integer. (The
367 `^` operator can only accept integer powers.)
369 This release also includes a couple of small tweaks to the [extended math
370 library][16], mostly to fix returning numbers with too high of `scale`.
374 This release is a production release which addresses inconsistencies in the
375 Portuguese locales. No `bc` code was changed.
377 The issues were that the ISO files used different naming, and also that the
378 files that should have been symlinks were not. I did not catch that because
379 GitHub rendered them the exact same way.
383 This release is a production release.
385 No code was changed, but the build system was changed to allow `CFLAGS` to be
386 given to `CC`, like this:
389 CC="gcc -O3 -march=native" ./configure.sh
392 If this happens, the flags are automatically put into `CFLAGS`, and the compiler
393 is set appropriately. In the example above this means that `CC` will be "gcc"
394 and `CFLAGS` will be "-O3 -march=native".
396 This behavior was added to conform to GNU autotools practices.
400 This is a production release which addresses portability concerns discovered
401 in the `bc` build system. No `bc` code was changed.
403 * Support for Solaris SPARC and AIX were added.
404 * Minor documentations edits were performed.
405 * An option for `configure.sh` was added to disable long options if
406 `getopt_long()` is missing.
410 This is a production release with new translations. No code changed.
412 The translations were contributed by [bugcrazy][15], and they are for
413 Portuguese, both Portugal and Brazil locales.
417 This is a production release primarily aimed at improving `dc`.
419 * A couple of copy and paste errors in the [`dc` manual][10] were fixed.
420 * `dc` startup was optimized by making sure it didn't have to set up `bc`-only
422 * The `bc` `&&` and `||` operators were made available to `dc` through the `M`
423 and `m` commands, respectively.
424 * `dc` macros were changed to be tail call-optimized.
426 The last item, tail call optimization, means that if the last thing in a macro
427 is a call to another macro, then the old macro is popped before executing the
428 new macro. This change was made to stop `dc` from consuming more and more memory
429 as macros are executed in a loop.
431 The `q` and `Q` commands still respect the "hidden" macros by way of recording
432 how many macros were removed by tail call optimization.
436 This is a production release meant to fix warnings in the Gentoo `ebuild` by
437 making it possible to disable binary stripping. Other users do *not* need to
442 This is a production release. It fixes a bug that caused `-1000000000 < -1` to
443 return `0`. This only happened with negative numbers and only if the value on
444 the left was more negative by a certain amount. That said, this bug *is* a bad
445 bug, and needs to be fixed.
447 **ALL USERS SHOULD UPDATE `bc`**.
451 This is a production release with changes to the build system.
455 This release is a production release. It only has new features and performance
458 1. The performance of `sqrt(x)` was improved.
459 2. The new function `root(x, n)` was added to the extended math library to
460 calculate `n`th roots.
461 3. The new function `cbrt(x)` was added to the extended math library to
462 calculate cube roots.
466 This is a non-critical release; it just changes the build system, and in
469 1. Linked locale files were changed to link to their sources with a relative
471 2. A bug in `configure.sh` that caused long option parsing to fail under `bash`
476 This release is not a critical release.
478 1. A few codes were added to history.
479 2. Multiplication was optimized a bit more.
480 3. Addition and subtraction were both optimized a bit more.
484 This release contains a fix for the test suite made for Linux from Scratch: now
485 the test suite prints `pass` when a test is passed.
487 Other than that, there is no change in this release, so distros and other users
488 do not need to upgrade.
492 This release is a production release.
494 The following bugs were fixed:
496 1. A `dc` bug that caused stack mishandling was fixed.
497 2. A warning on OpenBSD was fixed.
498 3. Bugs in `ctrl+arrow` operations in history were fixed.
499 4. The ability to paste multiple lines in history was added.
500 5. A `bc` bug, mishandling of array arguments to functions, was fixed.
501 6. A crash caused by freeing the wrong pointer was fixed.
