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*...]
42 bc(1) is an interactive processor for a language first standardized in 1991 by
43 POSIX. (The current standard is [here][1].) The language provides unlimited
44 precision decimal arithmetic and is somewhat C-like, but there are differences.
45 Such differences will be noted in this document.
47 After parsing and handling options, this bc(1) reads any files given on the
48 command line and executes them before reading from **stdin**.
51 This bc(1) is a drop-in replacement for *any* bc(1), including (and
52 especially) the GNU bc(1). It also has many extensions and extra features beyond
53 other implementations.
56 This bc(1) is a drop-in replacement for *any* bc(1), including (and
57 especially) the GNU bc(1).
62 The following are the options that bc(1) accepts.
64 **-g**, **--global-stacks**
66 {{ A H N P HN HP NP HNP }}
67 : Turns the globals **ibase**, **obase**, **scale**, and **seed** into stacks.
69 This has the effect that a copy of the current value of all four are pushed
71 {{ E EH EN EP EHN EHP ENP EHNP }}
72 Turns the globals **ibase**, **obase**, and **scale** into stacks.
74 This has the effect that a copy of the current value of all three are pushed
76 onto a stack for every function call, as well as popped when every function
77 returns. This means that functions can assign to any and all of those
78 globals without worrying that the change will affect other functions.
79 Thus, a hypothetical function named **output(x,b)** that simply printed
80 **x** in base **b** could be written like this:
82 define void output(x, b) {
89 define void output(x, b) {
97 This makes writing functions much easier.
99 {{ A H N P HN HP NP HNP }}
100 (**Note**: the function **output(x,b)** exists in the extended math library.
101 See the **LIBRARY** section.)
103 However, since using this flag means that functions cannot set **ibase**,
104 **obase**, **scale**, or **seed** globally, functions that are made to do so
105 cannot work anymore. There are two possible use cases for that, and each has
108 {{ E EH EN EP EHN EHP ENP EHNP }}
109 However, since using this flag means that functions cannot set **ibase**,
110 **obase**, or **scale** globally, functions that are made to do so cannot
111 work anymore. There are two possible use cases for that, and each has a
115 First, if a function is called on startup to turn bc(1) into a number
116 converter, it is possible to replace that capability with various shell
119 alias d2o="bc -e ibase=A -e obase=8"
120 alias h2b="bc -e ibase=G -e obase=2"
122 {{ A H N P HN HP NP HNP }}
123 Second, if the purpose of a function is to set **ibase**, **obase**,
124 **scale**, or **seed** globally for any other purpose, it could be split
125 into one to four functions (based on how many globals it sets) and each of
126 those functions could return the desired value for a global.
128 For functions that set **seed**, the value assigned to **seed** is not
129 propagated to parent functions. This means that the sequence of
130 pseudo-random numbers that they see will not be the same sequence of
131 pseudo-random numbers that any parent sees. This is only the case once
132 **seed** has been set.
134 If a function desires to not affect the sequence of pseudo-random numbers
135 of its parents, but wants to use the same **seed**, it can use the following
140 {{ E EH EN EP EHN EHP ENP EHNP }}
141 Second, if the purpose of a function is to set **ibase**, **obase**, or
142 **scale** globally for any other purpose, it could be split into one to
143 three functions (based on how many globals it sets) and each of those
144 functions could return the desired value for a global.
147 If the behavior of this option is desired for every run of bc(1), then users
148 could make sure to define **BC_ENV_ARGS** and include this option (see the
149 **ENVIRONMENT VARIABLES** section for more details).
151 If **-s**, **-w**, or any equivalents are used, this option is ignored.
153 This is a **non-portable extension**.
157 : Prints a usage message and quits.
159 **-i**, **--interactive**
161 : Forces interactive mode. (See the **INTERACTIVE MODE** section.)
163 This is a **non-portable extension**.
165 **-l**, **--mathlib**
167 : Sets **scale** (see the **SYNTAX** section) to **20** and loads the included
168 {{ A H N P HN HP NP HNP }}
169 math library and the extended math library before running any code,
170 including any expressions or files specified on the command line.
172 To learn what is in the libraries, see the **LIBRARY** section.
174 {{ E EH EN EP EHN EHP ENP EHNP }}
175 math library before running any code, including any expressions or files
176 specified on the command line.
178 To learn what is in the library, see the **LIBRARY** section.
181 **-P**, **--no-prompt**
183 {{ A E H N EH EN HN EHN }}
184 : Disables the prompt in TTY mode. (The prompt is only enabled in TTY mode.
185 See the **TTY MODE** section) This is mostly for those users that do not
186 want a prompt or are not used to having them in bc(1). Most of those users
187 would want to put this option in **BC_ENV_ARGS** (see the
188 **ENVIRONMENT VARIABLES** section).
190 {{ P EP HP NP EHP ENP HNP EHNP }}
191 : This option is a no-op.
194 This is a **non-portable extension**.
198 : This option is for compatibility with the [GNU bc(1)][2]; it is a no-op.
199 Without this option, GNU bc(1) prints a copyright header. This bc(1) only
200 prints the copyright header if one or more of the **-v**, **-V**, or
201 **--version** options are given.
203 This is a **non-portable extension**.
205 **-s**, **--standard**
207 : Process exactly the language defined by the [standard][1] and error if any
210 This is a **non-portable extension**.
212 **-v**, **-V**, **--version**
214 : Print the version information (copyright header) and exit.
216 This is a **non-portable extension**.
220 : Like **-s** and **--standard**, except that warnings (and not errors) are
221 printed for non-standard extensions and execution continues normally.
223 This is a **non-portable extension**.
225 **-e** *expr*, **--expression**=*expr*
227 : Evaluates *expr*. If multiple expressions are given, they are evaluated in
228 order. If files are given as well (see below), the expressions and files are
229 evaluated in the order given. This means that if a file is given before an
230 expression, the file is read in and evaluated first.
232 If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
233 see the **ENVIRONMENT VARIABLES** section), then after processing all
234 expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
235 as an argument at least once to **-f** or **--file**, whether on the
236 command-line or in **BC_ENV_ARGS**. However, if any other **-e**,
237 **--expression**, **-f**, or **--file** arguments are given after **-f-** or
238 equivalent is given, bc(1) will give a fatal error and exit.
240 This is a **non-portable extension**.
242 **-f** *file*, **--file**=*file*
244 : Reads in *file* and evaluates it, line by line, as though it were read
245 through **stdin**. If expressions are also given (see above), the
246 expressions are evaluated in the order given.
248 If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
249 see the **ENVIRONMENT VARIABLES** section), then after processing all
250 expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
251 as an argument at least once to **-f** or **--file**. However, if any other
252 **-e**, **--expression**, **-f**, or **--file** arguments are given after
253 **-f-** or equivalent is given, bc(1) will give a fatal error and exit.
