| 1 | .\" Automatically generated by Pod::Man v1.37, Pod::Parser v1.32 |
| 2 | .\" |
| 3 | .\" Standard preamble: |
| 4 | .\" ======================================================================== |
| 5 | .de Sh \" Subsection heading |
| 6 | .br |
| 7 | .if t .Sp |
| 8 | .ne 5 |
| 9 | .PP |
| 10 | \fB\\$1\fR |
| 11 | .PP |
| 12 | .. |
| 13 | .de Sp \" Vertical space (when we can't use .PP) |
| 14 | .if t .sp .5v |
| 15 | .if n .sp |
| 16 | .. |
| 17 | .de Vb \" Begin verbatim text |
| 18 | .ft CW |
| 19 | .nf |
| 20 | .ne \\$1 |
| 21 | .. |
| 22 | .de Ve \" End verbatim text |
| 23 | .ft R |
| 24 | .fi |
| 25 | .. |
| 26 | .\" Set up some character translations and predefined strings. \*(-- will |
| 27 | .\" give an unbreakable dash, \*(PI will give pi, \*(L" will give a left |
| 28 | .\" double quote, and \*(R" will give a right double quote. | will give a |
| 29 | .\" real vertical bar. \*(C+ will give a nicer C++. Capital omega is used to |
| 30 | .\" do unbreakable dashes and therefore won't be available. \*(C` and \*(C' |
| 31 | .\" expand to `' in nroff, nothing in troff, for use with C<>. |
| 32 | .tr \(*W-|\(bv\*(Tr |
| 33 | .ds C+ C\v'-.1v'\h'-1p'\s-2+\h'-1p'+\s0\v'.1v'\h'-1p' |
| 34 | .ie n \{\ |
| 35 | . ds -- \(*W- |
| 36 | . ds PI pi |
| 37 | . if (\n(.H=4u)&(1m=24u) .ds -- \(*W\h'-12u'\(*W\h'-12u'-\" diablo 10 pitch |
| 38 | . if (\n(.H=4u)&(1m=20u) .ds -- \(*W\h'-12u'\(*W\h'-8u'-\" diablo 12 pitch |
| 39 | . ds L" "" |
| 40 | . ds R" "" |
| 41 | . ds C` "" |
| 42 | . ds C' "" |
| 43 | 'br\} |
| 44 | .el\{\ |
| 45 | . ds -- \|\(em\| |
| 46 | . ds PI \(*p |
| 47 | . ds L" `` |
| 48 | . ds R" '' |
| 49 | 'br\} |
| 50 | .\" |
| 51 | .\" If the F register is turned on, we'll generate index entries on stderr for |
| 52 | .\" titles (.TH), headers (.SH), subsections (.Sh), items (.Ip), and index |
| 53 | .\" entries marked with X<> in POD. Of course, you'll have to process the |
| 54 | .\" output yourself in some meaningful fashion. |
| 55 | .if \nF \{\ |
| 56 | . de IX |
| 57 | . tm Index:\\$1\t\\n%\t"\\$2" |
| 58 | .. |
| 59 | . nr % 0 |
| 60 | . rr F |
| 61 | .\} |
| 62 | .\" |
| 63 | .\" For nroff, turn off justification. Always turn off hyphenation; it makes |
| 64 | .\" way too many mistakes in technical documents. |
| 65 | .hy 0 |
| 66 | .if n .na |
| 67 | .\" |
| 68 | .\" Accent mark definitions (@(#)ms.acc 1.5 88/02/08 SMI; from UCB 4.2). |
| 69 | .\" Fear. Run. Save yourself. No user-serviceable parts. |
| 70 | . \" fudge factors for nroff and troff |
| 71 | .if n \{\ |
| 72 | . ds #H 0 |
| 73 | . ds #V .8m |
| 74 | . ds #F .3m |
| 75 | . ds #[ \f1 |
| 76 | . ds #] \fP |
| 77 | .\} |
| 78 | .if t \{\ |
| 79 | . ds #H ((1u-(\\\\n(.fu%2u))*.13m) |
| 80 | . ds #V .6m |
| 81 | . ds #F 0 |
| 82 | . ds #[ \& |
| 83 | . ds #] \& |
| 84 | .\} |
| 85 | . \" simple accents for nroff and troff |
| 86 | .if n \{\ |
| 87 | . ds ' \& |
| 88 | . ds ` \& |
| 89 | . ds ^ \& |
| 90 | . ds , \& |
| 91 | . ds ~ ~ |
| 92 | . ds / |
| 93 | .\} |
| 94 | .if t \{\ |
| 95 | . ds ' \\k:\h'-(\\n(.wu*8/10-\*(#H)'\'\h"|\\n:u" |
| 96 | . ds ` \\k:\h'-(\\n(.wu*8/10-\*(#H)'\`\h'|\\n:u' |
| 97 | . ds ^ \\k:\h'-(\\n(.wu*10/11-\*(#H)'^\h'|\\n:u' |
| 98 | . ds , \\k:\h'-(\\n(.wu*8/10)',\h'|\\n:u' |
| 99 | . ds ~ \\k:\h'-(\\n(.wu-\*(#H-.1m)'~\h'|\\n:u' |
| 100 | . ds / \\k:\h'-(\\n(.wu*8/10-\*(#H)'\z\(sl\h'|\\n:u' |
| 101 | .\} |
| 102 | . \" troff and (daisy-wheel) nroff accents |
| 103 | .ds : \\k:\h'-(\\n(.wu*8/10-\*(#H+.1m+\*(#F)'\v'-\*(#V'\z.\h'.2m+\*(#F'.\h'|\\n:u'\v'\*(#V' |
| 104 | .ds 8 \h'\*(#H'\(*b\h'-\*(#H' |
| 105 | .ds o \\k:\h'-(\\n(.wu+\w'\(de'u-\*(#H)/2u'\v'-.3n'\*(#[\z\(de\v'.3n'\h'|\\n:u'\*(#] |
| 106 | .ds d- \h'\*(#H'\(pd\h'-\w'~'u'\v'-.25m'\f2\(hy\fP\v'.25m'\h'-\*(#H' |
| 107 | .ds D- D\\k:\h'-\w'D'u'\v'-.11m'\z\(hy\v'.11m'\h'|\\n:u' |
| 108 | .ds th \*(#[\v'.3m'\s+1I\s-1\v'-.3m'\h'-(\w'I'u*2/3)'\s-1o\s+1\*(#] |
| 109 | .ds Th \*(#[\s+2I\s-2\h'-\w'I'u*3/5'\v'-.3m'o\v'.3m'\*(#] |
| 110 | .ds ae a\h'-(\w'a'u*4/10)'e |
| 111 | .ds Ae A\h'-(\w'A'u*4/10)'E |
| 112 | . \" corrections for vroff |
| 113 | .if v .ds ~ \\k:\h'-(\\n(.wu*9/10-\*(#H)'\s-2\u~\d\s+2\h'|\\n:u' |
| 114 | .if v .ds ^ \\k:\h'-(\\n(.wu*10/11-\*(#H)'\v'-.4m'^\v'.4m'\h'|\\n:u' |
| 115 | . \" for low resolution devices (crt and lpr) |
| 116 | .if \n(.H>23 .if \n(.V>19 \ |
| 117 | \{\ |
| 118 | . ds : e |
| 119 | . ds 8 ss |
| 120 | . ds o a |
| 121 | . ds d- d\h'-1'\(ga |
| 122 | . ds D- D\h'-1'\(hy |
| 123 | . ds th \o'bp' |
| 124 | . ds Th \o'LP' |
| 125 | . ds ae ae |
| 126 | . ds Ae AE |
| 127 | .\} |
| 128 | .rm #[ #] #H #V #F C |
| 129 | .\" ======================================================================== |
| 130 | .\" |
| 131 | .IX Title "PERLPACKTUT 1" |
| 132 | .TH PERLPACKTUT 1 "2006-01-07" "perl v5.8.8" "Perl Programmers Reference Guide" |
| 133 | .SH "NAME" |
| 134 | perlpacktut \- tutorial on \f(CW\*(C`pack\*(C'\fR and \f(CW\*(C`unpack\*(C'\fR |
| 135 | .SH "DESCRIPTION" |
| 136 | .IX Header "DESCRIPTION" |
| 137 | \&\f(CW\*(C`pack\*(C'\fR and \f(CW\*(C`unpack\*(C'\fR are two functions for transforming data according |
| 138 | to a user-defined template, between the guarded way Perl stores values |
| 139 | and some well-defined representation as might be required in the |
| 140 | environment of a Perl program. Unfortunately, they're also two of |
| 141 | the most misunderstood and most often overlooked functions that Perl |
| 142 | provides. This tutorial will demystify them for you. |
| 143 | .SH "The Basic Principle" |
| 144 | .IX Header "The Basic Principle" |
| 145 | Most programming languages don't shelter the memory where variables are |
| 146 | stored. In C, for instance, you can take the address of some variable, |
| 147 | and the \f(CW\*(C`sizeof\*(C'\fR operator tells you how many bytes are allocated to |
| 148 | the variable. Using the address and the size, you may access the storage |
| 149 | to your heart's content. |
| 150 | .PP |
| 151 | In Perl, you just can't access memory at random, but the structural and |
| 152 | representational conversion provided by \f(CW\*(C`pack\*(C'\fR and \f(CW\*(C`unpack\*(C'\fR is an |
| 153 | excellent alternative. The \f(CW\*(C`pack\*(C'\fR function converts values to a byte |
| 154 | sequence containing representations according to a given specification, |
| 155 | the so-called \*(L"template\*(R" argument. \f(CW\*(C`unpack\*(C'\fR is the reverse process, |
| 156 | deriving some values from the contents of a string of bytes. (Be cautioned, |
| 157 | however, that not all that has been packed together can be neatly unpacked \- |
| 158 | a very common experience as seasoned travellers are likely to confirm.) |
| 159 | .PP |
| 160 | Why, you may ask, would you need a chunk of memory containing some values |
| 161 | in binary representation? One good reason is input and output accessing |
| 162 | some file, a device, or a network connection, whereby this binary |
| 163 | representation is either forced on you or will give you some benefit |
| 164 | in processing. Another cause is passing data to some system call that |
| 165 | is not available as a Perl function: \f(CW\*(C`syscall\*(C'\fR requires you to provide |
| 166 | parameters stored in the way it happens in a C program. Even text processing |
| 167 | (as shown in the next section) may be simplified with judicious usage |
| 168 | of these two functions. |
