3 SPDX-License-Identifier: BSD-2-Clause
5 Copyright (c) 2018-2021 Gavin D. Howard and contributors.
7 Redistribution and use in source and binary forms, with or without
8 modification, are permitted provided that the following conditions are met:
10 * Redistributions of source code must retain the above copyright notice, this
11 list of conditions and the following disclaimer.
13 * Redistributions in binary form must reproduce the above copyright notice,
14 this list of conditions and the following disclaimer in the documentation
15 and/or other materials provided with the distribution.
17 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
18 AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19 IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20 ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
21 LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
22 CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
23 SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
24 INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
25 CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
26 ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
27 POSSIBILITY OF SUCH DAMAGE.
33 bc - arbitrary-precision decimal arithmetic language and calculator
37 **bc** [**-ghilPqsvVw**] [**-\-global-stacks**] [**-\-help**] [**-\-interactive**] [**-\-mathlib**] [**-\-no-prompt**] [**-\-quiet**] [**-\-standard**] [**-\-warn**] [**-\-version**] [**-e** *expr*] [**-\-expression**=*expr*...] [**-f** *file*...] [**-\-file**=*file*...] [*file*...]
41 bc(1) is an interactive processor for a language first standardized in 1991 by
42 POSIX. (The current standard is [here][1].) The language provides unlimited
43 precision decimal arithmetic and is somewhat C-like, but there are differences.
44 Such differences will be noted in this document.
46 After parsing and handling options, this bc(1) reads any files given on the
47 command line and executes them before reading from **stdin**.
50 This bc(1) is a drop-in replacement for *any* bc(1), including (and
51 especially) the GNU bc(1). It also has many extensions and extra features beyond
52 other implementations.
55 This bc(1) is a drop-in replacement for *any* bc(1), including (and
56 especially) the GNU bc(1).
61 The following are the options that bc(1) accepts.
63 **-g**, **-\-global-stacks**
65 {{ A H N P HN HP NP HNP }}
66 : Turns the globals **ibase**, **obase**, **scale**, and **seed** into stacks.
68 This has the effect that a copy of the current value of all four are pushed
70 {{ E EH EN EP EHN EHP ENP EHNP }}
71 Turns the globals **ibase**, **obase**, and **scale** into stacks.
73 This has the effect that a copy of the current value of all three are pushed
75 onto a stack for every function call, as well as popped when every function
76 returns. This means that functions can assign to any and all of those
77 globals without worrying that the change will affect other functions.
78 Thus, a hypothetical function named **output(x,b)** that simply printed
79 **x** in base **b** could be written like this:
81 define void output(x, b) {
88 define void output(x, b) {
96 This makes writing functions much easier.
98 {{ A H N P HN HP NP HNP }}
99 (**Note**: the function **output(x,b)** exists in the extended math library.
100 See the **LIBRARY** section.)
102 However, since using this flag means that functions cannot set **ibase**,
103 **obase**, **scale**, or **seed** globally, functions that are made to do so
104 cannot work anymore. There are two possible use cases for that, and each has
107 {{ E EH EN EP EHN EHP ENP EHNP }}
108 However, since using this flag means that functions cannot set **ibase**,
109 **obase**, or **scale** globally, functions that are made to do so cannot
110 work anymore. There are two possible use cases for that, and each has a
114 First, if a function is called on startup to turn bc(1) into a number
115 converter, it is possible to replace that capability with various shell
118 alias d2o="bc -e ibase=A -e obase=8"
119 alias h2b="bc -e ibase=G -e obase=2"
121 {{ A H N P HN HP NP HNP }}
122 Second, if the purpose of a function is to set **ibase**, **obase**,
123 **scale**, or **seed** globally for any other purpose, it could be split
124 into one to four functions (based on how many globals it sets) and each of
125 those functions could return the desired value for a global.
127 For functions that set **seed**, the value assigned to **seed** is not
128 propagated to parent functions. This means that the sequence of
129 pseudo-random numbers that they see will not be the same sequence of
130 pseudo-random numbers that any parent sees. This is only the case once
131 **seed** has been set.
133 If a function desires to not affect the sequence of pseudo-random numbers
134 of its parents, but wants to use the same **seed**, it can use the following
139 {{ E EH EN EP EHN EHP ENP EHNP }}
140 Second, if the purpose of a function is to set **ibase**, **obase**, or
141 **scale** globally for any other purpose, it could be split into one to
142 three functions (based on how many globals it sets) and each of those
143 functions could return the desired value for a global.
146 If the behavior of this option is desired for every run of bc(1), then users
147 could make sure to define **BC_ENV_ARGS** and include this option (see the
148 **ENVIRONMENT VARIABLES** section for more details).
150 If **-s**, **-w**, or any equivalents are used, this option is ignored.
152 This is a **non-portable extension**.
156 : Prints a usage message and quits.
158 **-i**, **-\-interactive**
160 : Forces interactive mode. (See the **INTERACTIVE MODE** section.)
162 This is a **non-portable extension**.
164 **-l**, **-\-mathlib**
166 : Sets **scale** (see the **SYNTAX** section) to **20** and loads the included
167 {{ A H N P HN HP NP HNP }}
168 math library and the extended math library before running any code,
169 including any expressions or files specified on the command line.
171 To learn what is in the libraries, see the **LIBRARY** section.
173 {{ E EH EN EP EHN EHP ENP EHNP }}
174 math library before running any code, including any expressions or files
175 specified on the command line.
177 To learn what is in the library, see the **LIBRARY** section.
180 **-P**, **-\-no-prompt**
182 {{ A E H N EH EN HN EHN }}
183 : Disables the prompt in TTY mode. (The prompt is only enabled in TTY mode.
184 See the **TTY MODE** section) This is mostly for those users that do not
185 want a prompt or are not used to having them in bc(1). Most of those users
186 would want to put this option in **BC_ENV_ARGS** (see the
187 **ENVIRONMENT VARIABLES** section).
189 {{ P EP HP NP EHP ENP HNP EHNP }}
190 : This option is a no-op.
193 This is a **non-portable extension**.
197 : This option is for compatibility with the [GNU bc(1)][2]; it is a no-op.
198 Without this option, GNU bc(1) prints a copyright header. This bc(1) only
199 prints the copyright header if one or more of the **-v**, **-V**, or
200 **-\-version** options are given.
202 This is a **non-portable extension**.
204 **-s**, **-\-standard**
206 : Process exactly the language defined by the [standard][1] and error if any
209 This is a **non-portable extension**.
211 **-v**, **-V**, **-\-version**
213 : Print the version information (copyright header) and exit.
215 This is a **non-portable extension**.
219 : Like **-s** and **-\-standard**, except that warnings (and not errors) are
220 printed for non-standard extensions and execution continues normally.
222 This is a **non-portable extension**.
224 **-e** *expr*, **-\-expression**=*expr*
226 : Evaluates *expr*. If multiple expressions are given, they are evaluated in
227 order. If files are given as well (see below), the expressions and files are
228 evaluated in the order given. This means that if a file is given before an
229 expression, the file is read in and evaluated first.