255 This is a **non-portable extension**.
257 All long options are **non-portable extensions**.
261 Any non-error output is written to **stdout**. In addition, if history (see the
262 **HISTORY** section) and the prompt (see the **TTY MODE** section) are enabled,
263 both are output to **stdout**.
265 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
266 error (see the **EXIT STATUS** section) if it cannot write to **stdout**, so if
267 **stdout** is closed, as in **bc <file> >&-**, it will quit with an error. This
268 is done so that bc(1) can report problems when **stdout** is redirected to a
271 If there are scripts that depend on the behavior of other bc(1) implementations,
272 it is recommended that those scripts be changed to redirect **stdout** to
277 Any error output is written to **stderr**.
279 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
280 error (see the **EXIT STATUS** section) if it cannot write to **stderr**, so if
281 **stderr** is closed, as in **bc <file> 2>&-**, it will quit with an error. This
282 is done so that bc(1) can exit with an error code when **stderr** is redirected
285 If there are scripts that depend on the behavior of other bc(1) implementations,
286 it is recommended that those scripts be changed to redirect **stderr** to
291 The syntax for bc(1) programs is mostly C-like, with some differences. This
292 bc(1) follows the [POSIX standard][1], which is a much more thorough resource
293 for the language this bc(1) accepts. This section is meant to be a summary and a
294 listing of all the extensions to the standard.
296 In the sections below, **E** means expression, **S** means statement, and **I**
299 Identifiers (**I**) start with a lowercase letter and can be followed by any
300 number (up to **BC_NAME_MAX-1**) of lowercase letters (**a-z**), digits
301 (**0-9**), and underscores (**\_**). The regex is **\[a-z\]\[a-z0-9\_\]\***.
302 Identifiers with more than one character (letter) are a
303 **non-portable extension**.
305 **ibase** is a global variable determining how to interpret constant numbers. It
306 is the "input" base, or the number base used for interpreting input numbers.
307 **ibase** is initially **10**. If the **-s** (**--standard**) and **-w**
308 (**--warn**) flags were not given on the command line, the max allowable value
309 for **ibase** is **36**. Otherwise, it is **16**. The min allowable value for
310 **ibase** is **2**. The max allowable value for **ibase** can be queried in
311 bc(1) programs with the **maxibase()** built-in function.
313 **obase** is a global variable determining how to output results. It is the
314 "output" base, or the number base used for outputting numbers. **obase** is
315 initially **10**. The max allowable value for **obase** is **BC_BASE_MAX** and
316 can be queried in bc(1) programs with the **maxobase()** built-in function. The
317 {{ A H N P HN HP NP HNP }}
318 min allowable value for **obase** is **0**. If **obase** is **0**, values are
319 output in scientific notation, and if **obase** is **1**, values are output in
320 engineering notation. Otherwise, values are output in the specified base.
322 Outputting in scientific and engineering notations are **non-portable
325 {{ E EH EN EP EHN EHP ENP EHNP }}
326 min allowable value for **obase** is **2**. Values are output in the specified
330 The *scale* of an expression is the number of digits in the result of the
331 expression right of the decimal point, and **scale** is a global variable that
332 sets the precision of any operations, with exceptions. **scale** is initially
333 **0**. **scale** cannot be negative. The max allowable value for **scale** is
334 **BC_SCALE_MAX** and can be queried in bc(1) programs with the **maxscale()**
337 bc(1) has both *global* variables and *local* variables. All *local*
338 variables are local to the function; they are parameters or are introduced in
339 the **auto** list of a function (see the **FUNCTIONS** section). If a variable
340 is accessed which is not a parameter or in the **auto** list, it is assumed to
341 be *global*. If a parent function has a *local* variable version of a variable
342 that a child function considers *global*, the value of that *global* variable in
343 the child function is the value of the variable in the parent function, not the
344 value of the actual *global* variable.
346 All of the above applies to arrays as well.
348 The value of a statement that is an expression (i.e., any of the named
349 expressions or operands) is printed unless the lowest precedence operator is an
350 assignment operator *and* the expression is notsurrounded by parentheses.
352 The value that is printed is also assigned to the special variable **last**. A
353 single dot (**.**) may also be used as a synonym for **last**. These are
354 **non-portable extensions**.
356 Either semicolons or newlines may separate statements.
360 There are two kinds of comments:
362 1. Block comments are enclosed in **/\*** and **\*/**.
363 2. Line comments go from **#** until, and not including, the next newline. This
364 is a **non-portable extension**.
368 The following are named expressions in bc(1):
371 2. Array Elements: **I[E]**
375 {{ A H N P HN HP NP HNP }}
377 7. **last** or a single dot (**.**)
379 Numbers 6 and 7 are **non-portable extensions**.
381 The meaning of **seed** is dependent on the current pseudo-random number
382 generator but is guaranteed to not change except for new major versions.
384 The *scale* and sign of the value may be significant.
386 If a previously used **seed** value is assigned to **seed** and used again, the
387 pseudo-random number generator is guaranteed to produce the same sequence of
388 pseudo-random numbers as it did when the **seed** value was previously used.
390 The exact value assigned to **seed** is not guaranteed to be returned if
391 **seed** is queried again immediately. However, if **seed** *does* return a
392 different value, both values, when assigned to **seed**, are guaranteed to
393 produce the same sequence of pseudo-random numbers. This means that certain
394 values assigned to **seed** will *not* produce unique sequences of pseudo-random
395 numbers. The value of **seed** will change after any use of the **rand()** and
396 **irand(E)** operands (see the *Operands* subsection below), except if the
397 parameter passed to **irand(E)** is **0**, **1**, or negative.
399 There is no limit to the length (number of significant decimal digits) or
400 *scale* of the value that can be assigned to **seed**.
402 {{ E EH EN EP EHN EHP ENP EHNP }}
403 6. **last** or a single dot (**.**)
405 Number 6 is a **non-portable extension**.
408 Variables and arrays do not interfere; users can have arrays named the same as
409 variables. This also applies to functions (see the **FUNCTIONS** section), so a
410 user can have a variable, array, and function that all have the same name, and
411 they will not shadow each other, whether inside of functions or not.
413 Named expressions are required as the operand of **increment**/**decrement**
414 operators and as the left side of **assignment** operators (see the *Operators*
419 The following are valid operands in bc(1):
421 1. Numbers (see the *Numbers* subsection below).
422 2. Array indices (**I[E]**).
423 3. **(E)**: The value of **E** (used to change precedence).
424 4. **sqrt(E)**: The square root of **E**. **E** must be non-negative.
425 5. **length(E)**: The number of significant decimal digits in **E**.
426 6. **length(I[])**: The number of elements in the array **I**. This is a
427 **non-portable extension**.
428 7. **scale(E)**: The *scale* of **E**.