| 169 | .PP |
| 170 | To see how (un)packing works, we'll start with a simple template |
| 171 | code where the conversion is in low gear: between the contents of a byte |
| 172 | sequence and a string of hexadecimal digits. Let's use \f(CW\*(C`unpack\*(C'\fR, since |
| 173 | this is likely to remind you of a dump program, or some desperate last |
| 174 | message unfortunate programs are wont to throw at you before they expire |
| 175 | into the wild blue yonder. Assuming that the variable \f(CW$mem\fR holds a |
| 176 | sequence of bytes that we'd like to inspect without assuming anything |
| 177 | about its meaning, we can write |
| 178 | .PP |
| 179 | .Vb 2 |
| 180 | \& my( $hex ) = unpack( 'H*', $mem ); |
| 181 | \& print "$hex\en"; |
| 182 | .Ve |
| 183 | .PP |
| 184 | whereupon we might see something like this, with each pair of hex digits |
| 185 | corresponding to a byte: |
| 186 | .PP |
| 187 | .Vb 1 |
| 188 | \& 41204d414e204120504c414e20412043414e414c2050414e414d41 |
| 189 | .Ve |
| 190 | .PP |
| 191 | What was in this chunk of memory? Numbers, characters, or a mixture of |
| 192 | both? Assuming that we're on a computer where \s-1ASCII\s0 (or some similar) |
| 193 | encoding is used: hexadecimal values in the range \f(CW0x40\fR \- \f(CW0x5A\fR |
| 194 | indicate an uppercase letter, and \f(CW0x20\fR encodes a space. So we might |
| 195 | assume it is a piece of text, which some are able to read like a tabloid; |
| 196 | but others will have to get hold of an \s-1ASCII\s0 table and relive that |
| 197 | firstgrader feeling. Not caring too much about which way to read this, |
| 198 | we note that \f(CW\*(C`unpack\*(C'\fR with the template code \f(CW\*(C`H\*(C'\fR converts the contents |
| 199 | of a sequence of bytes into the customary hexadecimal notation. Since |
| 200 | \&\*(L"a sequence of\*(R" is a pretty vague indication of quantity, \f(CW\*(C`H\*(C'\fR has been |
| 201 | defined to convert just a single hexadecimal digit unless it is followed |
| 202 | by a repeat count. An asterisk for the repeat count means to use whatever |
| 203 | remains. |
| 204 | .PP |
| 205 | The inverse operation \- packing byte contents from a string of hexadecimal |
| 206 | digits \- is just as easily written. For instance: |
| 207 | .PP |
| 208 | .Vb 2 |
| 209 | \& my $s = pack( 'H2' x 10, map { "3$_" } ( 0..9 ) ); |
| 210 | \& print "$s\en"; |
| 211 | .Ve |
| 212 | .PP |
| 213 | Since we feed a list of ten 2\-digit hexadecimal strings to \f(CW\*(C`pack\*(C'\fR, the |
| 214 | pack template should contain ten pack codes. If this is run on a computer |
| 215 | with \s-1ASCII\s0 character coding, it will print \f(CW0123456789\fR. |
| 216 | .SH "Packing Text" |
| 217 | .IX Header "Packing Text" |
| 218 | Let's suppose you've got to read in a data file like this: |
| 219 | .PP |
| 220 | .Vb 4 |
| 221 | \& Date |Description | Income|Expenditure |
| 222 | \& 01/24/2001 Ahmed's Camel Emporium 1147.99 |
| 223 | \& 01/28/2001 Flea spray 24.99 |
| 224 | \& 01/29/2001 Camel rides to tourists 235.00 |
| 225 | .Ve |
| 226 | .PP |
| 227 | How do we do it? You might think first to use \f(CW\*(C`split\*(C'\fR; however, since |
| 228 | \&\f(CW\*(C`split\*(C'\fR collapses blank fields, you'll never know whether a record was |
| 229 | income or expenditure. Oops. Well, you could always use \f(CW\*(C`substr\*(C'\fR: |
| 230 | .PP |
| 231 | .Vb 7 |
| 232 | \& while (<>) { |
| 233 | \& my $date = substr($_, 0, 11); |
| 234 | \& my $desc = substr($_, 12, 27); |
| 235 | \& my $income = substr($_, 40, 7); |
| 236 | \& my $expend = substr($_, 52, 7); |
| 237 | \& ... |
| 238 | \& } |
| 239 | .Ve |
| 240 | .PP |
| 241 | It's not really a barrel of laughs, is it? In fact, it's worse than it |
| 242 | may seem; the eagle-eyed may notice that the first field should only be |
| 243 | 10 characters wide, and the error has propagated right through the other |
| 244 | numbers \- which we've had to count by hand. So it's error-prone as well |
| 245 | as horribly unfriendly. |
| 246 | .PP |
| 247 | Or maybe we could use regular expressions: |
| 248 | .PP |
| 249 | .Vb 5 |
| 250 | \& while (<>) { |
| 251 | \& my($date, $desc, $income, $expend) = |
| 252 | \& m|(\ed\ed/\ed\ed/\ed{4}) (.{27}) (.{7})(.*)|; |
| 253 | \& ... |
| 254 | \& } |
| 255 | .Ve |
| 256 | .PP |
| 257 | Urgh. Well, it's a bit better, but \- well, would you want to maintain |
| 258 | that? |
| 259 | .PP |
| 260 | Hey, isn't Perl supposed to make this sort of thing easy? Well, it does, |
| 261 | if you use the right tools. \f(CW\*(C`pack\*(C'\fR and \f(CW\*(C`unpack\*(C'\fR are designed to help |
| 262 | you out when dealing with fixed-width data like the above. Let's have a |
| 263 | look at a solution with \f(CW\*(C`unpack\*(C'\fR: |
| 264 | .PP |
| 265 | .Vb 4 |
| 266 | \& while (<>) { |
| 267 | \& my($date, $desc, $income, $expend) = unpack("A10xA27xA7A*", $_); |
| 268 | \& ... |
| 269 | \& } |
| 270 | .Ve |
| 271 | .PP |
| 272 | That looks a bit nicer; but we've got to take apart that weird template. |
| 273 | Where did I pull that out of? |
| 274 | .PP |
| 275 | \&\s-1OK\s0, let's have a look at some of our data again; in fact, we'll include |
| 276 | the headers, and a handy ruler so we can keep track of where we are. |
| 277 | .PP |
| 278 | .Vb 5 |
| 279 | \& 1 2 3 4 5 |
| 280 | \& 1234567890123456789012345678901234567890123456789012345678 |
| 281 | \& Date |Description | Income|Expenditure |
| 282 | \& 01/28/2001 Flea spray 24.99 |
| 283 | \& 01/29/2001 Camel rides to tourists 235.00 |
| 284 | .Ve |
| 285 | .PP |
| 286 | From this, we can see that the date column stretches from column 1 to |
| 287 | column 10 \- ten characters wide. The \f(CW\*(C`pack\*(C'\fR\-ese for \*(L"character\*(R" is |
| 288 | \&\f(CW\*(C`A\*(C'\fR, and ten of them are \f(CW\*(C`A10\*(C'\fR. So if we just wanted to extract the |
| 289 | dates, we could say this: |
| 290 | .PP |
| 291 | .Vb 1 |
| 292 | \& my($date) = unpack("A10", $_); |
| 293 | .Ve |
| 294 | .PP |
| 295 | \&\s-1OK\s0, what's next? Between the date and the description is a blank column; |
| 296 | we want to skip over that. The \f(CW\*(C`x\*(C'\fR template means \*(L"skip forward\*(R", so we |
| 297 | want one of those. Next, we have another batch of characters, from 12 to |
| 298 | 38. That's 27 more characters, hence \f(CW\*(C`A27\*(C'\fR. (Don't make the fencepost |
| 299 | error \- there are 27 characters between 12 and 38, not 26. Count 'em!) |
| 300 | .PP |
| 301 | Now we skip another character and pick up the next 7 characters: |
| 302 | .PP |
| 303 | .Vb 1 |
| 304 | \& my($date,$description,$income) = unpack("A10xA27xA7", $_); |
| 305 | .Ve |
| 306 | .PP |
| 307 | Now comes the clever bit. Lines in our ledger which are just income and |
| 308 | not expenditure might end at column 46. Hence, we don't want to tell our |
| 309 | \&\f(CW\*(C`unpack\*(C'\fR pattern that we \fBneed\fR to find another 12 characters; we'll |
| 310 | just say \*(L"if there's anything left, take it\*(R". As you might guess from |
| 311 | regular expressions, that's what the \f(CW\*(C`*\*(C'\fR means: \*(L"use everything |
| 312 | remaining\*(R". |
| 313 | .IP "\(bu" 3 |
| 314 | Be warned, though, that unlike regular expressions, if the \f(CW\*(C`unpack\*(C'\fR |
| 315 | template doesn't match the incoming data, Perl will scream and die. |
| 316 | .PP |
| 317 | Hence, putting it all together: |