231 If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
232 see the **ENVIRONMENT VARIABLES** section), then after processing all
233 expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
234 as an argument at least once to **-f** or **-\-file**, whether on the
235 command-line or in **BC_ENV_ARGS**. However, if any other **-e**,
236 **-\-expression**, **-f**, or **-\-file** arguments are given after **-f-**
237 or equivalent is given, bc(1) will give a fatal error and exit.
239 This is a **non-portable extension**.
241 **-f** *file*, **-\-file**=*file*
243 : Reads in *file* and evaluates it, line by line, as though it were read
244 through **stdin**. If expressions are also given (see above), the
245 expressions are evaluated in the order given.
247 If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
248 see the **ENVIRONMENT VARIABLES** section), then after processing all
249 expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
250 as an argument at least once to **-f** or **-\-file**. However, if any other
251 **-e**, **-\-expression**, **-f**, or **-\-file** arguments are given after
252 **-f-** or equivalent is given, bc(1) will give a fatal error and exit.
254 This is a **non-portable extension**.
256 All long options are **non-portable extensions**.
260 Any non-error output is written to **stdout**. In addition, if history (see the
261 **HISTORY** section) and the prompt (see the **TTY MODE** section) are enabled,
262 both are output to **stdout**.
264 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
265 error (see the **EXIT STATUS** section) if it cannot write to **stdout**, so if
266 **stdout** is closed, as in **bc <file> >&-**, it will quit with an error. This
267 is done so that bc(1) can report problems when **stdout** is redirected to a
270 If there are scripts that depend on the behavior of other bc(1) implementations,
271 it is recommended that those scripts be changed to redirect **stdout** to
276 Any error output is written to **stderr**.
278 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
279 error (see the **EXIT STATUS** section) if it cannot write to **stderr**, so if
280 **stderr** is closed, as in **bc <file> 2>&-**, it will quit with an error. This
281 is done so that bc(1) can exit with an error code when **stderr** is redirected
284 If there are scripts that depend on the behavior of other bc(1) implementations,
285 it is recommended that those scripts be changed to redirect **stderr** to
290 The syntax for bc(1) programs is mostly C-like, with some differences. This
291 bc(1) follows the [POSIX standard][1], which is a much more thorough resource
292 for the language this bc(1) accepts. This section is meant to be a summary and a
293 listing of all the extensions to the standard.
295 In the sections below, **E** means expression, **S** means statement, and **I**
298 Identifiers (**I**) start with a lowercase letter and can be followed by any
299 number (up to **BC_NAME_MAX-1**) of lowercase letters (**a-z**), digits
300 (**0-9**), and underscores (**\_**). The regex is **\[a-z\]\[a-z0-9\_\]\***.
301 Identifiers with more than one character (letter) are a
302 **non-portable extension**.
304 **ibase** is a global variable determining how to interpret constant numbers. It
305 is the "input" base, or the number base used for interpreting input numbers.
306 **ibase** is initially **10**. If the **-s** (**-\-standard**) and **-w**
307 (**-\-warn**) flags were not given on the command line, the max allowable value
308 for **ibase** is **36**. Otherwise, it is **16**. The min allowable value for
309 **ibase** is **2**. The max allowable value for **ibase** can be queried in
310 bc(1) programs with the **maxibase()** built-in function.
312 **obase** is a global variable determining how to output results. It is the
313 "output" base, or the number base used for outputting numbers. **obase** is
314 initially **10**. The max allowable value for **obase** is **BC_BASE_MAX** and
315 can be queried in bc(1) programs with the **maxobase()** built-in function. The
316 {{ A H N P HN HP NP HNP }}
317 min allowable value for **obase** is **0**. If **obase** is **0**, values are
318 output in scientific notation, and if **obase** is **1**, values are output in
319 engineering notation. Otherwise, values are output in the specified base.
321 Outputting in scientific and engineering notations are **non-portable
324 {{ E EH EN EP EHN EHP ENP EHNP }}
325 min allowable value for **obase** is **2**. Values are output in the specified
329 The *scale* of an expression is the number of digits in the result of the
330 expression right of the decimal point, and **scale** is a global variable that
331 sets the precision of any operations, with exceptions. **scale** is initially
332 **0**. **scale** cannot be negative. The max allowable value for **scale** is
333 **BC_SCALE_MAX** and can be queried in bc(1) programs with the **maxscale()**
336 bc(1) has both *global* variables and *local* variables. All *local*
337 variables are local to the function; they are parameters or are introduced in
338 the **auto** list of a function (see the **FUNCTIONS** section). If a variable
339 is accessed which is not a parameter or in the **auto** list, it is assumed to
340 be *global*. If a parent function has a *local* variable version of a variable
341 that a child function considers *global*, the value of that *global* variable in
342 the child function is the value of the variable in the parent function, not the
343 value of the actual *global* variable.
345 All of the above applies to arrays as well.
347 The value of a statement that is an expression (i.e., any of the named
348 expressions or operands) is printed unless the lowest precedence operator is an
349 assignment operator *and* the expression is notsurrounded by parentheses.
351 The value that is printed is also assigned to the special variable **last**. A
352 single dot (**.**) may also be used as a synonym for **last**. These are
353 **non-portable extensions**.
355 Either semicolons or newlines may separate statements.
359 There are two kinds of comments:
361 1. Block comments are enclosed in **/\*** and **\*/**.
362 2. Line comments go from **#** until, and not including, the next newline. This
363 is a **non-portable extension**.
367 The following are named expressions in bc(1):
370 2. Array Elements: **I[E]**
374 {{ A H N P HN HP NP HNP }}
376 7. **last** or a single dot (**.**)
378 Numbers 6 and 7 are **non-portable extensions**.
380 The meaning of **seed** is dependent on the current pseudo-random number
381 generator but is guaranteed to not change except for new major versions.
383 The *scale* and sign of the value may be significant.
385 If a previously used **seed** value is assigned to **seed** and used again, the
386 pseudo-random number generator is guaranteed to produce the same sequence of
387 pseudo-random numbers as it did when the **seed** value was previously used.
389 The exact value assigned to **seed** is not guaranteed to be returned if
390 **seed** is queried again immediately. However, if **seed** *does* return a
391 different value, both values, when assigned to **seed**, are guaranteed to
392 produce the same sequence of pseudo-random numbers. This means that certain
393 values assigned to **seed** will *not* produce unique sequences of pseudo-random
394 numbers. The value of **seed** will change after any use of the **rand()** and
395 **irand(E)** operands (see the *Operands* subsection below), except if the
396 parameter passed to **irand(E)** is **0**, **1**, or negative.
398 There is no limit to the length (number of significant decimal digits) or
399 *scale* of the value that can be assigned to **seed**.