502 7. A `dc` bug where strings, in a rare case, were mishandled in parsing was
505 In addition, the following changes were made:
507 1. Division was slightly optimized.
508 2. An option was added to the build to disable printing of prompts.
509 3. The special case of empty arguments is now handled. This is to prevent
510 errors in scripts that end up passing empty arguments.
511 4. A harmless bug was fixed. This bug was that, with the pop instructions
512 (mostly) removed (see below), `bc` would leave extra values on its stack for
513 `void` functions and in a few other cases. These extra items would not
514 affect anything put on the stack and would not cause any sort of crash or
515 even buggy behavior, but they would cause `bc` to take more memory than it
518 On top of the above changes, the following optimizations were added:
520 1. The need for pop instructions in `bc` was removed.
521 2. Extra tests on every iteration of the interpreter loop were removed.
522 3. Updating function and code pointers on every iteration of the interpreter
523 loop was changed to only updating them when necessary.
524 4. Extra assignments to pointers were removed.
526 Altogether, these changes sped up the interpreter by around 2x.
528 ***NOTE***: This is the last release with new features because this `bc` is now
529 considered complete. From now on, only bug fixes and new translations will be
534 This is a production, bug-fix release.
536 Two bugs were fixed in this release:
538 1. A rare and subtle signal handling bug was fixed.
539 2. A misbehavior on `0` to a negative power was fixed.
541 The last bug bears some mentioning.
543 When I originally wrote power, I did not thoroughly check its error cases;
544 instead, I had it check if the first number was `0` and then if so, just return
545 `0`. However, `0` to a negative power means that `1` will be divided by `0`,
548 I caught this, but only after I stopped being cocky. You see, sometime later, I
549 had noticed that GNU `bc` returned an error, correctly, but I thought it was
550 wrong simply because that's not what my `bc` did. I saw it again later and had a
551 double take. I checked for real, finally, and found out that my `bc` was wrong
554 That was bad on me. But the bug was easy to fix, so it is fixed now.
556 There are two other things in this release:
558 1. Subtraction was optimized by [Stefan Eßer][14].
559 2. Division was also optimized, also by Stefan Eßer.
563 This release contains a fix for a possible overflow in the signal handling. I
564 would be surprised if any users ran into it because it would only happen after 2
565 billion (`2^31-1`) `SIGINT`'s, but I saw it and had to fix it.
569 This release contains very few things that will apply to any users.
571 1. A slight bug in `dc`'s interactive mode was fixed.
572 2. A bug in the test suite that was only triggered on NetBSD was fixed.
573 3. **The `-P`/`--no-prompt` option** was added for users that do not want a
575 4. A `make check` target was added as an alias for `make test`.
576 5. `dc` got its own read prompt: `?> `.
580 This release is a production release.
582 This release is also a little different from previous releases. From here on
583 out, I do not plan on adding any more features to this `bc`; I believe that it
584 is complete. However, there may be bug fix releases in the future, if I or any
585 others manage to find bugs.
587 This release has only a few new features:
589 1. `atan2(y, x)` was added to the extended math library as both `a2(y, x)` and
591 2. Locales were fixed.
592 3. A **POSIX shell-compatible script was added as an alternative to compiling
593 `gen/strgen.c`** on a host machine. More details about making the choice
594 between the two can be found by running `./configure.sh --help` or reading
595 the [build manual][13].
596 4. Multiplication was optimized by using **diagonal multiplication**, rather
597 than straight brute force.
598 5. The `locale_install.sh` script was fixed.
599 6. `dc` was given the ability to **use the environment variable
601 7. `dc` was also given the ability to **use the `-i` or `--interactive`**
603 8. Printing the prompt was fixed so that it did not print when it shouldn't.
604 9. Signal handling was fixed.
605 10. **Handling of `SIGTERM` and `SIGQUIT`** was fixed.
