• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1<!---
2
3SPDX-License-Identifier: BSD-2-Clause
4
5Copyright (c) 2018-2023 Gavin D. Howard and contributors.
6
7Redistribution and use in source and binary forms, with or without
8modification, are permitted provided that the following conditions are met:
9
10* Redistributions of source code must retain the above copyright notice, this
11  list of conditions and the following disclaimer.
12
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.
16
17THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
18AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
21LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
22CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
23SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
24INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
25CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
26ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
27POSSIBILITY OF SUCH DAMAGE.
28
29-->
30
31# NAME
32
33bc - arbitrary-precision decimal arithmetic language and calculator
34
35# SYNOPSIS
36
37{{ A H N HN }}
38**bc** [**-cCghilPqRsvVw**] [**-\-digit-clamp**] [**-\-no-digit-clamp**] [**-\-global-stacks**] [**-\-help**] [**-\-interactive**] [**-\-mathlib**] [**-\-no-prompt**] [**-\-no-read-prompt**] [**-\-quiet**] [**-\-standard**] [**-\-warn**] [**-\-version**] [**-e** *expr*] [**-\-expression**=*expr*...] [**-f** *file*...] [**-\-file**=*file*...] [*file*...] [**-I** *ibase*] [**-\-ibase**=*ibase*] [**-O** *obase*] [**-\-obase**=*obase*] [**-S** *scale*] [**-\-scale**=*scale*] [**-E** *seed*] [**-\-seed**=*seed*]
39{{ end }}
40{{ E EH EN EHN }}
41**bc** [**-cCghilPqRsvVw**] [**-\-digit-clamp**] [**-\-no-digit-clamp**] [**-\-global-stacks**] [**-\-help**] [**-\-interactive**] [**-\-mathlib**] [**-\-no-prompt**] [**-\-no-read-prompt**] [**-\-quiet**] [**-\-standard**] [**-\-warn**] [**-\-version**] [**-e** *expr*] [**-\-expression**=*expr*...] [**-f** *file*...] [**-\-file**=*file*...] [*file*...]
42{{ end }}
43
44# DESCRIPTION
45
46bc(1) is an interactive processor for a language first standardized in 1991 by
47POSIX. (See the **STANDARDS** section.) The language provides unlimited
48precision decimal arithmetic and is somewhat C-like, but there are differences.
49Such differences will be noted in this document.
50
51After parsing and handling options, this bc(1) reads any files given on the
52command line and executes them before reading from **stdin**.
53
54This bc(1) is a drop-in replacement for *any* bc(1), including (and
55especially) the GNU bc(1). It also has many extensions and extra features beyond
56other implementations.
57
58**Note**: If running this bc(1) on *any* script meant for another bc(1) gives a
59parse error, it is probably because a word this bc(1) reserves as a keyword is
60used as the name of a function, variable, or array. To fix that, use the
61command-line option **-r** *keyword*, where *keyword* is the keyword that is
62used as a name in the script. For more information, see the **OPTIONS** section.
63
64If parsing scripts meant for other bc(1) implementations still does not work,
65that is a bug and should be reported. See the **BUGS** section.
66
67# OPTIONS
68
69The following are the options that bc(1) accepts.
70
71**-C**, **-\-no-digit-clamp**
72
73:   Disables clamping of digits greater than or equal to the current **ibase**
74    when parsing numbers.
75
76    This means that the value added to a number from a digit is always that
77    digit's value multiplied by the value of ibase raised to the power of the
78    digit's position, which starts from 0 at the least significant digit.
79
80    If this and/or the **-c** or **-\-digit-clamp** options are given multiple
81    times, the last one given is used.
82
83    This option overrides the **BC_DIGIT_CLAMP** environment variable (see the
84    **ENVIRONMENT VARIABLES** section) and the default, which can be queried
85    with the **-h** or **-\-help** options.
86
87    This is a **non-portable extension**.
88
89**-c**, **-\-digit-clamp**
90
91:   Enables clamping of digits greater than or equal to the current **ibase**
92    when parsing numbers.
93
94    This means that digits that the value added to a number from a digit that is
95    greater than or equal to the ibase is the value of ibase minus 1 all
96    multiplied by the value of ibase raised to the power of the digit's
97    position, which starts from 0 at the least significant digit.
98
99    If this and/or the **-C** or **-\-no-digit-clamp** options are given
100    multiple times, the last one given is used.
101
102    This option overrides the **BC_DIGIT_CLAMP** environment variable (see the
103    **ENVIRONMENT VARIABLES** section) and the default, which can be queried
104    with the **-h** or **-\-help** options.
105
106    This is a **non-portable extension**.
107{{ A H N HN }}
108
109**-E** *seed*, **-\-seed**=*seed*
110
111:   Sets the builtin variable **seed** to the value *seed* assuming that *seed*
112    is in base 10. It is a fatal error if *seed* is not a valid number.
113
114    If multiple instances of this option are given, the last is used.
115
116    This is a **non-portable extension**.
117{{ end }}
118
119**-e** *expr*, **-\-expression**=*expr*
120
121:   Evaluates *expr*. If multiple expressions are given, they are evaluated in
122    order. If files are given as well (see the **-f** and **-\-file** options),
123    the expressions and files are evaluated in the order given. This means that
124    if a file is given before an expression, the file is read in and evaluated
125    first.
126
127    If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
128    see the **ENVIRONMENT VARIABLES** section), then after processing all
129    expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
130    as an argument at least once to **-f** or **-\-file**, whether on the
131    command-line or in **BC_ENV_ARGS**. However, if any other **-e**,
132    **-\-expression**, **-f**, or **-\-file** arguments are given after **-f-**
133    or equivalent is given, bc(1) will give a fatal error and exit.
134
135    This is a **non-portable extension**.
136
137**-f** *file*, **-\-file**=*file*
138
139:   Reads in *file* and evaluates it, line by line, as though it were read
140    through **stdin**. If expressions are also given (see the **-e** and
141    **-\-expression** options), the expressions are evaluated in the order
142    given.
143
144    If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
145    see the **ENVIRONMENT VARIABLES** section), then after processing all
146    expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
147    as an argument at least once to **-f** or **-\-file**. However, if any other
148    **-e**, **-\-expression**, **-f**, or **-\-file** arguments are given after
149    **-f-** or equivalent is given, bc(1) will give a fatal error and exit.
150
151    This is a **non-portable extension**.
152
153**-g**, **-\-global-stacks**
154
155{{ A H N HN }}
156:   Turns the globals **ibase**, **obase**, **scale**, and **seed** into stacks.
157
158    This has the effect that a copy of the current value of all four are pushed
159{{ end }}
160{{ E EH EN EHN }}
161:   Turns the globals **ibase**, **obase**, and **scale** into stacks.
162
163    This has the effect that a copy of the current value of all three are pushed
164{{ end }}
165    onto a stack for every function call, as well as popped when every function
166    returns. This means that functions can assign to any and all of those
167    globals without worrying that the change will affect other functions.
168    Thus, a hypothetical function named **output(x,b)** that simply printed
169    **x** in base **b** could be written like this:
170
171        define void output(x, b) {
172            obase=b
173            x
174        }
175
176    instead of like this:
177
178        define void output(x, b) {
179            auto c
180            c=obase
181            obase=b
182            x
183            obase=c
184        }
185
186    This makes writing functions much easier.
187
188{{ A H N HN }}
189    (**Note**: the function **output(x,b)** exists in the extended math library.
190     See the **LIBRARY** section.)
191
192    However, since using this flag means that functions cannot set **ibase**,
193    **obase**, **scale**, or **seed** globally, functions that are made to do so
194    cannot work anymore. There are two possible use cases for that, and each has
195    a solution.
196{{ end }}
197{{ E EH EN EHN }}
198    However, since using this flag means that functions cannot set **ibase**,
199    **obase**, or **scale** globally, functions that are made to do so cannot
200    work anymore. There are two possible use cases for that, and each has a
201    solution.
202{{ end }}
203
204    First, if a function is called on startup to turn bc(1) into a number
205    converter, it is possible to replace that capability with various shell
206    aliases. Examples:
207
208        alias d2o="bc -e ibase=A -e obase=8"
209        alias h2b="bc -e ibase=G -e obase=2"
210
211{{ A H N HN }}
212    Second, if the purpose of a function is to set **ibase**, **obase**,
213    **scale**, or **seed** globally for any other purpose, it could be split
214    into one to four functions (based on how many globals it sets) and each of
215    those functions could return the desired value for a global.
216
217    For functions that set **seed**, the value assigned to **seed** is not
218    propagated to parent functions. This means that the sequence of
219    pseudo-random numbers that they see will not be the same sequence of
220    pseudo-random numbers that any parent sees. This is only the case once
221    **seed** has been set.
222
223    If a function desires to not affect the sequence of pseudo-random numbers
224    of its parents, but wants to use the same **seed**, it can use the following
225    line:
226
227        seed = seed
228{{ end }}
229{{ E EH EN EHN }}
230    Second, if the purpose of a function is to set **ibase**, **obase**, or
231    **scale** globally for any other purpose, it could be split into one to
232    three functions (based on how many globals it sets) and each of those
233    functions could return the desired value for a global.
234{{ end }}
235
236    If the behavior of this option is desired for every run of bc(1), then users
237    could make sure to define **BC_ENV_ARGS** and include this option (see the
238    **ENVIRONMENT VARIABLES** section for more details).
239
240    If **-s**, **-w**, or any equivalents are used, this option is ignored.
241
242    This is a **non-portable extension**.
243
244**-h**, **-\-help**
245
246:   Prints a usage message and exits.
247
248**-I** *ibase*, **-\-ibase**=*ibase*
249
250:   Sets the builtin variable **ibase** to the value *ibase* assuming that
251    *ibase* is in base 10. It is a fatal error if *ibase* is not a valid number.
252
253    If multiple instances of this option are given, the last is used.
254
255    This is a **non-portable extension**.
256
257**-i**, **-\-interactive**
258
259:   Forces interactive mode. (See the **INTERACTIVE MODE** section.)
260
261    This is a **non-portable extension**.
262
263**-L**, **-\-no-line-length**
264
265:   Disables line length checking and prints numbers without backslashes and
266    newlines. In other words, this option sets **BC_LINE_LENGTH** to **0** (see
267    the **ENVIRONMENT VARIABLES** section).
268
269    This is a **non-portable extension**.
270
271**-l**, **-\-mathlib**
272
273:   Sets **scale** (see the **SYNTAX** section) to **20** and loads the included
274{{ A H N HN }}
275    math library and the extended math library before running any code,
276    including any expressions or files specified on the command line.
277
278    To learn what is in the libraries, see the **LIBRARY** section.
279{{ end }}
280{{ E EH EN EHN }}
281    math library before running any code, including any expressions or files
282    specified on the command line.
283
284    To learn what is in the library, see the **LIBRARY** section.
285{{ end }}
286
287**-O** *obase*, **-\-obase**=*obase*
288
289:   Sets the builtin variable **obase** to the value *obase* assuming that
290    *obase* is in base 10. It is a fatal error if *obase* is not a valid number.
291
292    If multiple instances of this option are given, the last is used.
293
294    This is a **non-portable extension**.
295
296**-P**, **-\-no-prompt**
297
298:   Disables the prompt in TTY mode. (The prompt is only enabled in TTY mode.
299    See the **TTY MODE** section.) This is mostly for those users that do not
300    want a prompt or are not used to having them in bc(1). Most of those users
301    would want to put this option in **BC_ENV_ARGS** (see the
302    **ENVIRONMENT VARIABLES** section).
303
304    These options override the **BC_PROMPT** and **BC_TTY_MODE** environment
305    variables (see the **ENVIRONMENT VARIABLES** section).
306
307    This is a **non-portable extension**.
308
309**-q**, **-\-quiet**
310
311:   This option is for compatibility with the GNU bc(1)
312    (https://www.gnu.org/software/bc/); it is a no-op. Without this option, GNU
313    bc(1) prints a copyright header. This bc(1) only prints the copyright header
314    if one or more of the **-v**, **-V**, or **-\-version** options are given
315    unless the **BC_BANNER** environment variable is set and contains a non-zero
316    integer or if this bc(1) was built with the header displayed by default. If
317    *any* of that is the case, then this option *does* prevent bc(1) from
318    printing the header.