429 8. **abs(E)**: The absolute value of **E**. This is a **non-portable
431 9. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
432 a non-**void** function (see the *Void Functions* subsection of the
433 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
434 **I[]**, which will automatically be turned into array references (see the
435 *Array References* subsection of the **FUNCTIONS** section) if the
436 corresponding parameter in the function definition is an array reference.
437 10. **read()**: Reads a line from **stdin** and uses that as an expression. The
438 result of that expression is the result of the **read()** operand. This is a
439 **non-portable extension**.
440 11. **maxibase()**: The max allowable **ibase**. This is a **non-portable
442 12. **maxobase()**: The max allowable **obase**. This is a **non-portable
444 13. **maxscale()**: The max allowable **scale**. This is a **non-portable
446 {{ A H N P HN HP NP HNP }}
447 14. **rand()**: A pseudo-random integer between **0** (inclusive) and
448 **BC_RAND_MAX** (inclusive). Using this operand will change the value of
449 **seed**. This is a **non-portable extension**.
450 15. **irand(E)**: A pseudo-random integer between **0** (inclusive) and the
451 value of **E** (exclusive). If **E** is negative or is a non-integer
452 (**E**'s *scale* is not **0**), an error is raised, and bc(1) resets (see
453 the **RESET** section) while **seed** remains unchanged. If **E** is larger
454 than **BC_RAND_MAX**, the higher bound is honored by generating several
455 pseudo-random integers, multiplying them by appropriate powers of
456 **BC_RAND_MAX+1**, and adding them together. Thus, the size of integer that
457 can be generated with this operand is unbounded. Using this operand will
458 change the value of **seed**, unless the value of **E** is **0** or **1**.
459 In that case, **0** is returned, and **seed** is *not* changed. This is a
460 **non-portable extension**.
461 16. **maxrand()**: The max integer returned by **rand()**. This is a
462 **non-portable extension**.
464 The integers generated by **rand()** and **irand(E)** are guaranteed to be as
465 unbiased as possible, subject to the limitations of the pseudo-random number
468 **Note**: The values returned by the pseudo-random number generator with
469 **rand()** and **irand(E)** are guaranteed to *NOT* be cryptographically secure.
470 This is a consequence of using a seeded pseudo-random number generator. However,
471 they *are* guaranteed to be reproducible with identical **seed** values. This
472 means that the pseudo-random values from bc(1) should only be used where a
473 reproducible stream of pseudo-random numbers is *ESSENTIAL*. In any other case,
474 use a non-seeded pseudo-random number generator.
479 Numbers are strings made up of digits, uppercase letters, and at most **1**
480 period for a radix. Numbers can have up to **BC_NUM_MAX** digits. Uppercase
481 letters are equal to **9** + their position in the alphabet (i.e., **A** equals
482 **10**, or **9+1**). If a digit or letter makes no sense with the current value
483 of **ibase**, they are set to the value of the highest valid digit in **ibase**.
485 Single-character numbers (i.e., **A** alone) take the value that they would have
486 if they were valid digits, regardless of the value of **ibase**. This means that
487 **A** alone always equals decimal **10** and **Z** alone always equals decimal
490 {{ A H N P HN HP NP HNP }}
491 In addition, bc(1) accepts numbers in scientific notation. These have the form
492 **\<number\>e\<integer\>**. The exponent (the portion after the **e**) must be
493 an integer. An example is **1.89237e9**, which is equal to **1892370000**.
494 Negative exponents are also allowed, so **4.2890e-3** is equal to **0.0042890**.
496 Using scientific notation is an error or warning if the **-s** or **-w**,
497 respectively, command-line options (or equivalents) are given.
499 **WARNING**: Both the number and the exponent in scientific notation are
500 interpreted according to the current **ibase**, but the number is still
501 multiplied by **10\^exponent** regardless of the current **ibase**. For example,
502 if **ibase** is **16** and bc(1) is given the number string **FFeA**, the
503 resulting decimal number will be **2550000000000**, and if bc(1) is given the
504 number string **10e-4**, the resulting decimal number will be **0.0016**.
506 Accepting input as scientific notation is a **non-portable extension**.
511 The following arithmetic and logical operators can be used. They are listed in
512 order of decreasing precedence. Operators in the same group have the same
517 : Type: Prefix and Postfix
521 Description: **increment**, **decrement**
529 Description: **negation**, **boolean not**
531 {{ A H N P HN HP NP HNP }}
538 Description: **truncation**
546 Description: **set precision**
555 Description: **power**
563 Description: **multiply**, **divide**, **modulus**
571 Description: **add**, **subtract**
573 {{ A H N P HN HP NP HNP }}
580 Description: **shift left**, **shift right**
582 **=** **\<\<=** **\>\>=** **+=** **-=** **\*=** **/=** **%=** **\^=** **\@=**
584 {{ E EH EN EP EHN EHP ENP EHNP }}
585 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
592 Description: **assignment**
594 **==** **\<=** **\>=** **!=** **\<** **\>**
600 Description: **relational**
608 Description: **boolean and**
616 Description: **boolean or**
618 The operators will be described in more detail below.
622 : The prefix and postfix **increment** and **decrement** operators behave
623 exactly like they would in C. They require a named expression (see the
624 *Named Expressions* subsection) as an operand.
626 The prefix versions of these operators are more efficient; use them where
631 : The **negation** operator returns **0** if a user attempts to negate any
632 expression with the value **0**. Otherwise, a copy of the expression with
633 its sign flipped is returned.
637 : The **boolean not** operator returns **1** if the expression is **0**, or
640 This is a **non-portable extension**.
642 {{ A H N P HN HP NP HNP }}
645 : The **truncation** operator returns a copy of the given expression with all
646 of its *scale* removed.
648 This is a **non-portable extension**.
652 : The **set precision** operator takes two expressions and returns a copy of
653 the first with its *scale* equal to the value of the second expression. That
654 could either mean that the number is returned without change (if the
655 *scale* of the first expression matches the value of the second
656 expression), extended (if it is less), or truncated (if it is more).
658 The second expression must be an integer (no *scale*) and non-negative.
660 This is a **non-portable extension**.
665 : The **power** operator (not the **exclusive or** operator, as it would be in
666 C) takes two expressions and raises the first to the power of the value of
667 the second. The *scale* of the result is equal to **scale**.
669 The second expression must be an integer (no *scale*), and if it is
670 negative, the first value must be non-zero.
674 : The **multiply** operator takes two expressions, multiplies them, and
675 returns the product. If **a** is the *scale* of the first expression and
676 **b** is the *scale* of the second expression, the *scale* of the result is
677 equal to **min(a+b,max(scale,a,b))** where **min()** and **max()** return
682 : The **divide** operator takes two expressions, divides them, and returns the
683 quotient. The *scale* of the result shall be the value of **scale**.