| 318 | .PP |
| 319 | .Vb 1 |
| 320 | \& my($date,$description,$income,$expend) = unpack("A10xA27xA7xA*", $_); |
| 321 | .Ve |
| 322 | .PP |
| 323 | Now, that's our data parsed. I suppose what we might want to do now is |
| 324 | total up our income and expenditure, and add another line to the end of |
| 325 | our ledger \- in the same format \- saying how much we've brought in and |
| 326 | how much we've spent: |
| 327 | .PP |
| 328 | .Vb 5 |
| 329 | \& while (<>) { |
| 330 | \& my($date, $desc, $income, $expend) = unpack("A10xA27xA7xA*", $_); |
| 331 | \& $tot_income += $income; |
| 332 | \& $tot_expend += $expend; |
| 333 | \& } |
| 334 | .Ve |
| 335 | .PP |
| 336 | .Vb 2 |
| 337 | \& $tot_income = sprintf("%.2f", $tot_income); # Get them into |
| 338 | \& $tot_expend = sprintf("%.2f", $tot_expend); # "financial" format |
| 339 | .Ve |
| 340 | .PP |
| 341 | .Vb 1 |
| 342 | \& $date = POSIX::strftime("%m/%d/%Y", localtime); |
| 343 | .Ve |
| 344 | .PP |
| 345 | .Vb 1 |
| 346 | \& # OK, let's go: |
| 347 | .Ve |
| 348 | .PP |
| 349 | .Vb 1 |
| 350 | \& print pack("A10xA27xA7xA*", $date, "Totals", $tot_income, $tot_expend); |
| 351 | .Ve |
| 352 | .PP |
| 353 | Oh, hmm. That didn't quite work. Let's see what happened: |
| 354 | .PP |
| 355 | .Vb 4 |
| 356 | \& 01/24/2001 Ahmed's Camel Emporium 1147.99 |
| 357 | \& 01/28/2001 Flea spray 24.99 |
| 358 | \& 01/29/2001 Camel rides to tourists 1235.00 |
| 359 | \& 03/23/2001Totals 1235.001172.98 |
| 360 | .Ve |
| 361 | .PP |
| 362 | \&\s-1OK\s0, it's a start, but what happened to the spaces? We put \f(CW\*(C`x\*(C'\fR, didn't |
| 363 | we? Shouldn't it skip forward? Let's look at what \*(L"pack\*(R" in perlfunc says: |
| 364 | .PP |
| 365 | .Vb 1 |
| 366 | \& x A null byte. |
| 367 | .Ve |
| 368 | .PP |
| 369 | Urgh. No wonder. There's a big difference between \*(L"a null byte\*(R", |
| 370 | character zero, and \*(L"a space\*(R", character 32. Perl's put something |
| 371 | between the date and the description \- but unfortunately, we can't see |
| 372 | it! |
| 373 | .PP |
| 374 | What we actually need to do is expand the width of the fields. The \f(CW\*(C`A\*(C'\fR |
| 375 | format pads any non-existent characters with spaces, so we can use the |
| 376 | additional spaces to line up our fields, like this: |
| 377 | .PP |
| 378 | .Vb 1 |
| 379 | \& print pack("A11 A28 A8 A*", $date, "Totals", $tot_income, $tot_expend); |
| 380 | .Ve |
| 381 | .PP |
| 382 | (Note that you can put spaces in the template to make it more readable, |
| 383 | but they don't translate to spaces in the output.) Here's what we got |
| 384 | this time: |
| 385 | .PP |
| 386 | .Vb 4 |
| 387 | \& 01/24/2001 Ahmed's Camel Emporium 1147.99 |
| 388 | \& 01/28/2001 Flea spray 24.99 |
| 389 | \& 01/29/2001 Camel rides to tourists 1235.00 |
| 390 | \& 03/23/2001 Totals 1235.00 1172.98 |
| 391 | .Ve |
| 392 | .PP |
| 393 | That's a bit better, but we still have that last column which needs to |
| 394 | be moved further over. There's an easy way to fix this up: |
| 395 | unfortunately, we can't get \f(CW\*(C`pack\*(C'\fR to right-justify our fields, but we |
| 396 | can get \f(CW\*(C`sprintf\*(C'\fR to do it: |
| 397 | .PP |
| 398 | .Vb 4 |
| 399 | \& $tot_income = sprintf("%.2f", $tot_income); |
| 400 | \& $tot_expend = sprintf("%12.2f", $tot_expend); |
| 401 | \& $date = POSIX::strftime("%m/%d/%Y", localtime); |
| 402 | \& print pack("A11 A28 A8 A*", $date, "Totals", $tot_income, $tot_expend); |
| 403 | .Ve |
| 404 | .PP |
| 405 | This time we get the right answer: |
| 406 | .PP |
| 407 | .Vb 3 |
| 408 | \& 01/28/2001 Flea spray 24.99 |
| 409 | \& 01/29/2001 Camel rides to tourists 1235.00 |
| 410 | \& 03/23/2001 Totals 1235.00 1172.98 |
| 411 | .Ve |
| 412 | .PP |
| 413 | So that's how we consume and produce fixed-width data. Let's recap what |
| 414 | we've seen of \f(CW\*(C`pack\*(C'\fR and \f(CW\*(C`unpack\*(C'\fR so far: |
| 415 | .IP "\(bu" 3 |
| 416 | Use \f(CW\*(C`pack\*(C'\fR to go from several pieces of data to one fixed-width |
| 417 | version; use \f(CW\*(C`unpack\*(C'\fR to turn a fixed-width-format string into several |
| 418 | pieces of data. |
| 419 | .IP "\(bu" 3 |
| 420 | The pack format \f(CW\*(C`A\*(C'\fR means \*(L"any character\*(R"; if you're \f(CW\*(C`pack\*(C'\fRing and |
| 421 | you've run out of things to pack, \f(CW\*(C`pack\*(C'\fR will fill the rest up with |
| 422 | spaces. |
| 423 | .IP "\(bu" 3 |
| 424 | \&\f(CW\*(C`x\*(C'\fR means \*(L"skip a byte\*(R" when \f(CW\*(C`unpack\*(C'\fRing; when \f(CW\*(C`pack\*(C'\fRing, it means |
| 425 | \&\*(L"introduce a null byte\*(R" \- that's probably not what you mean if you're |
| 426 | dealing with plain text. |
| 427 | .IP "\(bu" 3 |
| 428 | You can follow the formats with numbers to say how many characters |
| 429 | should be affected by that format: \f(CW\*(C`A12\*(C'\fR means \*(L"take 12 characters\*(R"; |
| 430 | \&\f(CW\*(C`x6\*(C'\fR means \*(L"skip 6 bytes\*(R" or \*(L"character 0, 6 times\*(R". |
| 431 | .IP "\(bu" 3 |
| 432 | Instead of a number, you can use \f(CW\*(C`*\*(C'\fR to mean \*(L"consume everything else |
| 433 | left\*(R". |
| 434 | .Sp |
| 435 | \&\fBWarning\fR: when packing multiple pieces of data, \f(CW\*(C`*\*(C'\fR only means |
| 436 | \&\*(L"consume all of the current piece of data\*(R". That's to say |
| 437 | .Sp |
| 438 | .Vb 1 |
| 439 | \& pack("A*A*", $one, $two) |
| 440 | .Ve |
| 441 | .Sp |
| 442 | packs all of \f(CW$one\fR into the first \f(CW\*(C`A*\*(C'\fR and then all of \f(CW$two\fR into |
| 443 | the second. This is a general principle: each format character |
| 444 | corresponds to one piece of data to be \f(CW\*(C`pack\*(C'\fRed. |
| 445 | .SH "Packing Numbers" |
| 446 | .IX Header "Packing Numbers" |
| 447 | So much for textual data. Let's get onto the meaty stuff that \f(CW\*(C`pack\*(C'\fR |
| 448 | and \f(CW\*(C`unpack\*(C'\fR are best at: handling binary formats for numbers. There is, |
| 449 | of course, not just one binary format \- life would be too simple \- but |
| 450 | Perl will do all the finicky labor for you. |
| 451 | .Sh "Integers" |
| 452 | .IX Subsection "Integers" |
| 453 | Packing and unpacking numbers implies conversion to and from some |
| 454 | \&\fIspecific\fR binary representation. Leaving floating point numbers |
| 455 | aside for the moment, the salient properties of any such representation |
| 456 | are: |
| 457 | .IP "\(bu" 4 |
| 458 | the number of bytes used for storing the integer, |
| 459 | .IP "\(bu" 4 |
| 460 | whether the contents are interpreted as a signed or unsigned number, |
| 461 | .IP "\(bu" 4 |
| 462 | the byte ordering: whether the first byte is the least or most |
| 463 | significant byte (or: little-endian or big\-endian, respectively). |
| 464 | .PP |
| 465 | So, for instance, to pack 20302 to a signed 16 bit integer in your |
| 466 | computer's representation you write |
| 467 | .PP |
| 468 | .Vb 1 |
| 469 | \& my $ps = pack( 's', 20302 ); |
| 470 | .Ve |
| 471 | .PP |
| 472 | Again, the result is a string, now containing 2 bytes. If you print |
| 473 | this string (which is, generally, not recommended) you might see |
| 474 | \&\f(CW\*(C`ON\*(C'\fR or \f(CW\*(C`NO\*(C'\fR (depending on your system's byte ordering) \- or something |
| 475 | entirely different if your computer doesn't use \s-1ASCII\s0 character encoding. |