606 11. The **built-in functions `maxibase()`, `maxobase()`, and `maxscale()`** (the
607 commands `T`, `U`, `V` in `dc`, respectively) were added to allow scripts to
608 query for the max allowable values of those globals.
609 12. Some incompatibilities with POSIX were fixed.
611 In addition, this release is `2.0.0` for a big reason: the internal format for
612 numbers changed. They used to be a `char` array. Now, they are an array of
613 larger integers, packing more decimal digits into each integer. This has
614 delivered ***HUGE*** performance improvements, especially for multiplication,
617 This `bc` should now be the fastest `bc` available, but I may be wrong.
621 This release contains a fix for a harmless bug (it is harmless in that it still
622 works, but it just copies extra data) in the [`locale_install.sh`][12] script.
626 This version contains fixes for the build on Arch Linux.
630 This release removes the use of `local` in shell scripts because it's not POSIX
631 shell-compatible, and also updates a man page that should have been updated a
632 long time ago but was missed.
636 This release contains some missing locale `*.msg` files.
640 This release contains a few bug fixes and new French translations.
644 This release contains a fix for a bug: use of uninitialized data. Such data was
645 only used when outputting an error message, but I am striving for perfection. As
646 Michelangelo said, "Trifles make perfection, and perfection is no trifle."
650 This release contains fixes for OpenBSD.
654 This release contains bug fixes for some rare bugs.
658 This is a production release.
660 There have been several changes since `1.1.0`:
662 1. The build system had some changes.
663 2. Locale support has been added. (Patches welcome for translations.)
664 3. **The ability to turn `ibase`, `obase`, and `scale` into stacks** was added
665 with the `-g` command-line option. (See the [`bc` manual][9] for more
667 4. Support for compiling on Mac OSX out of the box was added.
668 5. The extended math library got `t(x)`, `ceil(x)`, and some aliases.
669 6. The extended math library also got `r2d(x)` (for converting from radians to
670 degrees) and `d2r(x)` (for converting from degrees to radians). This is to
671 allow using degrees with the standard library.
672 7. Both calculators now accept numbers in **scientific notation**. See the
673 [`bc` manual][9] and the [`dc` manual][10] for details.
674 8. Both calculators can **output in either scientific or engineering
675 notation**. See the [`bc` manual][9] and the [`dc` manual][10] for details.
676 9. Some inefficiencies were removed.
677 10. Some bugs were fixed.
678 11. Some bugs in the extended library were fixed.
679 12. Some defects from [Coverity Scan][11] were fixed.
683 This release contains a fix to the build system that allows it to build on older
688 This release contains a fix for a bug in the test suite where `bc` tests and
689 `dc` tests could not be run in parallel.
693 This release has a fix for a history bug; the down arrow did not work.
697 This release fixes a bug in the `1.1.0` build system. The source is exactly the
700 The bug that was fixed was a failure to install if no `EXECSUFFIX` was used.
704 This is a production release. However, many new features were added since `1.0`.
706 1. **The build system has been changed** to use a custom, POSIX
707 shell-compatible configure script ([`configure.sh`][6]) to generate a POSIX
708 make-compatible `Makefile`, which means that `bc` and `dc` now build out of
709 the box on any POSIX-compatible system.
710 2. Out-of-memory and output errors now cause the `bc` to report the error,
711 clean up, and die, rather than just reporting and trying to continue.
712 3. **Strings and constants are now garbage collected** when possible.
713 4. Signal handling and checking has been made more simple and more thorough.
714 5. `BcGlobals` was refactored into `BcVm` and `BcVm` was made global. Some
715 procedure names were changed to reflect its difference to everything else.
716 6. Addition got a speed improvement.
717 7. Some common code for addition and multiplication was refactored into its own
719 8. A bug was removed where `dc` could have been selected, but the internal
720 `#define` that returned `true` for a query about `dc` would not have
722 9. Useless calls to `bc_num_zero()` were removed.
723 10. **History support was added.** The history support is based off of a
724 [UTF-8 aware fork][7] of [`linenoise`][8], which has been customized with
725 `bc`'s own data structures and signal handling.