685 The second expression must be non-zero.
689 : The **modulus** operator takes two expressions, **a** and **b**, and
690 evaluates them by 1) Computing **a/b** to current **scale** and 2) Using the
691 result of step 1 to calculate **a-(a/b)\*b** to *scale*
692 **max(scale+scale(b),scale(a))**.
694 The second expression must be non-zero.
698 : The **add** operator takes two expressions, **a** and **b**, and returns the
699 sum, with a *scale* equal to the max of the *scale*s of **a** and **b**.
703 : The **subtract** operator takes two expressions, **a** and **b**, and
704 returns the difference, with a *scale* equal to the max of the *scale*s of
707 {{ A H N P HN HP NP HNP }}
710 : The **left shift** operator takes two expressions, **a** and **b**, and
711 returns a copy of the value of **a** with its decimal point moved **b**
714 The second expression must be an integer (no *scale*) and non-negative.
716 This is a **non-portable extension**.
720 : The **right shift** operator takes two expressions, **a** and **b**, and
721 returns a copy of the value of **a** with its decimal point moved **b**
724 The second expression must be an integer (no *scale*) and non-negative.
726 This is a **non-portable extension**.
729 {{ A H N P HN HP NP HNP }}
730 **=** **\<\<=** **\>\>=** **+=** **-=** **\*=** **/=** **%=** **\^=** **\@=**
732 {{ E EH EN EP EHN EHP ENP EHNP }}
733 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
736 : The **assignment** operators take two expressions, **a** and **b** where
737 **a** is a named expression (see the *Named Expressions* subsection).
739 For **=**, **b** is copied and the result is assigned to **a**. For all
740 others, **a** and **b** are applied as operands to the corresponding
741 arithmetic operator and the result is assigned to **a**.
743 {{ A H N P HN HP NP HNP }}
744 The **assignment** operators that correspond to operators that are
745 extensions are themselves **non-portable extensions**.
748 **==** **\<=** **\>=** **!=** **\<** **\>**
750 : The **relational** operators compare two expressions, **a** and **b**, and
751 if the relation holds, according to C language semantics, the result is
752 **1**. Otherwise, it is **0**.
754 Note that unlike in C, these operators have a lower precedence than the
755 **assignment** operators, which means that **a=b\>c** is interpreted as
758 Also, unlike the [standard][1] requires, these operators can appear anywhere
759 any other expressions can be used. This allowance is a
760 **non-portable extension**.
764 : The **boolean and** operator takes two expressions and returns **1** if both
765 expressions are non-zero, **0** otherwise.
767 This is *not* a short-circuit operator.
769 This is a **non-portable extension**.
773 : The **boolean or** operator takes two expressions and returns **1** if one
774 of the expressions is non-zero, **0** otherwise.
776 This is *not* a short-circuit operator.
778 This is a **non-portable extension**.
782 The following items are statements:
785 2. **{** **S** **;** ... **;** **S** **}**
786 3. **if** **(** **E** **)** **S**
787 4. **if** **(** **E** **)** **S** **else** **S**
788 5. **while** **(** **E** **)** **S**
789 6. **for** **(** **E** **;** **E** **;** **E** **)** **S**
790 7. An empty statement
796 13. A string of characters, enclosed in double quotes
797 14. **print** **E** **,** ... **,** **E**
798 15. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
799 a **void** function (see the *Void Functions* subsection of the
800 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
801 **I[]**, which will automatically be turned into array references (see the
802 *Array References* subsection of the **FUNCTIONS** section) if the
803 corresponding parameter in the function definition is an array reference.
805 Numbers 4, 9, 11, 12, 14, and 15 are **non-portable extensions**.
807 Also, as a **non-portable extension**, any or all of the expressions in the
808 header of a for loop may be omitted. If the condition (second expression) is
809 omitted, it is assumed to be a constant **1**.
811 The **break** statement causes a loop to stop iterating and resume execution
812 immediately following a loop. This is only allowed in loops.
814 The **continue** statement causes a loop iteration to stop early and returns to
815 the start of the loop, including testing the loop condition. This is only
818 The **if** **else** statement does the same thing as in C.
820 The **quit** statement causes bc(1) to quit, even if it is on a branch that will
821 not be executed (it is a compile-time command).
823 The **halt** statement causes bc(1) to quit, if it is executed. (Unlike **quit**
824 if it is on a branch of an **if** statement that is not executed, bc(1) does not
827 The **limits** statement prints the limits that this bc(1) is subject to. This
828 is like the **quit** statement in that it is a compile-time command.
830 An expression by itself is evaluated and printed, followed by a newline.
832 {{ A H N P HN HP NP HNP }}
833 Both scientific notation and engineering notation are available for printing the
834 results of expressions. Scientific notation is activated by assigning **0** to
835 **obase**, and engineering notation is activated by assigning **1** to
836 **obase**. To deactivate them, just assign a different value to **obase**.
838 Scientific notation and engineering notation are disabled if bc(1) is run with
839 either the **-s** or **-w** command-line options (or equivalents).
841 Printing numbers in scientific notation and/or engineering notation is a
842 **non-portable extension**.
847 The "expressions" in a **print** statement may also be strings. If they are, there
848 are backslash escape sequences that are interpreted specially. What those
849 sequences are, and what they cause to be printed, are shown below:
863 Any other character following a backslash causes the backslash and character to
866 Any non-string expression in a print statement shall be assigned to **last**,
867 like any other expression that is printed.
869 ## Order of Evaluation
871 All expressions in a statment are evaluated left to right, except as necessary
872 to maintain order of operations. This means, for example, assuming that **i** is
873 equal to **0**, in the expression
877 the first (or 0th) element of **a** is set to **1**, and **i** is equal to **2**
878 at the end of the expression.
880 This includes function arguments. Thus, assuming **i** is equal to **0**, this
881 means that in the expression
885 the first argument passed to **x()** is **0**, and the second argument is **1**,
886 while **i** is equal to **2** before the function starts executing.
890 Function definitions are as follows:
900 Any **I** in the parameter list or **auto** list may be replaced with **I[]** to
901 make a parameter or **auto** var an array, and any **I** in the parameter list
902 may be replaced with **\*I[]** to make a parameter an array reference. Callers
903 of functions that take array references should not put an asterisk in the call;
904 they must be called with just **I[]** like normal array parameters and will be
905 automatically converted into references.
907 As a **non-portable extension**, the opening brace of a **define** statement may
908 appear on the next line.
910 As a **non-portable extension**, the return statement may also be in one of the
914 2. **return** **(** **)**
917 The first two, or not specifying a **return** statement, is equivalent to
918 **return (0)**, unless the function is a **void** function (see the *Void
919 Functions* subsection below).
923 Functions can also be **void** functions, defined as follows:
926 define void I(I,...,I){
933 They can only be used as standalone expressions, where such an expression would
934 be printed alone, except in a print statement.