319
320    This is a **non-portable extension**.
321
322**-R**, **-\-no-read-prompt**
323
324:   Disables the read prompt in TTY mode. (The read prompt is only enabled in
325    TTY mode. See the **TTY MODE** section.) This is mostly for those users that
326    do not want a read prompt or are not used to having them in bc(1). Most of
327    those users would want to put this option in **BC_ENV_ARGS** (see the
328    **ENVIRONMENT VARIABLES** section). This option is also useful in hash bang
329    lines of bc(1) scripts that prompt for user input.
330
331    This option does not disable the regular prompt because the read prompt is
332    only used when the **read()** built-in function is called.
333
334    These options *do* override the **BC_PROMPT** and **BC_TTY_MODE**
335    environment variables (see the **ENVIRONMENT VARIABLES** section), but only
336    for the read prompt.
337
338    This is a **non-portable extension**.
339
340**-r** *keyword*, **-\-redefine**=*keyword*
341
342:   Redefines *keyword* in order to allow it to be used as a function, variable,
343    or array name. This is useful when this bc(1) gives parse errors when
344    parsing scripts meant for other bc(1) implementations.
345
346    The keywords this bc(1) allows to be redefined are:
347
348    * **abs**
349    * **asciify**
350    * **continue**
351    * **divmod**
352    * **else**
353    * **halt**
354{{ A H N HN }}
355    * **irand**
356{{ end }}
357    * **last**
358    * **limits**
359    * **maxibase**
360    * **maxobase**
361{{ A H N HN }}
362    * **maxrand**
363{{ end }}
364    * **maxscale**
365    * **modexp**
366    * **print**
367{{ A H N HN }}
368    * **rand**
369{{ end }}
370    * **read**
371{{ A H N HN }}
372    * **seed**
373{{ end }}
374	* **stream**
375
376    If any of those keywords are used as a function, variable, or array name in
377    a script, use this option with the keyword as the argument. If multiple are
378    used, use this option for all of them; it can be used multiple times.
379
380    Keywords are *not* redefined when parsing the builtin math library (see the
381    **LIBRARY** section).
382
383    It is a fatal error to redefine keywords mandated by the POSIX standard (see
384    the **STANDARDS** section). It is a fatal error to attempt to redefine words
385    that this bc(1) does not reserve as keywords.
386
387**-S** *scale*, **-\-scale**=*scale*
388
389:   Sets the builtin variable **scale** to the value *scale* assuming that
390    *scale* is in base 10. It is a fatal error if *scale* is not a valid number.
391
392    If multiple instances of this option are given, the last is used.
393
394    This is a **non-portable extension**.
395
396**-s**, **-\-standard**
397
398:   Process exactly the language defined by the standard (see the **STANDARDS**
399    section) and error if any extensions are used.
400
401    This is a **non-portable extension**.
402
403**-v**, **-V**, **-\-version**
404
405:   Print the version information (copyright header) and exits.
406
407    This is a **non-portable extension**.
408
409**-w**, **-\-warn**
410
411:   Like **-s** and **-\-standard**, except that warnings (and not errors) are
412    printed for non-standard extensions and execution continues normally.
413
414    This is a **non-portable extension**.
415
416**-z**, **-\-leading-zeroes**
417
418:   Makes bc(1) print all numbers greater than **-1** and less than **1**, and
419    not equal to **0**, with a leading zero.
420
421    This can be set for individual numbers with the **plz(x)**, **plznl(x)**,
422    **pnlz(x)**, and **pnlznl(x)** functions in the extended math library (see
423    the **LIBRARY** section).
424
425    This is a **non-portable extension**.
426
427All long options are **non-portable extensions**.
428
429# STDIN
430
431If no files or expressions are given by the **-f**, **-\-file**, **-e**, or
432**-\-expression** options, then bc(1) reads from **stdin**.
433
434However, there are a few caveats to this.
435
436First, **stdin** is evaluated a line at a time. The only exception to this is if
437the parse cannot complete. That means that starting a string without ending it
438or starting a function, **if** statement, or loop without ending it will also
439cause bc(1) to not execute.
440
441Second, after an **if** statement, bc(1) doesn't know if an **else** statement
442will follow, so it will not execute until it knows there will not be an **else**
443statement.
444
445# STDOUT
446
447Any non-error output is written to **stdout**. In addition, if history (see the
448**HISTORY** section) and the prompt (see the **TTY MODE** section) are enabled,
449both are output to **stdout**.
450
451**Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
452error (see the **EXIT STATUS** section) if it cannot write to **stdout**, so if
453**stdout** is closed, as in **bc <file> >&-**, it will quit with an error. This
454is done so that bc(1) can report problems when **stdout** is redirected to a
455file.
456
457If there are scripts that depend on the behavior of other bc(1) implementations,
458it is recommended that those scripts be changed to redirect **stdout** to
459**/dev/null**.
460
461# STDERR
462
463Any error output is written to **stderr**.
464
465**Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
466error (see the **EXIT STATUS** section) if it cannot write to **stderr**, so if
467**stderr** is closed, as in **bc <file> 2>&-**, it will quit with an error. This
468is done so that bc(1) can exit with an error code when **stderr** is redirected
469to a file.
470
471If there are scripts that depend on the behavior of other bc(1) implementations,
472it is recommended that those scripts be changed to redirect **stderr** to
473**/dev/null**.
474
475# SYNTAX
476
477The syntax for bc(1) programs is mostly C-like, with some differences. This
478bc(1) follows the POSIX standard (see the **STANDARDS** section), which is a
479much more thorough resource for the language this bc(1) accepts. This section is
480meant to be a summary and a listing of all the extensions to the standard.
481
482In the sections below, **E** means expression, **S** means statement, and **I**
483means identifier.
484
485Identifiers (**I**) start with a lowercase letter and can be followed by any
486number (up to **BC_NAME_MAX-1**) of lowercase letters (**a-z**), digits
487(**0-9**), and underscores (**\_**). The regex is **\[a-z\]\[a-z0-9\_\]\***.
488Identifiers with more than one character (letter) are a
489**non-portable extension**.
490
491**ibase** is a global variable determining how to interpret constant numbers. It
492is the "input" base, or the number base used for interpreting input numbers.
493**ibase** is initially **10**. If the **-s** (**-\-standard**) and **-w**
494(**-\-warn**) flags were not given on the command line, the max allowable value
495for **ibase** is **36**. Otherwise, it is **16**. The min allowable value for
496**ibase** is **2**. The max allowable value for **ibase** can be queried in
497bc(1) programs with the **maxibase()** built-in function.
498
499**obase** is a global variable determining how to output results. It is the
500"output" base, or the number base used for outputting numbers. **obase** is
501initially **10**. The max allowable value for **obase** is **BC_BASE_MAX** and
502can be queried in bc(1) programs with the **maxobase()** built-in function. The
503{{ A H N HN }}
504min allowable value for **obase** is **0**. If **obase** is **0**, values are
505output in scientific notation, and if **obase** is **1**, values are output in
506engineering notation. Otherwise, values are output in the specified base.
507
508Outputting in scientific and engineering notations are **non-portable
509extensions**.
510{{ end }}
511{{ E EH EN EHN }}
512min allowable value for **obase** is **2**. Values are output in the specified
513base.
514{{ end }}
515
516The *scale* of an expression is the number of digits in the result of the
517expression right of the decimal point, and **scale** is a global variable that
518sets the precision of any operations, with exceptions. **scale** is initially
519**0**. **scale** cannot be negative. The max allowable value for **scale** is
520**BC_SCALE_MAX** and can be queried in bc(1) programs with the **maxscale()**
521built-in function.
522
523bc(1) has both *global* variables and *local* variables. All *local*
524variables are local to the function; they are parameters or are introduced in
525the **auto** list of a function (see the **FUNCTIONS** section). If a variable
526is accessed which is not a parameter or in the **auto** list, it is assumed to
527be *global*. If a parent function has a *local* variable version of a variable
528that a child function considers *global*, the value of that *global* variable in
529the child function is the value of the variable in the parent function, not the
530value of the actual *global* variable.
531
532All of the above applies to arrays as well.
533
534The value of a statement that is an expression (i.e., any of the named
535expressions or operands) is printed unless the lowest precedence operator is an
536assignment operator *and* the expression is notsurrounded by parentheses.
537
538The value that is printed is also assigned to the special variable **last**. A
539single dot (**.**) may also be used as a synonym for **last**. These are
540**non-portable extensions**.
541
542Either semicolons or newlines may separate statements.
543
544## Comments
545
546There are two kinds of comments:
547
5481.	Block comments are enclosed in **/\*** and **\*/**.
5492.	Line comments go from **#** until, and not including, the next newline. This
550	is a **non-portable extension**.
551
552## Named Expressions
553
554The following are named expressions in bc(1):
555
5561.	Variables: **I**
5572.	Array Elements: **I[E]**
5583.	**ibase**
5594.	**obase**
5605.	**scale**
561{{ A H N HN }}
5626.	**seed**
5637.	**last** or a single dot (**.**)
564
565Numbers 6 and 7 are **non-portable extensions**.
566
567The meaning of **seed** is dependent on the current pseudo-random number
568generator but is guaranteed to not change except for new major versions.
569
570The *scale* and sign of the value may be significant.
571
572If a previously used **seed** value is assigned to **seed** and used again, the
573pseudo-random number generator is guaranteed to produce the same sequence of
574pseudo-random numbers as it did when the **seed** value was previously used.
575
576The exact value assigned to **seed** is not guaranteed to be returned if
577**seed** is queried again immediately. However, if **seed** *does* return a
578different value, both values, when assigned to **seed**, are guaranteed to
579produce the same sequence of pseudo-random numbers. This means that certain
580values assigned to **seed** will *not* produce unique sequences of pseudo-random
581numbers. The value of **seed** will change after any use of the **rand()** and
582**irand(E)** operands (see the *Operands* subsection below), except if the
583parameter passed to **irand(E)** is **0**, **1**, or negative.
584
585There is no limit to the length (number of significant decimal digits) or
586*scale* of the value that can be assigned to **seed**.
587{{ end }}
588{{ E EH EN EHN }}
5896.	**last** or a single dot (**.**)
590
591Number 6 is a **non-portable extension**.
592{{ end }}
593
594Variables and arrays do not interfere; users can have arrays named the same as
595variables. This also applies to functions (see the **FUNCTIONS** section), so a
596user can have a variable, array, and function that all have the same name, and
597they will not shadow each other, whether inside of functions or not.
598
599Named expressions are required as the operand of **increment**/**decrement**
600operators  and as the left side of **assignment** operators (see the *Operators*
601subsection).
602
603## Operands
604
605The following are valid operands in bc(1):
606
6071.	Numbers (see the *Numbers* subsection below).
6082.	Array indices (**I[E]**).
6093.	**(E)**: The value of **E** (used to change precedence).
6104.	**sqrt(E)**: The square root of **E**. **E** must be non-negative.
6115.	**length(E)**: The number of significant decimal digits in **E**. Returns
612	**1** for **0** with no decimal places. If given a string, the length of the
613	string is returned. Passing a string to **length(E)** is a **non-portable
614	extension**.
6156.	**length(I[])**: The number of elements in the array **I**. This is a
616	**non-portable extension**.
6177.	**scale(E)**: The *scale* of **E**.
6188.	**abs(E)**: The absolute value of **E**. This is a **non-portable
619	extension**.
6209.	**is_number(E)**: **1** if the given argument is a number, **0** if it is a
621	string. This is a **non-portable extension**.
62210.	**is_string(E)**: **1** if the given argument is a string, **0** if it is a
623	number. This is a **non-portable extension**.
62411.	**modexp(E, E, E)**: Modular exponentiation, where the first expression is
625	the base, the second is the exponent, and the third is the modulus. All
626	three values must be integers. The second argument must be non-negative. The
627	third argument must be non-zero. This is a **non-portable extension**.
62811.	**divmod(E, E, I[])**: Division and modulus in one operation. This is for
629	optimization. The first expression is the dividend, and the second is the
630	divisor, which must be non-zero. The return value is the quotient, and the
631	modulus is stored in index **0** of the provided array (the last argument).