401 {{ E EH EN EP EHN EHP ENP EHNP }}
402 6. **last** or a single dot (**.**)
404 Number 6 is a **non-portable extension**.
407 Variables and arrays do not interfere; users can have arrays named the same as
408 variables. This also applies to functions (see the **FUNCTIONS** section), so a
409 user can have a variable, array, and function that all have the same name, and
410 they will not shadow each other, whether inside of functions or not.
412 Named expressions are required as the operand of **increment**/**decrement**
413 operators and as the left side of **assignment** operators (see the *Operators*
418 The following are valid operands in bc(1):
420 1. Numbers (see the *Numbers* subsection below).
421 2. Array indices (**I[E]**).
422 3. **(E)**: The value of **E** (used to change precedence).
423 4. **sqrt(E)**: The square root of **E**. **E** must be non-negative.
424 5. **length(E)**: The number of significant decimal digits in **E**.
425 6. **length(I[])**: The number of elements in the array **I**. This is a
426 **non-portable extension**.
427 7. **scale(E)**: The *scale* of **E**.
428 8. **abs(E)**: The absolute value of **E**. This is a **non-portable
430 9. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
431 a non-**void** function (see the *Void Functions* subsection of the
432 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
433 **I[]**, which will automatically be turned into array references (see the
434 *Array References* subsection of the **FUNCTIONS** section) if the
435 corresponding parameter in the function definition is an array reference.
436 10. **read()**: Reads a line from **stdin** and uses that as an expression. The
437 result of that expression is the result of the **read()** operand. This is a
438 **non-portable extension**.
439 11. **maxibase()**: The max allowable **ibase**. This is a **non-portable
441 12. **maxobase()**: The max allowable **obase**. This is a **non-portable
443 13. **maxscale()**: The max allowable **scale**. This is a **non-portable
445 {{ A H N P HN HP NP HNP }}
446 14. **rand()**: A pseudo-random integer between **0** (inclusive) and
447 **BC_RAND_MAX** (inclusive). Using this operand will change the value of
448 **seed**. This is a **non-portable extension**.
449 15. **irand(E)**: A pseudo-random integer between **0** (inclusive) and the
450 value of **E** (exclusive). If **E** is negative or is a non-integer
451 (**E**'s *scale* is not **0**), an error is raised, and bc(1) resets (see
452 the **RESET** section) while **seed** remains unchanged. If **E** is larger
453 than **BC_RAND_MAX**, the higher bound is honored by generating several
454 pseudo-random integers, multiplying them by appropriate powers of
455 **BC_RAND_MAX+1**, and adding them together. Thus, the size of integer that
456 can be generated with this operand is unbounded. Using this operand will
457 change the value of **seed**, unless the value of **E** is **0** or **1**.
458 In that case, **0** is returned, and **seed** is *not* changed. This is a
459 **non-portable extension**.
460 16. **maxrand()**: The max integer returned by **rand()**. This is a
461 **non-portable extension**.
463 The integers generated by **rand()** and **irand(E)** are guaranteed to be as
464 unbiased as possible, subject to the limitations of the pseudo-random number
467 **Note**: The values returned by the pseudo-random number generator with
468 **rand()** and **irand(E)** are guaranteed to *NOT* be cryptographically secure.
469 This is a consequence of using a seeded pseudo-random number generator. However,
470 they *are* guaranteed to be reproducible with identical **seed** values. This
471 means that the pseudo-random values from bc(1) should only be used where a
472 reproducible stream of pseudo-random numbers is *ESSENTIAL*. In any other case,
473 use a non-seeded pseudo-random number generator.
478 Numbers are strings made up of digits, uppercase letters, and at most **1**
479 period for a radix. Numbers can have up to **BC_NUM_MAX** digits. Uppercase
480 letters are equal to **9** + their position in the alphabet (i.e., **A** equals
481 **10**, or **9+1**). If a digit or letter makes no sense with the current value
482 of **ibase**, they are set to the value of the highest valid digit in **ibase**.
484 Single-character numbers (i.e., **A** alone) take the value that they would have
485 if they were valid digits, regardless of the value of **ibase**. This means that
486 **A** alone always equals decimal **10** and **Z** alone always equals decimal
489 {{ A H N P HN HP NP HNP }}
490 In addition, bc(1) accepts numbers in scientific notation. These have the form
491 **\<number\>e\<integer\>**. The exponent (the portion after the **e**) must be
492 an integer. An example is **1.89237e9**, which is equal to **1892370000**.
493 Negative exponents are also allowed, so **4.2890e-3** is equal to **0.0042890**.
495 Using scientific notation is an error or warning if the **-s** or **-w**,
496 respectively, command-line options (or equivalents) are given.
498 **WARNING**: Both the number and the exponent in scientific notation are
499 interpreted according to the current **ibase**, but the number is still
500 multiplied by **10\^exponent** regardless of the current **ibase**. For example,
501 if **ibase** is **16** and bc(1) is given the number string **FFeA**, the
502 resulting decimal number will be **2550000000000**, and if bc(1) is given the
503 number string **10e-4**, the resulting decimal number will be **0.0016**.
505 Accepting input as scientific notation is a **non-portable extension**.
510 The following arithmetic and logical operators can be used. They are listed in
511 order of decreasing precedence. Operators in the same group have the same
516 : Type: Prefix and Postfix
520 Description: **increment**, **decrement**
528 Description: **negation**, **boolean not**
530 {{ A H N P HN HP NP HNP }}
537 Description: **truncation**
545 Description: **set precision**
554 Description: **power**
562 Description: **multiply**, **divide**, **modulus**
570 Description: **add**, **subtract**
572 {{ A H N P HN HP NP HNP }}
579 Description: **shift left**, **shift right**
581 **=** **\<\<=** **\>\>=** **+=** **-=** **\*=** **/=** **%=** **\^=** **\@=**
583 {{ E EH EN EP EHN EHP ENP EHNP }}
584 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
591 Description: **assignment**
593 **==** **\<=** **\>=** **!=** **\<** **\>**
599 Description: **relational**
607 Description: **boolean and**
615 Description: **boolean or**
617 The operators will be described in more detail below.
621 : The prefix and postfix **increment** and **decrement** operators behave
622 exactly like they would in C. They require a named expression (see the
623 *Named Expressions* subsection) as an operand.
625 The prefix versions of these operators are more efficient; use them where
630 : The **negation** operator returns **0** if a user attempts to negate any
631 expression with the value **0**. Otherwise, a copy of the expression with
632 its sign flipped is returned.
636 : The **boolean not** operator returns **1** if the expression is **0**, or
639 This is a **non-portable extension**.
641 {{ A H N P HN HP NP HNP }}
644 : The **truncation** operator returns a copy of the given expression with all
645 of its *scale* removed.
647 This is a **non-portable extension**.
651 : The **set precision** operator takes two expressions and returns a copy of
652 the first with its *scale* equal to the value of the second expression. That
653 could either mean that the number is returned without change (if the
654 *scale* of the first expression matches the value of the second
655 expression), extended (if it is less), or truncated (if it is more).
