1 <html><head><title>toybox source code walkthrough</title></head> 2 <!--#include file="header.html" --> 3 4 <p><h1><a name="style" /><a href="#style">Code style</a></h1></p> 5 6 <p>The primary goal of toybox is _simple_ code. Keeping the code small is 7 second, with speed and lots of features coming in somewhere after that. 8 (For more on that, see the <a href=design.html>design</a> page.)</p> 9 10 <p>A simple implementation usually takes up fewer lines of source code, 11 meaning more code can fit on the screen at once, meaning the programmer can 12 see more of it on the screen and thus keep more if in their head at once. 13 This helps code auditing and thus reduces bugs. That said, sometimes being 14 more explicit is preferable to being clever enough to outsmart yourself: 15 don't be so terse your code is unreadable.</p> 16 17 <p>Toybox has an actual coding style guide over on 18 <a href=design.html#codestyle>the design page</a>, but in general we just 19 want the code to be consistent.</p> 20 21 <p><h1><a name="building" /><a href="#building">Building Toybox</a></h1></p> 22 23 <p>Toybox is configured using the Kconfig language pioneered by the Linux 24 kernel, and adopted by many other projects (uClibc, OpenEmbedded, etc). 25 This generates a ".config" file containing the selected options, which 26 controls which features are included when compiling toybox.</p> 27 28 <p>Each configuration option has a default value. The defaults indicate the 29 "maximum sane configuration", I.E. if the feature defaults to "n" then it 30 either isn't complete or is a special-purpose option (such as debugging 31 code) that isn't intended for general purpose use.</p> 32 33 <p>The standard build invocation is:</p> 34 35 <ul> 36 <li>make defconfig #(or menuconfig)</li> 37 <li>make</li> 38 <li>make install</li> 39 </ul> 40 41 <p>Type "make help" to see all available build options.</p> 42 43 <p>The file "configure" contains a number of environment variable definitions 44 which influence the build, such as specifying which compiler to use or where 45 to install the resulting binaries. This file is included by the build, but 46 accepts existing definitions of the environment variables, so it may be sourced 47 or modified by the developer before building and the definitions exported 48 to the environment will take precedence.</p> 49 50 <p>(To clarify: "configure" describes the build and installation environment, 51 ".config" lists the features selected by defconfig/menuconfig.)</p> 52 53 <p><h1><a name="running"><a href="#running">Running a command</a></h1></p> 54 55 <h2>main</h2> 56 57 <p>The toybox main() function is at the end of main.c at the top level. It has 58 two possible codepaths, only one of which is configured into any given build 59 of toybox.</p> 60 61 <p>If CONFIG_SINGLE is selected, toybox is configured to contain only a single 62 command, so most of the normal setup can be skipped. In this case the 63 multiplexer isn't used, instead main() calls toy_singleinit() (also in main.c) 64 to set up global state and parse command line arguments, calls the command's 65 main function out of toy_list (in the CONFIG_SINGLE case the array has a single entry, no need to search), and if the function returns instead of exiting 66 it flushes stdout (detecting error) and returns toys.exitval.</p> 67 68 <p>When CONFIG_SINGLE is not selected, main() uses basename() to find the 69 name it was run as, shifts its argument list one to the right so it lines up 70 with where the multiplexer function expects it, and calls toybox_main(). This 71 leverages the multiplexer command's infrastructure to find and run the 72 appropriate command. (A command name starting with "toybox" will 73 recursively call toybox_main(); you can go "./toybox toybox toybox toybox ls" 74 if you want to...)</p> 75 76 <h2>toybox_main</h2> 77 78 <p>The toybox_main() function is also in main,c. It handles a possible 79 --help option ("toybox --help ls"), prints the list of available commands if no 80 arguments were provided to the multiplexer (or with full path names if any 81 other option is provided before a command name, ala "toybox --list"). 82 Otherwise it calls toy_exec() on its argument list.</p> 83 84 <p>Note that the multiplexer is the first entry in toy_list (the rest of the 85 list is sorted alphabetically to allow binary search), so toybox_main can 86 cheat and just grab the first entry to quickly set up its context without 87 searching. Since all command names go through the multiplexer at least once 88 in the non-TOYBOX_SINGLE case, this avoids a redundant search of 89 the list.</p> 90 91 <p>The toy_exec() function is also in main.c. It performs toy_find() to 92 perform a binary search on the toy_list array to look up the command's 93 entry by name and saves it in the global variable which, calls toy_init() 94 to parse command line arguments and set up global state (using which->options), 95 and calls the appropriate command's main() function (which->toy_main). On 96 return it flushes all pending ansi FILE * I/O, detects if stdout had an 97 error, and then calls xexit() (which uses toys.exitval).</p> 98 99 <p><h1><a name="infrastructure" /><a href="#infrastructure">Infrastructure</a></h1></p> 100 101 <p>The toybox source code is in following directories:</p> 102 <ul> 103 <li>The <a href="#top">top level directory</a> contains the file main.c (were 104 execution starts), the header file toys.h (included by every command), and 105 other global infrastructure.</li> 106 <li>The <a href="#lib">lib directory</a> contains common functions shared by 107 multiple commands:</li> 108 <ul> 109 <li><a href="#lib_lib">lib/lib.c</a></li> 110 <li><a href="#lib_xwrap">lib/xwrap.c</a></li> 111 <li><a href="#lib_llist">lib/llist.c</a></li> 112 <li><a href="#lib_args">lib/args.c</a></li> 113 <li><a href="#lib_dirtree">lib/dirtree.c</a></li> 114 </ul> 115 <li>The <a href="#toys">toys directory</a> contains the C files implementating 116 each command. Currently it contains five subdirectories categorizing the 117 commands: posix, lsb, other, example, and pending.</li> 118 <li>The <a href="#scripts">scripts directory</a> contains the build and 119 test infrastructure.</li> 120 <li>The <a href="#kconfig">kconfig directory</a> contains the configuration 121 infrastructure implementing menuconfig (copied from the Linux kernel).</li> 122 <li>The <a href="#generated">generated directory</a> contains intermediate 123 files generated from other parts of the source code.</li> 124 </ul> 125 126 <a name="adding" /> 127 <p><h1><a href="#adding">Adding a new command</a></h1></p> 128 <p>To add a new command to toybox, add a C file implementing that command to 129 one of the subdirectories under the toys directory. No other files need to 130 be modified; the build extracts all the information it needs (such as command 131 line arguments) from specially formatted comments and macros in the C file. 132 (See the description of the <a href="#generated">"generated" directory</a> 133 for details.)</p> 134 135 <p>Currently there are five subdirectories under "toys", one for commands 136 defined by the POSIX standard, one for commands defined by the Linux Standard 137 Base, an "other" directory for commands not covered by an obvious standard, 138 a directory of example commands (templates to use when starting new commands), 139 and a "pending" directory of commands that need further review/cleanup 140 before moving to one of the other directories (run these at your own risk, 141 cleanup patches welcome). 142 These directories are just for developer convenience sorting the commands, 143 the directories are otherwise functionally identical. To add a new category, 144 create the appropriate directory with a README file in it whose first line 145 is the description menuconfig should use for the directory.)</p> 146 147 <p>An easy way to start a new command is copy the file "toys/example/hello.c" 148 to the name of the new command, and modify this copy to implement the new 149 command (more or less by turning every instance of "hello" into the 150 name of your command, updating the command line arguments, globals, and 151 help data, and then filling out its "main" function with code that does 152 something interesting).</p> 153 154 <p>You could also start with "toys/example/skeleton.c", which provides a lot 155 more example code (showing several variants of command line option 156 parsing, how to implement multiple commands in the same file, and so on). 157 But usually it's just more stuff to delete.</p> 158 159 <p>Here's a checklist of steps to turn hello.c into another command:</p> 160 161 <ul> 162 <li><p>First "cp toys/example/hello.c toys/other/yourcommand.c" and open 163 the new file in your preferred text editor.</p> 164 <ul><li><p>Note that the 165 name of the new file is significant: it's the name of the new command you're 166 adding to toybox. The build includes all *.c files under toys/*/ whose 167 names are a case insensitive match for an enabled config symbol. So 168 toys/posix/cat.c only gets included if you have "CAT=y" in ".config".