| 476 | Unpacking \f(CW$ps\fR with the same template returns the original integer value: |
| 477 | .PP |
| 478 | .Vb 1 |
| 479 | \& my( $s ) = unpack( 's', $ps ); |
| 480 | .Ve |
| 481 | .PP |
| 482 | This is true for all numeric template codes. But don't expect miracles: |
| 483 | if the packed value exceeds the allotted byte capacity, high order bits |
| 484 | are silently discarded, and unpack certainly won't be able to pull them |
| 485 | back out of some magic hat. And, when you pack using a signed template |
| 486 | code such as \f(CW\*(C`s\*(C'\fR, an excess value may result in the sign bit |
| 487 | getting set, and unpacking this will smartly return a negative value. |
| 488 | .PP |
| 489 | 16 bits won't get you too far with integers, but there is \f(CW\*(C`l\*(C'\fR and \f(CW\*(C`L\*(C'\fR |
| 490 | for signed and unsigned 32\-bit integers. And if this is not enough and |
| 491 | your system supports 64 bit integers you can push the limits much closer |
| 492 | to infinity with pack codes \f(CW\*(C`q\*(C'\fR and \f(CW\*(C`Q\*(C'\fR. A notable exception is provided |
| 493 | by pack codes \f(CW\*(C`i\*(C'\fR and \f(CW\*(C`I\*(C'\fR for signed and unsigned integers of the |
| 494 | \&\*(L"local custom\*(R" variety: Such an integer will take up as many bytes as |
| 495 | a local C compiler returns for \f(CW\*(C`sizeof(int)\*(C'\fR, but it'll use \fIat least\fR |
| 496 | 32 bits. |
| 497 | .PP |
| 498 | Each of the integer pack codes \f(CW\*(C`sSlLqQ\*(C'\fR results in a fixed number of bytes, |
| 499 | no matter where you execute your program. This may be useful for some |
| 500 | applications, but it does not provide for a portable way to pass data |
| 501 | structures between Perl and C programs (bound to happen when you call |
| 502 | \&\s-1XS\s0 extensions or the Perl function \f(CW\*(C`syscall\*(C'\fR), or when you read or |
| 503 | write binary files. What you'll need in this case are template codes that |
| 504 | depend on what your local C compiler compiles when you code \f(CW\*(C`short\*(C'\fR or |
| 505 | \&\f(CW\*(C`unsigned long\*(C'\fR, for instance. These codes and their corresponding |
| 506 | byte lengths are shown in the table below. Since the C standard leaves |
| 507 | much leeway with respect to the relative sizes of these data types, actual |
| 508 | values may vary, and that's why the values are given as expressions in |
| 509 | C and Perl. (If you'd like to use values from \f(CW%Config\fR in your program |
| 510 | you have to import it with \f(CW\*(C`use Config\*(C'\fR.) |
| 511 | .PP |
| 512 | .Vb 5 |
| 513 | \& signed unsigned byte length in C byte length in Perl |
| 514 | \& s! S! sizeof(short) $Config{shortsize} |
| 515 | \& i! I! sizeof(int) $Config{intsize} |
| 516 | \& l! L! sizeof(long) $Config{longsize} |
| 517 | \& q! Q! sizeof(long long) $Config{longlongsize} |
| 518 | .Ve |
| 519 | .PP |
| 520 | The \f(CW\*(C`i!\*(C'\fR and \f(CW\*(C`I!\*(C'\fR codes aren't different from \f(CW\*(C`i\*(C'\fR and \f(CW\*(C`I\*(C'\fR; they are |
| 521 | tolerated for completeness' sake. |
| 522 | .Sh "Unpacking a Stack Frame" |
| 523 | .IX Subsection "Unpacking a Stack Frame" |
| 524 | Requesting a particular byte ordering may be necessary when you work with |
| 525 | binary data coming from some specific architecture whereas your program could |
| 526 | run on a totally different system. As an example, assume you have 24 bytes |
| 527 | containing a stack frame as it happens on an Intel 8086: |
| 528 | .PP |
| 529 | .Vb 11 |
| 530 | \& +---------+ +----+----+ +---------+ |
| 531 | \& TOS: | IP | TOS+4:| FL | FH | FLAGS TOS+14:| SI | |
| 532 | \& +---------+ +----+----+ +---------+ |
| 533 | \& | CS | | AL | AH | AX | DI | |
| 534 | \& +---------+ +----+----+ +---------+ |
| 535 | \& | BL | BH | BX | BP | |
| 536 | \& +----+----+ +---------+ |
| 537 | \& | CL | CH | CX | DS | |
| 538 | \& +----+----+ +---------+ |
| 539 | \& | DL | DH | DX | ES | |
| 540 | \& +----+----+ +---------+ |
| 541 | .Ve |
| 542 | .PP |
| 543 | First, we note that this time-honored 16\-bit \s-1CPU\s0 uses little-endian order, |
| 544 | and that's why the low order byte is stored at the lower address. To |
| 545 | unpack such a (signed) short we'll have to use code \f(CW\*(C`v\*(C'\fR. A repeat |
| 546 | count unpacks all 12 shorts: |
| 547 | .PP |
| 548 | .Vb 2 |
| 549 | \& my( $ip, $cs, $flags, $ax, $bx, $cd, $dx, $si, $di, $bp, $ds, $es ) = |
| 550 | \& unpack( 'v12', $frame ); |
| 551 | .Ve |
| 552 | .PP |
| 553 | Alternatively, we could have used \f(CW\*(C`C\*(C'\fR to unpack the individually |
| 554 | accessible byte registers \s-1FL\s0, \s-1FH\s0, \s-1AL\s0, \s-1AH\s0, etc.: |
| 555 | .PP |
| 556 | .Vb 2 |
| 557 | \& my( $fl, $fh, $al, $ah, $bl, $bh, $cl, $ch, $dl, $dh ) = |
| 558 | \& unpack( 'C10', substr( $frame, 4, 10 ) ); |
| 559 | .Ve |
| 560 | .PP |
| 561 | It would be nice if we could do this in one fell swoop: unpack a short, |
| 562 | back up a little, and then unpack 2 bytes. Since Perl \fIis\fR nice, it |
| 563 | proffers the template code \f(CW\*(C`X\*(C'\fR to back up one byte. Putting this all |
| 564 | together, we may now write: |
| 565 | .PP |
| 566 | .Vb 5 |
| 567 | \& my( $ip, $cs, |
| 568 | \& $flags,$fl,$fh, |
| 569 | \& $ax,$al,$ah, $bx,$bl,$bh, $cx,$cl,$ch, $dx,$dl,$dh, |
| 570 | \& $si, $di, $bp, $ds, $es ) = |
| 571 | \& unpack( 'v2' . ('vXXCC' x 5) . 'v5', $frame ); |
| 572 | .Ve |
| 573 | .PP |
| 574 | (The clumsy construction of the template can be avoided \- just read on!) |
| 575 | .PP |
| 576 | We've taken some pains to construct the template so that it matches |
| 577 | the contents of our frame buffer. Otherwise we'd either get undefined values, |
| 578 | or \f(CW\*(C`unpack\*(C'\fR could not unpack all. If \f(CW\*(C`pack\*(C'\fR runs out of items, it will |
| 579 | supply null strings (which are coerced into zeroes whenever the pack code |
| 580 | says so). |
| 581 | .Sh "How to Eat an Egg on a Net" |
| 582 | .IX Subsection "How to Eat an Egg on a Net" |
| 583 | The pack code for big-endian (high order byte at the lowest address) is |
| 584 | \&\f(CW\*(C`n\*(C'\fR for 16 bit and \f(CW\*(C`N\*(C'\fR for 32 bit integers. You use these codes |
| 585 | if you know that your data comes from a compliant architecture, but, |
| 586 | surprisingly enough, you should also use these pack codes if you |
| 587 | exchange binary data, across the network, with some system that you |
| 588 | know next to nothing about. The simple reason is that this |
| 589 | order has been chosen as the \fInetwork order\fR, and all standard-fearing |
| 590 | programs ought to follow this convention. (This is, of course, a stern |
| 591 | backing for one of the Lilliputian parties and may well influence the |
| 592 | political development there.) So, if the protocol expects you to send |
| 593 | a message by sending the length first, followed by just so many bytes, |
| 594 | you could write: |
| 595 | .PP |
| 596 | .Vb 1 |
| 597 | \& my $buf = pack( 'N', length( $msg ) ) . $msg; |
| 598 | .Ve |
| 599 | .PP |
| 600 | or even: |
| 601 | .PP |
| 602 | .Vb 1 |
| 603 | \& my $buf = pack( 'NA*', length( $msg ), $msg ); |
| 604 | .Ve |
| 605 | .PP |
| 606 | and pass \f(CW$buf\fR to your send routine. Some protocols demand that the |
| 607 | count should include the length of the count itself: then just add 4 |
| 608 | to the data length. (But make sure to read \*(L"Lengths and Widths\*(R" before |