657 The second expression must be an integer (no *scale*) and non-negative.
659 This is a **non-portable extension**.
664 : The **power** operator (not the **exclusive or** operator, as it would be in
665 C) takes two expressions and raises the first to the power of the value of
666 the second. The *scale* of the result is equal to **scale**.
668 The second expression must be an integer (no *scale*), and if it is
669 negative, the first value must be non-zero.
673 : The **multiply** operator takes two expressions, multiplies them, and
674 returns the product. If **a** is the *scale* of the first expression and
675 **b** is the *scale* of the second expression, the *scale* of the result is
676 equal to **min(a+b,max(scale,a,b))** where **min()** and **max()** return
681 : The **divide** operator takes two expressions, divides them, and returns the
682 quotient. The *scale* of the result shall be the value of **scale**.
684 The second expression must be non-zero.
688 : The **modulus** operator takes two expressions, **a** and **b**, and
689 evaluates them by 1) Computing **a/b** to current **scale** and 2) Using the
690 result of step 1 to calculate **a-(a/b)\*b** to *scale*
691 **max(scale+scale(b),scale(a))**.
693 The second expression must be non-zero.
697 : The **add** operator takes two expressions, **a** and **b**, and returns the
698 sum, with a *scale* equal to the max of the *scale*s of **a** and **b**.
702 : The **subtract** operator takes two expressions, **a** and **b**, and
703 returns the difference, with a *scale* equal to the max of the *scale*s of
706 {{ A H N P HN HP NP HNP }}
709 : The **left shift** operator takes two expressions, **a** and **b**, and
710 returns a copy of the value of **a** with its decimal point moved **b**
713 The second expression must be an integer (no *scale*) and non-negative.
715 This is a **non-portable extension**.
719 : The **right shift** operator takes two expressions, **a** and **b**, and
720 returns a copy of the value of **a** with its decimal point moved **b**
723 The second expression must be an integer (no *scale*) and non-negative.
725 This is a **non-portable extension**.
728 {{ A H N P HN HP NP HNP }}
729 **=** **\<\<=** **\>\>=** **+=** **-=** **\*=** **/=** **%=** **\^=** **\@=**
731 {{ E EH EN EP EHN EHP ENP EHNP }}
732 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
735 : The **assignment** operators take two expressions, **a** and **b** where
736 **a** is a named expression (see the *Named Expressions* subsection).
738 For **=**, **b** is copied and the result is assigned to **a**. For all
739 others, **a** and **b** are applied as operands to the corresponding
740 arithmetic operator and the result is assigned to **a**.
742 {{ A H N P HN HP NP HNP }}
743 The **assignment** operators that correspond to operators that are
744 extensions are themselves **non-portable extensions**.
747 **==** **\<=** **\>=** **!=** **\<** **\>**
749 : The **relational** operators compare two expressions, **a** and **b**, and
750 if the relation holds, according to C language semantics, the result is
751 **1**. Otherwise, it is **0**.
753 Note that unlike in C, these operators have a lower precedence than the
754 **assignment** operators, which means that **a=b\>c** is interpreted as
757 Also, unlike the [standard][1] requires, these operators can appear anywhere
758 any other expressions can be used. This allowance is a
759 **non-portable extension**.
763 : The **boolean and** operator takes two expressions and returns **1** if both
764 expressions are non-zero, **0** otherwise.
766 This is *not* a short-circuit operator.
768 This is a **non-portable extension**.
772 : The **boolean or** operator takes two expressions and returns **1** if one
773 of the expressions is non-zero, **0** otherwise.
775 This is *not* a short-circuit operator.
777 This is a **non-portable extension**.
781 The following items are statements:
784 2. **{** **S** **;** ... **;** **S** **}**
785 3. **if** **(** **E** **)** **S**
786 4. **if** **(** **E** **)** **S** **else** **S**
787 5. **while** **(** **E** **)** **S**
788 6. **for** **(** **E** **;** **E** **;** **E** **)** **S**
789 7. An empty statement
795 13. A string of characters, enclosed in double quotes
796 14. **print** **E** **,** ... **,** **E**
797 15. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
798 a **void** function (see the *Void Functions* subsection of the
799 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
800 **I[]**, which will automatically be turned into array references (see the
801 *Array References* subsection of the **FUNCTIONS** section) if the
802 corresponding parameter in the function definition is an array reference.
804 Numbers 4, 9, 11, 12, 14, and 15 are **non-portable extensions**.
806 Also, as a **non-portable extension**, any or all of the expressions in the
807 header of a for loop may be omitted. If the condition (second expression) is
808 omitted, it is assumed to be a constant **1**.
810 The **break** statement causes a loop to stop iterating and resume execution
811 immediately following a loop. This is only allowed in loops.
813 The **continue** statement causes a loop iteration to stop early and returns to
814 the start of the loop, including testing the loop condition. This is only
817 The **if** **else** statement does the same thing as in C.
819 The **quit** statement causes bc(1) to quit, even if it is on a branch that will
820 not be executed (it is a compile-time command).
822 The **halt** statement causes bc(1) to quit, if it is executed. (Unlike **quit**
823 if it is on a branch of an **if** statement that is not executed, bc(1) does not
826 The **limits** statement prints the limits that this bc(1) is subject to. This
827 is like the **quit** statement in that it is a compile-time command.
829 An expression by itself is evaluated and printed, followed by a newline.
831 {{ A H N P HN HP NP HNP }}
832 Both scientific notation and engineering notation are available for printing the
833 results of expressions. Scientific notation is activated by assigning **0** to
834 **obase**, and engineering notation is activated by assigning **1** to
835 **obase**. To deactivate them, just assign a different value to **obase**.
837 Scientific notation and engineering notation are disabled if bc(1) is run with
838 either the **-s** or **-w** command-line options (or equivalents).
840 Printing numbers in scientific notation and/or engineering notation is a
841 **non-portable extension**.
846 The "expressions" in a **print** statement may also be strings. If they are, there
847 are backslash escape sequences that are interpreted specially. What those
848 sequences are, and what they cause to be printed, are shown below:
862 Any other character following a backslash causes the backslash and character to
865 Any non-string expression in a print statement shall be assigned to **last**,
866 like any other expression that is printed.
868 ## Order of Evaluation
870 All expressions in a statment are evaluated left to right, except as necessary
871 to maintain order of operations. This means, for example, assuming that **i** is
872 equal to **0**, in the expression
876 the first (or 0th) element of **a** is set to **1**, and **i** is equal to **2**
877 at the end of the expression.
879 This includes function arguments. Thus, assuming **i** is equal to **0**, this
880 means that in the expression
884 the first argument passed to **x()** is **0**, and the second argument is **1**,
885 while **i** is equal to **2** before the function starts executing.
889 Function definitions are as follows:
899 Any **I** in the parameter list or **auto** list may be replaced with **I[]** to
900 make a parameter or **auto** var an array, and any **I** in the parameter list
901 may be replaced with **\*I[]** to make a parameter an array reference. Callers
902 of functions that take array references should not put an asterisk in the call;
903 they must be called with just **I[]** like normal array parameters and will be
904 automatically converted into references.