</p></li> 169 </ul></p></li> 170 171 <li><p>Change the one line comment at the top of the file (currently 172 "hello.c - A hello world program") to describe your new file.</p></li> 173 174 <li><p>Change the copyright notice to your name, email, and the current 175 year.</p></li> 176 177 <li><p>Give a URL to the relevant standards document, where applicable. 178 (Sample links to SUSv4 and LSB are provided, feel free to link to other 179 documentation or standards as appropriate.)</p></li> 180 181 <li><p>Update the USE_YOURCOMMAND(NEWTOY(yourcommand,"blah",0)) line. 182 The NEWTOY macro fills out this command's <a href="#toy_list">toy_list</a> 183 structure. The arguments to the NEWTOY macro are:</p> 184 185 <ol> 186 <li><p>the name used to run your command</p></li> 187 <li><p>the command line argument <a href="#lib_args">option parsing string</a> (0 if none)</p></li> 188 <li><p>a bitfield of TOYFLAG values 189 (defined in toys.h) providing additional information such as where your 190 command should be installed on a running system, whether to blank umask 191 before running, whether or not the command must run as root (and thus should 192 retain root access if installed SUID), and so on.</p></li> 193 </ol> 194 </li> 195 196 <li><p>Change the kconfig data (from "config YOURCOMMAND" to the end of the 197 comment block) to supply your command's configuration and help 198 information. The uppper case config symbols are used by menuconfig, and are 199 also what the CFG_ and USE_() macros are generated from (see [TODO]). The 200 help information here is used by menuconfig, and also by the "help" command to 201 describe your new command. (See [TODO] for details.) By convention, 202 unfinished commands default to "n" and finished commands default to "y", 203 so "make defconfig" selects all finished commands. (Note, "finished" means 204 "ready to be used", not that it'll never change again.)<p> 205 206 <p>Each help block should start with a "usage: yourcommand" line explaining 207 any command line arguments added by this config option. The "help" command 208 outputs this text, and scripts/config2help.c in the build infrastructure 209 collates these usage lines for commands with multiple configuration 210 options when producing generated/help.h.</p> 211 </li> 212 213 <li><p>Change the "#define FOR_hello" line to "#define FOR_yourcommand" right 214 before the "#include <toys.h>". (This selects the appropriate FLAG_ macros and 215 does a "#define TT this.yourcommand" so you can access the global variables 216 out of the space-saving union of structures. If you aren't using any command 217 flag bits and aren't defining a GLOBAL block, you can delete this line.)</p></li> 218 219 <li><p>Update the GLOBALS() macro to contain your command's global 220 variables. If your command has no global variables, delete this macro.</p> 221 222 <p>Variables in the GLOBALS() block are are stored in a space saving 223 <a href="#toy_union">union of structures</a> format, which may be accessed 224 using the TT macro as if TT were a global structure (so TT.membername). 225 If you specified two-character command line arguments in 226 NEWTOY(), the first few global variables will be initialized by the automatic 227 argument parsing logic, and the type and order of these variables must 228 correspond to the arguments specified in NEWTOY(). 229 (See <a href="#lib_args">lib/args.c</a> for details.)</p></li> 230 231 <li><p>Rename hello_main() to yourcommand_main(). This is the main() function 232 where execution of your command starts. Your command line options are 233 already sorted into this.optflags, this.optargs, this.optc, and the GLOBALS() 234 as appropriate by the time this function is called. (See 235 <a href="#lib_args">get_optflags()</a> for details.)</p></li> 236 </ul> 237 238 <a name="headers" /><h2><a href="#headers">Headers.</a></h2> 239 240 <p>Commands generally don't have their own headers. If it's common code 241 it can live in lib/, if it isn't put it in the command's .c file. (The line 242 between implementing multiple commands in a C file via OLDTOY() to share 243 infrastructure and moving that shared infrastructure to lib/ is a judgement 244 call. Try to figure out which is simplest.)</p> 245 246 <p>The top level toys.h should #include all the standard (posix) headers 247 that any command uses. (Partly this is friendly to ccache and partly this 248 makes the command implementations shorter.) Individual commands should only 249 need to include nonstandard headers that might prevent that command from 250 building in some context we'd care about (and thus requiring that command to 251 be disabled to avoid a build break).</p> 252 253 <p>Target-specific stuff (differences between compiler versions, libc versions, 254 or operating systems) should be confined to lib/portability.h and 255 lib/portability.c. (There's even some minimal compile-time environment probing 256 that writes data to generated/portability.h, see scripts/genconfig.sh.)</p> 257 258 <p>Only include linux/*.h headers from individual commands (not from other 259 headers), and only if you really need to. Data that varies per architecture 260 is a good reason to include a header. If you just need a couple constants 261 that haven't changed since the 1990's, it's ok to #define them yourself or 262 just use the constant inline with a comment explaining what it is. (A 263 #define that's only used once isn't really helping.)</p> 264 265 <p><a name="top" /><h1><a href="#top">Top level directory.</a></h1></p> 266 267 <p>This directory contains global infrastructure.</p> 268 269 <h3>toys.h</h3> 270 <p>Each command #includes "toys.h" as part of its standard prolog. It 271 may "#define FOR_commandname" before doing so to get some extra entries 272 specific to this command.</p> 273 274 <p>This file sucks in most of the commonly used standard #includes, so 275 individual files can just #include "toys.h" and not have to worry about 276 stdargs.h and so on. Individual commands still need to #include 277 special-purpose headers that may not be present on all systems (and thus would 278 prevent toybox from building that command on such a system with that command 279 enabled). Examples include regex support, any "linux/" or "asm/" headers, mtab 280 support (mntent.h and sys/mount.h), and so on.</p> 281 282 <p>The toys.h header also defines structures for most of the global variables 283 provided to each command by toybox_main(). These are described in 284 detail in the description for main.c, where they are initialized.</p> 285 286 <p>The global variables are grouped into structures (and a union) for space 287 savings, to more easily track the amount of memory consumed by them, 288 so that they may be automatically cleared/initialized as needed, and so 289 that access to global variables is more easily distinguished from access to 290 local variables.</p> 291 292 <h3>main.c</h3> 293 <p>Contains the main() function where execution starts, plus 294 common infrastructure to initialize global variables and select which command 295 to run. The "toybox" multiplexer command also lives here. (This is the 296 only command defined outside of the toys directory.)</p> 297 298 <p>Execution starts in main() which trims any path off of the first command 299 name and calls toybox_main(), which calls toy_exec(), which calls toy_find() 300 and toy_init() before calling the appropriate command's function from 301 toy_list[] (via toys.which->toy_main()). 302 If the command is "toybox", execution recurses into toybox_main(), otherwise 303 the call goes to the appropriate commandname_main() from a C file in the toys 304 directory.</p> 305 306 <p>The following global variables are defined in main.c:</p> 307 <ul> 308 <a name="toy_list" /> 309 <li><p><b>struct toy_list toy_list[]</b> - array describing all the 310 commands currently configured into toybox. The first entry (toy_list[0]) is 311 for the "toybox" multiplexer command, which runs all the other built-in commands 312 without symlinks by using its first argument as the name of the command to 313 run and the rest as that command's argument list (ala "./toybox echo hello"). 314 The remaining entries are the commands in alphabetical order (for efficient 315 binary search).</p> 316 317 <p>This is a read-only array initialized at compile time by 318 defining macros and #including generated/newtoys.h.</p> 319 320 <p>Members of struct toy_list (defined in "toys.h") include:</p> 321 <ul> 322 <li><p>char *<b>name</b> - the name of this command.</p></li> 323 <li><p>void (*<b>toy_main</b>)(void) - function pointer to run this 324 command.</p></li> 325 <li><p>char *<b>options</b> - command line option string (used by 326 get_optflags() in lib/args.c to intialize toys.optflags, toys.optargs, and 327 entries in the toy's GLOBALS struct). When this is NULL, no option 328 parsing is done before calling toy_main().