| 609 | you really code this!) |
| 610 | .Sh "Floating point Numbers" |
| 611 | .IX Subsection "Floating point Numbers" |
| 612 | For packing floating point numbers you have the choice between the |
| 613 | pack codes \f(CW\*(C`f\*(C'\fR and \f(CW\*(C`d\*(C'\fR which pack into (or unpack from) single-precision or |
| 614 | double-precision representation as it is provided by your system. (There |
| 615 | is no such thing as a network representation for reals, so if you want |
| 616 | to send your real numbers across computer boundaries, you'd better stick |
| 617 | to \s-1ASCII\s0 representation, unless you're absolutely sure what's on the other |
| 618 | end of the line.) |
| 619 | .SH "Exotic Templates" |
| 620 | .IX Header "Exotic Templates" |
| 621 | .Sh "Bit Strings" |
| 622 | .IX Subsection "Bit Strings" |
| 623 | Bits are the atoms in the memory world. Access to individual bits may |
| 624 | have to be used either as a last resort or because it is the most |
| 625 | convenient way to handle your data. Bit string (un)packing converts |
| 626 | between strings containing a series of \f(CW0\fR and \f(CW1\fR characters and |
| 627 | a sequence of bytes each containing a group of 8 bits. This is almost |
| 628 | as simple as it sounds, except that there are two ways the contents of |
| 629 | a byte may be written as a bit string. Let's have a look at an annotated |
| 630 | byte: |
| 631 | .PP |
| 632 | .Vb 5 |
| 633 | \& 7 6 5 4 3 2 1 0 |
| 634 | \& +-----------------+ |
| 635 | \& | 1 0 0 0 1 1 0 0 | |
| 636 | \& +-----------------+ |
| 637 | \& MSB LSB |
| 638 | .Ve |
| 639 | .PP |
| 640 | It's egg-eating all over again: Some think that as a bit string this should |
| 641 | be written \*(L"10001100\*(R" i.e. beginning with the most significant bit, others |
| 642 | insist on \*(L"00110001\*(R". Well, Perl isn't biased, so that's why we have two bit |
| 643 | string codes: |
| 644 | .PP |
| 645 | .Vb 2 |
| 646 | \& $byte = pack( 'B8', '10001100' ); # start with MSB |
| 647 | \& $byte = pack( 'b8', '00110001' ); # start with LSB |
| 648 | .Ve |
| 649 | .PP |
| 650 | It is not possible to pack or unpack bit fields \- just integral bytes. |
| 651 | \&\f(CW\*(C`pack\*(C'\fR always starts at the next byte boundary and \*(L"rounds up\*(R" to the |
| 652 | next multiple of 8 by adding zero bits as required. (If you do want bit |
| 653 | fields, there is \*(L"vec\*(R" in perlfunc. Or you could implement bit field |
| 654 | handling at the character string level, using split, substr, and |
| 655 | concatenation on unpacked bit strings.) |
| 656 | .PP |
| 657 | To illustrate unpacking for bit strings, we'll decompose a simple |
| 658 | status register (a \*(L"\-\*(R" stands for a \*(L"reserved\*(R" bit): |
| 659 | .PP |
| 660 | .Vb 4 |
| 661 | \& +-----------------+-----------------+ |
| 662 | \& | S Z - A - P - C | - - - - O D I T | |
| 663 | \& +-----------------+-----------------+ |
| 664 | \& MSB LSB MSB LSB |
| 665 | .Ve |
| 666 | .PP |
| 667 | Converting these two bytes to a string can be done with the unpack |
| 668 | template \f(CW'b16'\fR. To obtain the individual bit values from the bit |
| 669 | string we use \f(CW\*(C`split\*(C'\fR with the \*(L"empty\*(R" separator pattern which dissects |
| 670 | into individual characters. Bit values from the \*(L"reserved\*(R" positions are |
| 671 | simply assigned to \f(CW\*(C`undef\*(C'\fR, a convenient notation for \*(L"I don't care where |
| 672 | this goes\*(R". |
| 673 | .PP |
| 674 | .Vb 3 |
| 675 | \& ($carry, undef, $parity, undef, $auxcarry, undef, $zero, $sign, |
| 676 | \& $trace, $interrupt, $direction, $overflow) = |
| 677 | \& split( //, unpack( 'b16', $status ) ); |
| 678 | .Ve |
| 679 | .PP |
| 680 | We could have used an unpack template \f(CW'b12'\fR just as well, since the |
| 681 | last 4 bits can be ignored anyway. |
| 682 | .Sh "Uuencoding" |
| 683 | .IX Subsection "Uuencoding" |
| 684 | Another odd-man-out in the template alphabet is \f(CW\*(C`u\*(C'\fR, which packs an |
| 685 | \&\*(L"uuencoded string\*(R". (\*(L"uu\*(R" is short for Unix\-to\-Unix.) Chances are that |
| 686 | you won't ever need this encoding technique which was invented to overcome |
| 687 | the shortcomings of old-fashioned transmission mediums that do not support |
| 688 | other than simple \s-1ASCII\s0 data. The essential recipe is simple: Take three |
| 689 | bytes, or 24 bits. Split them into 4 six\-packs, adding a space (0x20) to |
| 690 | each. Repeat until all of the data is blended. Fold groups of 4 bytes into |
| 691 | lines no longer than 60 and garnish them in front with the original byte count |
| 692 | (incremented by 0x20) and a \f(CW"\en"\fR at the end. \- The \f(CW\*(C`pack\*(C'\fR chef will |
| 693 | prepare this for you, a la minute, when you select pack code \f(CW\*(C`u\*(C'\fR on the menu: |
| 694 | .PP |
| 695 | .Vb 1 |
| 696 | \& my $uubuf = pack( 'u', $bindat ); |
| 697 | .Ve |
| 698 | .PP |
| 699 | A repeat count after \f(CW\*(C`u\*(C'\fR sets the number of bytes to put into an |
| 700 | uuencoded line, which is the maximum of 45 by default, but could be |
| 701 | set to some (smaller) integer multiple of three. \f(CW\*(C`unpack\*(C'\fR simply ignores |
| 702 | the repeat count. |
| 703 | .Sh "Doing Sums" |
| 704 | .IX Subsection "Doing Sums" |
| 705 | An even stranger template code is \f(CW\*(C`%\*(C'\fR<\fInumber\fR>. First, because |
| 706 | it's used as a prefix to some other template code. Second, because it |
| 707 | cannot be used in \f(CW\*(C`pack\*(C'\fR at all, and third, in \f(CW\*(C`unpack\*(C'\fR, doesn't return the |
| 708 | data as defined by the template code it precedes. Instead it'll give you an |
| 709 | integer of \fInumber\fR bits that is computed from the data value by |
| 710 | doing sums. For numeric unpack codes, no big feat is achieved: |
| 711 | .PP |
| 712 | .Vb 2 |
| 713 | \& my $buf = pack( 'iii', 100, 20, 3 ); |
| 714 | \& print unpack( '%32i3', $buf ), "\en"; # prints 123 |
| 715 | .Ve |
| 716 | .PP |
| 717 | For string values, \f(CW\*(C`%\*(C'\fR returns the sum of the byte values saving |
| 718 | you the trouble of a sum loop with \f(CW\*(C`substr\*(C'\fR and \f(CW\*(C`ord\*(C'\fR: |
| 719 | .PP |
| 720 | .Vb 1 |
| 721 | \& print unpack( '%32A*', "\ex01\ex10" ), "\en"; # prints 17 |
| 722 | .Ve |
| 723 | .PP |
| 724 | Although the \f(CW\*(C`%\*(C'\fR code is documented as returning a \*(L"checksum\*(R": |
| 725 | don't put your trust in such values! Even when applied to a small number |
| 726 | of bytes, they won't guarantee a noticeable Hamming distance. |
| 727 | .PP |
| 728 | In connection with \f(CW\*(C`b\*(C'\fR or \f(CW\*(C`B\*(C'\fR, \f(CW\*(C`%\*(C'\fR simply adds bits, and this can be put |
| 729 | to good use to count set bits efficiently: |
| 730 | .PP |
| 731 | .Vb 1 |
| 732 | \& my $bitcount = unpack( '%32b*', $mask ); |
| 733 | .Ve |
| 734 | .PP |
| 735 | And an even parity bit can be determined like this: |
| 736 | .PP |
| 737 | .Vb 1 |
| 738 | \& my $evenparity = unpack( '%1b*', $mask ); |
| 739 | .Ve |
| 740 | .Sh "Unicode" |
| 741 | .IX Subsection "Unicode" |
| 742 | Unicode is a character set that can represent most characters in most of |
| 743 | the world's languages, providing room for over one million different |
| 744 | characters. Unicode 3.1 specifies 94,140 characters: The Basic Latin |