906 As a **non-portable extension**, the opening brace of a **define** statement may
907 appear on the next line.
909 As a **non-portable extension**, the return statement may also be in one of the
913 2. **return** **(** **)**
916 The first two, or not specifying a **return** statement, is equivalent to
917 **return (0)**, unless the function is a **void** function (see the *Void
918 Functions* subsection below).
922 Functions can also be **void** functions, defined as follows:
925 define void I(I,...,I){
932 They can only be used as standalone expressions, where such an expression would
933 be printed alone, except in a print statement.
935 Void functions can only use the first two **return** statements listed above.
936 They can also omit the return statement entirely.
938 The word "void" is not treated as a keyword; it is still possible to have
939 variables, arrays, and functions named **void**. The word "void" is only
940 treated specially right after the **define** keyword.
942 This is a **non-portable extension**.
946 For any array in the parameter list, if the array is declared in the form
952 it is a **reference**. Any changes to the array in the function are reflected,
953 when the function returns, to the array that was passed in.
955 Other than this, all function arguments are passed by value.
957 This is a **non-portable extension**.
961 {{ A H N P HN HP NP HNP }}
962 All of the functions below, including the functions in the extended math
963 library (see the *Extended Library* subsection below), are available when the
964 **-l** or **-\-mathlib** command-line flags are given, except that the extended
965 math library is not available when the **-s** option, the **-w** option, or
966 equivalents are given.
968 {{ E EH EN EP EHN EHP ENP EHNP }}
969 All of the functions below are available when the **-l** or **-\-mathlib**
970 command-line flags are given.
975 The [standard][1] defines the following functions for the math library:
979 : Returns the sine of **x**, which is assumed to be in radians.
981 This is a transcendental function (see the *Transcendental Functions*
986 : Returns the cosine of **x**, which is assumed to be in radians.
988 This is a transcendental function (see the *Transcendental Functions*
993 : Returns the arctangent of **x**, in radians.
995 This is a transcendental function (see the *Transcendental Functions*
1000 : Returns the natural logarithm of **x**.
1002 This is a transcendental function (see the *Transcendental Functions*
1007 : Returns the mathematical constant **e** raised to the power of **x**.
1009 This is a transcendental function (see the *Transcendental Functions*
1014 : Returns the bessel integer order **n** (truncated) of **x**.
1016 This is a transcendental function (see the *Transcendental Functions*
1019 {{ A H N P HN HP NP HNP }}
1022 The extended library is *not* loaded when the **-s**/**-\-standard** or
1023 **-w**/**-\-warn** options are given since they are not part of the library
1024 defined by the [standard][1].
1026 The extended library is a **non-portable extension**.
1030 : Calculates **x** to the power of **y**, even if **y** is not an integer, and
1031 returns the result to the current **scale**.
1033 It is an error if **y** is negative and **x** is **0**.
1035 This is a transcendental function (see the *Transcendental Functions*
1040 : Returns **x** rounded to **p** decimal places according to the rounding mode
1041 [round half away from **0**][3].
1045 : Returns **x** rounded to **p** decimal places according to the rounding mode
1046 [round away from **0**][6].
1050 : Returns the factorial of the truncated absolute value of **x**.
1054 : Returns the permutation of the truncated absolute value of **n** of the
1055 truncated absolute value of **k**, if **k \<= n**. If not, it returns **0**.
1059 : Returns the combination of the truncated absolute value of **n** of the
1060 truncated absolute value of **k**, if **k \<= n**. If not, it returns **0**.
1064 : Returns the logarithm base **2** of **x**.
1066 This is a transcendental function (see the *Transcendental Functions*
1071 : Returns the logarithm base **10** of **x**.
1073 This is a transcendental function (see the *Transcendental Functions*
1078 : Returns the logarithm base **b** of **x**.
1080 This is a transcendental function (see the *Transcendental Functions*
1085 : Returns the cube root of **x**.
1089 : Calculates the truncated value of **n**, **r**, and returns the **r**th root
1090 of **x** to the current **scale**.
1092 If **r** is **0** or negative, this raises an error and causes bc(1) to
1093 reset (see the **RESET** section). It also raises an error and causes bc(1)
1094 to reset if **r** is even and **x** is negative.
1098 : Returns **pi** to **p** decimal places.
1100 This is a transcendental function (see the *Transcendental Functions*
1105 : Returns the tangent of **x**, which is assumed to be in radians.
1107 This is a transcendental function (see the *Transcendental Functions*
1112 : Returns the arctangent of **y/x**, in radians. If both **y** and **x** are
1113 equal to **0**, it raises an error and causes bc(1) to reset (see the
1114 **RESET** section). Otherwise, if **x** is greater than **0**, it returns
1115 **a(y/x)**. If **x** is less than **0**, and **y** is greater than or equal
1116 to **0**, it returns **a(y/x)+pi**. If **x** is less than **0**, and **y**
1117 is less than **0**, it returns **a(y/x)-pi**. If **x** is equal to **0**,
1118 and **y** is greater than **0**, it returns **pi/2**. If **x** is equal to
1119 **0**, and **y** is less than **0**, it returns **-pi/2**.
1121 This function is the same as the **atan2()** function in many programming
1124 This is a transcendental function (see the *Transcendental Functions*
1129 : Returns the sine of **x**, which is assumed to be in radians.
1131 This is an alias of **s(x)**.
1133 This is a transcendental function (see the *Transcendental Functions*
1138 : Returns the cosine of **x**, which is assumed to be in radians.
1140 This is an alias of **c(x)**.
1142 This is a transcendental function (see the *Transcendental Functions*
1147 : Returns the tangent of **x**, which is assumed to be in radians.
1149 If **x** is equal to **1** or **-1**, this raises an error and causes bc(1)
1150 to reset (see the **RESET** section).
1152 This is an alias of **t(x)**.
1154 This is a transcendental function (see the *Transcendental Functions*
1159 : Returns the arctangent of **x**, in radians.
1161 This is an alias of **a(x)**.
1163 This is a transcendental function (see the *Transcendental Functions*
1168 : Returns the arctangent of **y/x**, in radians. If both **y** and **x** are
1169 equal to **0**, it raises an error and causes bc(1) to reset (see the
1170 **RESET** section). Otherwise, if **x** is greater than **0**, it returns
1171 **a(y/x)**. If **x** is less than **0**, and **y** is greater than or equal
1172 to **0**, it returns **a(y/x)+pi**. If **x** is less than **0**, and **y**
1173 is less than **0**, it returns **a(y/x)-pi**. If **x** is equal to **0**,
1174 and **y** is greater than **0**, it returns **pi/2**. If **x** is equal to
1175 **0**, and **y** is less than **0**, it returns **-pi/2**.