632	This is a **non-portable extension**.
63312.	**asciify(E)**: If **E** is a string, returns a string that is the first
634	letter of its argument. If it is a number, calculates the number mod **256**
635	and returns that number as a one-character string. This is a **non-portable
636	extension**.
63713.	**asciify(I[])**: A string that is made up of the characters that would
638	result from running **asciify(E)** on each element of the array identified
639	by the argument. This allows creating multi-character strings and storing
640	them. This is a **non-portable extension**.
64114.	**I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
642	a non-**void** function (see the *Void Functions* subsection of the
643	**FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
644	**I[]**, which will automatically be turned into array references (see the
645	*Array References* subsection of the **FUNCTIONS** section) if the
646	corresponding parameter in the function definition is an array reference.
64715.	**read()**: Reads a line from **stdin** and uses that as an expression. The
648	result of that expression is the result of the **read()** operand. This is a
649	**non-portable extension**.
65016.	**maxibase()**: The max allowable **ibase**. This is a **non-portable
651	extension**.
65217.	**maxobase()**: The max allowable **obase**. This is a **non-portable
653	extension**.
65418.	**maxscale()**: The max allowable **scale**. This is a **non-portable
655	extension**.
65619.	**line_length()**: The line length set with **BC_LINE_LENGTH** (see the
657	**ENVIRONMENT VARIABLES** section). This is a **non-portable extension**.
65820.	**global_stacks()**: **0** if global stacks are not enabled with the **-g**
659	or **-\-global-stacks** options, non-zero otherwise. See the **OPTIONS**
660	section. This is a **non-portable extension**.
66121.	**leading_zero()**: **0** if leading zeroes are not enabled with the **-z**
662	or **--leading-zeroes** options, non-zero otherwise. See the **OPTIONS**
663	section. This is a **non-portable extension**.
664{{ A H N HN }}
66522.	**rand()**: A pseudo-random integer between **0** (inclusive) and
666	**BC_RAND_MAX** (inclusive). Using this operand will change the value of
667	**seed**. This is a **non-portable extension**.
66823.	**irand(E)**: A pseudo-random integer between **0** (inclusive) and the
669	value of **E** (exclusive). If **E** is negative or is a non-integer
670	(**E**'s *scale* is not **0**), an error is raised, and bc(1) resets (see
671	the **RESET** section) while **seed** remains unchanged. If **E** is larger
672	than **BC_RAND_MAX**, the higher bound is honored by generating several
673	pseudo-random integers, multiplying them by appropriate powers of
674	**BC_RAND_MAX+1**, and adding them together. Thus, the size of integer that
675	can be generated with this operand is unbounded. Using this operand will
676	change the value of **seed**, unless the value of **E** is **0** or **1**.
677	In that case, **0** is returned, and **seed** is *not* changed. This is a
678	**non-portable extension**.
67924.	**maxrand()**: The max integer returned by **rand()**. This is a
680	**non-portable extension**.
681
682The integers generated by **rand()** and **irand(E)** are guaranteed to be as
683unbiased as possible, subject to the limitations of the pseudo-random number
684generator.
685
686**Note**: The values returned by the pseudo-random number generator with
687**rand()** and **irand(E)** are guaranteed to *NOT* be cryptographically secure.
688This is a consequence of using a seeded pseudo-random number generator. However,
689they *are* guaranteed to be reproducible with identical **seed** values. This
690means that the pseudo-random values from bc(1) should only be used where a
691reproducible stream of pseudo-random numbers is *ESSENTIAL*. In any other case,
692use a non-seeded pseudo-random number generator.
693{{ end }}
694
695## Numbers
696
697Numbers are strings made up of digits, uppercase letters, and at most **1**
698period for a radix. Numbers can have up to **BC_NUM_MAX** digits. Uppercase
699letters are equal to **9** plus their position in the alphabet, starting from
700**1** (i.e., **A** equals **10**, or **9+1**).
701
702If a digit or letter makes no sense with the current value of **ibase** (i.e.,
703they are greater than or equal to the current value of **ibase**), then the
704behavior depends on the existence of the **-c**/**-\-digit-clamp** or
705**-C**/**-\-no-digit-clamp** options (see the **OPTIONS** section), the
706existence and setting of the **BC_DIGIT_CLAMP** environment variable (see the
707**ENVIRONMENT VARIABLES** section), or the default, which can be queried with
708the **-h**/**-\-help** option.
709
710If clamping is off, then digits or letters that are greater than or equal to the
711current value of **ibase** are not changed. Instead, their given value is
712multiplied by the appropriate power of **ibase** and added into the number. This
713means that, with an **ibase** of **3**, the number **AB** is equal to
714**3\^1\*A+3\^0\*B**, which is **3** times **10** plus **11**, or **41**.
715
716If clamping is on, then digits or letters that are greater than or equal to the
717current value of **ibase** are set to the value of the highest valid digit in
718**ibase** before being multiplied by the appropriate power of **ibase** and
719added into the number. This means that, with an **ibase** of **3**, the number
720**AB** is equal to **3\^1\*2+3\^0\*2**, which is **3** times **2** plus **2**,
721or **8**.
722
723There is one exception to clamping: single-character numbers (i.e., **A**
724alone). Such numbers are never clamped and always take the value they would have
725in the highest possible **ibase**. This means that **A** alone always equals
726decimal **10** and **Z** alone always equals decimal **35**. This behavior is
727mandated by the standard (see the STANDARDS section) and is meant to provide an
728easy way to set the current **ibase** (with the **i** command) regardless of the
729current value of **ibase**.
730
731If clamping is on, and the clamped value of a character is needed, use a leading
732zero, i.e., for **A**, use **0A**.
733
734{{ A H N HN }}
735In addition, bc(1) accepts numbers in scientific notation. These have the form
736**\<number\>e\<integer\>**. The exponent (the portion after the **e**) must be
737an integer. An example is **1.89237e9**, which is equal to **1892370000**.
738Negative exponents are also allowed, so **4.2890e-3** is equal to **0.0042890**.
739
740Using scientific notation is an error or warning if the **-s** or **-w**,
741respectively, command-line options (or equivalents) are given.
742
743**WARNING**: Both the number and the exponent in scientific notation are
744interpreted according to the current **ibase**, but the number is still
745multiplied by **10\^exponent** regardless of the current **ibase**. For example,
746if **ibase** is **16** and bc(1) is given the number string **FFeA**, the
747resulting decimal number will be **2550000000000**, and if bc(1) is given the
748number string **10e-4**, the resulting decimal number will be **0.0016**.
749
750Accepting input as scientific notation is a **non-portable extension**.
751{{ end }}
752
753## Operators
754
755The following arithmetic and logical operators can be used. They are listed in
756order of decreasing precedence. Operators in the same group have the same
757precedence.
758
759**++** **-\-**
760
761:   Type: Prefix and Postfix
762
763    Associativity: None
764
765    Description: **increment**, **decrement**
766
767**-** **!**
768
769:   Type: Prefix
770
771    Associativity: None
772
773    Description: **negation**, **boolean not**
774
775{{ A H N HN }}
776**\$**
777
778:   Type: Postfix
779
780    Associativity: None
781
782    Description: **truncation**
783
784**\@**
785
786:   Type: Binary
787
788    Associativity: Right
789
790    Description: **set precision**
791{{ end }}
792
793**\^**
794
795:   Type: Binary
796
797    Associativity: Right
798
799    Description: **power**
800
801**\*** **/** **%**
802
803:   Type: Binary
804
805    Associativity: Left
806
807    Description: **multiply**, **divide**, **modulus**
808
809**+** **-**
810
811:   Type: Binary
812
813    Associativity: Left
814
815    Description: **add**, **subtract**
816
817{{ A H N HN }}
818**\<\<** **\>\>**
819
820:   Type: Binary
821
822    Associativity: Left
823
824    Description: **shift left**, **shift right**
825
826**=** **\<\<=** **\>\>=** **+=** **-=** **\*=** **/=** **%=** **\^=** **\@=**
827{{ end }}
828{{ E EH EN EHN }}
829**=** **+=** **-=** **\*=** **/=** **%=** **\^=**
830{{ end }}
831
832:   Type: Binary
833
834    Associativity: Right
835
836    Description: **assignment**
837
838**==** **\<=** **\>=** **!=** **\<** **\>**
839
840:   Type: Binary
841
842    Associativity: Left
843
844    Description: **relational**
845
846**&&**
847
848:   Type: Binary
849
850    Associativity: Left
851
852    Description: **boolean and**
853
854**||**
855
856:   Type: Binary
857
858    Associativity: Left
859
860    Description: **boolean or**
861
862The operators will be described in more detail below.
863
864**++** **-\-**
865
866:   The prefix and postfix **increment** and **decrement** operators behave
867    exactly like they would in C. They require a named expression (see the
868    *Named Expressions* subsection) as an operand.
869
870    The prefix versions of these operators are more efficient; use them where
871    possible.
872
873**-**
874
875:   The **negation** operator returns **0** if a user attempts to negate any
876    expression with the value **0**. Otherwise, a copy of the expression with
877    its sign flipped is returned.
878
879**!**
880
881:   The **boolean not** operator returns **1** if the expression is **0**, or
882    **0** otherwise.
883
884    This is a **non-portable extension**.
885
886{{ A H N HN }}
887**\$**
888
889:   The **truncation** operator returns a copy of the given expression with all
890    of its *scale* removed.
891
892    This is a **non-portable extension**.
893
894**\@**
895
896:   The **set precision** operator takes two expressions and returns a copy of
897    the first with its *scale* equal to the value of the second expression. That
898    could either mean that the number is returned without change (if the
899    *scale* of the first expression matches the value of the second
900    expression), extended (if it is less), or truncated (if it is more).
901
902    The second expression must be an integer (no *scale*) and non-negative.
903
904    This is a **non-portable extension**.
905{{ end }}
906
907**\^**
908
909:   The **power** operator (not the **exclusive or** operator, as it would be in
910    C) takes two expressions and raises the first to the power of the value of
911    the second. The *scale* of the result is equal to **scale**.
912
913    The second expression must be an integer (no *scale*), and if it is
914    negative, the first value must be non-zero.
915
916**\***
917
918:   The **multiply** operator takes two expressions, multiplies them, and
919    returns the product. If **a** is the *scale* of the first expression and
920    **b** is the *scale* of the second expression, the *scale* of the result is
921    equal to **min(a+b,max(scale,a,b))** where **min()** and **max()** return
922    the obvious values.
923
924**/**
925
926:   The **divide** operator takes two expressions, divides them, and returns the
927    quotient. The *scale* of the result shall be the value of **scale**.
928
929    The second expression must be non-zero.
930
931**%**
932
933:   The **modulus** operator takes two expressions, **a** and **b**, and
934    evaluates them by 1) Computing **a/b** to current **scale** and 2) Using the
935    result of step 1 to calculate **a-(a/b)\*b** to *scale*
936    **max(scale+scale(b),scale(a))**.
937
938    The second expression must be non-zero.
939
940**+**
941
942:   The **add** operator takes two expressions, **a** and **b**, and returns the
943    sum, with a *scale* equal to the max of the *scale*s of **a** and **b**.
944
945**-**
946
947:   The **subtract** operator takes two expressions, **a** and **b**, and
948    returns the difference, with a *scale* equal to the max of the *scale*s of
949    **a** and **b**.
950
951{{ A H N HN }}
952**\<\<**
953
954:   The **left shift** operator takes two expressions, **a** and **b**, and
955    returns a copy of the value of **a** with its decimal point moved **b**
956    places to the right.
957
958    The second expression must be an integer (no *scale*) and non-negative.
959
960    This is a **non-portable extension**.
961
962**\>\>**
963
964:   The **right shift** operator takes two expressions, **a** and **b**, and
965    returns a copy of the value of **a** with its decimal point moved **b**
966    places to the left.
967
968    The second expression must be an integer (no *scale*) and non-negative.
969
970    This is a **non-portable extension**.
971{{ end }}
972
973{{ A H N HN }}
974**=** **\<\<=** **\>\>=** **+=** **-=** **\*=** **/=** **%=** **\^=** **\@=**
975{{ end }}
976{{ E EH EN EHN }}
977**=** **+=** **-=** **\*=** **/=** **%=** **\^=**
978{{ end }}
979
980:   The **assignment** operators take two expressions, **a** and **b** where
981    **a** is a named expression (see the *Named Expressions* subsection).