| 745 | characters are assigned to the numbers 0 \- 127. The Latin\-1 Supplement with |
| 746 | characters that are used in several European languages is in the next |
| 747 | range, up to 255. After some more Latin extensions we find the character |
| 748 | sets from languages using non-Roman alphabets, interspersed with a |
| 749 | variety of symbol sets such as currency symbols, Zapf Dingbats or Braille. |
| 750 | (You might want to visit www.unicode.org for a look at some of |
| 751 | them \- my personal favourites are Telugu and Kannada.) |
| 752 | .PP |
| 753 | The Unicode character sets associates characters with integers. Encoding |
| 754 | these numbers in an equal number of bytes would more than double the |
| 755 | requirements for storing texts written in Latin alphabets. |
| 756 | The \s-1UTF\-8\s0 encoding avoids this by storing the most common (from a western |
| 757 | point of view) characters in a single byte while encoding the rarer |
| 758 | ones in three or more bytes. |
| 759 | .PP |
| 760 | So what has this got to do with \f(CW\*(C`pack\*(C'\fR? Well, if you want to convert |
| 761 | between a Unicode number and its \s-1UTF\-8\s0 representation you can do so by |
| 762 | using template code \f(CW\*(C`U\*(C'\fR. As an example, let's produce the \s-1UTF\-8\s0 |
| 763 | representation of the Euro currency symbol (code number 0x20AC): |
| 764 | .PP |
| 765 | .Vb 1 |
| 766 | \& $UTF8{Euro} = pack( 'U', 0x20AC ); |
| 767 | .Ve |
| 768 | .PP |
| 769 | Inspecting \f(CW$UTF8{Euro}\fR shows that it contains 3 bytes: \*(L"\exe2\ex82\exac\*(R". The |
| 770 | round trip can be completed with \f(CW\*(C`unpack\*(C'\fR: |
| 771 | .PP |
| 772 | .Vb 1 |
| 773 | \& $Unicode{Euro} = unpack( 'U', $UTF8{Euro} ); |
| 774 | .Ve |
| 775 | .PP |
| 776 | Usually you'll want to pack or unpack \s-1UTF\-8\s0 strings: |
| 777 | .PP |
| 778 | .Vb 3 |
| 779 | \& # pack and unpack the Hebrew alphabet |
| 780 | \& my $alefbet = pack( 'U*', 0x05d0..0x05ea ); |
| 781 | \& my @hebrew = unpack( 'U*', $utf ); |
| 782 | .Ve |
| 783 | .Sh "Another Portable Binary Encoding" |
| 784 | .IX Subsection "Another Portable Binary Encoding" |
| 785 | The pack code \f(CW\*(C`w\*(C'\fR has been added to support a portable binary data |
| 786 | encoding scheme that goes way beyond simple integers. (Details can |
| 787 | be found at Casbah.org, the Scarab project.) A \s-1BER\s0 (Binary Encoded |
| 788 | Representation) compressed unsigned integer stores base 128 |
| 789 | digits, most significant digit first, with as few digits as possible. |
| 790 | Bit eight (the high bit) is set on each byte except the last. There |
| 791 | is no size limit to \s-1BER\s0 encoding, but Perl won't go to extremes. |
| 792 | .PP |
| 793 | .Vb 1 |
| 794 | \& my $berbuf = pack( 'w*', 1, 128, 128+1, 128*128+127 ); |
| 795 | .Ve |
| 796 | .PP |
| 797 | A hex dump of \f(CW$berbuf\fR, with spaces inserted at the right places, |
| 798 | shows 01 8100 8101 81807F. Since the last byte is always less than |
| 799 | 128, \f(CW\*(C`unpack\*(C'\fR knows where to stop. |
| 800 | .SH "Template Grouping" |
| 801 | .IX Header "Template Grouping" |
| 802 | Prior to Perl 5.8, repetitions of templates had to be made by |
| 803 | \&\f(CW\*(C`x\*(C'\fR\-multiplication of template strings. Now there is a better way as |
| 804 | we may use the pack codes \f(CW\*(C`(\*(C'\fR and \f(CW\*(C`)\*(C'\fR combined with a repeat count. |
| 805 | The \f(CW\*(C`unpack\*(C'\fR template from the Stack Frame example can simply |
| 806 | be written like this: |
| 807 | .PP |
| 808 | .Vb 1 |
| 809 | \& unpack( 'v2 (vXXCC)5 v5', $frame ) |
| 810 | .Ve |
| 811 | .PP |
| 812 | Let's explore this feature a little more. We'll begin with the equivalent of |
| 813 | .PP |
| 814 | .Vb 1 |
| 815 | \& join( '', map( substr( $_, 0, 1 ), @str ) ) |
| 816 | .Ve |
| 817 | .PP |
| 818 | which returns a string consisting of the first character from each string. |
| 819 | Using pack, we can write |
| 820 | .PP |
| 821 | .Vb 1 |
| 822 | \& pack( '(A)'.@str, @str ) |
| 823 | .Ve |
| 824 | .PP |
| 825 | or, because a repeat count \f(CW\*(C`*\*(C'\fR means \*(L"repeat as often as required\*(R", |
| 826 | simply |
| 827 | .PP |
| 828 | .Vb 1 |
| 829 | \& pack( '(A)*', @str ) |
| 830 | .Ve |
| 831 | .PP |
| 832 | (Note that the template \f(CW\*(C`A*\*(C'\fR would only have packed \f(CW$str[0]\fR in full |
| 833 | length.) |
| 834 | .PP |
| 835 | To pack dates stored as triplets ( day, month, year ) in an array \f(CW@dates\fR |
| 836 | into a sequence of byte, byte, short integer we can write |
| 837 | .PP |
| 838 | .Vb 1 |
| 839 | \& $pd = pack( '(CCS)*', map( @$_, @dates ) ); |
| 840 | .Ve |
| 841 | .PP |
| 842 | To swap pairs of characters in a string (with even length) one could use |
| 843 | several techniques. First, let's use \f(CW\*(C`x\*(C'\fR and \f(CW\*(C`X\*(C'\fR to skip forward and back: |
| 844 | .PP |
| 845 | .Vb 1 |
| 846 | \& $s = pack( '(A)*', unpack( '(xAXXAx)*', $s ) ); |
| 847 | .Ve |
| 848 | .PP |
| 849 | We can also use \f(CW\*(C`@\*(C'\fR to jump to an offset, with 0 being the position where |
| 850 | we were when the last \f(CW\*(C`(\*(C'\fR was encountered: |
| 851 | .PP |
| 852 | .Vb 1 |
| 853 | \& $s = pack( '(A)*', unpack( '(@1A @0A @2)*', $s ) ); |
| 854 | .Ve |
| 855 | .PP |
| 856 | Finally, there is also an entirely different approach by unpacking big |
| 857 | endian shorts and packing them in the reverse byte order: |
| 858 | .PP |
| 859 | .Vb 1 |
| 860 | \& $s = pack( '(v)*', unpack( '(n)*', $s ); |
| 861 | .Ve |
| 862 | .SH "Lengths and Widths" |
| 863 | .IX Header "Lengths and Widths" |
| 864 | .Sh "String Lengths" |
| 865 | .IX Subsection "String Lengths" |
| 866 | In the previous section we've seen a network message that was constructed |
| 867 | by prefixing the binary message length to the actual message. You'll find |
| 868 | that packing a length followed by so many bytes of data is a |
| 869 | frequently used recipe since appending a null byte won't work |
| 870 | if a null byte may be part of the data. Here is an example where both |
| 871 | techniques are used: after two null terminated strings with source and |
| 872 | destination address, a Short Message (to a mobile phone) is sent after |
| 873 | a length byte: |
| 874 | .PP |
| 875 | .Vb 1 |
| 876 | \& my $msg = pack( 'Z*Z*CA*', $src, $dst, length( $sm ), $sm ); |
| 877 | .Ve |
| 878 | .PP |
| 879 | Unpacking this message can be done with the same template: |
| 880 | .PP |
| 881 | .Vb 1 |
| 882 | \& ( $src, $dst, $len, $sm ) = unpack( 'Z*Z*CA*', $msg ); |
| 883 | .Ve |
| 884 | .PP |
| 885 | There's a subtle trap lurking in the offing: Adding another field after |
| 886 | the Short Message (in variable \f(CW$sm\fR) is all right when packing, but this |
| 887 | cannot be unpacked naively: |
| 888 | .PP |
| 889 | .Vb 2 |
| 890 | \& # pack a message |
| 891 | \& my $msg = pack( 'Z*Z*CA*C', $src, $dst, length( $sm ), $sm, $prio ); |
| 892 | .Ve |
| 893 | .PP |
| 894 | .Vb 2 |
| 895 | \& # unpack fails - $prio remains undefined! |
| 896 | \& ( $src, $dst, $len, $sm, $prio ) = unpack( 'Z*Z*CA*C', $msg ); |
| 897 | .Ve |
| 898 | .PP |
| 899 | The pack code \f(CW\*(C`A*\*(C'\fR gobbles up all remaining bytes, and \f(CW$prio\fR remains |
| 900 | undefined! Before we let disappointment dampen the morale: Perl's got |
| 901 | the trump card to make this trick too, just a little further up the sleeve. |