</p></li> 329 <li><p>int <b>flags</b> - Behavior flags for this command. The following flags are currently understood:</p> 330 331 <ul> 332 <li><b>TOYFLAG_USR</b> - Install this command under /usr</li> 333 <li><b>TOYFLAG_BIN</b> - Install this command under /bin</li> 334 <li><b>TOYFLAG_SBIN</b> - Install this command under /sbin</li> 335 <li><b>TOYFLAG_NOFORK</b> - This command can be used as a shell builtin.</li> 336 <li><b>TOYFLAG_UMASK</b> - Call umask(0) before running this command.</li> 337 <li><b>TOYFLAG_STAYROOT</b> - Don't drop permissions for this command if toybox is installed SUID root.</li> 338 <li><b>TOYFLAG_NEEDROOT</b> - This command cannot function unless run with root access.</li> 339 </ul> 340 <br> 341 342 <p>These flags are combined with | (or). For example, to install a command 343 in /usr/bin, or together TOYFLAG_USR|TOYFLAG_BIN.</p> 344 </ul> 345 </li> 346 347 <li><p><b>struct toy_context toys</b> - global structure containing information 348 common to all commands, initializd by toy_init() and defined in "toys.h". 349 Members of this structure include:</p> 350 <ul> 351 <li><p>struct toy_list *<b>which</b> - a pointer to this command's toy_list 352 structure. Mostly used to grab the name of the running command 353 (toys->which.name).</p> 354 </li> 355 <li><p>int <b>exitval</b> - Exit value of this command. Defaults to zero. The 356 error_exit() functions will return 1 if this is zero, otherwise they'll 357 return this value.</p></li> 358 <li><p>char **<b>argv</b> - "raw" command line options, I.E. the original 359 unmodified string array passed in to main(). Note that modifying this changes 360 "ps" output, and is not recommended. This array is null terminated; a NULL 361 entry indicates the end of the array.</p> 362 <p>Most commands don't use this field, instead the use optargs, optflags, 363 and the fields in the GLOBALS struct initialized by get_optflags().</p> 364 </li> 365 <li><p>unsigned <b>optflags</b> - Command line option flags, set by 366 <a href="#lib_args">get_optflags()</a>. Indicates which of the command line options listed in 367 toys->which.options occurred this time.</p> 368 369 <p>The rightmost command line argument listed in toys->which.options sets bit 370 1, the next one sets bit 2, and so on. This means the bits are set in the same 371 order the binary digits would be listed if typed out as a string. For example, 372 the option string "abcd" would parse the command line "-c" to set optflags to 2, 373 "-a" would set optflags to 8, and "-bd" would set optflags to 6 (4|2).</p> 374 375 <p>Only letters are relevant to optflags. In the string "a*b:c#d", d=1, c=2, 376 b=4, a=8. Punctuation after a letter initializes global variables at the 377 start of the GLOBALS() block (see <a href="#toy_union">union toy_union this</a> 378 for details).</p> 379 380 <p>The build infrastructure creates FLAG_ macros for each option letter, 381 corresponding to the bit position, so you can check (toys.optflags & FLAG_x) 382 to see if a flag was specified. (The correct set of FLAG_ macros is selected 383 by defining FOR_mycommand before #including toys.h. The macros live in 384 toys/globals.h which is generated by scripts/make.sh.)</p> 385 386 <p>For more information on option parsing, see <a href="#lib_args">get_optflags()</a>.</p> 387 388 </li> 389 <li><p>char **<b>optargs</b> - Null terminated array of arguments left over 390 after get_optflags() removed all the ones it understood. Note: optarg[0] is 391 the first argument, not the command name. Use toys.which->name for the command 392 name.</p></li> 393 <li><p>int <b>optc</b> - Optarg count, equivalent to argc but for 394 optargs[].<p></li> 395 <li><p>int <b>exithelp</b> - Whether error_exit() should print a usage message 396 via help_main() before exiting. (True during option parsing, defaults to 397 false afterwards.)</p></li> 398 </ul> 399 400 <a name="toy_union" /> 401 <li><p><b>union toy_union this</b> - Union of structures containing each 402 command's global variables.</p> 403 404 <p>Global variables are useful: they reduce the overhead of passing extra 405 command line arguments between functions, they conveniently start prezeroed to 406 save initialization costs, and the command line argument parsing infrastructure 407 can also initialize global variables with its results.</p> 408 409 <p>But since each toybox process can only run one command at a time, allocating 410 space for global variables belonging to other commands you aren't currently 411 running would be wasteful.</p> 412 413 <p>Toybox handles this by encapsulating each command's global variables in 414 a structure, and declaring a union of those structures with a single global 415 instance (called "this"). The GLOBALS() macro contains the global 416 variables that should go in the current command's global structure. Each 417 variable can then be accessed as "this.commandname.varname". 418 If you #defined FOR_commandname before including toys.h, the macro TT is 419 #defined to this.commandname so the variable can then be accessed as 420 "TT.variable". See toys/hello.c for an example.</p> 421 422 <p>A command that needs global variables should declare a structure to 423 contain them all, and add that structure to this union. A command should never 424 declare global variables outside of this, because such global variables would 425 allocate memory when running other commands that don't use those global 426 variables.</p> 427 428 <p>The first few fields of this structure can be intialized by <a href="#lib_args">get_optargs()</a>, 429 as specified by the options field off this command's toy_list entry. See 430 the get_optargs() description in lib/args.c for details.</p> 431 </li> 432 433 <li><b>char toybuf[4096]</b> - a common scratch space buffer so 434 commands don't need to allocate their own. Any command is free to use this, 435 and it should never be directly referenced by functions in lib/ (although 436 commands are free to pass toybuf in to a library function as an argument).</li> 437 </ul> 438 439 <p>The following functions are defined in main.c:</p> 440 <ul> 441 <li><p>struct toy_list *<b>toy_find</b>(char *name) - Return the toy_list 442 structure for this command name, or NULL if not found.</p></li> 443 <li><p>void <b>toy_init</b>(struct toy_list *which, char *argv[]) - fill out 444 the global toys structure, calling get_optargs() if necessary.</p></li> 445 <li><p>void <b>toy_exec</b>(char *argv[]) - Run a built-in command with 446 arguments.</p> 447 <p>Calls toy_find() on argv[0] (which must be just a command name 448 without path). Returns if it can't find this command, otherwise calls 449 toy_init(), toys->which.toy_main(), and exit() instead of returning.</p> 450 451 <p>Use the library function xexec() to fall back to external executables 452 in $PATH if toy_exec() can't find a built-in command. Note that toy_exec() 453 does not strip paths before searching for a command, so "./command" will 454 never match an internal command.</li> 455 456 <li><p>void <b>toybox_main</b>(void) - the main function for the multiplexer 457 command (I.E. "toybox"). Given a command name as its first argument, calls 458 toy_exec() on its arguments. With no arguments, it lists available commands. 459 If the first argument starts with "-" it lists each command with its default 460 install path prepended.</p></li> 461 462 </ul> 463 464 <h3>Config.in</h3> 465 466 <p>Top level configuration file in a stylized variant of 467 <a href=http://kernel.org/doc/Documentation/kbuild/kconfig-language.txt>kconfig</a> format. Includes generated/Config.in.</p> 468 469 <p>These files are directly used by "make menuconfig" to select which commands 470 to build into toybox (thus generating a .config file), and by 471 scripts/config2help.py to create generated/help.h.</p> 472 473 <a name="generated" /> 474 <h1><a href="#generated">Temporary files:</a></h1> 475 476 <p>There is one temporary file in the top level source directory:</p> 477 <ul> 478 <li><p><b>.config</b> - Configuration file generated by kconfig, indicating 479 which commands (and options to commands) are currently enabled. Used 480 to make generated/config.h and determine which toys/*/*.c files to build.</p> 481 482 <p>You can create a human readable "miniconfig" version of this file using 483 <a href=http://landley.net/aboriginal/new_platform.html#miniconfig>these 484 instructions</a>.</p> 485 </li> 486 </ul> 487 488 <p><h2>Directory generated/</h2></p> 489 490 <p>The remaining temporary files live in the "generated/" directory, 491 which is for files generated at build time from other source files.</p> 492 493 <ul> 494 <li><p><b>generated/Config.in</b> - Kconfig entries for each command, included 495 from the top level Config.in. The help text here is used to generate 496 help.h.