1177 This function is the same as the **atan2()** function in many programming
1180 This is an alias of **a2(y, x)**.
1182 This is a transcendental function (see the *Transcendental Functions*
1187 : Converts **x** from radians to degrees and returns the result.
1189 This is a transcendental function (see the *Transcendental Functions*
1194 : Converts **x** from degrees to radians and returns the result.
1196 This is a transcendental function (see the *Transcendental Functions*
1201 : Generates a pseudo-random number between **0** (inclusive) and **1**
1202 (exclusive) with the number of decimal digits after the decimal point equal
1203 to the truncated absolute value of **p**. If **p** is not **0**, then
1204 calling this function will change the value of **seed**. If **p** is **0**,
1205 then **0** is returned, and **seed** is *not* changed.
1209 : Generates a pseudo-random number that is between **0** (inclusive) and the
1210 truncated absolute value of **i** (exclusive) with the number of decimal
1211 digits after the decimal point equal to the truncated absolute value of
1212 **p**. If the absolute value of **i** is greater than or equal to **2**, and
1213 **p** is not **0**, then calling this function will change the value of
1214 **seed**; otherwise, **0** is returned and **seed** is not changed.
1218 : Returns **x** with its sign flipped with probability **0.5**. In other
1219 words, it randomizes the sign of **x**.
1223 : Returns a random boolean value (either **0** or **1**).
1227 : Returns the numbers of unsigned integer bytes required to hold the truncated
1228 absolute value of **x**.
1232 : Returns the numbers of signed, two's-complement integer bytes required to
1233 hold the truncated value of **x**.
1237 : Outputs the hexadecimal (base **16**) representation of **x**.
1239 This is a **void** function (see the *Void Functions* subsection of the
1240 **FUNCTIONS** section).
1244 : Outputs the binary (base **2**) representation of **x**.
1246 This is a **void** function (see the *Void Functions* subsection of the
1247 **FUNCTIONS** section).
1251 : Outputs the base **b** representation of **x**.
1253 This is a **void** function (see the *Void Functions* subsection of the
1254 **FUNCTIONS** section).
1258 : Outputs the representation, in binary and hexadecimal, of **x** as an
1259 unsigned integer in as few power of two bytes as possible. Both outputs are
1260 split into bytes separated by spaces.
1262 If **x** is not an integer or is negative, an error message is printed
1263 instead, but bc(1) is not reset (see the **RESET** section).
1265 This is a **void** function (see the *Void Functions* subsection of the
1266 **FUNCTIONS** section).
1270 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1271 two's-complement integer in as few power of two bytes as possible. Both
1272 outputs are split into bytes separated by spaces.
1274 If **x** is not an integer, an error message is printed instead, but bc(1)
1275 is not reset (see the **RESET** section).
1277 This is a **void** function (see the *Void Functions* subsection of the
1278 **FUNCTIONS** section).
1282 : Outputs the representation, in binary and hexadecimal, of **x** as an
1283 unsigned integer in **n** bytes. Both outputs are split into bytes separated
1286 If **x** is not an integer, is negative, or cannot fit into **n** bytes, an
1287 error message is printed instead, but bc(1) is not reset (see the **RESET**
1290 This is a **void** function (see the *Void Functions* subsection of the
1291 **FUNCTIONS** section).
1295 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1296 two's-complement integer in **n** bytes. Both outputs are split into bytes
1297 separated by spaces.
1299 If **x** is not an integer or cannot fit into **n** bytes, an error message
1300 is printed instead, but bc(1) is not reset (see the **RESET** section).
1302 This is a **void** function (see the *Void Functions* subsection of the
1303 **FUNCTIONS** section).
1307 : Outputs the representation, in binary and hexadecimal, of **x** as an
1308 unsigned integer in **1** byte. Both outputs are split into bytes separated
1311 If **x** is not an integer, is negative, or cannot fit into **1** byte, an
1312 error message is printed instead, but bc(1) is not reset (see the **RESET**
1315 This is a **void** function (see the *Void Functions* subsection of the
1316 **FUNCTIONS** section).
1320 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1321 two's-complement integer in **1** byte. Both outputs are split into bytes
1322 separated by spaces.
1324 If **x** is not an integer or cannot fit into **1** byte, an error message
1325 is printed instead, but bc(1) is not reset (see the **RESET** section).
1327 This is a **void** function (see the *Void Functions* subsection of the
1328 **FUNCTIONS** section).
1332 : Outputs the representation, in binary and hexadecimal, of **x** as an
1333 unsigned integer in **2** bytes. Both outputs are split into bytes separated
1336 If **x** is not an integer, is negative, or cannot fit into **2** bytes, an
1337 error message is printed instead, but bc(1) is not reset (see the **RESET**
1340 This is a **void** function (see the *Void Functions* subsection of the
1341 **FUNCTIONS** section).
1345 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1346 two's-complement integer in **2** bytes. Both outputs are split into bytes
1347 separated by spaces.
1349 If **x** is not an integer or cannot fit into **2** bytes, an error message
1350 is printed instead, but bc(1) is not reset (see the **RESET** section).
1352 This is a **void** function (see the *Void Functions* subsection of the
1353 **FUNCTIONS** section).
1357 : Outputs the representation, in binary and hexadecimal, of **x** as an
1358 unsigned integer in **4** bytes. Both outputs are split into bytes separated
1361 If **x** is not an integer, is negative, or cannot fit into **4** bytes, an
1362 error message is printed instead, but bc(1) is not reset (see the **RESET**
1365 This is a **void** function (see the *Void Functions* subsection of the
1366 **FUNCTIONS** section).
1370 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1371 two's-complement integer in **4** bytes. Both outputs are split into bytes
1372 separated by spaces.
1374 If **x** is not an integer or cannot fit into **4** bytes, an error message
1375 is printed instead, but bc(1) is not reset (see the **RESET** section).
1377 This is a **void** function (see the *Void Functions* subsection of the
1378 **FUNCTIONS** section).
1382 : Outputs the representation, in binary and hexadecimal, of **x** as an
1383 unsigned integer in **8** bytes. Both outputs are split into bytes separated
1386 If **x** is not an integer, is negative, or cannot fit into **8** bytes, an
1387 error message is printed instead, but bc(1) is not reset (see the **RESET**
1390 This is a **void** function (see the *Void Functions* subsection of the
1391 **FUNCTIONS** section).
1395 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1396 two's-complement integer in **8** bytes. Both outputs are split into bytes
1397 separated by spaces.
1399 If **x** is not an integer or cannot fit into **8** bytes, an error message
1400 is printed instead, but bc(1) is not reset (see the **RESET** section).
1402 This is a **void** function (see the *Void Functions* subsection of the
1403 **FUNCTIONS** section).
1407 : Outputs the representation of the truncated absolute value of **x** as an
1408 unsigned integer in hexadecimal using **n** bytes. Not all of the value will
1409 be output if **n** is too small.
1411 This is a **void** function (see the *Void Functions* subsection of the
1412 **FUNCTIONS** section).