982
983    For **=**, **b** is copied and the result is assigned to **a**. For all
984    others, **a** and **b** are applied as operands to the corresponding
985    arithmetic operator and the result is assigned to **a**.
986
987{{ A H N HN }}
988    The **assignment** operators that correspond to operators that are
989    extensions are themselves **non-portable extensions**.
990{{ end }}
991
992**==** **\<=** **\>=** **!=** **\<** **\>**
993
994:   The **relational** operators compare two expressions, **a** and **b**, and
995    if the relation holds, according to C language semantics, the result is
996    **1**. Otherwise, it is **0**.
997
998    Note that unlike in C, these operators have a lower precedence than the
999    **assignment** operators, which means that **a=b\>c** is interpreted as
1000    **(a=b)\>c**.
1001
1002    Also, unlike the standard (see the **STANDARDS** section) requires, these
1003    operators can appear anywhere any other expressions can be used. This
1004    allowance is a **non-portable extension**.
1005
1006**&&**
1007
1008:   The **boolean and** operator takes two expressions and returns **1** if both
1009    expressions are non-zero, **0** otherwise.
1010
1011    This is *not* a short-circuit operator.
1012
1013    This is a **non-portable extension**.
1014
1015**||**
1016
1017:   The **boolean or** operator takes two expressions and returns **1** if one
1018    of the expressions is non-zero, **0** otherwise.
1019
1020    This is *not* a short-circuit operator.
1021
1022    This is a **non-portable extension**.
1023
1024## Statements
1025
1026The following items are statements:
1027
10281.	**E**
10292.	**{** **S** **;** ... **;** **S** **}**
10303.	**if** **(** **E** **)** **S**
10314.	**if** **(** **E** **)** **S** **else** **S**
10325.	**while** **(** **E** **)** **S**
10336.	**for** **(** **E** **;** **E** **;** **E** **)** **S**
10347.	An empty statement
10358.	**break**
10369.	**continue**
103710.	**quit**
103811.	**halt**
103912.	**limits**
104013.	A string of characters, enclosed in double quotes
104114.	**print** **E** **,** ... **,** **E**
104215.	**stream** **E** **,** ... **,** **E**
104316.	**I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
1044	a **void** function (see the *Void Functions* subsection of the
1045	**FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
1046	**I[]**, which will automatically be turned into array references (see the
1047	*Array References* subsection of the **FUNCTIONS** section) if the
1048	corresponding parameter in the function definition is an array reference.
1049
1050Numbers 4, 9, 11, 12, 14, 15, and 16 are **non-portable extensions**.
1051
1052Also, as a **non-portable extension**, any or all of the expressions in the
1053header of a for loop may be omitted. If the condition (second expression) is
1054omitted, it is assumed to be a constant **1**.
1055
1056The **break** statement causes a loop to stop iterating and resume execution
1057immediately following a loop. This is only allowed in loops.
1058
1059The **continue** statement causes a loop iteration to stop early and returns to
1060the start of the loop, including testing the loop condition. This is only
1061allowed in loops.
1062
1063The **if** **else** statement does the same thing as in C.
1064
1065The **quit** statement causes bc(1) to quit, even if it is on a branch that will
1066not be executed (it is a compile-time command).
1067
1068**Warning**: The behavior of this bc(1) on **quit** is slightly different from
1069other bc(1) implementations. Other bc(1) implementations will exit as soon as
1070they finish parsing the line that a **quit** command is on. This bc(1) will
1071execute any completed and executable statements that occur before the **quit**
1072statement before exiting.
1073
1074In other words, for the bc(1) code below:
1075
1076    for (i = 0; i < 3; ++i) i; quit
1077
1078Other bc(1) implementations will print nothing, and this bc(1) will print **0**,
1079**1**, and **2** on successive lines before exiting.
1080
1081The **halt** statement causes bc(1) to quit, if it is executed. (Unlike **quit**
1082if it is on a branch of an **if** statement that is not executed, bc(1) does not
1083quit.)
1084
1085The **limits** statement prints the limits that this bc(1) is subject to. This
1086is like the **quit** statement in that it is a compile-time command.
1087
1088An expression by itself is evaluated and printed, followed by a newline.
1089
1090{{ A H N HN }}
1091Both scientific notation and engineering notation are available for printing the
1092results of expressions. Scientific notation is activated by assigning **0** to
1093**obase**, and engineering notation is activated by assigning **1** to
1094**obase**. To deactivate them, just assign a different value to **obase**.
1095
1096Scientific notation and engineering notation are disabled if bc(1) is run with
1097either the **-s** or **-w** command-line options (or equivalents).
1098
1099Printing numbers in scientific notation and/or engineering notation is a
1100**non-portable extension**.
1101{{ end }}
1102
1103## Strings
1104
1105If strings appear as a statement by themselves, they are printed without a
1106trailing newline.
1107
1108In addition to appearing as a lone statement by themselves, strings can be
1109assigned to variables and array elements. They can also be passed to functions
1110in variable parameters.
1111
1112If any statement that expects a string is given a variable that had a string
1113assigned to it, the statement acts as though it had received a string.
1114
1115If any math operation is attempted on a string or a variable or array element
1116that has been assigned a string, an error is raised, and bc(1) resets (see the
1117**RESET** section).
1118
1119Assigning strings to variables and array elements and passing them to functions
1120are **non-portable extensions**.
1121
1122## Print Statement
1123
1124The "expressions" in a **print** statement may also be strings. If they are, there
1125are backslash escape sequences that are interpreted specially. What those
1126sequences are, and what they cause to be printed, are shown below:
1127
1128**\\a**:   **\\a**
1129
1130**\\b**:   **\\b**
1131
1132**\\\\**:   **\\**
1133
1134**\\e**:   **\\**
1135
1136**\\f**:   **\\f**
1137
1138**\\n**:   **\\n**
1139
1140**\\q**:   **"**
1141
1142**\\r**:   **\\r**
1143
1144**\\t**:   **\\t**
1145
1146Any other character following a backslash causes the backslash and character to
1147be printed as-is.
1148
1149Any non-string expression in a print statement shall be assigned to **last**,
1150like any other expression that is printed.
1151
1152## Stream Statement
1153
1154The "expressions in a **stream** statement may also be strings.
1155
1156If a **stream** statement is given a string, it prints the string as though the
1157string had appeared as its own statement. In other words, the **stream**
1158statement prints strings normally, without a newline.
1159
1160If a **stream** statement is given a number, a copy of it is truncated and its
1161absolute value is calculated. The result is then printed as though **obase** is
1162**256** and each digit is interpreted as an 8-bit ASCII character, making it a
1163byte stream.
1164
1165## Order of Evaluation
1166
1167All expressions in a statment are evaluated left to right, except as necessary
1168to maintain order of operations. This means, for example, assuming that **i** is
1169equal to **0**, in the expression
1170
1171    a[i++] = i++
1172
1173the first (or 0th) element of **a** is set to **1**, and **i** is equal to **2**
1174at the end of the expression.
1175
1176This includes function arguments. Thus, assuming **i** is equal to **0**, this
1177means that in the expression
1178
1179    x(i++, i++)
1180
1181the first argument passed to **x()** is **0**, and the second argument is **1**,
1182while **i** is equal to **2** before the function starts executing.
1183
1184# FUNCTIONS
1185
1186Function definitions are as follows:
1187
1188```
1189define I(I,...,I){
1190	auto I,...,I
1191	S;...;S
1192	return(E)
1193}
1194```
1195
1196Any **I** in the parameter list or **auto** list may be replaced with **I[]** to
1197make a parameter or **auto** var an array, and any **I** in the parameter list
1198may be replaced with **\*I[]** to make a parameter an array reference. Callers
1199of functions that take array references should not put an asterisk in the call;
1200they must be called with just **I[]** like normal array parameters and will be
1201automatically converted into references.
1202
1203As a **non-portable extension**, the opening brace of a **define** statement may
1204appear on the next line.
1205
1206As a **non-portable extension**, the return statement may also be in one of the
1207following forms:
1208
12091.	**return**
12102.	**return** **(** **)**
12113.	**return** **E**
1212
1213The first two, or not specifying a **return** statement, is equivalent to
1214**return (0)**, unless the function is a **void** function (see the *Void
1215Functions* subsection below).
1216
1217## Void Functions
1218
1219Functions can also be **void** functions, defined as follows:
1220
1221```
1222define void I(I,...,I){
1223	auto I,...,I
1224	S;...;S
1225	return
1226}
1227```
1228
1229They can only be used as standalone expressions, where such an expression would
1230be printed alone, except in a print statement.
1231
1232Void functions can only use the first two **return** statements listed above.
1233They can also omit the return statement entirely.
1234
1235The word "void" is not treated as a keyword; it is still possible to have
1236variables, arrays, and functions named **void**. The word "void" is only
1237treated specially right after the **define** keyword.
1238
1239This is a **non-portable extension**.
1240
1241## Array References
1242
1243For any array in the parameter list, if the array is declared in the form
1244
1245```
1246*I[]
1247```
1248
1249it is a **reference**. Any changes to the array in the function are reflected,
1250when the function returns, to the array that was passed in.
1251
1252Other than this, all function arguments are passed by value.
1253
1254This is a **non-portable extension**.
1255
1256# LIBRARY
1257
1258{{ A H N HN }}
1259All of the functions below, including the functions in the extended math
1260library (see the *Extended Library* subsection below), are available when the
1261**-l** or **-\-mathlib** command-line flags are given, except that the extended
1262math library is not available when the **-s** option, the **-w** option, or
1263equivalents are given.
1264{{ end }}
1265{{ E EH EN EHN }}
1266All of the functions below  are available when the **-l** or **-\-mathlib**
1267command-line flags are given.
1268{{ end }}
1269
1270## Standard Library
1271
1272The standard (see the **STANDARDS** section) defines the following functions for
1273the math library:
1274
1275**s(x)**
1276
1277:   Returns the sine of **x**, which is assumed to be in radians.
1278
1279    This is a transcendental function (see the *Transcendental Functions*
1280    subsection below).
1281
1282**c(x)**
1283
1284:   Returns the cosine of **x**, which is assumed to be in radians.
1285
1286    This is a transcendental function (see the *Transcendental Functions*
1287    subsection below).
1288
1289**a(x)**
1290
1291:   Returns the arctangent of **x**, in radians.
1292
1293    This is a transcendental function (see the *Transcendental Functions*
1294    subsection below).
1295
1296**l(x)**
1297
1298:   Returns the natural logarithm of **x**.
1299
1300    This is a transcendental function (see the *Transcendental Functions*
1301    subsection below).
1302
1303**e(x)**
1304
1305:   Returns the mathematical constant **e** raised to the power of **x**.
1306
1307    This is a transcendental function (see the *Transcendental Functions*
1308    subsection below).
1309
1310**j(x, n)**
1311
1312:   Returns the bessel integer order **n** (truncated) of **x**.
1313
1314    This is a transcendental function (see the *Transcendental Functions*
1315    subsection below).
1316
1317{{ A H N HN }}
1318## Extended Library
1319
1320The extended library is *not* loaded when the **-s**/**-\-standard** or
1321**-w**/**-\-warn** options are given since they are not part of the library
1322defined by the standard (see the **STANDARDS** section).
1323
1324The extended library is a **non-portable extension**.
1325
1326**p(x, y)**
1327
1328:   Calculates **x** to the power of **y**, even if **y** is not an integer, and
1329    returns the result to the current **scale**.
1330
1331    It is an error if **y** is negative and **x** is **0**.
1332
1333    This is a transcendental function (see the *Transcendental Functions*
1334    subsection below).
1335
1336**r(x, p)**
1337
1338:   Returns **x** rounded to **p** decimal places according to the rounding mode
1339    round half away from **0**
1340    (https://en.wikipedia.org/wiki/Rounding#Round_half_away_from_zero).
1341
1342**ceil(x, p)**
1343
1344:   Returns **x** rounded to **p** decimal places according to the rounding mode
1345    round away from **0**
1346    (https://en.wikipedia.org/wiki/Rounding#Rounding_away_from_zero).
1347
1348**f(x)**
1349
1350:   Returns the factorial of the truncated absolute value of **x**.