</p> 497 498 <p>Each command has a configuration entry with an upper case version of 499 the command name. Options to commands start with the command 500 name followed by an underscore and the option name. Global options are attached 501 to the "toybox" command, and thus use the prefix "TOYBOX_". This organization 502 is used by scripts/cfg2files to select which toys/*/*.c files to compile for a 503 given .config.</p> 504 </li> 505 506 <li><p><b>generated/config.h</b> - list of CFG_SYMBOL and USE_SYMBOL() macros, 507 generated from .config by a sed invocation in scripts/make.sh.</p> 508 509 <p>CFG_SYMBOL is a comple time constant set to 1 for enabled symbols and 0 for 510 disabled symbols. This allows the use of normal if() statements to remove 511 code at compile time via the optimizer's dead code elimination (which removes 512 from the binary any code that cannot be reached). This saves space without 513 cluttering the code with #ifdefs or leading to configuration dependent build 514 breaks. (See the 1992 Usenix paper 515 <a href=http://doc.cat-v.org/henry_spencer/ifdef_considered_harmful.pdf>#ifdef 516 Considered Harmful</a> for more information.)</p> 517 518 <p>When you can't entirely avoid an #ifdef, the USE_SYMBOL(code) macro 519 provides a less intrusive alternative, evaluating to the code in parentheses 520 when the symbol is enabled, and nothing when the symbol is disabled. This 521 is most commonly used around NEWTOY() declarations (so only the enabled 522 commands show up in toy_list), and in option strings. This can also be used 523 for things like varargs or structure members which can't always be 524 eliminated by a simple test on CFG_SYMBOL. Remember, unlike CFG_SYMBOL 525 this is really just a variant of #ifdef, and can still result in configuration 526 dependent build breaks. Use with caution.</p> 527 </li> 528 529 <li><p><b>generated/flags.h</b> - FLAG_? macros indicating which command 530 line options were seen. The option parsing in lib/args.c sets bits in 531 toys.optflags, which can be tested by anding with the appropriate FLAG_ 532 macro. (Bare longopts, which have no corresponding short option, will 533 have the longopt name after FLAG_. All others use the single letter short 534 option.)</p> 535 536 <p>To get the appropriate macros for your command, #define FOR_commandname 537 before #including toys.h. To switch macro sets (because you have an OLDTOY() 538 with different options in the same .c file), #define CLEANUP_oldcommand 539 and also #define FOR_newcommand, then #include "generated/flags.h" to switch. 540 </p> 541 </li> 542 543 <li><p><b>generated/globals.h</b> - 544 Declares structures to hold the contents of each command's GLOBALS(), 545 and combines them into "global_union this". (Yes, the name was 546 chosen to piss off C++ developers who think that C 547 is merely a subset of C++, not a language in its own right.)</p> 548 549 <p>The union reuses the same memory for each command's global struct: 550 since only one command's globals are in use at any given time, collapsing 551 them together saves space. The headers #define TT to the appropriate 552 "this.commandname", so you can refer to the current command's global 553 variables out of "this" as TT.variablename.</p> 554 555 <p>The globals start zeroed, and the first few are filled out by the 556 lib/args.c argument parsing code called from main.c.</p> 557 </li> 558 559 <li><p><b>toys/help.h</b> - Help strings for use by the "help" command and 560 --help options. This file #defines a help_symbolname string for each 561 symbolname, but only the symbolnames matching command names get used 562 by show_help() in lib/help.c to display help for commands.</p> 563 564 <p>This file is created by scripts/make.sh, which compiles scripts/config2help.c 565 into the binary generated/config2help, and then runs it against the top 566 level .config and Config.in files to extract the help text from each config 567 entry and collate together dependent options.</p> 568 569 <p>This file contains help text for all commands, regardless of current 570 configuration, but only the ones currently enabled in the .config file 571 wind up in the help_data[] array, and only the enabled dependent options 572 have their help text added to the command they depend on.</p> 573 </li> 574 575 <li><p><b>generated/newtoys.h</b> - 576 All the NEWTOY() and OLDTOY() macros from toys/*/*.c. The "toybox" multiplexer 577 is the first entry, the rest are in alphabetical order. Each line should be 578 inside an appropriate USE_ macro, so code that #includes this file only sees 579 the currently enabled commands.</p> 580 581 <p>By #definining NEWTOY() to various things before #including this file, 582 it may be used to create function prototypes (in toys.h), initialize the 583 help_data array (in lib/help.c), initialize the toy_list array (in main.c, 584 the alphabetical order lets toy_find() do a binary search, the exception to 585 the alphabetical order lets it use the multiplexer without searching), and so 586 on. (It's even used to initialize the NEED_OPTIONS macro, which produces a 1 587 or 0 for each command using command line option parsing, which is ORed together 588 to allow compile-time dead code elimination to remove the whole of 589 lib/args.c if nothing currently enabled is using it.)<p> 590 591 <p>Each NEWTOY and OLDTOY macro contains the command name, command line 592 option string (telling lib/args.c how to parse command line options for 593 this command), recommended install location, and miscelaneous data such 594 as whether this command should retain root permissions if installed suid.</p> 595 </li> 596 597 <li><p><b>toys/oldtoys.h</b> - Macros with the command line option parsing 598 string for each NEWTOY. This allows an OLDTOY that's just an alias for an 599 existing command to refer to the existing option string instead of 600 having to repeat it.</p> 601 </li> 602 </ul> 603 604 <a name="lib"> 605 <h2>Directory lib/</h2> 606 607 <p>TODO: document lots more here.</p> 608 609 <p>lib: getmountlist(), error_msg/error_exit, xmalloc(), 610 strlcpy(), xexec(), xopen()/xread(), xgetcwd(), xabspath(), find_in_path(), 611 itoa().</p> 612 613 614 615 <a name="lib_xwrap"><h3>lib/xwrap.c</h3> 616 617 <p>Functions prefixed with the letter x call perror_exit() when they hit 618 errors, to eliminate common error checking. This prints an error message 619 and the strerror() string for the errno encountered.</p> 620 621 <p>You can intercept this exit by assigning a setjmp/longjmp buffer to 622 toys.rebound (set it back to zero to restore the default behavior). 623 If you do this, cleaning up resource leaks is your problem.</p> 624 625 <ul> 626 <li><b>void xstrncpy(char *dest, char *src, size_t size)</b></li> 627 <li><b>void xexit(void)</b></li> 628 <li><b>void *xmalloc(size_t size)</b></li> 629 <li><b>void *xzalloc(size_t size)</b></li> 630 <li><b>void *xrealloc(void *ptr, size_t size)</b></li> 631 <li><b>char *xstrndup(char *s, size_t n)</b></li> 632 <li><b>char *xstrdup(char *s)</b></li> 633 <li><b>char *xmprintf(char *format, ...)</b></li> 634 <li><b>void xprintf(char *format, ...)</b></li> 635 <li><b>void xputs(char *s)</b></li> 636 <li><b>void xputc(char c)</b></li> 637 <li><b>void xflush(void)</b></li> 638 <li><b>pid_t xfork(void)</b></li> 639 <li><b>void xexec_optargs(int skip)</b></li> 640 <li><b>void xexec(char **argv)</b></li> 641 <li><b>pid_t xpopen(char **argv, int *pipes)</b></li> 642 <li><b>int xpclose(pid_t pid, int *pipes)</b></li> 643 <li><b>void xaccess(char *path, int flags)</b></li> 644 <li><b>void xunlink(char *path)</b></li> 645 <li><p><b>int xcreate(char *path, int flags, int mode)<br /> 646 int xopen(char *path, int flags)</b></p> 647 648 <p>The xopen() and xcreate() functions open an existing file (exiting if 649 it's not there) and create a new file (exiting if it can't).</p> 650 651 <p>They default to O_CLOEXEC so the filehandles aren't passed on to child 652 processes. Feed in O_CLOEXEC to disable this.</p> 653 </li> 654 <li><p><b>void xclose(int fd)</b></p> 655 656 <p>Because NFS is broken, and won't necessarily perform the requested 657 operation (and report the error) until you close the file. Of course, this 658 being NFS, it's not guaranteed to report the error there either, but it 659 _can_.</p> 660 661 <p>Nothing else ever reports an error on close, everywhere else it's just a 662 VFS operation freeing some resources. NFS is _special_, in a way that 663 other network filesystems like smbfs and v9fs aren't..</p> 664 </li> 665 <li><b>int xdup(int fd)</b></li> 666 <li><p><b>size_t xread(int fd, void *buf, size_t len)</b></p> 667 668 <p>Can return 0, but not -1.</p> 669 </li> 670 <li><p><b>void xreadall(int fd, void *buf, size_t len)</b></p> 671 672 <p>Reads the entire len-sized buffer, retrying to complete short 673 reads. Exits if it can't get enough data.</p></li> 674 675 <li><p><b>void xwrite(int fd, void *buf, size_t len)</b></p> 676 677 <p>Retries short writes, exits if can't write the entire buffer.