936 Void functions can only use the first two **return** statements listed above.
937 They can also omit the return statement entirely.
939 The word "void" is not treated as a keyword; it is still possible to have
940 variables, arrays, and functions named **void**. The word "void" is only
941 treated specially right after the **define** keyword.
943 This is a **non-portable extension**.
947 For any array in the parameter list, if the array is declared in the form
953 it is a **reference**. Any changes to the array in the function are reflected,
954 when the function returns, to the array that was passed in.
956 Other than this, all function arguments are passed by value.
958 This is a **non-portable extension**.
962 {{ A H N P HN HP NP HNP }}
963 All of the functions below, including the functions in the extended math
964 library (see the *Extended Library* subsection below), are available when the
965 **-l** or **--mathlib** command-line flags are given, except that the extended
966 math library is not available when the **-s** option, the **-w** option, or
967 equivalents are given.
969 {{ E EH EN EP EHN EHP ENP EHNP }}
970 All of the functions below are available when the **-l** or **--mathlib**
971 command-line flags are given.
976 The [standard][1] defines the following functions for the math library:
980 : Returns the sine of **x**, which is assumed to be in radians.
982 This is a transcendental function (see the *Transcendental Functions*
987 : Returns the cosine of **x**, which is assumed to be in radians.
989 This is a transcendental function (see the *Transcendental Functions*
994 : Returns the arctangent of **x**, in radians.
996 This is a transcendental function (see the *Transcendental Functions*
1001 : Returns the natural logarithm of **x**.
1003 This is a transcendental function (see the *Transcendental Functions*
1008 : Returns the mathematical constant **e** raised to the power of **x**.
1010 This is a transcendental function (see the *Transcendental Functions*
1015 : Returns the bessel integer order **n** (truncated) of **x**.
1017 This is a transcendental function (see the *Transcendental Functions*
1020 {{ A H N P HN HP NP HNP }}
1023 The extended library is *not* loaded when the **-s**/**--standard** or
1024 **-w**/**--warn** options are given since they are not part of the library
1025 defined by the [standard][1].
1027 The extended library is a **non-portable extension**.
1031 : Calculates **x** to the power of **y**, even if **y** is not an integer, and
1032 returns the result to the current **scale**.
1034 It is an error if **y** is negative and **x** is **0**.
1036 This is a transcendental function (see the *Transcendental Functions*
1041 : Returns **x** rounded to **p** decimal places according to the rounding mode
1042 [round half away from **0**][3].
1046 : Returns **x** rounded to **p** decimal places according to the rounding mode
1047 [round away from **0**][6].
1051 : Returns the factorial of the truncated absolute value of **x**.
1055 : Returns the permutation of the truncated absolute value of **n** of the
1056 truncated absolute value of **k**, if **k \<= n**. If not, it returns **0**.
1060 : Returns the combination of the truncated absolute value of **n** of the
1061 truncated absolute value of **k**, if **k \<= n**. If not, it returns **0**.
1065 : Returns the logarithm base **2** of **x**.
1067 This is a transcendental function (see the *Transcendental Functions*
1072 : Returns the logarithm base **10** of **x**.
1074 This is a transcendental function (see the *Transcendental Functions*
1079 : Returns the logarithm base **b** of **x**.
1081 This is a transcendental function (see the *Transcendental Functions*
1086 : Returns the cube root of **x**.
1090 : Calculates the truncated value of **n**, **r**, and returns the **r**th root
1091 of **x** to the current **scale**.
1093 If **r** is **0** or negative, this raises an error and causes bc(1) to
1094 reset (see the **RESET** section). It also raises an error and causes bc(1)
1095 to reset if **r** is even and **x** is negative.
1099 : Returns **pi** to **p** decimal places.
1101 This is a transcendental function (see the *Transcendental Functions*
1106 : Returns the tangent of **x**, which is assumed to be in radians.
1108 This is a transcendental function (see the *Transcendental Functions*
1113 : Returns the arctangent of **y/x**, in radians. If both **y** and **x** are
1114 equal to **0**, it raises an error and causes bc(1) to reset (see the
1115 **RESET** section). Otherwise, if **x** is greater than **0**, it returns
1116 **a(y/x)**. If **x** is less than **0**, and **y** is greater than or equal
1117 to **0**, it returns **a(y/x)+pi**. If **x** is less than **0**, and **y**
1118 is less than **0**, it returns **a(y/x)-pi**. If **x** is equal to **0**,
1119 and **y** is greater than **0**, it returns **pi/2**. If **x** is equal to
1120 **0**, and **y** is less than **0**, it returns **-pi/2**.
1122 This function is the same as the **atan2()** function in many programming
1125 This is a transcendental function (see the *Transcendental Functions*
1130 : Returns the sine of **x**, which is assumed to be in radians.
1132 This is an alias of **s(x)**.
1134 This is a transcendental function (see the *Transcendental Functions*
1139 : Returns the cosine of **x**, which is assumed to be in radians.
1141 This is an alias of **c(x)**.
1143 This is a transcendental function (see the *Transcendental Functions*
1148 : Returns the tangent of **x**, which is assumed to be in radians.
1150 If **x** is equal to **1** or **-1**, this raises an error and causes bc(1)
1151 to reset (see the **RESET** section).
1153 This is an alias of **t(x)**.
1155 This is a transcendental function (see the *Transcendental Functions*
1160 : Returns the arctangent of **x**, in radians.
1162 This is an alias of **a(x)**.
1164 This is a transcendental function (see the *Transcendental Functions*
1169 : Returns the arctangent of **y/x**, in radians. If both **y** and **x** are
1170 equal to **0**, it raises an error and causes bc(1) to reset (see the
1171 **RESET** section). Otherwise, if **x** is greater than **0**, it returns
1172 **a(y/x)**. If **x** is less than **0**, and **y** is greater than or equal
1173 to **0**, it returns **a(y/x)+pi**. If **x** is less than **0**, and **y**
1174 is less than **0**, it returns **a(y/x)-pi**. If **x** is equal to **0**,
1175 and **y** is greater than **0**, it returns **pi/2**. If **x** is equal to
1176 **0**, and **y** is less than **0**, it returns **-pi/2**.
1178 This function is the same as the **atan2()** function in many programming
1181 This is an alias of **a2(y, x)**.
1183 This is a transcendental function (see the *Transcendental Functions*
1188 : Converts **x** from radians to degrees and returns the result.
1190 This is a transcendental function (see the *Transcendental Functions*
1195 : Converts **x** from degrees to radians and returns the result.