1351
1352**perm(n, k)**
1353
1354:   Returns the permutation of the truncated absolute value of **n** of the
1355    truncated absolute value of **k**, if **k \<= n**. If not, it returns **0**.
1356
1357**comb(n, k)**
1358
1359:   Returns the combination of the truncated absolute value of **n** of the
1360    truncated absolute value of **k**, if **k \<= n**. If not, it returns **0**.
1361
1362**fib(n)**
1363
1364:   Returns the Fibonacci number of the truncated absolute value of **n**.
1365
1366**l2(x)**
1367
1368:   Returns the logarithm base **2** of **x**.
1369
1370    This is a transcendental function (see the *Transcendental Functions*
1371    subsection below).
1372
1373**l10(x)**
1374
1375:   Returns the logarithm base **10** of **x**.
1376
1377    This is a transcendental function (see the *Transcendental Functions*
1378    subsection below).
1379
1380**log(x, b)**
1381
1382:   Returns the logarithm base **b** of **x**.
1383
1384    This is a transcendental function (see the *Transcendental Functions*
1385    subsection below).
1386
1387**cbrt(x)**
1388
1389:   Returns the cube root of **x**.
1390
1391**root(x, n)**
1392
1393:   Calculates the truncated value of **n**, **r**, and returns the **r**th root
1394    of **x** to the current **scale**.
1395
1396    If **r** is **0** or negative, this raises an error and causes bc(1) to
1397    reset (see the **RESET** section). It also raises an error and causes bc(1)
1398    to reset if **r** is even and **x** is negative.
1399
1400**gcd(a, b)**
1401
1402:   Returns the greatest common divisor (factor) of the truncated absolute value
1403    of **a** and the truncated absolute value of **b**.
1404
1405**lcm(a, b)**
1406
1407:   Returns the least common multiple of the truncated absolute value of **a**
1408    and the truncated absolute value of **b**.
1409
1410**pi(p)**
1411
1412:   Returns **pi** to **p** decimal places.
1413
1414    This is a transcendental function (see the *Transcendental Functions*
1415    subsection below).
1416
1417**t(x)**
1418
1419:   Returns the tangent of **x**, which is assumed to be in radians.
1420
1421    This is a transcendental function (see the *Transcendental Functions*
1422    subsection below).
1423
1424**a2(y, x)**
1425
1426:   Returns the arctangent of **y/x**, in radians. If both **y** and **x** are
1427    equal to **0**, it raises an error and causes bc(1) to reset (see the
1428    **RESET** section). Otherwise, if **x** is greater than **0**, it returns
1429    **a(y/x)**. If **x** is less than **0**, and **y** is greater than or equal
1430    to **0**, it returns **a(y/x)+pi**. If **x** is less than **0**, and **y**
1431    is less than **0**, it returns **a(y/x)-pi**. If **x** is equal to **0**,
1432    and **y** is greater than **0**, it returns **pi/2**. If **x** is equal to
1433    **0**, and **y** is less than **0**, it returns **-pi/2**.
1434
1435    This function is the same as the **atan2()** function in many programming
1436    languages.
1437
1438    This is a transcendental function (see the *Transcendental Functions*
1439    subsection below).
1440
1441**sin(x)**
1442
1443:   Returns the sine of **x**, which is assumed to be in radians.
1444
1445    This is an alias of **s(x)**.
1446
1447    This is a transcendental function (see the *Transcendental Functions*
1448    subsection below).
1449
1450**cos(x)**
1451
1452:   Returns the cosine of **x**, which is assumed to be in radians.
1453
1454    This is an alias of **c(x)**.
1455
1456    This is a transcendental function (see the *Transcendental Functions*
1457    subsection below).
1458
1459**tan(x)**
1460
1461:   Returns the tangent of **x**, which is assumed to be in radians.
1462
1463    If **x** is equal to **1** or **-1**, this raises an error and causes bc(1)
1464    to reset (see the **RESET** section).
1465
1466    This is an alias of **t(x)**.
1467
1468    This is a transcendental function (see the *Transcendental Functions*
1469    subsection below).
1470
1471**atan(x)**
1472
1473:   Returns the arctangent of **x**, in radians.
1474
1475    This is an alias of **a(x)**.
1476
1477    This is a transcendental function (see the *Transcendental Functions*
1478    subsection below).
1479
1480**atan2(y, x)**
1481
1482:   Returns the arctangent of **y/x**, in radians. If both **y** and **x** are
1483    equal to **0**, it raises an error and causes bc(1) to reset (see the
1484    **RESET** section). Otherwise, if **x** is greater than **0**, it returns
1485    **a(y/x)**. If **x** is less than **0**, and **y** is greater than or equal
1486    to **0**, it returns **a(y/x)+pi**. If **x** is less than **0**, and **y**
1487    is less than **0**, it returns **a(y/x)-pi**. If **x** is equal to **0**,
1488    and **y** is greater than **0**, it returns **pi/2**. If **x** is equal to
1489    **0**, and **y** is less than **0**, it returns **-pi/2**.
1490
1491    This function is the same as the **atan2()** function in many programming
1492    languages.
1493
1494    This is an alias of **a2(y, x)**.
1495
1496    This is a transcendental function (see the *Transcendental Functions*
1497    subsection below).
1498
1499**r2d(x)**
1500
1501:   Converts **x** from radians to degrees and returns the result.
1502
1503    This is a transcendental function (see the *Transcendental Functions*
1504    subsection below).
1505
1506**d2r(x)**
1507
1508:   Converts **x** from degrees to radians and returns the result.
1509
1510    This is a transcendental function (see the *Transcendental Functions*
1511    subsection below).
1512
1513**frand(p)**
1514
1515:   Generates a pseudo-random number between **0** (inclusive) and **1**
1516    (exclusive) with the number of decimal digits after the decimal point equal
1517    to the truncated absolute value of **p**. If **p** is not **0**, then
1518    calling this function will change the value of **seed**. If **p** is **0**,
1519    then **0** is returned, and **seed** is *not* changed.
1520
1521**ifrand(i, p)**
1522
1523:   Generates a pseudo-random number that is between **0** (inclusive) and the
1524    truncated absolute value of **i** (exclusive) with the number of decimal
1525    digits after the decimal point equal to the truncated absolute value of
1526    **p**. If the absolute value of **i** is greater than or equal to **2**, and
1527    **p** is not **0**, then calling this function will change the value of
1528    **seed**; otherwise, **0** is returned and **seed** is not changed.
1529
1530**srand(x)**
1531
1532:   Returns **x** with its sign flipped with probability **0.5**. In other
1533    words, it randomizes the sign of **x**.
1534
1535**brand()**
1536
1537:   Returns a random boolean value (either **0** or **1**).
1538
1539**band(a, b)**
1540
1541:   Takes the truncated absolute value of both **a** and **b** and calculates
1542    and returns the result of the bitwise **and** operation between them.
1543
1544    If you want to use signed two's complement arguments, use **s2u(x)** to
1545    convert.
1546
1547**bor(a, b)**
1548
1549:   Takes the truncated absolute value of both **a** and **b** and calculates
1550    and returns the result of the bitwise **or** operation between them.
1551
1552    If you want to use signed two's complement arguments, use **s2u(x)** to
1553    convert.
1554
1555**bxor(a, b)**
1556
1557:   Takes the truncated absolute value of both **a** and **b** and calculates
1558    and returns the result of the bitwise **xor** operation between them.
1559
1560    If you want to use signed two's complement arguments, use **s2u(x)** to
1561    convert.
1562
1563**bshl(a, b)**
1564
1565:   Takes the truncated absolute value of both **a** and **b** and calculates
1566    and returns the result of **a** bit-shifted left by **b** places.
1567
1568    If you want to use signed two's complement arguments, use **s2u(x)** to
1569    convert.
1570
1571**bshr(a, b)**
1572
1573:   Takes the truncated absolute value of both **a** and **b** and calculates
1574    and returns the truncated result of **a** bit-shifted right by **b** places.
1575
1576    If you want to use signed two's complement arguments, use **s2u(x)** to
1577    convert.
1578
1579**bnotn(x, n)**
1580
1581:   Takes the truncated absolute value of **x** and does a bitwise not as though
1582    it has the same number of bytes as the truncated absolute value of **n**.
1583
1584    If you want to a use signed two's complement argument, use **s2u(x)** to
1585    convert.
1586
1587**bnot8(x)**
1588
1589:   Does a bitwise not of the truncated absolute value of **x** as though it has
1590    **8** binary digits (1 unsigned byte).
1591
1592    If you want to a use signed two's complement argument, use **s2u(x)** to
1593    convert.
1594
1595**bnot16(x)**
1596
1597:   Does a bitwise not of the truncated absolute value of **x** as though it has
1598    **16** binary digits (2 unsigned bytes).
1599
1600    If you want to a use signed two's complement argument, use **s2u(x)** to
1601    convert.
1602
1603**bnot32(x)**
1604
1605:   Does a bitwise not of the truncated absolute value of **x** as though it has
1606    **32** binary digits (4 unsigned bytes).
1607
1608    If you want to a use signed two's complement argument, use **s2u(x)** to
1609    convert.
1610
1611**bnot64(x)**
1612
1613:   Does a bitwise not of the truncated absolute value of **x** as though it has
1614    **64** binary digits (8 unsigned bytes).
1615
1616    If you want to a use signed two's complement argument, use **s2u(x)** to
1617    convert.
1618
1619**bnot(x)**
1620
1621:   Does a bitwise not of the truncated absolute value of **x** as though it has
1622    the minimum number of power of two unsigned bytes.
1623
1624    If you want to a use signed two's complement argument, use **s2u(x)** to
1625    convert.
1626
1627**brevn(x, n)**
1628
1629:   Runs a bit reversal on the truncated absolute value of **x** as though it
1630    has the same number of 8-bit bytes as the truncated absolute value of **n**.
1631
1632    If you want to a use signed two's complement argument, use **s2u(x)** to
1633    convert.
1634
1635**brev8(x)**
1636
1637:   Runs a bit reversal on the truncated absolute value of **x** as though it
1638    has 8 binary digits (1 unsigned byte).
1639
1640    If you want to a use signed two's complement argument, use **s2u(x)** to
1641    convert.
1642
1643**brev16(x)**
1644
1645:   Runs a bit reversal on the truncated absolute value of **x** as though it
1646    has 16 binary digits (2 unsigned bytes).
1647
1648    If you want to a use signed two's complement argument, use **s2u(x)** to
1649    convert.
1650
1651**brev32(x)**
1652
1653:   Runs a bit reversal on the truncated absolute value of **x** as though it
1654    has 32 binary digits (4 unsigned bytes).
1655
1656    If you want to a use signed two's complement argument, use **s2u(x)** to
1657    convert.
1658
1659**brev64(x)**
1660
1661:   Runs a bit reversal on the truncated absolute value of **x** as though it
1662    has 64 binary digits (8 unsigned bytes).
1663
1664    If you want to a use signed two's complement argument, use **s2u(x)** to
1665    convert.
1666
1667**brev(x)**
1668
1669:   Runs a bit reversal on the truncated absolute value of **x** as though it
1670    has the minimum number of power of two unsigned bytes.
1671
1672    If you want to a use signed two's complement argument, use **s2u(x)** to
1673    convert.
1674
1675**broln(x, p, n)**
1676
1677:   Does a left bitwise rotatation of the truncated absolute value of **x**, as
1678    though it has the same number of unsigned 8-bit bytes as the truncated
1679    absolute value of **n**, by the number of places equal to the truncated
1680    absolute value of **p** modded by the **2** to the power of the number of
1681    binary digits in **n** 8-bit bytes.
1682
1683    If you want to a use signed two's complement argument, use **s2u(x)** to
1684    convert.
1685
1686**brol8(x, p)**
1687
1688:   Does a left bitwise rotatation of the truncated absolute value of **x**, as
1689    though it has **8** binary digits (**1** unsigned byte), by the number of
1690    places equal to the truncated absolute value of **p** modded by **2** to the
1691    power of **8**.
1692
1693    If you want to a use signed two's complement argument, use **s2u(x)** to
1694    convert.
1695
1696**brol16(x, p)**
1697
1698:   Does a left bitwise rotatation of the truncated absolute value of **x**, as
1699    though it has **16** binary digits (**2** unsigned bytes), by the number of
1700    places equal to the truncated absolute value of **p** modded by **2** to the
1701    power of **16**.