</p></li> 678 679 <li><b>off_t xlseek(int fd, off_t offset, int whence)</b></li> 680 <li><b>char *xgetcwd(void)</b></li> 681 <li><b>void xstat(char *path, struct stat *st)</b></li> 682 <li><p><b>char *xabspath(char *path, int exact) </b></p> 683 684 <p>After several years of 685 <a href=http://landley.net/notes-2007.html#18-06-2007>wrestling</a> 686 <a href=http://landley.net/notes-2008.html#19-01-2008>with</a> realpath(), 687 I broke down and <a href=http://landley.net/notes-2012.html#20-11-2012>wrote 688 my own</a> implementation that doesn't use the one in libc. As I explained: 689 690 <blockquote><p>If the path ends with a broken link, 691 readlink -f should show where the link points to, not where the broken link 692 lives. (The point of readlink -f is "if I write here, where would it attempt 693 to create a file".) The problem is, realpath() returns NULL for a path ending 694 with a broken link, and I can't beat different behavior out of code locked 695 away in libc.</p></blockquote> 696 697 <p> 698 </li> 699 <li><b>void xchdir(char *path)</b></li> 700 <li><b>void xchroot(char *path)</b></li> 701 702 <li><p><b>struct passwd *xgetpwuid(uid_t uid)<br /> 703 struct group *xgetgrgid(gid_t gid)<br /> 704 struct passwd *xgetpwnam(char *name)</b></p> 705 706 <p></p> 707 </li> 708 709 710 711 <li><b>void xsetuser(struct passwd *pwd)</b></li> 712 <li><b>char *xreadlink(char *name)</b></li> 713 <li><b>char *xreadfile(char *name, char *buf, off_t len)</b></li> 714 <li><b>int xioctl(int fd, int request, void *data)</b></li> 715 <li><b>void xpidfile(char *name)</b></li> 716 <li><b>void xsendfile(int in, int out)</b></li> 717 <li><b>long xparsetime(char *arg, long units, long *fraction)</b></li> 718 <li><b>void xregcomp(regex_t *preg, char *regex, int cflags)</b></li> 719 </ul> 720 721 <a name="lib_lib"><h3>lib/lib.c</h3> 722 <p>Eight gazillion common functions:</p> 723 724 <ul> 725 <li><b>void verror_msg(char *msg, int err, va_list va)</b></li> 726 <li><b>void error_msg(char *msg, ...)</b></li> 727 <li><b>void perror_msg(char *msg, ...)</b></li> 728 <li><b>void error_exit(char *msg, ...)</b></li> 729 <li><b>void perror_exit(char *msg, ...)</b></li> 730 <li><b>ssize_t readall(int fd, void *buf, size_t len)</b></li> 731 <li><b>ssize_t writeall(int fd, void *buf, size_t len)</b></li> 732 <li><b>off_t lskip(int fd, off_t offset)</b></li> 733 <li><b>int mkpathat(int atfd, char *dir, mode_t lastmode, int flags)</b></li> 734 <li><b>struct string_list **splitpath(char *path, struct string_list **list)</b></li> 735 <li><b>struct string_list *find_in_path(char *path, char *filename)</b></li> 736 <li><b>long atolx(char *numstr)</b></li> 737 <li><b>long atolx_range(char *numstr, long low, long high)</b></li> 738 <li><b>int numlen(long l)</b></li> 739 <li><b>int stridx(char *haystack, char needle)</b></li> 740 <li><b>int strstart(char **a, char *b)</b></li> 741 <li><b>off_t fdlength(int fd)</b></li> 742 <li><b>char *readfile(char *name, char *ibuf, off_t len)</b></li> 743 <li><b>void msleep(long miliseconds)</b></li> 744 <li><b>int64_t peek_le(void *ptr, unsigned size)</b></li> 745 <li><b>int64_t peek_be(void *ptr, unsigned size)</b></li> 746 <li><b>int64_t peek(void *ptr, unsigned size)</b></li> 747 <li><b>void poke(void *ptr, uint64_t val, int size)</b></li> 748 <li><b>void loopfiles_rw(char **argv, int flags, int permissions, int failok,</b></li> 749 <li><b>void loopfiles(char **argv, void (*function)(int fd, char *name))</b></li> 750 <li><b>char *get_rawline(int fd, long *plen, char end)</b></li> 751 <li><b>char *get_line(int fd)</b></li> 752 <li><b>int wfchmodat(int fd, char *name, mode_t mode)</b></li> 753 <li><b>static void tempfile_handler(int i)</b></li> 754 <li><b>int copy_tempfile(int fdin, char *name, char **tempname)</b></li> 755 <li><b>void delete_tempfile(int fdin, int fdout, char **tempname)</b></li> 756 <li><b>void replace_tempfile(int fdin, int fdout, char **tempname)</b></li> 757 <li><b>void crc_init(unsigned int *crc_table, int little_endian)</b></li> 758 <li><b>int terminal_size(unsigned *xx, unsigned *yy)</b></li> 759 <li><b>int yesno(char *prompt, int def)</b></li> 760 <li><b>void generic_signal(int sig)</b></li> 761 <li><b>void sigatexit(void *handler)</b></li> 762 <li><b>int sig_to_num(char *pidstr)</b></li> 763 <li><b>char *num_to_sig(int sig)</b></li> 764 <li><b>mode_t string_to_mode(char *modestr, mode_t mode)</b></li> 765 <li><b>void mode_to_string(mode_t mode, char *buf)</b></li> 766 <li><b>void names_to_pid(char **names, int (*callback)(pid_t pid, char *name))</b></li> 767 <li><b>int human_readable(char *buf, unsigned long long num)</b></li> 768 </ul> 769 770 <h3>lib/portability.h</h3> 771 772 <p>This file is automatically included from the top of toys.h, and smooths 773 over differences between platforms (hardware targets, compilers, C libraries, 774 operating systems, etc).</p> 775 776 <p>This file provides SWAP macros (SWAP_BE16(x) and SWAP_LE32(x) and so on).</p> 777 778 <p>A macro like SWAP_LE32(x) means "The value in x is stored as a little 779 endian 32 bit value, so perform the translation to/from whatever the native 780 32-bit format is". You do the swap once on the way in, and once on the way 781 out. If your target is already little endian, the macro is a NOP.</p> 782 783 <p>The SWAP macros come in BE and LE each with 16, 32, and 64 bit versions. 784 In each case, the name of the macro refers to the _external_ representation, 785 and converts to/from whatever your native representation happens to be (which 786 can vary depending on what you're currently compiling for).</p> 787 788 <a name="lib_llist"><h3>lib/llist.c</h3> 789 790 <p>Some generic single and doubly linked list functions, which take 791 advantage of a couple properties of C:</p> 792 793 <ul> 794 <li><p>Structure elements are laid out in memory in the order listed, and 795 the first element has no padding. This means you can always treat (typecast) 796 a pointer to a structure as a pointer to the first element of the structure, 797 even if you don't know anything about the data following it.</p></li> 798 799 <li><p>An array of length zero at the end of a structure adds no space 800 to the sizeof() the structure, but if you calculate how much extra space 801 you want when you malloc() the structure it will be available at the end. 802 Since C has no bounds checking, this means each struct can have one variable 803 length array.</p></li> 804 </ul> 805 806 <p>Toybox's list structures always have their <b>next</b> pointer as 807 the first entry of each struct, and singly linked lists end with a NULL pointer. 808 This allows generic code to traverse such lists without knowing anything 809 else about the specific structs composing them: if your pointer isn't NULL 810 typecast it to void ** and dereference once to get the next entry.</p> 811 812 <p><b>lib/lib.h</b> defines three structure types:</p> 813 <ul> 814 <li><p><b>struct string_list</b> - stores a single string (<b>char str[0]</b>), 815 memory for which is allocated as part of the node. (I.E. llist_traverse(list, 816 free); can clean up after this type of list.)</p></li> 817 818 <li><p><b>struct arg_list</b> - stores a pointer to a single string 819 (<b>char *arg</b>) which is stored in a separate chunk of memory.</p></li> 820 821 <li><p><b>struct double_list</b> - has a second pointer (<b>struct double_list 822 *prev</b> along with a <b>char *data</b> for payload.</p></li> 823 </ul> 824 825 <b>List Functions</b> 826 827 <ul> 828 <li><p>void *<b>llist_pop</b>(void **list) - advances through a list ala 829 <b>node = llist_pop(&list);</b> This doesn't modify the list contents, 830 but does advance the pointer you feed it (which is why you pass the _address_ 831 of that pointer, not the pointer itself).</p></li> 832 833 <li><p>void <b>llist_traverse</b>(void *list, void (*using)(void *data)) - 834 iterate through a list calling a function on each node.</p></li> 835 836 <li><p>struct double_list *<b>dlist_add</b>(struct double_list **llist, char *data) 837 - append an entry to a circular linked list. 838 This function allocates a new struct double_list wrapper and returns the 839 pointer to the new entry (which you can usually ignore since it's llist->prev, 840 but if llist was NULL you need it). The argument is the ->data field for the 841 new node.</p></li> 842 <ul><li><p>void <b>dlist_add_nomalloc</b>(struct double_list **llist, 843 struct double_list *new) - append existing struct double_list to 844 list, does not allocate anything.</p></li></ul> 845 </ul> 846 847 <b>List code trivia questions:</b> 848 849 <ul> 850 <li><p><b>Why do arg_list and double_list contain a char * payload instead of 851 a void *?</b> - Because you always have to typecast a void * to use it, and 852 typecasting a char * does no harm. Since strings are the most common 853 payload, and doing math on the pointer ala 854 "(type *)(ptr+sizeof(thing)+sizeof(otherthing))" requires ptr to be char * 855 anyway (at least according to the C standard), defaulting to char * saves 856 a typecast.