| 902 | Watch this: |
| 903 | .PP |
| 904 | .Vb 2 |
| 905 | \& # pack a message: ASCIIZ, ASCIIZ, length/string, byte |
| 906 | \& my $msg = pack( 'Z* Z* C/A* C', $src, $dst, $sm, $prio ); |
| 907 | .Ve |
| 908 | .PP |
| 909 | .Vb 2 |
| 910 | \& # unpack |
| 911 | \& ( $src, $dst, $sm, $prio ) = unpack( 'Z* Z* C/A* C', $msg ); |
| 912 | .Ve |
| 913 | .PP |
| 914 | Combining two pack codes with a slash (\f(CW\*(C`/\*(C'\fR) associates them with a single |
| 915 | value from the argument list. In \f(CW\*(C`pack\*(C'\fR, the length of the argument is |
| 916 | taken and packed according to the first code while the argument itself |
| 917 | is added after being converted with the template code after the slash. |
| 918 | This saves us the trouble of inserting the \f(CW\*(C`length\*(C'\fR call, but it is |
| 919 | in \f(CW\*(C`unpack\*(C'\fR where we really score: The value of the length byte marks the |
| 920 | end of the string to be taken from the buffer. Since this combination |
| 921 | doesn't make sense except when the second pack code isn't \f(CW\*(C`a*\*(C'\fR, \f(CW\*(C`A*\*(C'\fR |
| 922 | or \f(CW\*(C`Z*\*(C'\fR, Perl won't let you. |
| 923 | .PP |
| 924 | The pack code preceding \f(CW\*(C`/\*(C'\fR may be anything that's fit to represent a |
| 925 | number: All the numeric binary pack codes, and even text codes such as |
| 926 | \&\f(CW\*(C`A4\*(C'\fR or \f(CW\*(C`Z*\*(C'\fR: |
| 927 | .PP |
| 928 | .Vb 4 |
| 929 | \& # pack/unpack a string preceded by its length in ASCII |
| 930 | \& my $buf = pack( 'A4/A*', "Humpty-Dumpty" ); |
| 931 | \& # unpack $buf: '13 Humpty-Dumpty' |
| 932 | \& my $txt = unpack( 'A4/A*', $buf ); |
| 933 | .Ve |
| 934 | .PP |
| 935 | \&\f(CW\*(C`/\*(C'\fR is not implemented in Perls before 5.6, so if your code is required to |
| 936 | work on older Perls you'll need to \f(CW\*(C`unpack( 'Z* Z* C')\*(C'\fR to get the length, |
| 937 | then use it to make a new unpack string. For example |
| 938 | .PP |
| 939 | .Vb 2 |
| 940 | \& # pack a message: ASCIIZ, ASCIIZ, length, string, byte (5.005 compatible) |
| 941 | \& my $msg = pack( 'Z* Z* C A* C', $src, $dst, length $sm, $sm, $prio ); |
| 942 | .Ve |
| 943 | .PP |
| 944 | .Vb 3 |
| 945 | \& # unpack |
| 946 | \& ( undef, undef, $len) = unpack( 'Z* Z* C', $msg ); |
| 947 | \& ($src, $dst, $sm, $prio) = unpack ( "Z* Z* x A$len C", $msg ); |
| 948 | .Ve |
| 949 | .PP |
| 950 | But that second \f(CW\*(C`unpack\*(C'\fR is rushing ahead. It isn't using a simple literal |
| 951 | string for the template. So maybe we should introduce... |
| 952 | .Sh "Dynamic Templates" |
| 953 | .IX Subsection "Dynamic Templates" |
| 954 | So far, we've seen literals used as templates. If the list of pack |
| 955 | items doesn't have fixed length, an expression constructing the |
| 956 | template is required (whenever, for some reason, \f(CW\*(C`()*\*(C'\fR cannot be used). |
| 957 | Here's an example: To store named string values in a way that can be |
| 958 | conveniently parsed by a C program, we create a sequence of names and |
| 959 | null terminated \s-1ASCII\s0 strings, with \f(CW\*(C`=\*(C'\fR between the name and the value, |
| 960 | followed by an additional delimiting null byte. Here's how: |
| 961 | .PP |
| 962 | .Vb 2 |
| 963 | \& my $env = pack( '(A*A*Z*)' . keys( %Env ) . 'C', |
| 964 | \& map( { ( $_, '=', $Env{$_} ) } keys( %Env ) ), 0 ); |
| 965 | .Ve |
| 966 | .PP |
| 967 | Let's examine the cogs of this byte mill, one by one. There's the \f(CW\*(C`map\*(C'\fR |
| 968 | call, creating the items we intend to stuff into the \f(CW$env\fR buffer: |
| 969 | to each key (in \f(CW$_\fR) it adds the \f(CW\*(C`=\*(C'\fR separator and the hash entry value. |
| 970 | Each triplet is packed with the template code sequence \f(CW\*(C`A*A*Z*\*(C'\fR that |
| 971 | is repeated according to the number of keys. (Yes, that's what the \f(CW\*(C`keys\*(C'\fR |
| 972 | function returns in scalar context.) To get the very last null byte, |
| 973 | we add a \f(CW0\fR at the end of the \f(CW\*(C`pack\*(C'\fR list, to be packed with \f(CW\*(C`C\*(C'\fR. |
| 974 | (Attentive readers may have noticed that we could have omitted the 0.) |
| 975 | .PP |
| 976 | For the reverse operation, we'll have to determine the number of items |
| 977 | in the buffer before we can let \f(CW\*(C`unpack\*(C'\fR rip it apart: |
| 978 | .PP |
| 979 | .Vb 2 |
| 980 | \& my $n = $env =~ tr/\e0// - 1; |
| 981 | \& my %env = map( split( /=/, $_ ), unpack( "(Z*)$n", $env ) ); |
| 982 | .Ve |
| 983 | .PP |
| 984 | The \f(CW\*(C`tr\*(C'\fR counts the null bytes. The \f(CW\*(C`unpack\*(C'\fR call returns a list of |
| 985 | name-value pairs each of which is taken apart in the \f(CW\*(C`map\*(C'\fR block. |
| 986 | .Sh "Counting Repetitions" |
| 987 | .IX Subsection "Counting Repetitions" |
| 988 | Rather than storing a sentinel at the end of a data item (or a list of items), |
| 989 | we could precede the data with a count. Again, we pack keys and values of |
| 990 | a hash, preceding each with an unsigned short length count, and up front |
| 991 | we store the number of pairs: |
| 992 | .PP |
| 993 | .Vb 1 |
| 994 | \& my $env = pack( 'S(S/A* S/A*)*', scalar keys( %Env ), %Env ); |
| 995 | .Ve |
| 996 | .PP |
| 997 | This simplifies the reverse operation as the number of repetitions can be |
| 998 | unpacked with the \f(CW\*(C`/\*(C'\fR code: |
| 999 | .PP |
| 1000 | .Vb 1 |
| 1001 | \& my %env = unpack( 'S/(S/A* S/A*)', $env ); |
| 1002 | .Ve |
| 1003 | .PP |
| 1004 | Note that this is one of the rare cases where you cannot use the same |
| 1005 | template for \f(CW\*(C`pack\*(C'\fR and \f(CW\*(C`unpack\*(C'\fR because \f(CW\*(C`pack\*(C'\fR can't determine |
| 1006 | a repeat count for a \f(CW\*(C`()\*(C'\fR\-group. |
| 1007 | .SH "Packing and Unpacking C Structures" |
| 1008 | .IX Header "Packing and Unpacking C Structures" |
| 1009 | In previous sections we have seen how to pack numbers and character |
| 1010 | strings. If it were not for a couple of snags we could conclude this |
| 1011 | section right away with the terse remark that C structures don't |
| 1012 | contain anything else, and therefore you already know all there is to it. |
| 1013 | Sorry, no: read on, please. |
| 1014 | .Sh "The Alignment Pit" |
| 1015 | .IX Subsection "The Alignment Pit" |
| 1016 | In the consideration of speed against memory requirements the balance |
| 1017 | has been tilted in favor of faster execution. This has influenced the |
| 1018 | way C compilers allocate memory for structures: On architectures |
| 1019 | where a 16\-bit or 32\-bit operand can be moved faster between places in |
| 1020 | memory, or to or from a \s-1CPU\s0 register, if it is aligned at an even or |
| 1021 | multiple-of-four or even at a multiple-of eight address, a C compiler |
| 1022 | will give you this speed benefit by stuffing extra bytes into structures. |
| 1023 | If you don't cross the C shoreline this is not likely to cause you any |
| 1024 | grief (although you should care when you design large data structures, |
| 1025 | or you want your code to be portable between architectures (you do want |
| 1026 | that, don't you?)). |
| 1027 | .PP |
| 1028 | To see how this affects \f(CW\*(C`pack\*(C'\fR and \f(CW\*(C`unpack\*(C'\fR, we'll compare these two |
| 1029 | C structures: |
| 1030 | .PP |
| 1031 | .Vb 6 |