1702
1703    If you want to a use signed two's complement argument, use **s2u(x)** to
1704    convert.
1705
1706**brol32(x, p)**
1707
1708:   Does a left bitwise rotatation of the truncated absolute value of **x**, as
1709    though it has **32** binary digits (**2** unsigned bytes), by the number of
1710    places equal to the truncated absolute value of **p** modded by **2** to the
1711    power of **32**.
1712
1713    If you want to a use signed two's complement argument, use **s2u(x)** to
1714    convert.
1715
1716**brol64(x, p)**
1717
1718:   Does a left bitwise rotatation of the truncated absolute value of **x**, as
1719    though it has **64** binary digits (**2** unsigned bytes), by the number of
1720    places equal to the truncated absolute value of **p** modded by **2** to the
1721    power of **64**.
1722
1723    If you want to a use signed two's complement argument, use **s2u(x)** to
1724    convert.
1725
1726**brol(x, p)**
1727
1728:   Does a left bitwise rotatation of the truncated absolute value of **x**, as
1729    though it has the minimum number of power of two unsigned 8-bit bytes, by
1730    the number of places equal to the truncated absolute value of **p** modded
1731    by 2 to the power of the number of binary digits in the minimum number of
1732    8-bit bytes.
1733
1734    If you want to a use signed two's complement argument, use **s2u(x)** to
1735    convert.
1736
1737**brorn(x, p, n)**
1738
1739:   Does a right bitwise rotatation of the truncated absolute value of **x**, as
1740    though it has the same number of unsigned 8-bit bytes as the truncated
1741    absolute value of **n**, by the number of places equal to the truncated
1742    absolute value of **p** modded by the **2** to the power of the number of
1743    binary digits in **n** 8-bit bytes.
1744
1745    If you want to a use signed two's complement argument, use **s2u(x)** to
1746    convert.
1747
1748**bror8(x, p)**
1749
1750:   Does a right bitwise rotatation of the truncated absolute value of **x**, as
1751    though it has **8** binary digits (**1** unsigned byte), by the number of
1752    places equal to the truncated absolute value of **p** modded by **2** to the
1753    power of **8**.
1754
1755    If you want to a use signed two's complement argument, use **s2u(x)** to
1756    convert.
1757
1758**bror16(x, p)**
1759
1760:   Does a right bitwise rotatation of the truncated absolute value of **x**, as
1761    though it has **16** binary digits (**2** unsigned bytes), by the number of
1762    places equal to the truncated absolute value of **p** modded by **2** to the
1763    power of **16**.
1764
1765    If you want to a use signed two's complement argument, use **s2u(x)** to
1766    convert.
1767
1768**bror32(x, p)**
1769
1770:   Does a right bitwise rotatation of the truncated absolute value of **x**, as
1771    though it has **32** binary digits (**2** unsigned bytes), by the number of
1772    places equal to the truncated absolute value of **p** modded by **2** to the
1773    power of **32**.
1774
1775    If you want to a use signed two's complement argument, use **s2u(x)** to
1776    convert.
1777
1778**bror64(x, p)**
1779
1780:   Does a right bitwise rotatation of the truncated absolute value of **x**, as
1781    though it has **64** binary digits (**2** unsigned bytes), by the number of
1782    places equal to the truncated absolute value of **p** modded by **2** to the
1783    power of **64**.
1784
1785    If you want to a use signed two's complement argument, use **s2u(x)** to
1786    convert.
1787
1788**bror(x, p)**
1789
1790:   Does a right bitwise rotatation of the truncated absolute value of **x**, as
1791    though it has the minimum number of power of two unsigned 8-bit bytes, by
1792    the number of places equal to the truncated absolute value of **p** modded
1793    by 2 to the power of the number of binary digits in the minimum number of
1794    8-bit bytes.
1795
1796    If you want to a use signed two's complement argument, use **s2u(x)** to
1797    convert.
1798
1799**bmodn(x, n)**
1800
1801:   Returns the modulus of the truncated absolute value of **x** by **2** to the
1802    power of the multiplication of the truncated absolute value of **n** and
1803    **8**.
1804
1805    If you want to a use signed two's complement argument, use **s2u(x)** to
1806    convert.
1807
1808**bmod8(x, n)**
1809
1810:   Returns the modulus of the truncated absolute value of **x** by **2** to the
1811    power of **8**.
1812
1813    If you want to a use signed two's complement argument, use **s2u(x)** to
1814    convert.
1815
1816**bmod16(x, n)**
1817
1818:   Returns the modulus of the truncated absolute value of **x** by **2** to the
1819    power of **16**.
1820
1821    If you want to a use signed two's complement argument, use **s2u(x)** to
1822    convert.
1823
1824**bmod32(x, n)**
1825
1826:   Returns the modulus of the truncated absolute value of **x** by **2** to the
1827    power of **32**.
1828
1829    If you want to a use signed two's complement argument, use **s2u(x)** to
1830    convert.
1831
1832**bmod64(x, n)**
1833
1834:   Returns the modulus of the truncated absolute value of **x** by **2** to the
1835    power of **64**.
1836
1837    If you want to a use signed two's complement argument, use **s2u(x)** to
1838    convert.
1839
1840**bunrev(t)**
1841
1842:   Assumes **t** is a bitwise-reversed number with an extra set bit one place
1843    more significant than the real most significant bit (which was the least
1844    significant bit in the original number). This number is reversed and
1845    returned without the extra set bit.
1846
1847    This function is used to implement other bitwise functions; it is not meant
1848    to be used by users, but it can be.
1849
1850**plz(x)**
1851
1852:   If **x** is not equal to **0** and greater that **-1** and less than **1**,
1853    it is printed with a leading zero, regardless of the use of the **-z**
1854    option (see the **OPTIONS** section) and without a trailing newline.
1855
1856    Otherwise, **x** is printed normally, without a trailing newline.
1857
1858**plznl(x)**
1859
1860:   If **x** is not equal to **0** and greater that **-1** and less than **1**,
1861    it is printed with a leading zero, regardless of the use of the **-z**
1862    option (see the **OPTIONS** section) and with a trailing newline.
1863
1864    Otherwise, **x** is printed normally, with a trailing newline.
1865
1866**pnlz(x)**
1867
1868:   If **x** is not equal to **0** and greater that **-1** and less than **1**,
1869    it is printed without a leading zero, regardless of the use of the **-z**
1870    option (see the **OPTIONS** section) and without a trailing newline.
1871
1872    Otherwise, **x** is printed normally, without a trailing newline.
1873
1874**pnlznl(x)**
1875
1876:   If **x** is not equal to **0** and greater that **-1** and less than **1**,
1877    it is printed without a leading zero, regardless of the use of the **-z**
1878    option (see the **OPTIONS** section) and with a trailing newline.
1879
1880    Otherwise, **x** is printed normally, with a trailing newline.
1881
1882**ubytes(x)**
1883
1884:   Returns the numbers of unsigned integer bytes required to hold the truncated
1885    absolute value of **x**.
1886
1887**sbytes(x)**
1888
1889:   Returns the numbers of signed, two's-complement integer bytes required to
1890    hold the truncated value of **x**.
1891
1892**s2u(x)**
1893
1894:   Returns **x** if it is non-negative. If it *is* negative, then it calculates
1895    what **x** would be as a 2's-complement signed integer and returns the
1896    non-negative integer that would have the same representation in binary.
1897
1898**s2un(x,n)**
1899
1900:   Returns **x** if it is non-negative. If it *is* negative, then it calculates
1901    what **x** would be as a 2's-complement signed integer with **n** bytes and
1902    returns the non-negative integer that would have the same representation in
1903    binary. If **x** cannot fit into **n** 2's-complement signed bytes, it is
1904    truncated to fit.
1905
1906**hex(x)**
1907
1908:   Outputs the hexadecimal (base **16**) representation of **x**.
1909
1910    This is a **void** function (see the *Void Functions* subsection of the
1911    **FUNCTIONS** section).
1912
1913**binary(x)**
1914
1915:   Outputs the binary (base **2**) representation of **x**.
1916
1917    This is a **void** function (see the *Void Functions* subsection of the
1918    **FUNCTIONS** section).
1919
1920**output(x, b)**
1921
1922:   Outputs the base **b** representation of **x**.
1923
1924    This is a **void** function (see the *Void Functions* subsection of the
1925    **FUNCTIONS** section).
1926
1927**uint(x)**
1928
1929:   Outputs the representation, in binary and hexadecimal, of **x** as an
1930    unsigned integer in as few power of two bytes as possible. Both outputs are
1931    split into bytes separated by spaces.
1932
1933    If **x** is not an integer or is negative, an error message is printed
1934    instead, but bc(1) is not reset (see the **RESET** section).
1935
1936    This is a **void** function (see the *Void Functions* subsection of the
1937    **FUNCTIONS** section).
1938
1939**int(x)**
1940
1941:   Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1942    two's-complement integer in as few power of two bytes as possible. Both
1943    outputs are split into bytes separated by spaces.
1944
1945    If **x** is not an integer, an error message is printed instead, but bc(1)
1946    is not reset (see the **RESET** section).
1947
1948    This is a **void** function (see the *Void Functions* subsection of the
1949    **FUNCTIONS** section).
1950
1951**uintn(x, n)**
1952
1953:   Outputs the representation, in binary and hexadecimal, of **x** as an
1954    unsigned integer in **n** bytes. Both outputs are split into bytes separated
1955    by spaces.
1956
1957    If **x** is not an integer, is negative, or cannot fit into **n** bytes, an
1958    error message is printed instead, but bc(1) is not reset (see the **RESET**
1959    section).
1960
1961    This is a **void** function (see the *Void Functions* subsection of the
1962    **FUNCTIONS** section).
1963
1964**intn(x, n)**
1965
1966:   Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1967    two's-complement integer in **n** bytes. Both outputs are split into bytes
1968    separated by spaces.
1969
1970    If **x** is not an integer or cannot fit into **n** bytes, an error message
1971    is printed instead, but bc(1) is not reset (see the **RESET** section).
1972
1973    This is a **void** function (see the *Void Functions* subsection of the
1974    **FUNCTIONS** section).
1975
1976**uint8(x)**
1977
1978:   Outputs the representation, in binary and hexadecimal, of **x** as an
1979    unsigned integer in **1** byte. Both outputs are split into bytes separated
1980    by spaces.
1981
1982    If **x** is not an integer, is negative, or cannot fit into **1** byte, an
1983    error message is printed instead, but bc(1) is not reset (see the **RESET**
1984    section).
1985
1986    This is a **void** function (see the *Void Functions* subsection of the
1987    **FUNCTIONS** section).
1988
1989**int8(x)**
1990
1991:   Outputs the representation, in binary and hexadecimal, of **x** as a signed,
1992    two's-complement integer in **1** byte. Both outputs are split into bytes
1993    separated by spaces.
1994
1995    If **x** is not an integer or cannot fit into **1** byte, an error message
1996    is printed instead, but bc(1) is not reset (see the **RESET** section).
1997
1998    This is a **void** function (see the *Void Functions* subsection of the
1999    **FUNCTIONS** section).
2000
2001**uint16(x)**
2002
2003:   Outputs the representation, in binary and hexadecimal, of **x** as an
2004    unsigned integer in **2** bytes. Both outputs are split into bytes separated
2005    by spaces.
2006
2007    If **x** is not an integer, is negative, or cannot fit into **2** bytes, an
2008    error message is printed instead, but bc(1) is not reset (see the **RESET**
2009    section).
2010
2011    This is a **void** function (see the *Void Functions* subsection of the
2012    **FUNCTIONS** section).
2013
2014**int16(x)**
2015
2016:   Outputs the representation, in binary and hexadecimal, of **x** as a signed,
2017    two's-complement integer in **2** bytes. Both outputs are split into bytes
2018    separated by spaces.
2019
2020    If **x** is not an integer or cannot fit into **2** bytes, an error message
2021    is printed instead, but bc(1) is not reset (see the **RESET** section).
2022
2023    This is a **void** function (see the *Void Functions* subsection of the
2024    **FUNCTIONS** section).
2025
2026**uint32(x)**
2027
2028:   Outputs the representation, in binary and hexadecimal, of **x** as an
2029    unsigned integer in **4** bytes. Both outputs are split into bytes separated
2030    by spaces.