726 11. Generating C source from the math library now removes tabs from the library,
727 shrinking the size of the executable.
728 12. The math library was shrunk.
729 13. Error handling and reporting was improved.
730 14. Reallocations were reduced by giving access to the request size for each
732 15. **`abs()` (`b` command for `dc`) was added as a builtin.**
733 16. Both calculators were tested on FreeBSD.
734 17. Many obscure parse bugs were fixed.
735 18. Markdown and man page manuals were added, and the man pages are installed by
737 19. Executable size was reduced, though the added features probably made the
738 executable end up bigger.
739 20. **GNU-style array references were added as a supported feature.**
740 21. Allocations were reduced.
741 22. **New operators were added**: `$` (`$` for `dc`), `@` (`@` for `dc`), `@=`,
742 `<<` (`H` for `dc`), `<<=`, `>>` (`h` for `dc`), and `>>=`. See the
743 [`bc` manual][9] and the [`dc` manual][10] for more details.
744 23. **An extended math library was added.** This library contains code that
745 makes it so I can replace my desktop calculator with this `bc`. See the
746 [`bc` manual][3] for more details.
747 24. Support for all capital letters as numbers was added.
748 25. **Support for GNU-style void functions was added.**
749 26. A bug fix for improper handling of function parameters was added.
750 27. Precedence for the or (`||`) operator was changed to match GNU `bc`.
751 28. `dc` was given an explicit negation command.
752 29. `dc` was changed to be able to handle strings in arrays.
754 ## 1.1 Release Candidate 3
756 This release is the eighth release candidate for 1.1, though it is the third
757 release candidate meant as a general release candidate. The new code has not
758 been tested as thoroughly as it should for release.
760 ## 1.1 Release Candidate 2
762 This release is the seventh release candidate for 1.1, though it is the second
763 release candidate meant as a general release candidate. The new code has not
764 been tested as thoroughly as it should for release.
766 ## 1.1 FreeBSD Beta 5
768 This release is the sixth release candidate for 1.1, though it is the fifth
769 release candidate meant specifically to test if `bc` works on FreeBSD. The new
770 code has not been tested as thoroughly as it should for release.
772 ## 1.1 FreeBSD Beta 4
774 This release is the fifth release candidate for 1.1, though it is the fourth
775 release candidate meant specifically to test if `bc` works on FreeBSD. The new
776 code has not been tested as thoroughly as it should for release.
778 ## 1.1 FreeBSD Beta 3
780 This release is the fourth release candidate for 1.1, though it is the third
781 release candidate meant specifically to test if `bc` works on FreeBSD. The new
782 code has not been tested as thoroughly as it should for release.
784 ## 1.1 FreeBSD Beta 2
786 This release is the third release candidate for 1.1, though it is the second
787 release candidate meant specifically to test if `bc` works on FreeBSD. The new
788 code has not been tested as thoroughly as it should for release.
790 ## 1.1 FreeBSD Beta 1
792 This release is the second release candidate for 1.1, though it is meant
793 specifically to test if `bc` works on FreeBSD. The new code has not been tested as
794 thoroughly as it should for release.
796 ## 1.1 Release Candidate 1
798 This is the first release candidate for 1.1. The new code has not been tested as
799 thoroughly as it should for release.
803 This is the first non-beta release. `bc` is ready for production use.
805 As such, a lot has changed since 0.5.
807 1. `dc` has been added. It has been tested even more thoroughly than `bc` was
808 for `0.5`. It does not have the `!` command, and for security reasons, it
809 never will, so it is complete.
810 2. `bc` has been more thoroughly tested. An entire section of the test suite
811 (for both programs) has been added to test for errors.
812 3. A prompt (`>>> `) has been added for interactive mode, making it easier to
813 see inputs and outputs.
814 4. Interrupt handling has been improved, including elimination of race
815 conditions (as much as possible).