</p> 857 </li> 858 859 <li><p><b>Why do the names ->str, ->arg, and ->data differ?</b> - To force 860 you to keep track of which one you're using, calling free(node->str) would 861 be bad, and _failing_ to free(node->arg) leaks memory.</p></li> 862 863 <li><p><b>Why does llist_pop() take a void * instead of void **?</b> - 864 because the stupid compiler complains about "type punned pointers" when 865 you typecast and dereference on the same line, 866 due to insane FSF developers hardwiring limitations of their optimizer 867 into gcc's warning system. Since C automatically typecasts any other 868 pointer type to and from void *, the current code works fine. It's sad that it 869 won't warn you if you forget the &, but the code crashes pretty quickly in 870 that case.</p></li> 871 872 <li><p><b>How do I assemble a singly-linked-list in order?</b> - use 873 a double_list, dlist_add() your entries, and then break the circle with 874 <b>list->prev->next = NULL;</b> when done.</li> 875 </ul> 876 877 <a name="lib_args"><h3>lib/args.c</h3> 878 879 <p>Toybox's main.c automatically parses command line options before calling the 880 command's main function. Option parsing starts in get_optflags(), which stores 881 results in the global structures "toys" (optflags and optargs) and "this".</p> 882 883 <p>The option parsing infrastructure stores a bitfield in toys.optflags to 884 indicate which options the current command line contained, and defines FLAG 885 macros code can use to check whether each argument's bit is set. Arguments 886 attached to those options are saved into the command's global structure 887 ("this"). Any remaining command line arguments are collected together into 888 the null-terminated array toys.optargs, with the length in toys.optc. (Note 889 that toys.optargs does not contain the current command name at position zero, 890 use "toys.which->name" for that.) The raw command line arguments get_optflags() 891 parsed are retained unmodified in toys.argv[].</p> 892 893 <p>Toybox's option parsing logic is controlled by an "optflags" string, using 894 a format reminiscent of getopt's optargs but with several important differences. 895 Toybox does not use the getopt() 896 function out of the C library, get_optflags() is an independent implementation 897 which doesn't permute the original arguments (and thus doesn't change how the 898 command is displayed in ps and top), and has many features not present in 899 libc optargs() (such as the ability to describe long options in the same string 900 as normal options).</p> 901 902 <p>Each command's NEWTOY() macro has an optflags string as its middle argument, 903 which sets toy_list.options for that command to tell get_optflags() what 904 command line arguments to look for, and what to do with them. 905 If a command has no option 906 definition string (I.E. the argument is NULL), option parsing is skipped 907 for that command, which must look at the raw data in toys.argv to parse its 908 own arguments. (If no currently enabled command uses option parsing, 909 get_optflags() is optimized out of the resulting binary by the compiler's 910 --gc-sections option.)</p> 911 912 <p>You don't have to free the option strings, which point into the environment 913 space (I.E. the string data is not copied). A TOYFLAG_NOFORK command 914 that uses the linked list type "*" should free the list objects but not 915 the data they point to, via "llist_free(TT.mylist, NULL);". (If it's not 916 NOFORK, exit() will free all the malloced data anyway unless you want 917 to implement a CONFIG_TOYBOX_FREE cleanup for it.)</p> 918 919 <h4>Optflags format string</h4> 920 921 <p>Note: the optflags option description string format is much more 922 concisely described by a large comment at the top of lib/args.c.</p> 923 924 <p>The general theory is that letters set optflags, and punctuation describes 925 other actions the option parsing logic should take.</p> 926 927 <p>For example, suppose the command line <b>command -b fruit -d walrus -a 42</b> 928 is parsed using the optflags string "<b>a#b:c:d</b>". (I.E. 929 toys.which->options="a#b:c:d" and argv = ["command", "-b", "fruit", "-d", 930 "walrus", "-a", "42"]). When get_optflags() returns, the following data is 931 available to command_main(): 932 933 <ul> 934 <li><p>In <b>struct toys</b>: 935 <ul> 936 <li>toys.optflags = 13; // FLAG_a = 8 | FLAG_b = 4 | FLAG_d = 1</li> 937 <li>toys.optargs[0] = "walrus"; // leftover argument</li> 938 <li>toys.optargs[1] = NULL; // end of list</li> 939 <li>toys.optc = 1; // there was 1 leftover argument</li> 940 <li>toys.argv[] = {"-b", "fruit", "-d", "walrus", "-a", "42"}; // The original command line arguments 941 </ul> 942 <p></li> 943 944 <li><p>In <b>union this</b> (treated as <b>long this[]</b>): 945 <ul> 946 <li>this[0] = NULL; // -c didn't get an argument this time, so get_optflags() didn't change it and toys_init() zeroed "this" during setup.)</li> 947 <li>this[1] = (long)"fruit"; // argument to -b</li> 948 <li>this[2] = 42; // argument to -a</li> 949 </ul> 950 </p></li> 951 </ul> 952 953 <p>If the command's globals are:</p> 954 955 <blockquote><pre> 956 GLOBALS( 957 char *c; 958 char *b; 959 long a; 960 ) 961 </pre></blockquote> 962 963 <p>That would mean TT.c == NULL, TT.b == "fruit", and TT.a == 42. (Remember, 964 each entry that receives an argument must be a long or pointer, to line up 965 with the array position. Right to left in the optflags string corresponds to 966 top to bottom in GLOBALS().</p> 967 968 <p>Put globals not filled out by the option parsing logic at the end of the 969 GLOBALS block. Common practice is to list the options one per line (to 970 make the ordering explicit, first to last in globals corresponds to right 971 to left in the option string), then leave a blank line before any non-option 972 globals.</p> 973 974 <p><b>long toys.optflags</b></p> 975 976 <p>Each option in the optflags string corresponds to a bit position in 977 toys.optflags, with the same value as a corresponding binary digit. The 978 rightmost argument is (1<<0), the next to last is (1<<1) and so on. If 979 the option isn't encountered while parsing argv[], its bit remains 0.</p> 980 981 <p>Each option -x has a FLAG_x macro for the command letter. Bare --longopts 982 with no corresponding short option have a FLAG_longopt macro for the long 983 optionname. Commands enable these macros by #defining FOR_commandname before 984 #including <toys.h>. When multiple commands are implemented in the same 985 source file, you can switch flag contexts later in the file by 986 #defining CLEANUP_oldcommand and #defining FOR_newcommand, then 987 #including <generated/flags.h>.</p> 988 989 <p>Options disabled in the current configuration (wrapped in 990 a USE_BLAH() macro for a CONFIG_BLAH that's switched off) have their 991 corresponding FLAG macro set to zero, so code checking them ala 992 if (toys.optargs & FLAG_x) gets optimized out via dead code elimination. 993 #defining FORCE_FLAGS when switching flag context disables this 994 behavior: the flag is never zero even if the config is disabled. This 995 allows code shared between multiple commands to use the same flag 996 values, as long as the common flags match up right to left in both option 997 strings.</p> 998 999 <p>For example, 1000 the optflags string "abcd" would parse the command line argument "-c" to set 1001 optflags to 2, "-a" would set optflags to 8, "-bd" would set optflags to 1002 6 (I.E. 4|2), and "-a -c" would set optflags to 10 (2|8). To check if -c 1003 was encountered, code could test: if (toys.optflags & FLAG_c) printf("yup"); 1004 (See the toys/examples directory for more.)</p> 1005 1006 <p>Only letters are relevant to optflags, punctuation is skipped: in the 1007 string "a*b:c#d", d=1, c=2, b=4, a=8. The punctuation after a letter 1008 usually indicate that the option takes an argument.</p> 1009 1010 <p>Since toys.optflags is an unsigned int, it only stores 32 bits. (Which is 1011 the amount a long would have on 32-bit platforms anyway; 64 bit code on 1012 32 bit platforms is too expensive to require in common code used by almost 1013 all commands.) Bit positions beyond the 1<<31 aren't recorded, but 1014 parsing higher options can still set global variables.</p> 1015 1016 <p><b>Automatically setting global variables from arguments (union this)</b></p> 1017 1018 <p>The following punctuation characters may be appended to an optflags 1019 argument letter, indicating the option takes an additional argument:</p> 1020 1021 <ul> 1022 <li><b>:</b> - plus a string argument, keep most recent if more than one.</li> 1023 <li><b>*</b> - plus a string argument, appended to a linked list.</li> 1024 <li><b>@</b> - plus an occurrence counter (stored in a long)</li> 1025 <li><b>#</b> - plus a signed long argument. 1026 <li><b>-</b> - plus a signed long argument defaulting to negative (start argument with + to force a positive value).</li> 1027 <li><b>.