2031
2032    If **x** is not an integer, is negative, or cannot fit into **4** bytes, an
2033    error message is printed instead, but bc(1) is not reset (see the **RESET**
2034    section).
2035
2036    This is a **void** function (see the *Void Functions* subsection of the
2037    **FUNCTIONS** section).
2038
2039**int32(x)**
2040
2041:   Outputs the representation, in binary and hexadecimal, of **x** as a signed,
2042    two's-complement integer in **4** bytes. Both outputs are split into bytes
2043    separated by spaces.
2044
2045    If **x** is not an integer or cannot fit into **4** bytes, an error message
2046    is printed instead, but bc(1) is not reset (see the **RESET** section).
2047
2048    This is a **void** function (see the *Void Functions* subsection of the
2049    **FUNCTIONS** section).
2050
2051**uint64(x)**
2052
2053:   Outputs the representation, in binary and hexadecimal, of **x** as an
2054    unsigned integer in **8** bytes. Both outputs are split into bytes separated
2055    by spaces.
2056
2057    If **x** is not an integer, is negative, or cannot fit into **8** bytes, an
2058    error message is printed instead, but bc(1) is not reset (see the **RESET**
2059    section).
2060
2061    This is a **void** function (see the *Void Functions* subsection of the
2062    **FUNCTIONS** section).
2063
2064**int64(x)**
2065
2066:   Outputs the representation, in binary and hexadecimal, of **x** as a signed,
2067    two's-complement integer in **8** bytes. Both outputs are split into bytes
2068    separated by spaces.
2069
2070    If **x** is not an integer or cannot fit into **8** bytes, an error message
2071    is printed instead, but bc(1) is not reset (see the **RESET** section).
2072
2073    This is a **void** function (see the *Void Functions* subsection of the
2074    **FUNCTIONS** section).
2075
2076**hex_uint(x, n)**
2077
2078:   Outputs the representation of the truncated absolute value of **x** as an
2079    unsigned integer in hexadecimal using **n** bytes. Not all of the value will
2080    be output if **n** is too small.
2081
2082    This is a **void** function (see the *Void Functions* subsection of the
2083    **FUNCTIONS** section).
2084
2085**binary_uint(x, n)**
2086
2087:   Outputs the representation of the truncated absolute value of **x** as an
2088    unsigned integer in binary using **n** bytes. Not all of the value will be
2089    output if **n** is too small.
2090
2091    This is a **void** function (see the *Void Functions* subsection of the
2092    **FUNCTIONS** section).
2093
2094**output_uint(x, n)**
2095
2096:   Outputs the representation of the truncated absolute value of **x** as an
2097    unsigned integer in the current **obase** (see the **SYNTAX** section) using
2098    **n** bytes. Not all of the value will be output if **n** is too small.
2099
2100    This is a **void** function (see the *Void Functions* subsection of the
2101    **FUNCTIONS** section).
2102
2103**output_byte(x, i)**
2104
2105:   Outputs byte **i** of the truncated absolute value of **x**, where **0** is
2106    the least significant byte and **number_of_bytes - 1** is the most
2107    significant byte.
2108
2109    This is a **void** function (see the *Void Functions* subsection of the
2110    **FUNCTIONS** section).
2111{{ end }}
2112
2113## Transcendental Functions
2114
2115All transcendental functions can return slightly inaccurate results, up to 1 ULP
2116(https://en.wikipedia.org/wiki/Unit_in_the_last_place). This is unavoidable, and
2117the  article at https://people.eecs.berkeley.edu/~wkahan/LOG10HAF.TXT explains
2118why it is impossible and unnecessary to calculate exact results for the
2119transcendental functions.
2120
2121Because of the possible inaccuracy, I recommend that users call those functions
2122with the precision (**scale**) set to at least 1 higher than is necessary. If
2123exact results are *absolutely* required, users can double the precision
2124(**scale**) and then truncate.
2125
2126The transcendental functions in the standard math library are:
2127
2128* **s(x)**
2129* **c(x)**
2130* **a(x)**
2131* **l(x)**
2132* **e(x)**
2133* **j(x, n)**
2134
2135{{ A H N HN }}
2136The transcendental functions in the extended math library are:
2137
2138* **l2(x)**
2139* **l10(x)**
2140* **log(x, b)**
2141* **pi(p)**
2142* **t(x)**
2143* **a2(y, x)**
2144* **sin(x)**
2145* **cos(x)**
2146* **tan(x)**
2147* **atan(x)**
2148* **atan2(y, x)**
2149* **r2d(x)**
2150* **d2r(x)**
2151{{ end }}
2152
2153# RESET
2154
2155When bc(1) encounters an error or a signal that it has a non-default handler
2156for, it resets. This means that several things happen.
2157
2158First, any functions that are executing are stopped and popped off the stack.
2159The behavior is not unlike that of exceptions in programming languages. Then
2160the execution point is set so that any code waiting to execute (after all
2161functions returned) is skipped.
2162
2163Thus, when bc(1) resets, it skips any remaining code waiting to be executed.
2164Then, if it is interactive mode, and the error was not a fatal error (see the
2165**EXIT STATUS** section), it asks for more input; otherwise, it exits with the
2166appropriate return code.
2167
2168Note that this reset behavior is different from the GNU bc(1), which attempts to
2169start executing the statement right after the one that caused an error.
2170
2171# PERFORMANCE
2172
2173Most bc(1) implementations use **char** types to calculate the value of **1**
2174decimal digit at a time, but that can be slow. This bc(1) does something
2175different.
2176
2177It uses large integers to calculate more than **1** decimal digit at a time. If
2178built in a environment where **BC_LONG_BIT** (see the **LIMITS** section) is
2179**64**, then each integer has **9** decimal digits. If built in an environment
2180where **BC_LONG_BIT** is **32** then each integer has **4** decimal digits. This
2181value (the number of decimal digits per large integer) is called
2182**BC_BASE_DIGS**.
2183
2184The actual values of **BC_LONG_BIT** and **BC_BASE_DIGS** can be queried with
2185the **limits** statement.
2186
2187In addition, this bc(1) uses an even larger integer for overflow checking. This
2188integer type depends on the value of **BC_LONG_BIT**, but is always at least
2189twice as large as the integer type used to store digits.
2190
2191# LIMITS
2192
2193The following are the limits on bc(1):
2194
2195**BC_LONG_BIT**
2196
2197:   The number of bits in the **long** type in the environment where bc(1) was
2198    built. This determines how many decimal digits can be stored in a single
2199    large integer (see the **PERFORMANCE** section).
2200
2201**BC_BASE_DIGS**
2202
2203:   The number of decimal digits per large integer (see the **PERFORMANCE**
2204    section). Depends on **BC_LONG_BIT**.
2205
2206**BC_BASE_POW**
2207
2208:   The max decimal number that each large integer can store (see
2209    **BC_BASE_DIGS**) plus **1**. Depends on **BC_BASE_DIGS**.
2210
2211**BC_OVERFLOW_MAX**
2212
2213:   The max number that the overflow type (see the **PERFORMANCE** section) can
2214    hold. Depends on **BC_LONG_BIT**.
2215
2216**BC_BASE_MAX**
2217
2218:   The maximum output base. Set at **BC_BASE_POW**.
2219
2220**BC_DIM_MAX**
2221
2222:   The maximum size of arrays. Set at **SIZE_MAX-1**.
2223
2224**BC_SCALE_MAX**
2225
2226:   The maximum **scale**. Set at **BC_OVERFLOW_MAX-1**.
2227
2228**BC_STRING_MAX**
2229
2230:   The maximum length of strings. Set at **BC_OVERFLOW_MAX-1**.
2231
2232**BC_NAME_MAX**
2233
2234:   The maximum length of identifiers. Set at **BC_OVERFLOW_MAX-1**.
2235
2236**BC_NUM_MAX**
2237
2238:   The maximum length of a number (in decimal digits), which includes digits
2239    after the decimal point. Set at **BC_OVERFLOW_MAX-1**.
2240
2241{{ A H N HN }}
2242**BC_RAND_MAX**
2243
2244:   The maximum integer (inclusive) returned by the **rand()** operand. Set at
2245    **2\^BC_LONG_BIT-1**.
2246{{ end }}
2247
2248Exponent
2249
2250:   The maximum allowable exponent (positive or negative). Set at
2251    **BC_OVERFLOW_MAX**.
2252
2253Number of vars
2254
2255:   The maximum number of vars/arrays. Set at **SIZE_MAX-1**.
2256
2257The actual values can be queried with the **limits** statement.
2258
2259These limits are meant to be effectively non-existent; the limits are so large
2260(at least on 64-bit machines) that there should not be any point at which they
2261become a problem. In fact, memory should be exhausted before these limits should
2262be hit.
2263
2264# ENVIRONMENT VARIABLES
2265
2266As **non-portable extensions**, bc(1) recognizes the following environment
2267variables:
2268
2269**POSIXLY_CORRECT**
2270
2271:   If this variable exists (no matter the contents), bc(1) behaves as if
2272    the **-s** option was given.
2273
2274**BC_ENV_ARGS**
2275
2276:   This is another way to give command-line arguments to bc(1). They should be
2277    in the same format as all other command-line arguments. These are always
2278    processed first, so any files given in **BC_ENV_ARGS** will be processed
2279    before arguments and files given on the command-line. This gives the user
2280    the ability to set up "standard" options and files to be used at every
2281    invocation. The most useful thing for such files to contain would be useful
2282    functions that the user might want every time bc(1) runs.
2283
2284    The code that parses **BC_ENV_ARGS** will correctly handle quoted arguments,
2285    but it does not understand escape sequences. For example, the string
2286    **"/home/gavin/some bc file.bc"** will be correctly parsed, but the string
2287    **"/home/gavin/some \"bc\" file.bc"** will include the backslashes.
2288
2289    The quote parsing will handle either kind of quotes, **'** or **"**. Thus,
2290    if you have a file with any number of single quotes in the name, you can use
2291    double quotes as the outside quotes, as in **"some 'bc' file.bc"**, and vice
2292    versa if you have a file with double quotes. However, handling a file with
2293    both kinds of quotes in **BC_ENV_ARGS** is not supported due to the
2294    complexity of the parsing, though such files are still supported on the
2295    command-line where the parsing is done by the shell.
2296
2297**BC_LINE_LENGTH**
2298
2299:   If this environment variable exists and contains an integer that is greater
2300    than **1** and is less than **UINT16_MAX** (**2\^16-1**), bc(1) will output
2301    lines to that length, including the backslash (**\\**). The default line
2302    length is **70**.
2303
2304    The special value of **0** will disable line length checking and print
2305    numbers without regard to line length and without backslashes and newlines.
2306
2307**BC_BANNER**
2308
2309:   If this environment variable exists and contains an integer, then a non-zero
2310    value activates the copyright banner when bc(1) is in interactive mode,
2311    while zero deactivates it.
2312
2313    If bc(1) is not in interactive mode (see the **INTERACTIVE MODE** section),
2314    then this environment variable has no effect because bc(1) does not print
2315    the banner when not in interactive mode.
2316
2317    This environment variable overrides the default, which can be queried with
2318    the **-h** or **-\-help** options.
2319
2320**BC_SIGINT_RESET**
2321
2322:   If bc(1) is not in interactive mode (see the **INTERACTIVE MODE** section),
2323    then this environment variable has no effect because bc(1) exits on
2324    **SIGINT** when not in interactive mode.
2325
2326    However, when bc(1) is in interactive mode, then if this environment
2327    variable exists and contains an integer, a non-zero value makes bc(1) reset
2328    on **SIGINT**, rather than exit, and zero makes bc(1) exit. If this
2329    environment variable exists and is *not* an integer, then bc(1) will exit on
2330    **SIGINT**.
2331
2332    This environment variable overrides the default, which can be queried with
2333    the **-h** or **-\-help** options.
2334
2335**BC_TTY_MODE**
2336
2337:   If TTY mode is *not* available (see the **TTY MODE** section), then this
2338    environment variable has no effect.
2339
2340    However, when TTY mode is available, then if this environment variable
2341    exists and contains an integer, then a non-zero value makes bc(1) use TTY
2342    mode, and zero makes bc(1) not use TTY mode.
2343
2344    This environment variable overrides the default, which can be queried with
2345    the **-h** or **-\-help** options.