1197 This is a transcendental function (see the *Transcendental Functions*
1202 : Generates a pseudo-random number between **0** (inclusive) and **1**
1203 (exclusive) with the number of decimal digits after the decimal point equal
1204 to the truncated absolute value of **p**. If **p** is not **0**, then
1205 calling this function will change the value of **seed**. If **p** is **0**,
1206 then **0** is returned, and **seed** is *not* changed.
1210 : Generates a pseudo-random number that is between **0** (inclusive) and the
1211 truncated absolute value of **i** (exclusive) with the number of decimal
1212 digits after the decimal point equal to the truncated absolute value of
1213 **p**. If the absolute value of **i** is greater than or equal to **2**, and
1214 **p** is not **0**, then calling this function will change the value of
1215 **seed**; otherwise, **0** is returned and **seed** is not changed.
1219 : Returns **x** with its sign flipped with probability **0.5**. In other
1220 words, it randomizes the sign of **x**.
1224 : Returns a random boolean value (either **0** or **1**).
1228 : Returns the numbers of unsigned integer bytes required to hold the truncated
1229 absolute value of **x**.
1233 : Returns the numbers of signed, two's-complement integer bytes required to
1234 hold the truncated value of **x**.
1238 : Outputs the hexadecimal (base **16**) representation of **x**.
1240 This is a **void** function (see the *Void Functions* subsection of the
1241 **FUNCTIONS** section).
1245 : Outputs the binary (base **2**) representation of **x**.
1247 This is a **void** function (see the *Void Functions* subsection of the
1248 **FUNCTIONS** section).
1252 : Outputs the base **b** representation of **x**.
1254 This is a **void** function (see the *Void Functions* subsection of the
1255 **FUNCTIONS** section).
1259 : Outputs the representation, in binary and hexadecimal, of **x** as an
1260 unsigned integer in as few power of two bytes as possible. Both outputs are
1261 split into bytes separated by spaces.
1263 If **x** is not an integer or is negative, an error message is printed
1264 instead, but bc(1) is not reset (see the **RESET** section).
1266 This is a **void** function (see the *Void Functions* subsection of the
1267 **FUNCTIONS** section).
1271 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1272 two's-complement integer in as few power of two bytes as possible. Both
1273 outputs are split into bytes separated by spaces.
1275 If **x** is not an integer, an error message is printed instead, but bc(1)
1276 is not reset (see the **RESET** section).
1278 This is a **void** function (see the *Void Functions* subsection of the
1279 **FUNCTIONS** section).
1283 : Outputs the representation, in binary and hexadecimal, of **x** as an
1284 unsigned integer in **n** bytes. Both outputs are split into bytes separated
1287 If **x** is not an integer, is negative, or cannot fit into **n** bytes, an
1288 error message is printed instead, but bc(1) is not reset (see the **RESET**
1291 This is a **void** function (see the *Void Functions* subsection of the
1292 **FUNCTIONS** section).
1296 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1297 two's-complement integer in **n** bytes. Both outputs are split into bytes
1298 separated by spaces.
1300 If **x** is not an integer or cannot fit into **n** bytes, an error message
1301 is printed instead, but bc(1) is not reset (see the **RESET** section).
1303 This is a **void** function (see the *Void Functions* subsection of the
1304 **FUNCTIONS** section).
1308 : Outputs the representation, in binary and hexadecimal, of **x** as an
1309 unsigned integer in **1** byte. Both outputs are split into bytes separated
1312 If **x** is not an integer, is negative, or cannot fit into **1** byte, an
1313 error message is printed instead, but bc(1) is not reset (see the **RESET**
1316 This is a **void** function (see the *Void Functions* subsection of the
1317 **FUNCTIONS** section).
1321 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1322 two's-complement integer in **1** byte. Both outputs are split into bytes
1323 separated by spaces.
1325 If **x** is not an integer or cannot fit into **1** byte, an error message
1326 is printed instead, but bc(1) is not reset (see the **RESET** section).
1328 This is a **void** function (see the *Void Functions* subsection of the
1329 **FUNCTIONS** section).
1333 : Outputs the representation, in binary and hexadecimal, of **x** as an
1334 unsigned integer in **2** bytes. Both outputs are split into bytes separated
1337 If **x** is not an integer, is negative, or cannot fit into **2** bytes, an
1338 error message is printed instead, but bc(1) is not reset (see the **RESET**
1341 This is a **void** function (see the *Void Functions* subsection of the
1342 **FUNCTIONS** section).
1346 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1347 two's-complement integer in **2** bytes. Both outputs are split into bytes
1348 separated by spaces.
1350 If **x** is not an integer or cannot fit into **2** bytes, an error message
1351 is printed instead, but bc(1) is not reset (see the **RESET** section).
1353 This is a **void** function (see the *Void Functions* subsection of the
1354 **FUNCTIONS** section).
1358 : Outputs the representation, in binary and hexadecimal, of **x** as an
1359 unsigned integer in **4** bytes. Both outputs are split into bytes separated
1362 If **x** is not an integer, is negative, or cannot fit into **4** bytes, an
1363 error message is printed instead, but bc(1) is not reset (see the **RESET**
1366 This is a **void** function (see the *Void Functions* subsection of the
1367 **FUNCTIONS** section).
1371 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1372 two's-complement integer in **4** bytes. Both outputs are split into bytes
1373 separated by spaces.
1375 If **x** is not an integer or cannot fit into **4** bytes, an error message
1376 is printed instead, but bc(1) is not reset (see the **RESET** section).
1378 This is a **void** function (see the *Void Functions* subsection of the
1379 **FUNCTIONS** section).
1383 : Outputs the representation, in binary and hexadecimal, of **x** as an
1384 unsigned integer in **8** bytes. Both outputs are split into bytes separated
1387 If **x** is not an integer, is negative, or cannot fit into **8** bytes, an
1388 error message is printed instead, but bc(1) is not reset (see the **RESET**
1391 This is a **void** function (see the *Void Functions* subsection of the
1392 **FUNCTIONS** section).
1396 : Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1397 two's-complement integer in **8** bytes. Both outputs are split into bytes
1398 separated by spaces.
1400 If **x** is not an integer or cannot fit into **8** bytes, an error message
1401 is printed instead, but bc(1) is not reset (see the **RESET** section).
1403 This is a **void** function (see the *Void Functions* subsection of the
1404 **FUNCTIONS** section).
1408 : Outputs the representation of the truncated absolute value of **x** as an
1409 unsigned integer in hexadecimal using **n** bytes. Not all of the value will
1410 be output if **n** is too small.
1412 This is a **void** function (see the *Void Functions* subsection of the
1413 **FUNCTIONS** section).
1415 **binary_uint(x, n)**
1417 : Outputs the representation of the truncated absolute value of **x** as an
1418 unsigned integer in binary using **n** bytes. Not all of the value will be
1419 output if **n** is too small.
1421 This is a **void** function (see the *Void Functions* subsection of the
1422 **FUNCTIONS** section).