816 5. MinGW and [Windows Subsystem for Linux][1] support has been added (see
817 [xstatic][2] for binaries).
818 6. Memory leaks and errors have been eliminated (as far as ASan and Valgrind
820 7. Crashes have been eliminated (as far as [afl][3] can tell).
821 8. Karatsuba multiplication was added (and thoroughly) tested, speeding up
822 multiplication and power by orders of magnitude.
823 9. Performance was further enhanced by using a "divmod" function to reduce
824 redundant divisions and by removing superfluous `memset()` calls.
825 10. To switch between Karatsuba and `O(n^2)` multiplication, the config variable
826 `BC_NUM_KARATSUBA_LEN` was added. It is set to a sane default, but the
827 optimal number can be found with [`karatsuba.py`][4] (requires Python 3)
828 and then configured through `make`.
829 11. The random math test generator script was changed to Python 3 and improved.
830 `bc` and `dc` have together been run through 30+ million random tests.
831 12. All known math bugs have been fixed, including out of control memory
832 allocations in `sine` and `cosine` (that was actually a parse bug), certain
833 cases of infinite loop on square root, and slight inaccuracies (as much as
834 possible; see the [README][5]) in transcendental functions.
835 13. Parsing has been fixed as much as possible.
836 14. Test coverage was improved to 94.8%. The only paths not covered are ones
837 that happen when `malloc()` or `realloc()` fails.
838 15. An extension to get the length of an array was added.
839 16. The boolean not (`!`) had its precedence change to match negation.
840 17. Data input was hardened.
841 18. `bc` was made fully compliant with POSIX when the `-s` flag is used or
842 `POSIXLY_CORRECT` is defined.
843 19. Error handling was improved.
844 20. `bc` now checks that files it is given are not directories.
846 ## 1.0 Release Candidate 7
848 This is the seventh release candidate for 1.0. It fixes a few bugs in 1.0
851 ## 1.0 Release Candidate 6
853 This is the sixth release candidate for 1.0. It fixes a few bugs in 1.0 Release
856 ## 1.0 Release Candidate 5
858 This is the fifth release candidate for 1.0. It fixes a few bugs in 1.0 Release
861 ## 1.0 Release Candidate 4
863 This is the fourth release candidate for 1.0. It fixes a few bugs in 1.0 Release
866 ## 1.0 Release Candidate 3
868 This is the third release candidate for 1.0. It fixes a few bugs in 1.0 Release
871 ## 1.0 Release Candidate 2
873 This is the second release candidate for 1.0. It fixes a few bugs in 1.0 Release
876 ## 1.0 Release Candidate 1
878 This is the first Release Candidate for 1.0. `bc` is complete, with `dc`, but it
883 This beta release completes more features, but it is still not complete nor
884 tested as thoroughly as necessary.
888 This beta release fixes a few bugs in 0.4.
892 This is a beta release. It does not have the complete set of features, and it is
893 not thoroughly tested.
895 [1]: https://docs.microsoft.com/en-us/windows/wsl/install-win10
896 [2]: https://pkg.musl.cc/bc/
897 [3]: http://lcamtuf.coredump.cx/afl/
901 [7]: https://github.com/rain-1/linenoise-mob
902 [8]: https://github.com/antirez/linenoise
903 [9]: ./manuals/bc/A.1.md
904 [10]: ./manuals/dc/A.1.md
905 [11]: https://scan.coverity.com/projects/gavinhoward-bc
906 [12]: ./locale_install.sh
907 [13]: ./manuals/build.md
908 [14]: https://github.com/stesser
909 [15]: https://github.com/bugcrazy
910 [16]: ./manuals/bc/A.1.md#extended-library
911 [17]: https://github.com/skeeto/optparse
912 [18]: https://www.deepl.com/translator
913 [19]: ./manuals/benchmarks.md
914 [20]: https://github.com/apjanke/ronn-ng
915 [21]: https://pandoc.org/