</b> - plus a floating point argument (if CFG_TOYBOX_FLOAT).</li> 1028 <ul>The following can be appended to a float or double: 1029 <li><b><123</b> - error if argument is less than this</li> 1030 <li><b>>123</b> - error if argument is greater than this</li> 1031 <li><b>=123</b> - default value if argument not supplied</li> 1032 </ul> 1033 </ul> 1034 1035 <p><b>GLOBALS</b></p> 1036 1037 <p>Options which have an argument fill in the corresponding slot in the global 1038 union "this" (see generated/globals.h), treating it as an array of longs 1039 with the rightmost saved in this[0]. As described above, using "a*b:c#d", 1040 "-c 42" would set this[0] = 42; and "-b 42" would set this[1] = "42"; each 1041 slot is left NULL if the corresponding argument is not encountered.</p> 1042 1043 <p>This behavior is useful because the LP64 standard ensures long and pointer 1044 are the same size. C99 guarantees structure members will occur in memory 1045 in the same order they're declared, and that padding won't be inserted between 1046 consecutive variables of register size. Thus the first few entries can 1047 be longs or pointers corresponding to the saved arguments.</p> 1048 1049 <p>See toys/example/*.c for longer examples of parsing options into the 1050 GLOBALS block.</p> 1051 1052 <p><b>char *toys.optargs[]</b></p> 1053 1054 <p>Command line arguments in argv[] which are not consumed by option parsing 1055 (I.E. not recognized either as -flags or arguments to -flags) will be copied 1056 to toys.optargs[], with the length of that array in toys.optc. 1057 (When toys.optc is 0, no unrecognized command line arguments remain.) 1058 The order of entries is preserved, and as with argv[] this new array is also 1059 terminated by a NULL entry.</p> 1060 1061 <p>Option parsing can require a minimum or maximum number of optargs left 1062 over, by adding "<1" (read "at least one") or ">9" ("at most nine") to the 1063 start of the optflags string.</p> 1064 1065 <p>The special argument "--" terminates option parsing, storing all remaining 1066 arguments in optargs. The "--" itself is consumed.</p> 1067 1068 <p><b>Other optflags control characters</b></p> 1069 1070 <p>The following characters may occur at the start of each command's 1071 optflags string, before any options that would set a bit in toys.optflags:</p> 1072 1073 <ul> 1074 <li><b>^</b> - stop at first nonoption argument (for nice, xargs...)</li> 1075 <li><b>?</b> - allow unknown arguments (pass non-option arguments starting 1076 with - through to optargs instead of erroring out).</li> 1077 <li><b>&</b> - the first argument has imaginary dash (ala tar/ps. If given twice, all arguments have imaginary dash.)</li> 1078 <li><b><</b> - must be followed by a decimal digit indicating at least this many leftover arguments are needed in optargs (default 0)</li> 1079 <li><b>></b> - must be followed by a decimal digit indicating at most this many leftover arguments allowed (default MAX_INT)</li> 1080 </ul> 1081 1082 <p>The following characters may be appended to an option character, but do 1083 not by themselves indicate an extra argument should be saved in this[]. 1084 (Technically any character not recognized as a control character sets an 1085 optflag, but letters are never control characters.)</p> 1086 1087 <ul> 1088 <li><b>^</b> - stop parsing options after encountering this option, everything else goes into optargs.</li> 1089 <li><b>|</b> - this option is required. If more than one marked, only one is required.</li> 1090 </ul> 1091 1092 <p>The following may be appended to a float or double:</p> 1093 1094 <ul> 1095 <li><b><123</b> - error if argument is less than this</li> 1096 <li><b>>123</b> - error if argument is greater than this</li> 1097 <li><b>=123</b> - default value if argument not supplied</li> 1098 </ul> 1099 1100 <p>Option parsing only understands <>= after . when CFG_TOYBOX_FLOAT 1101 is enabled. (Otherwise the code to determine where floating point constants 1102 end drops out. When disabled, it can reserve a global data slot for the 1103 argument so offsets won't change, but will never fill it out.) You can handle 1104 this by using the USE_BLAH() macros with C string concatenation, ala:</p> 1105 1106 <blockquote>"abc." USE_TOYBOX_FLOAT("<1.23>4.56=7.89") "def"</blockquote> 1107 1108 <p><b>--longopts</b></p> 1109 1110 <p>The optflags string can contain long options, which are enclosed in 1111 parentheses. They may be appended to an existing option character, in 1112 which case the --longopt is a synonym for that option, ala "a:(--fred)" 1113 which understands "-a blah" or "--fred blah" as synonyms.</p> 1114 1115 <p>Longopts may also appear before any other options in the optflags string, 1116 in which case they have no corresponding short argument, but instead set 1117 their own bit based on position. So for "(walrus)#(blah)xy:z", "command 1118 --walrus 42" would set toys.optflags = 16 (-z = 1, -y = 2, -x = 4, --blah = 8) 1119 and would assign this[1] = 42;</p> 1120 1121 <p>A short option may have multiple longopt synonyms, "a(one)(two)", but 1122 each "bare longopt" (ala "(one)(two)abc" before any option characters) 1123 always sets its own bit (although you can group them with +X).</p> 1124 1125 <p>Only bare longopts have a FLAG_ macro with the longopt name 1126 (ala --fred would #define FLAG_fred). Other longopts use the short 1127 option's FLAG macro to test the toys.optflags bit.</p> 1128 1129 <p>Options with a semicolon ";" after their data type can only set their 1130 corresponding GLOBALS() entry via "--longopt=value". For example, option 1131 string "x(boing): y" would set TT.x if it saw "--boing=value", but would 1132 treat "--boing value" as setting FLAG_x in toys.optargs, leaving TT.x NULL, 1133 and keeping "value" in toys.optargs[]. (This lets "ls --color" and 1134 "ls --color=auto" both work.)</p> 1135 1136 <p><b>[groups]</b></p> 1137 1138 <p>At the end of the option string, square bracket groups can define 1139 relationships between existing options. (This only applies to short 1140 options, bare --longopts can't participate.)</p> 1141 1142 <p>The first character of the group defines the type, the remaining 1143 characters are options it applies to:</p> 1144 1145 <ul> 1146 <li><b>-</b> - Exclusive, switch off all others in this group.</li> 1147 <li><b>+</b> - Inclusive, switch on all others in this group.</li> 1148 <li><b>!</b> - Error, fail if more than one defined.</li> 1149 </ul> 1150 1151 <p>So "abc[-abc]" means -ab = -b, -ba = -a, -abc = -c. "abc[+abc]" 1152 means -ab=-abc, -c=-abc, and "abc[!abc] means -ab calls error_exit("no -b 1153 with -a"). Note that [-] groups clear the GLOBALS option slot of 1154 options they're switching back off, but [+] won't set options it didn't see 1155 (just the optflags).</p> 1156 1157 <p><b>whitespace</b></p> 1158 1159 <p>Arguments may occur with or without a space (I.E. "-a 42" or "-a42"). 1160 The command line argument "-abc" may be interepreted many different ways: 1161 the optflags string "cba" sets toys.optflags = 7, "c:ba" sets toys.optflags=4 1162 and saves "ba" as the argument to -c, and "cb:a" sets optflags to 6 and saves 1163 "c" as the argument to -b.</p> 1164 1165 <p>Note that & changes whitespace handling, so that the command line 1166 "tar cvfCj outfile.tar.bz2 topdir filename" is parsed the same as 1167 "tar filename -c -v -j -f outfile.tar.bz2 -C topdir". Note that "tar -cvfCj 1168 one two three" would equal "tar -c -v -f Cj one two three". (This matches 1169 historical usage.)</p> 1170 1171 <p>Appending a space to the option in the option string ("a: b") makes it 1172 require a space, I.E. "-ab" is interpreted as "-a" "-b". That way "kill -stop" 1173 differs from "kill -s top".</p> 1174 1175 <p>Appending ; to a longopt in the option string makes its argument optional, 1176 and only settable with =, so in ls "(color):;" can accept "ls --color" and 1177 "ls --color=auto" without complaining that the first has no argument.</p> 1178 1179 <a name="lib_dirtree"><h3>lib/dirtree.c</h3> 1180 1181 <p>The directory tree traversal code should be sufficiently generic 1182 that commands never need to use readdir(), scandir(), or the fts.h family 1183 of functions.</p> 1184 1185 <p>These functions do not call chdir() or rely on PATH_MAX. Instead they 1186 use openat() and friends, using one filehandle per directory level to 1187 recurseinto subdirectories. (I.E. they can descend 1000 directories deep 1188 if setrlimit(RLIMIT_NOFILE) allows enough open filehandles, and the default 1189 in /proc/self/limits is generally 1024.)</p> 1190 1191 <p>The basic dirtree functions are:</p> 1192 1193 <ul> 1194 <li><p><b>dirtree_read(char *path, int (*callback)(struct dirtree node))</b> - 1195 recursively read directories, either applying callback() or returning 1196 a tree of struct dirtree if callback is NULL.</p></li> 1197 1198 <li><p><b>dirtree_path(struct dirtree *node, int *plen)</b> - malloc() a 1199 string containing the path from the root of this tree to this node. If 1200 plen isn't NULL then *plen is how many extra bytes to malloc at the end 1201 of string.