1414 **binary_uint(x, n)**
1416 : Outputs the representation of the truncated absolute value of **x** as an
1417 unsigned integer in binary using **n** bytes. Not all of the value will be
1418 output if **n** is too small.
1420 This is a **void** function (see the *Void Functions* subsection of the
1421 **FUNCTIONS** section).
1423 **output_uint(x, n)**
1425 : Outputs the representation of the truncated absolute value of **x** as an
1426 unsigned integer in the current **obase** (see the **SYNTAX** section) using
1427 **n** bytes. Not all of the value will be output if **n** is too small.
1429 This is a **void** function (see the *Void Functions* subsection of the
1430 **FUNCTIONS** section).
1432 **output_byte(x, i)**
1434 : Outputs byte **i** of the truncated absolute value of **x**, where **0** is
1435 the least significant byte and **number_of_bytes - 1** is the most
1438 This is a **void** function (see the *Void Functions* subsection of the
1439 **FUNCTIONS** section).
1442 ## Transcendental Functions
1444 All transcendental functions can return slightly inaccurate results (up to 1
1445 [ULP][4]). This is unavoidable, and [this article][5] explains why it is
1446 impossible and unnecessary to calculate exact results for the transcendental
1449 Because of the possible inaccuracy, I recommend that users call those functions
1450 with the precision (**scale**) set to at least 1 higher than is necessary. If
1451 exact results are *absolutely* required, users can double the precision
1452 (**scale**) and then truncate.
1454 The transcendental functions in the standard math library are:
1463 {{ A H N P HN HP NP HNP }}
1464 The transcendental functions in the extended math library are:
1483 When bc(1) encounters an error or a signal that it has a non-default handler
1484 for, it resets. This means that several things happen.
1486 First, any functions that are executing are stopped and popped off the stack.
1487 The behavior is not unlike that of exceptions in programming languages. Then
1488 the execution point is set so that any code waiting to execute (after all
1489 functions returned) is skipped.
1491 Thus, when bc(1) resets, it skips any remaining code waiting to be executed.
1492 Then, if it is interactive mode, and the error was not a fatal error (see the
1493 **EXIT STATUS** section), it asks for more input; otherwise, it exits with the
1494 appropriate return code.
1496 Note that this reset behavior is different from the GNU bc(1), which attempts to
1497 start executing the statement right after the one that caused an error.
1501 Most bc(1) implementations use **char** types to calculate the value of **1**
1502 decimal digit at a time, but that can be slow. This bc(1) does something
1505 It uses large integers to calculate more than **1** decimal digit at a time. If
1506 built in a environment where **BC_LONG_BIT** (see the **LIMITS** section) is
1507 **64**, then each integer has **9** decimal digits. If built in an environment
1508 where **BC_LONG_BIT** is **32** then each integer has **4** decimal digits. This
1509 value (the number of decimal digits per large integer) is called
1512 The actual values of **BC_LONG_BIT** and **BC_BASE_DIGS** can be queried with
1513 the **limits** statement.
1515 In addition, this bc(1) uses an even larger integer for overflow checking. This
1516 integer type depends on the value of **BC_LONG_BIT**, but is always at least
1517 twice as large as the integer type used to store digits.
1521 The following are the limits on bc(1):
1525 : The number of bits in the **long** type in the environment where bc(1) was
1526 built. This determines how many decimal digits can be stored in a single
1527 large integer (see the **PERFORMANCE** section).
1531 : The number of decimal digits per large integer (see the **PERFORMANCE**
1532 section). Depends on **BC_LONG_BIT**.
1536 : The max decimal number that each large integer can store (see
1537 **BC_BASE_DIGS**) plus **1**. Depends on **BC_BASE_DIGS**.
1541 : The max number that the overflow type (see the **PERFORMANCE** section) can
1542 hold. Depends on **BC_LONG_BIT**.
1546 : The maximum output base. Set at **BC_BASE_POW**.
1550 : The maximum size of arrays. Set at **SIZE_MAX-1**.
1554 : The maximum **scale**. Set at **BC_OVERFLOW_MAX-1**.
1558 : The maximum length of strings. Set at **BC_OVERFLOW_MAX-1**.
1562 : The maximum length of identifiers. Set at **BC_OVERFLOW_MAX-1**.
1566 : The maximum length of a number (in decimal digits), which includes digits
1567 after the decimal point. Set at **BC_OVERFLOW_MAX-1**.
1569 {{ A H N P HN HP NP HNP }}
1572 : The maximum integer (inclusive) returned by the **rand()** operand. Set at
1573 **2\^BC_LONG_BIT-1**.
1578 : The maximum allowable exponent (positive or negative). Set at
1579 **BC_OVERFLOW_MAX**.
1583 : The maximum number of vars/arrays. Set at **SIZE_MAX-1**.
1585 The actual values can be queried with the **limits** statement.
1587 These limits are meant to be effectively non-existent; the limits are so large
1588 (at least on 64-bit machines) that there should not be any point at which they
1589 become a problem. In fact, memory should be exhausted before these limits should
1592 # ENVIRONMENT VARIABLES
1594 bc(1) recognizes the following environment variables:
1598 : If this variable exists (no matter the contents), bc(1) behaves as if
1599 the **-s** option was given.
1603 : This is another way to give command-line arguments to bc(1). They should be
1604 in the same format as all other command-line arguments. These are always
1605 processed first, so any files given in **BC_ENV_ARGS** will be processed
1606 before arguments and files given on the command-line. This gives the user
1607 the ability to set up "standard" options and files to be used at every
1608 invocation. The most useful thing for such files to contain would be useful
1609 functions that the user might want every time bc(1) runs.
1611 The code that parses **BC_ENV_ARGS** will correctly handle quoted arguments,
1612 but it does not understand escape sequences. For example, the string
1613 **"/home/gavin/some bc file.bc"** will be correctly parsed, but the string
1614 **"/home/gavin/some \"bc\" file.bc"** will include the backslashes.
1616 The quote parsing will handle either kind of quotes, **'** or **"**. Thus,
1617 if you have a file with any number of single quotes in the name, you can use
1618 double quotes as the outside quotes, as in **"some 'bc' file.bc"**, and vice
1619 versa if you have a file with double quotes. However, handling a file with
1620 both kinds of quotes in **BC_ENV_ARGS** is not supported due to the
1621 complexity of the parsing, though such files are still supported on the
1622 command-line where the parsing is done by the shell.
1626 : If this environment variable exists and contains an integer that is greater
1627 than **1** and is less than **UINT16_MAX** (**2\^16-1**), bc(1) will output
1628 lines to that length, including the backslash (**\\**). The default line
1633 bc(1) returns the following exit statuses:
1641 : A math error occurred. This follows standard practice of using **1** for
1642 expected errors, since math errors will happen in the process of normal
1645 {{ A H N P HN HP NP HNP }}
1646 Math errors include divide by **0**, taking the square root of a negative
1647 number, using a negative number as a bound for the pseudo-random number
1648 generator, attempting to convert a negative number to a hardware integer,
1649 overflow when converting a number to a hardware integer, and attempting to
1650 use a non-integer where an integer is required.