2346
2347**BC_PROMPT**
2348
2349:   If TTY mode is *not* available (see the **TTY MODE** section), then this
2350    environment variable has no effect.
2351
2352    However, when TTY mode is available, then if this environment variable
2353    exists and contains an integer, a non-zero value makes bc(1) use a prompt,
2354    and zero or a non-integer makes bc(1) not use a prompt. If this environment
2355    variable does not exist and **BC_TTY_MODE** does, then the value of the
2356    **BC_TTY_MODE** environment variable is used.
2357
2358    This environment variable and the **BC_TTY_MODE** environment variable
2359    override the default, which can be queried with the **-h** or **-\-help**
2360    options.
2361
2362**BC_EXPR_EXIT**
2363
2364:   If any expressions or expression files are given on the command-line with
2365    **-e**, **-\-expression**, **-f**, or **-\-file**, then if this environment
2366    variable exists and contains an integer, a non-zero value makes bc(1) exit
2367    after executing the expressions and expression files, and a zero value makes
2368    bc(1) not exit.
2369
2370    This environment variable overrides the default, which can be queried with
2371    the **-h** or **-\-help** options.
2372
2373**BC_DIGIT_CLAMP**
2374
2375:   When parsing numbers and if this environment variable exists and contains an
2376    integer, a non-zero value makes bc(1) clamp digits that are greater than or
2377    equal to the current **ibase** so that all such digits are considered equal
2378    to the **ibase** minus 1, and a zero value disables such clamping so that
2379    those digits are always equal to their value, which is multiplied by the
2380    power of the **ibase**.
2381
2382    This never applies to single-digit numbers, as per the standard (see the
2383    **STANDARDS** section).
2384
2385    This environment variable overrides the default, which can be queried with
2386    the **-h** or **-\-help** options.
2387
2388# EXIT STATUS
2389
2390bc(1) returns the following exit statuses:
2391
2392**0**
2393
2394:   No error.
2395
2396**1**
2397
2398:   A math error occurred. This follows standard practice of using **1** for
2399    expected errors, since math errors will happen in the process of normal
2400    execution.
2401
2402    Math errors include divide by **0**, taking the square root of a negative
2403{{ A H N HN }}
2404    number, using a negative number as a bound for the pseudo-random number
2405    generator, attempting to convert a negative number to a hardware integer,
2406    overflow when converting a number to a hardware integer, overflow when
2407    calculating the size of a number, and attempting to use a non-integer where
2408    an integer is required.
2409
2410    Converting to a hardware integer happens for the second operand of the power
2411    (**\^**), places (**\@**), left shift (**\<\<**), and right shift (**\>\>**)
2412    operators and their corresponding assignment operators.
2413{{ end }}
2414{{ E EH EN EHN }}
2415    number, attempting to convert a negative number to a hardware integer,
2416    overflow when converting a number to a hardware integer, overflow when
2417    calculating the size of a number, and attempting to use a non-integer where
2418    an integer is required.
2419
2420    Converting to a hardware integer happens for the second operand of the power
2421    (**\^**) operator and the corresponding assignment operator.
2422{{ end }}
2423
2424**2**
2425
2426:   A parse error occurred.
2427
2428    Parse errors include unexpected **EOF**, using an invalid character, failing
2429    to find the end of a string or comment, using a token where it is invalid,
2430    giving an invalid expression, giving an invalid print statement, giving an
2431    invalid function definition, attempting to assign to an expression that is
2432    not a named expression (see the *Named Expressions* subsection of the
2433    **SYNTAX** section), giving an invalid **auto** list, having a duplicate
2434    **auto**/function parameter, failing to find the end of a code block,
2435    attempting to return a value from a **void** function, attempting to use a
2436    variable as a reference, and using any extensions when the option **-s** or
2437    any equivalents were given.
2438
2439**3**
2440
2441:   A runtime error occurred.
2442
2443    Runtime errors include assigning an invalid number to any global (**ibase**,
2444    **obase**, or **scale**), giving a bad expression to a **read()** call,
2445    calling **read()** inside of a **read()** call, type errors, passing the
2446    wrong number of arguments to functions, attempting to call an undefined
2447    function, and attempting to use a **void** function call as a value in an
2448    expression.
2449
2450**4**
2451
2452:   A fatal error occurred.
2453
2454    Fatal errors include memory allocation errors, I/O errors, failing to open
2455    files, attempting to use files that do not have only ASCII characters (bc(1)
2456    only accepts ASCII characters), attempting to open a directory as a file,
2457    and giving invalid command-line options.
2458
2459The exit status **4** is special; when a fatal error occurs, bc(1) always exits
2460and returns **4**, no matter what mode bc(1) is in.
2461
2462The other statuses will only be returned when bc(1) is not in interactive mode
2463(see the **INTERACTIVE MODE** section), since bc(1) resets its state (see the
2464**RESET** section) and accepts more input when one of those errors occurs in
2465interactive mode. This is also the case when interactive mode is forced by the
2466**-i** flag or **-\-interactive** option.
2467
2468These exit statuses allow bc(1) to be used in shell scripting with error
2469checking, and its normal behavior can be forced by using the **-i** flag or
2470**-\-interactive** option.
2471
2472# INTERACTIVE MODE
2473
2474Per the standard (see the **STANDARDS** section), bc(1) has an interactive mode
2475and a non-interactive mode. Interactive mode is turned on automatically when
2476both **stdin** and **stdout** are hooked to a terminal, but the **-i** flag and
2477**-\-interactive** option can turn it on in other situations.
2478
2479In interactive mode, bc(1) attempts to recover from errors (see the **RESET**
2480section), and in normal execution, flushes **stdout** as soon as execution is
2481done for the current input. bc(1) may also reset on **SIGINT** instead of exit,
2482depending on the contents of, or default for, the **BC_SIGINT_RESET**
2483environment variable (see the **ENVIRONMENT VARIABLES** section).
2484
2485# TTY MODE
2486
2487If **stdin**, **stdout**, and **stderr** are all connected to a TTY, then "TTY
2488mode" is considered to be available, and thus, bc(1) can turn on TTY mode,
2489subject to some settings.
2490
2491If there is the environment variable **BC_TTY_MODE** in the environment (see the
2492**ENVIRONMENT VARIABLES** section), then if that environment variable contains a
2493non-zero integer, bc(1) will turn on TTY mode when **stdin**, **stdout**, and
2494**stderr** are all connected to a TTY. If the **BC_TTY_MODE** environment
2495variable exists but is *not* a non-zero integer, then bc(1) will not turn TTY
2496mode on.
2497
2498If the environment variable **BC_TTY_MODE** does *not* exist, the default
2499setting is used. The default setting can be queried with the **-h** or
2500**-\-help** options.
2501
2502TTY mode is different from interactive mode because interactive mode is required
2503in the bc(1) standard (see the **STANDARDS** section), and interactive mode
2504requires only **stdin** and **stdout** to be connected to a terminal.
2505
2506{{ A E N EN }}
2507## Command-Line History
2508
2509Command-line history is only enabled if TTY mode is, i.e., that **stdin**,
2510**stdout**, and **stderr** are connected to a TTY and the **BC_TTY_MODE**
2511environment variable (see the **ENVIRONMENT VARIABLES** section) and its default
2512do not disable TTY mode. See the **COMMAND LINE HISTORY** section for more
2513information.
2514{{ end }}
2515
2516## Prompt
2517
2518If TTY mode is available, then a prompt can be enabled. Like TTY mode itself, it
2519can be turned on or off with an environment variable: **BC_PROMPT** (see the
2520**ENVIRONMENT VARIABLES** section).
2521
2522If the environment variable **BC_PROMPT** exists and is a non-zero integer, then
2523the prompt is turned on when **stdin**, **stdout**, and **stderr** are connected
2524to a TTY and the **-P** and **-\-no-prompt** options were not used. The read
2525prompt will be turned on under the same conditions, except that the **-R** and
2526**-\-no-read-prompt** options must also not be used.
2527
2528However, if **BC_PROMPT** does not exist, the prompt can be enabled or disabled
2529with the **BC_TTY_MODE** environment variable, the **-P** and **-\-no-prompt**
2530options, and the **-R** and **-\-no-read-prompt** options. See the **ENVIRONMENT
2531VARIABLES** and **OPTIONS** sections for more details.
2532
2533# SIGNAL HANDLING
2534
2535Sending a **SIGINT** will cause bc(1) to do one of two things.
2536
2537If bc(1) is not in interactive mode (see the **INTERACTIVE MODE** section), or
2538the **BC_SIGINT_RESET** environment variable (see the **ENVIRONMENT VARIABLES**
2539section), or its default, is either not an integer or it is zero, bc(1) will
2540exit.
2541
2542However, if bc(1) is in interactive mode, and the **BC_SIGINT_RESET** or its
2543default is an integer and non-zero, then bc(1) will stop executing the current
2544input and reset (see the **RESET** section) upon receiving a **SIGINT**.
2545
2546Note that "current input" can mean one of two things. If bc(1) is processing
2547input from **stdin** in interactive mode, it will ask for more input. If bc(1)
2548is processing input from a file in interactive mode, it will stop processing the
2549file and start processing the next file, if one exists, or ask for input from
2550**stdin** if no other file exists.
2551
2552This means that if a **SIGINT** is sent to bc(1) as it is executing a file, it
2553can seem as though bc(1) did not respond to the signal since it will immediately
2554start executing the next file. This is by design; most files that users execute
2555when interacting with bc(1) have function definitions, which are quick to parse.
2556If a file takes a long time to execute, there may be a bug in that file. The
2557rest of the files could still be executed without problem, allowing the user to
2558continue.
2559
2560**SIGTERM** and **SIGQUIT** cause bc(1) to clean up and exit, and it uses the
2561{{ A E N EN }}
2562default handler for all other signals. The one exception is **SIGHUP**; in that
2563case, and only when bc(1) is in TTY mode (see the **TTY MODE** section), a
2564**SIGHUP** will cause bc(1) to clean up and exit.
2565{{ end }}
2566{{ H EH HN EHN }}
2567default handler for all other signals.
2568{{ end }}
2569
2570{{ A E N EN }}
2571# COMMAND LINE HISTORY
2572
2573bc(1) supports interactive command-line editing.
2574
2575If bc(1) can be in TTY mode (see the **TTY MODE** section), history can be
2576enabled. This means that command-line history can only be enabled when
2577**stdin**, **stdout**, and **stderr** are all connected to a TTY.
2578
2579Like TTY mode itself, it can be turned on or off with the environment variable
2580**BC_TTY_MODE** (see the **ENVIRONMENT VARIABLES** section).
2581
2582If history is enabled, previous lines can be recalled and edited with the arrow
2583keys.
2584
2585**Note**: tabs are converted to 8 spaces.
2586{{ end }}
2587
2588{{ A E H EH }}
2589# LOCALES
2590
2591This bc(1) ships with support for adding error messages for different locales
2592and thus, supports **LC_MESSAGES**.
2593{{ end }}
2594
2595# SEE ALSO
2596
2597dc(1)
2598
2599# STANDARDS
2600
2601bc(1) is compliant with the IEEE Std 1003.1-2017 (“POSIX.1-2017”) specification
2602at https://pubs.opengroup.org/onlinepubs/9699919799/utilities/bc.html . The
2603flags **-efghiqsvVw**, all long options, and the extensions noted above are
2604extensions to that specification.
2605
2606In addition, the behavior of the **quit** implements an interpretation of that
2607specification that is different from all known implementations. For more
2608information see the **Statements** subsection of the **SYNTAX** section.
2609
2610Note that the specification explicitly says that bc(1) only accepts numbers that
2611use a period (**.**) as a radix point, regardless of the value of
2612**LC_NUMERIC**.
2613
2614{{ A E H EH }}
2615This bc(1) supports error messages for different locales, and thus, it supports
2616**LC_MESSAGES**.
2617{{ end }}
2618
2619# BUGS
2620
2621Before version **6.1.0**, this bc(1) had incorrect behavior for the **quit**
2622statement.
2623
2624No other bugs are known. Report bugs at https://git.gavinhoward.com/gavin/bc .
2625
2626# AUTHORS
2627
2628Gavin D. Howard <gavin@gavinhoward.com> and contributors.
2629