</p></li> 1202 1203 <li><p><b>dirtree_parentfd(struct dirtree *node)</b> - return fd of 1204 containing directory, for use with openat() and such.</p></li> 1205 </ul> 1206 1207 <p>The <b>dirtree_read()</b> function takes two arguments, a starting path for 1208 the root of the tree, and a callback function. The callback takes a 1209 <b>struct dirtree *</b> (from lib/lib.h) as its argument. If the callback is 1210 NULL, the traversal uses a default callback (dirtree_notdotdot()) which 1211 recursively assembles a tree of struct dirtree nodes for all files under 1212 this directory and subdirectories (filtering out "." and ".." entries), 1213 after which dirtree_read() returns the pointer to the root node of this 1214 snapshot tree.</p> 1215 1216 <p>Otherwise the callback() is called on each entry in the directory, 1217 with struct dirtree * as its argument. This includes the initial 1218 node created by dirtree_read() at the top of the tree.</p> 1219 1220 <p><b>struct dirtree</b></p> 1221 1222 <p>Each struct dirtree node contains <b>char name[]</b> and <b>struct stat 1223 st</b> entries describing a file, plus a <b>char *symlink</b> 1224 which is NULL for non-symlinks.</p> 1225 1226 <p>During a callback function, the <b>int data</b> field of directory nodes 1227 contains a dirfd (for use with the openat() family of functions). This is 1228 generally used by calling dirtree_parentfd() on the callback's node argument. 1229 For symlinks, data contains the length of the symlink string. On the second 1230 callback from DIRTREE_COMEAGAIN (depth-first traversal) data = -1 for 1231 all nodes (that's how you can tell it's the second callback).</p> 1232 1233 <p>Users of this code may put anything they like into the <b>long extra</b> 1234 field. For example, "cp" and "mv" use this to store a dirfd for the destination 1235 directory (and use DIRTREE_COMEAGAIN to get the second callback so they can 1236 close(node->extra) to avoid running out of filehandles). 1237 This field is not directly used by the dirtree code, and 1238 thanks to LP64 it's large enough to store a typecast pointer to an 1239 arbitrary struct.</p> 1240 1241 <p>The return value of the callback combines flags (with boolean or) to tell 1242 the traversal infrastructure how to behave:</p> 1243 1244 <ul> 1245 <li><p><b>DIRTREE_SAVE</b> - Save this node, assembling a tree. (Without 1246 this the struct dirtree is freed after the callback returns. Filtering out 1247 siblings is fine, but discarding a parent while keeping its child leaks 1248 memory.)</p></li> 1249 <li><p><b>DIRTREE_ABORT</b> - Do not examine any more entries in this 1250 directory. (Does not propagate up tree: to abort entire traversal, 1251 return DIRTREE_ABORT from parent callbacks too.)</p></li> 1252 <li><p><b>DIRTREE_RECURSE</b> - Examine directory contents. Ignored for 1253 non-directory entries. The remaining flags only take effect when 1254 recursing into the children of a directory.</p></li> 1255 <li><p><b>DIRTREE_COMEAGAIN</b> - Call the callback a second time after 1256 examining all directory contents, allowing depth-first traversal. 1257 On the second call, dirtree->data = -1.</p></li> 1258 <li><p><b>DIRTREE_SYMFOLLOW</b> - follow symlinks when populating children's 1259 <b>struct stat st</b> (by feeding a nonzero value to the symfollow argument of 1260 dirtree_add_node()), which means DIRTREE_RECURSE treats symlinks to 1261 directories as directories. (Avoiding infinite recursion is the callback's 1262 problem: the non-NULL dirtree->symlink can still distinguish between 1263 them.)</p></li> 1264 </ul> 1265 1266 <p>Each struct dirtree contains three pointers (next, parent, and child) 1267 to other struct dirtree.</p> 1268 1269 <p>The <b>parent</b> pointer indicates the directory 1270 containing this entry; even when not assembling a persistent tree of 1271 nodes the parent entries remain live up to the root of the tree while 1272 child nodes are active. At the top of the tree the parent pointer is 1273 NULL, meaning the node's name[] is either an absolute path or relative 1274 to cwd. The function dirtree_parentfd() gets the directory file descriptor 1275 for use with openat() and friends, returning AT_FDCWD at the top of tree.</p> 1276 1277 <p>The <b>child</b> pointer points to the first node of the list of contents of 1278 this directory. If the directory contains no files, or the entry isn't 1279 a directory, child is NULL.</p> 1280 1281 <p>The <b>next</b> pointer indicates sibling nodes in the same directory as this 1282 node, and since it's the first entry in the struct the llist.c traversal 1283 mechanisms work to iterate over sibling nodes. Each dirtree node is a 1284 single malloc() (even char *symlink points to memory at the end of the node), 1285 so llist_free() works but its callback must descend into child nodes (freeing 1286 a tree, not just a linked list), plus whatever the user stored in extra.</p> 1287 1288 <p>The <b>dirtree_read</b>() function is a simple wrapper, calling <b>dirtree_add_node</b>() 1289 to create a root node relative to the current directory, then calling 1290 <b>handle_callback</b>() on that node (which recurses as instructed by the callback 1291 return flags). Some commands (such as chgrp) bypass this wrapper, for example 1292 to control whether or not to follow symlinks to the root node; symlinks 1293 listed on the command line are often treated differently than symlinks 1294 encountered during recursive directory traversal). 1295 1296 <p>The ls command not only bypasses the wrapper, but never returns 1297 <b>DIRTREE_RECURSE</b> from the callback, instead calling <b>dirtree_recurse</b>() manually 1298 from elsewhere in the program. This gives ls -lR manual control 1299 of traversal order, which is neither depth first nor breadth first but 1300 instead a sort of FIFO order requried by the ls standard.</p> 1301 1302 <a name="toys"> 1303 <h1><a href="#toys">Directory toys/</a></h1> 1304 1305 <p>This directory contains command implementations. Each command is a single 1306 self-contained file. Adding a new command involves adding a single 1307 file, and removing a command involves removing that file. Commands use 1308 shared infrastructure from the lib/ and generated/ directories.</p> 1309 1310 <p>Currently there are five subdirectories under "toys/" containing "posix" 1311 commands described in POSIX-2008, "lsb" commands described in the Linux 1312 Standard Base 4.1, "other" commands not described by either standard, 1313 "pending" commands awaiting cleanup (which default to "n" in menuconfig 1314 because they don't necessarily work right yet), and "example" code showing 1315 how toybox infrastructure works and providing template/skeleton files to 1316 start new commands.</p> 1317 1318 <p>The only difference directory location makes is which menu the command 1319 shows up in during "make menuconfig", the directories are otherwise identical. 1320 Note that the commands exist within a single namespace at runtime, so you can't 1321 have the same command in multiple subdirectories. (The build tries to fail 1322 informatively when you do that.)</p> 1323 1324 <p>There is one more sub-menus in "make menuconfig" containing global 1325 configuration options for toybox. This menu is defined in the top level 1326 Config.in.</p> 1327 1328 <p>See <a href="#adding">adding a new command</a> for details on the 1329 layout of a command file.</p> 1330 1331 <h2>Directory scripts/</h2> 1332 1333 <p>Build infrastructure. The makefile calls scripts/make.sh for "make" 1334 and scripts/install.sh for "make install".</p> 1335 1336 <p>There's also a test suite, "make test" calls make/test.sh, which runs all 1337 the tests in make/test/*. You can run individual tests via 1338 "scripts/test.sh command", or "TEST_HOST=1 scripts/test.sh command" to run 1339 that test against the host implementation instead of the toybox one.</p> 1340 1341 <h3>scripts/cfg2files.sh</h3> 1342 1343 <p>Run .config through this filter to get a list of enabled commands, which 1344 is turned into a list of files in toys via a sed invocation in the top level 1345 Makefile. 1346 </p> 1347 1348 <h2>Directory kconfig/</h2> 1349 1350 <p>Menuconfig infrastructure copied from the Linux kernel. See the 1351 Linux kernel's Documentation/kbuild/kconfig-language.txt</p> 1352 1353 <!-- todo 1354 1355 Better OLDTOY and multiple command explanation. From Config.in: 1356 1357 <p>A command with multiple names (or multiple similar commands implemented in 1358 the same .c file) should have config symbols prefixed with the name of their 1359 C file. I.E. config symbol prefixes are NEWTOY() names. If OLDTOY() names 1360 have config symbols they must be options (symbols with an underscore and 1361 suffix) to the NEWTOY() name. (See generated/toylist.h)</p> 1362 --> 1363 1364 <!--#include file="footer.html" --> 1365