1652 Converting to a hardware integer happens for the second operand of the power
1653 (**\^**), places (**\@**), left shift (**\<\<**), and right shift (**\>\>**)
1654 operators and their corresponding assignment operators.
1656 {{ E EH EN EP EHN EHP ENP EHNP }}
1657 Math errors include divide by **0**, taking the square root of a negative
1658 number, attempting to convert a negative number to a hardware integer,
1659 overflow when converting a number to a hardware integer, and attempting to
1660 use a non-integer where an integer is required.
1662 Converting to a hardware integer happens for the second operand of the power
1663 (**\^**) operator and the corresponding assignment operator.
1668 : A parse error occurred.
1670 Parse errors include unexpected **EOF**, using an invalid character, failing
1671 to find the end of a string or comment, using a token where it is invalid,
1672 giving an invalid expression, giving an invalid print statement, giving an
1673 invalid function definition, attempting to assign to an expression that is
1674 not a named expression (see the *Named Expressions* subsection of the
1675 **SYNTAX** section), giving an invalid **auto** list, having a duplicate
1676 **auto**/function parameter, failing to find the end of a code block,
1677 attempting to return a value from a **void** function, attempting to use a
1678 variable as a reference, and using any extensions when the option **-s** or
1679 any equivalents were given.
1683 : A runtime error occurred.
1685 Runtime errors include assigning an invalid number to **ibase**, **obase**,
1686 or **scale**; give a bad expression to a **read()** call, calling **read()**
1687 inside of a **read()** call, type errors, passing the wrong number of
1688 arguments to functions, attempting to call an undefined function, and
1689 attempting to use a **void** function call as a value in an expression.
1693 : A fatal error occurred.
1695 Fatal errors include memory allocation errors, I/O errors, failing to open
1696 files, attempting to use files that do not have only ASCII characters (bc(1)
1697 only accepts ASCII characters), attempting to open a directory as a file,
1698 and giving invalid command-line options.
1700 The exit status **4** is special; when a fatal error occurs, bc(1) always exits
1701 and returns **4**, no matter what mode bc(1) is in.
1703 The other statuses will only be returned when bc(1) is not in interactive mode
1704 (see the **INTERACTIVE MODE** section), since bc(1) resets its state (see the
1705 **RESET** section) and accepts more input when one of those errors occurs in
1706 interactive mode. This is also the case when interactive mode is forced by the
1707 **-i** flag or **-\-interactive** option.
1709 These exit statuses allow bc(1) to be used in shell scripting with error
1710 checking, and its normal behavior can be forced by using the **-i** flag or
1711 **-\-interactive** option.
1715 Per the [standard][1], bc(1) has an interactive mode and a non-interactive mode.
1716 Interactive mode is turned on automatically when both **stdin** and **stdout**
1717 are hooked to a terminal, but the **-i** flag and **-\-interactive** option can
1718 turn it on in other cases.
1720 In interactive mode, bc(1) attempts to recover from errors (see the **RESET**
1721 section), and in normal execution, flushes **stdout** as soon as execution is
1722 done for the current input.
1726 If **stdin**, **stdout**, and **stderr** are all connected to a TTY, bc(1) turns
1729 {{ A E N P EN EP NP ENP }}
1730 TTY mode is required for history to be enabled (see the **COMMAND LINE HISTORY**
1731 section). It is also required to enable special handling for **SIGINT** signals.
1734 {{ A E H N EH EN HN EHN }}
1735 The prompt is enabled in TTY mode.
1738 TTY mode is different from interactive mode because interactive mode is required
1739 in the [bc(1) specification][1], and interactive mode requires only **stdin**
1740 and **stdout** to be connected to a terminal.
1744 Sending a **SIGINT** will cause bc(1) to stop execution of the current input. If
1745 bc(1) is in TTY mode (see the **TTY MODE** section), it will reset (see the
1746 **RESET** section). Otherwise, it will clean up and exit.
1748 Note that "current input" can mean one of two things. If bc(1) is processing
1749 input from **stdin** in TTY mode, it will ask for more input. If bc(1) is
1750 processing input from a file in TTY mode, it will stop processing the file and
1751 start processing the next file, if one exists, or ask for input from **stdin**
1752 if no other file exists.
1754 This means that if a **SIGINT** is sent to bc(1) as it is executing a file, it
1755 can seem as though bc(1) did not respond to the signal since it will immediately
1756 start executing the next file. This is by design; most files that users execute
1757 when interacting with bc(1) have function definitions, which are quick to parse.
1758 If a file takes a long time to execute, there may be a bug in that file. The
1759 rest of the files could still be executed without problem, allowing the user to
1762 **SIGTERM** and **SIGQUIT** cause bc(1) to clean up and exit, and it uses the
1763 {{ A E N P EN EP NP ENP }}
1764 default handler for all other signals. The one exception is **SIGHUP**; in that
1765 case, when bc(1) is in TTY mode, a **SIGHUP** will cause bc(1) to clean up and
1768 {{ H EH HN HP EHN EHP HNP EHNP }}
1769 default handler for all other signals.
1772 {{ A E N P EN EP NP ENP }}
1773 # COMMAND LINE HISTORY
1775 bc(1) supports interactive command-line editing. If bc(1) is in TTY mode (see
1776 the **TTY MODE** section), history is enabled. Previous lines can be recalled
1777 and edited with the arrow keys.
1779 **Note**: tabs are converted to 8 spaces.
1782 {{ A E H P EH EP HP EHP }}
1785 This bc(1) ships with support for adding error messages for different locales
1786 and thus, supports **LC_MESSAGES**.
1795 bc(1) is compliant with the [IEEE Std 1003.1-2017 (“POSIX.1-2017”)][1]
1796 specification. The flags **-efghiqsvVw**, all long options, and the extensions
1797 noted above are extensions to that specification.
1799 Note that the specification explicitly says that bc(1) only accepts numbers that
1800 use a period (**.**) as a radix point, regardless of the value of
1803 {{ A E H P EH EP HP EHP }}
1804 This bc(1) supports error messages for different locales, and thus, it supports
1810 None are known. Report bugs at https://git.yzena.com/gavin/bc.
1814 Gavin D. Howard <gavin@yzena.com> and contributors.
1816 [1]: https://pubs.opengroup.org/onlinepubs/9699919799/utilities/bc.html
1817 [2]: https://www.gnu.org/software/bc/
1818 [3]: https://en.wikipedia.org/wiki/Rounding#Round_half_away_from_zero
1819 [4]: https://en.wikipedia.org/wiki/Unit_in_the_last_place
1820 [5]: https://people.eecs.berkeley.edu/~wkahan/LOG10HAF.TXT
1821 [6]: https://en.wikipedia.org/wiki/Rounding#Rounding_away_from_zero