1424 **output_uint(x, n)**
1426 : Outputs the representation of the truncated absolute value of **x** as an
1427 unsigned integer in the current **obase** (see the **SYNTAX** section) using
1428 **n** bytes. Not all of the value will be output if **n** is too small.
1430 This is a **void** function (see the *Void Functions* subsection of the
1431 **FUNCTIONS** section).
1433 **output_byte(x, i)**
1435 : Outputs byte **i** of the truncated absolute value of **x**, where **0** is
1436 the least significant byte and **number_of_bytes - 1** is the most
1439 This is a **void** function (see the *Void Functions* subsection of the
1440 **FUNCTIONS** section).
1443 ## Transcendental Functions
1445 All transcendental functions can return slightly inaccurate results (up to 1
1446 [ULP][4]). This is unavoidable, and [this article][5] explains why it is
1447 impossible and unnecessary to calculate exact results for the transcendental
1450 Because of the possible inaccuracy, I recommend that users call those functions
1451 with the precision (**scale**) set to at least 1 higher than is necessary. If
1452 exact results are *absolutely* required, users can double the precision
1453 (**scale**) and then truncate.
1455 The transcendental functions in the standard math library are:
1464 {{ A H N P HN HP NP HNP }}
1465 The transcendental functions in the extended math library are:
1484 When bc(1) encounters an error or a signal that it has a non-default handler
1485 for, it resets. This means that several things happen.
1487 First, any functions that are executing are stopped and popped off the stack.
1488 The behavior is not unlike that of exceptions in programming languages. Then
1489 the execution point is set so that any code waiting to execute (after all
1490 functions returned) is skipped.
1492 Thus, when bc(1) resets, it skips any remaining code waiting to be executed.
1493 Then, if it is interactive mode, and the error was not a fatal error (see the
1494 **EXIT STATUS** section), it asks for more input; otherwise, it exits with the
1495 appropriate return code.
1497 Note that this reset behavior is different from the GNU bc(1), which attempts to
1498 start executing the statement right after the one that caused an error.
1502 Most bc(1) implementations use **char** types to calculate the value of **1**
1503 decimal digit at a time, but that can be slow. This bc(1) does something
1506 It uses large integers to calculate more than **1** decimal digit at a time. If
1507 built in a environment where **BC_LONG_BIT** (see the **LIMITS** section) is
1508 **64**, then each integer has **9** decimal digits. If built in an environment
1509 where **BC_LONG_BIT** is **32** then each integer has **4** decimal digits. This
1510 value (the number of decimal digits per large integer) is called
1513 The actual values of **BC_LONG_BIT** and **BC_BASE_DIGS** can be queried with
1514 the **limits** statement.
1516 In addition, this bc(1) uses an even larger integer for overflow checking. This
1517 integer type depends on the value of **BC_LONG_BIT**, but is always at least
1518 twice as large as the integer type used to store digits.
1522 The following are the limits on bc(1):
1526 : The number of bits in the **long** type in the environment where bc(1) was
1527 built. This determines how many decimal digits can be stored in a single
1528 large integer (see the **PERFORMANCE** section).
1532 : The number of decimal digits per large integer (see the **PERFORMANCE**
1533 section). Depends on **BC_LONG_BIT**.
1537 : The max decimal number that each large integer can store (see
1538 **BC_BASE_DIGS**) plus **1**. Depends on **BC_BASE_DIGS**.
1542 : The max number that the overflow type (see the **PERFORMANCE** section) can
1543 hold. Depends on **BC_LONG_BIT**.
1547 : The maximum output base. Set at **BC_BASE_POW**.
1551 : The maximum size of arrays. Set at **SIZE_MAX-1**.
1555 : The maximum **scale**. Set at **BC_OVERFLOW_MAX-1**.
1559 : The maximum length of strings. Set at **BC_OVERFLOW_MAX-1**.
1563 : The maximum length of identifiers. Set at **BC_OVERFLOW_MAX-1**.
1567 : The maximum length of a number (in decimal digits), which includes digits
1568 after the decimal point. Set at **BC_OVERFLOW_MAX-1**.
1570 {{ A H N P HN HP NP HNP }}
1573 : The maximum integer (inclusive) returned by the **rand()** operand. Set at
1574 **2\^BC_LONG_BIT-1**.
1579 : The maximum allowable exponent (positive or negative). Set at
1580 **BC_OVERFLOW_MAX**.
1584 : The maximum number of vars/arrays. Set at **SIZE_MAX-1**.
1586 The actual values can be queried with the **limits** statement.
1588 These limits are meant to be effectively non-existent; the limits are so large
1589 (at least on 64-bit machines) that there should not be any point at which they
1590 become a problem. In fact, memory should be exhausted before these limits should
1593 # ENVIRONMENT VARIABLES
1595 bc(1) recognizes the following environment variables:
1599 : If this variable exists (no matter the contents), bc(1) behaves as if
1600 the **-s** option was given.
1604 : This is another way to give command-line arguments to bc(1). They should be
1605 in the same format as all other command-line arguments. These are always
1606 processed first, so any files given in **BC_ENV_ARGS** will be processed
1607 before arguments and files given on the command-line. This gives the user
1608 the ability to set up "standard" options and files to be used at every
1609 invocation. The most useful thing for such files to contain would be useful
1610 functions that the user might want every time bc(1) runs.
1612 The code that parses **BC_ENV_ARGS** will correctly handle quoted arguments,
1613 but it does not understand escape sequences. For example, the string
1614 **"/home/gavin/some bc file.bc"** will be correctly parsed, but the string
1615 **"/home/gavin/some \"bc\" file.bc"** will include the backslashes.
1617 The quote parsing will handle either kind of quotes, **'** or **"**. Thus,
1618 if you have a file with any number of single quotes in the name, you can use
1619 double quotes as the outside quotes, as in **"some 'bc' file.bc"**, and vice
1620 versa if you have a file with double quotes. However, handling a file with
1621 both kinds of quotes in **BC_ENV_ARGS** is not supported due to the
1622 complexity of the parsing, though such files are still supported on the
1623 command-line where the parsing is done by the shell.
1627 : If this environment variable exists and contains an integer that is greater
1628 than **1** and is less than **UINT16_MAX** (**2\^16-1**), bc(1) will output
1629 lines to that length, including the backslash (**\\**). The default line
1634 bc(1) returns the following exit statuses:
1642 : A math error occurred. This follows standard practice of using **1** for
1643 expected errors, since math errors will happen in the process of normal
1646 {{ A H N P HN HP NP HNP }}
1647 Math errors include divide by **0**, taking the square root of a negative
1648 number, using a negative number as a bound for the pseudo-random number
1649 generator, attempting to convert a negative number to a hardware integer,
1650 overflow when converting a number to a hardware integer, and attempting to
1651 use a non-integer where an integer is required.
1653 Converting to a hardware integer happens for the second operand of the power
1654 (**\^**), places (**\@**), left shift (**\<\<**), and right shift (**\>\>**)
1655 operators and their corresponding assignment operators.
1657 {{ E EH EN EP EHN EHP ENP EHNP }}
1658 Math errors include divide by **0**, taking the square root of a negative
1659 number, attempting to convert a negative number to a hardware integer,
1660 overflow when converting a number to a hardware integer, and attempting to
1661 use a non-integer where an integer is required.
1663 Converting to a hardware integer happens for the second operand of the power
1664 (**\^**) operator and the corresponding assignment operator.
1669 : A parse error occurred.
1671 Parse errors include unexpected **EOF**, using an invalid character, failing
1672 to find the end of a string or comment, using a token where it is invalid,
1673 giving an invalid expression, giving an invalid print statement, giving an
1674 invalid function definition, attempting to assign to an expression that is
1675 not a named expression (see the *Named Expressions* subsection of the
1676 **SYNTAX** section), giving an invalid **auto** list, having a duplicate
1677 **auto**/function parameter, failing to find the end of a code block,
1678 attempting to return a value from a **void** function, attempting to use a
1679 variable as a reference, and using any extensions when the option **-s** or
1680 any equivalents were given.
1684 : A runtime error occurred.
1686 Runtime errors include assigning an invalid number to **ibase**, **obase**,
1687 or **scale**; give a bad expression to a **read()** call, calling **read()**
1688 inside of a **read()** call, type errors, passing the wrong number of
1689 arguments to functions, attempting to call an undefined function, and
1690 attempting to use a **void** function call as a value in an expression.
1694 : A fatal error occurred.
1696 Fatal errors include memory allocation errors, I/O errors, failing to open
1697 files, attempting to use files that do not have only ASCII characters (bc(1)
1698 only accepts ASCII characters), attempting to open a directory as a file,
1699 and giving invalid command-line options.
1701 The exit status **4** is special; when a fatal error occurs, bc(1) always exits
1702 and returns **4**, no matter what mode bc(1) is in.
1704 The other statuses will only be returned when bc(1) is not in interactive mode
1705 (see the **INTERACTIVE MODE** section), since bc(1) resets its state (see the
1706 **RESET** section) and accepts more input when one of those errors occurs in
1707 interactive mode. This is also the case when interactive mode is forced by the
1708 **-i** flag or **--interactive** option.
1710 These exit statuses allow bc(1) to be used in shell scripting with error
1711 checking, and its normal behavior can be forced by using the **-i** flag or
1712 **--interactive** option.
1716 Per the [standard][1], bc(1) has an interactive mode and a non-interactive mode.
1717 Interactive mode is turned on automatically when both **stdin** and **stdout**
1718 are hooked to a terminal, but the **-i** flag and **--interactive** option can
1719 turn it on in other cases.
1721 In interactive mode, bc(1) attempts to recover from errors (see the **RESET**
1722 section), and in normal execution, flushes **stdout** as soon as execution is
1723 done for the current input.
1727 If **stdin**, **stdout**, and **stderr** are all connected to a TTY, bc(1) turns
1730 {{ A E N P EN EP NP ENP }}
1731 TTY mode is required for history to be enabled (see the **COMMAND LINE HISTORY**
1732 section). It is also required to enable special handling for **SIGINT** signals.
1735 {{ A E H N EH EN HN EHN }}
1736 The prompt is enabled in TTY mode.
1739 TTY mode is different from interactive mode because interactive mode is required
1740 in the [bc(1) specification][1], and interactive mode requires only **stdin**
1741 and **stdout** to be connected to a terminal.
1745 Sending a **SIGINT** will cause bc(1) to stop execution of the current input. If
1746 bc(1) is in TTY mode (see the **TTY MODE** section), it will reset (see the
1747 **RESET** section). Otherwise, it will clean up and exit.
1749 Note that "current input" can mean one of two things. If bc(1) is processing
1750 input from **stdin** in TTY mode, it will ask for more input. If bc(1) is
1751 processing input from a file in TTY mode, it will stop processing the file and
1752 start processing the next file, if one exists, or ask for input from **stdin**
1753 if no other file exists.
1755 This means that if a **SIGINT** is sent to bc(1) as it is executing a file, it
1756 can seem as though bc(1) did not respond to the signal since it will immediately
1757 start executing the next file. This is by design; most files that users execute
1758 when interacting with bc(1) have function definitions, which are quick to parse.
1759 If a file takes a long time to execute, there may be a bug in that file. The
1760 rest of the files could still be executed without problem, allowing the user to
1763 **SIGTERM** and **SIGQUIT** cause bc(1) to clean up and exit, and it uses the
1764 {{ A E N P EN EP NP ENP }}
1765 default handler for all other signals. The one exception is **SIGHUP**; in that
1766 case, when bc(1) is in TTY mode, a **SIGHUP** will cause bc(1) to clean up and
1769 {{ H EH HN HP EHN EHP HNP EHNP }}
1770 default handler for all other signals.
1773 {{ A E N P EN EP NP ENP }}
1774 # COMMAND LINE HISTORY
1776 bc(1) supports interactive command-line editing. If bc(1) is in TTY mode (see
1777 the **TTY MODE** section), history is enabled. Previous lines can be recalled
1778 and edited with the arrow keys.
1780 **Note**: tabs are converted to 8 spaces.
1783 {{ A E H P EH EP HP EHP }}
1786 This bc(1) ships with support for adding error messages for different locales
1787 and thus, supports **LC_MESSAGES**.
1796 bc(1) is compliant with the [IEEE Std 1003.1-2017 (“POSIX.1-2017”)][1]
1797 specification. The flags **-efghiqsvVw**, all long options, and the extensions
1798 noted above are extensions to that specification.
1800 Note that the specification explicitly says that bc(1) only accepts numbers that
1801 use a period (**.**) as a radix point, regardless of the value of
1804 {{ A E H P EH EP HP EHP }}
1805 This bc(1) supports error messages for different locales, and thus, it supports
1811 None are known. Report bugs at https://git.yzena.com/gavin/bc.
1815 Gavin D. Howard <gavin@yzena.com> and contributors.
1817 [1]: https://pubs.opengroup.org/onlinepubs/9699919799/utilities/bc.html
1818 [2]: https://www.gnu.org/software/bc/
1819 [3]: https://en.wikipedia.org/wiki/Rounding#Round_half_away_from_zero
1820 [4]: https://en.wikipedia.org/wiki/Unit_in_the_last_place
1821 [5]: https://people.eecs.berkeley.edu/~wkahan/LOG10HAF.TXT
1822 [6]: https://en.wikipedia.org/wiki/Rounding#Rounding_away_from_zero