| 1032 | \& typedef struct { |
| 1033 | \& char c1; |
| 1034 | \& short s; |
| 1035 | \& char c2; |
| 1036 | \& long l; |
| 1037 | \& } gappy_t; |
| 1038 | .Ve |
| 1039 | .PP |
| 1040 | .Vb 6 |
| 1041 | \& typedef struct { |
| 1042 | \& long l; |
| 1043 | \& short s; |
| 1044 | \& char c1; |
| 1045 | \& char c2; |
| 1046 | \& } dense_t; |
| 1047 | .Ve |
| 1048 | .PP |
| 1049 | Typically, a C compiler allocates 12 bytes to a \f(CW\*(C`gappy_t\*(C'\fR variable, but |
| 1050 | requires only 8 bytes for a \f(CW\*(C`dense_t\*(C'\fR. After investigating this further, |
| 1051 | we can draw memory maps, showing where the extra 4 bytes are hidden: |
| 1052 | .PP |
| 1053 | .Vb 5 |
| 1054 | \& 0 +4 +8 +12 |
| 1055 | \& +--+--+--+--+--+--+--+--+--+--+--+--+ |
| 1056 | \& |c1|xx| s |c2|xx|xx|xx| l | xx = fill byte |
| 1057 | \& +--+--+--+--+--+--+--+--+--+--+--+--+ |
| 1058 | \& gappy_t |
| 1059 | .Ve |
| 1060 | .PP |
| 1061 | .Vb 5 |
| 1062 | \& 0 +4 +8 |
| 1063 | \& +--+--+--+--+--+--+--+--+ |
| 1064 | \& | l | h |c1|c2| |
| 1065 | \& +--+--+--+--+--+--+--+--+ |
| 1066 | \& dense_t |
| 1067 | .Ve |
| 1068 | .PP |
| 1069 | And that's where the first quirk strikes: \f(CW\*(C`pack\*(C'\fR and \f(CW\*(C`unpack\*(C'\fR |
| 1070 | templates have to be stuffed with \f(CW\*(C`x\*(C'\fR codes to get those extra fill bytes. |
| 1071 | .PP |
| 1072 | The natural question: \*(L"Why can't Perl compensate for the gaps?\*(R" warrants |
| 1073 | an answer. One good reason is that C compilers might provide (non\-ANSI) |
| 1074 | extensions permitting all sorts of fancy control over the way structures |
| 1075 | are aligned, even at the level of an individual structure field. And, if |
| 1076 | this were not enough, there is an insidious thing called \f(CW\*(C`union\*(C'\fR where |
| 1077 | the amount of fill bytes cannot be derived from the alignment of the next |
| 1078 | item alone. |
| 1079 | .PP |
| 1080 | \&\s-1OK\s0, so let's bite the bullet. Here's one way to get the alignment right |
| 1081 | by inserting template codes \f(CW\*(C`x\*(C'\fR, which don't take a corresponding item |
| 1082 | from the list: |
| 1083 | .PP |
| 1084 | .Vb 1 |
| 1085 | \& my $gappy = pack( 'cxs cxxx l!', $c1, $s, $c2, $l ); |
| 1086 | .Ve |
| 1087 | .PP |
| 1088 | Note the \f(CW\*(C`!\*(C'\fR after \f(CW\*(C`l\*(C'\fR: We want to make sure that we pack a long |
| 1089 | integer as it is compiled by our C compiler. And even now, it will only |
| 1090 | work for the platforms where the compiler aligns things as above. |
| 1091 | And somebody somewhere has a platform where it doesn't. |
| 1092 | [Probably a Cray, where \f(CW\*(C`short\*(C'\fRs, \f(CW\*(C`int\*(C'\fRs and \f(CW\*(C`long\*(C'\fRs are all 8 bytes. :\-)] |
| 1093 | .PP |
| 1094 | Counting bytes and watching alignments in lengthy structures is bound to |
| 1095 | be a drag. Isn't there a way we can create the template with a simple |
| 1096 | program? Here's a C program that does the trick: |
| 1097 | .PP |
| 1098 | .Vb 2 |
| 1099 | \& #include <stdio.h> |
| 1100 | \& #include <stddef.h> |
| 1101 | .Ve |
| 1102 | .PP |
| 1103 | .Vb 6 |
| 1104 | \& typedef struct { |
| 1105 | \& char fc1; |
| 1106 | \& short fs; |
| 1107 | \& char fc2; |
| 1108 | \& long fl; |
| 1109 | \& } gappy_t; |
| 1110 | .Ve |
| 1111 | .PP |
| 1112 | .Vb 2 |
| 1113 | \& #define Pt(struct,field,tchar) \e |
| 1114 | \& printf( "@%d%s ", offsetof(struct,field), # tchar ); |
| 1115 | .Ve |
| 1116 | .PP |
| 1117 | .Vb 7 |
| 1118 | \& int main() { |
| 1119 | \& Pt( gappy_t, fc1, c ); |
| 1120 | \& Pt( gappy_t, fs, s! ); |
| 1121 | \& Pt( gappy_t, fc2, c ); |
| 1122 | \& Pt( gappy_t, fl, l! ); |
| 1123 | \& printf( "\en" ); |
| 1124 | \& } |
| 1125 | .Ve |
| 1126 | .PP |
| 1127 | The output line can be used as a template in a \f(CW\*(C`pack\*(C'\fR or \f(CW\*(C`unpack\*(C'\fR call: |
| 1128 | .PP |
| 1129 | .Vb 1 |
| 1130 | \& my $gappy = pack( '@0c @2s! @4c @8l!', $c1, $s, $c2, $l ); |
| 1131 | .Ve |
| 1132 | .PP |
| 1133 | Gee, yet another template code \- as if we hadn't plenty. But |
| 1134 | \&\f(CW\*(C`@\*(C'\fR saves our day by enabling us to specify the offset from the beginning |
| 1135 | of the pack buffer to the next item: This is just the value |
| 1136 | the \f(CW\*(C`offsetof\*(C'\fR macro (defined in \f(CW\*(C`<stddef.h>\*(C'\fR) returns when |
| 1137 | given a \f(CW\*(C`struct\*(C'\fR type and one of its field names (\*(L"member\-designator\*(R" in |
| 1138 | C standardese). |
| 1139 | .PP |
| 1140 | Neither using offsets nor adding \f(CW\*(C`x\*(C'\fR's to bridge the gaps is satisfactory. |
| 1141 | (Just imagine what happens if the structure changes.) What we really need |
| 1142 | is a way of saying \*(L"skip as many bytes as required to the next multiple of N\*(R". |
| 1143 | In fluent Templatese, you say this with \f(CW\*(C`x!N\*(C'\fR where N is replaced by the |
| 1144 | appropriate value. Here's the next version of our struct packaging: |
| 1145 | .PP |
| 1146 | .Vb 1 |
| 1147 | \& my $gappy = pack( 'c x!2 s c x!4 l!', $c1, $s, $c2, $l ); |
| 1148 | .Ve |
| 1149 | .PP |
| 1150 | That's certainly better, but we still have to know how long all the |
| 1151 | integers are, and portability is far away. Rather than \f(CW2\fR, |
| 1152 | for instance, we want to say \*(L"however long a short is\*(R". But this can be |
| 1153 | done by enclosing the appropriate pack code in brackets: \f(CW\*(C`[s]\*(C'\fR. So, here's |
| 1154 | the very best we can do: |
| 1155 | .PP |
| 1156 | .Vb 1 |
| 1157 | \& my $gappy = pack( 'c x![s] s c x![l!] l!', $c1, $s, $c2, $l ); |
| 1158 | .Ve |
| 1159 | .Sh "Alignment, Take 2" |
| 1160 | .IX Subsection "Alignment, Take 2" |
| 1161 | I'm afraid that we're not quite through with the alignment catch yet. The |
| 1162 | hydra raises another ugly head when you pack arrays of structures: |
| 1163 | .PP |
| 1164 | .Vb 4 |
| 1165 | \& typedef struct { |
| 1166 | \& short count; |
| 1167 | \& char glyph; |
| 1168 | \& } cell_t; |
| 1169 | .Ve |
| 1170 | .PP |
| 1171 | .Vb 1 |
| 1172 | \& typedef cell_t buffer_t[BUFLEN]; |
| 1173 | .Ve |
| 1174 | .PP |
| 1175 | Where's the catch? Padding is neither required before the first field \f(CW\*(C`count\*(C'\fR, |
| 1176 | nor between this and the next field \f(CW\*(C`glyph\*(C'\fR, so why can't we simply pack |
| 1177 | like this: |
| 1178 | .PP |
| 1179 | .Vb 3 |
| 1180 | \& # something goes wrong here: |
| 1181 | \& pack( 's!a' x @buffer, |
| 1182 | \& map{ ( $_->{count}, $_->{glyph} ) } @buffer ); |
| 1183 | .Ve |
| 1184 | .PP |
| 1185 | This packs \f(CW\*(C`3*@buffer\*(C'\fR bytes, but it turns out that the size of |
| 1186 | \&\f(CW\*(C`buffer_t\*(C'\fR is four times \f(CW\*(C`BUFLEN\*(C'\fR! The moral of the story is that |
| 1187 | the required alignment of a structure or array is propagated to the |
| 1188 | next higher level where we have to consider padding \fIat the end\fR |
| 1189 | of each component as well. Thus the correct template is: |
| 1190 | .PP |
| 1191 | .Vb 2 |
| 1192 | \& pack( 's!ax' x @buffer, |
| 1193 | \& map{ ( $_->{count}, $_->{glyph} ) } @buffer ); |
| 1194 | .Ve |
| 1195 | .Sh "Alignment, Take 3" |
| 1196 | .IX Subsection "Alignment, Take 3" |
| 1197 | And even if you take all the above into account, \s-1ANSI\s0 still lets this: |
| 1198 | .PP |
| 1199 | .Vb 3 |
| 1200 | \& typedef struct { |
| 1201 | \& char foo[2]; |
| 1202 | \& } foo_t; |
| 1203 | .Ve |
| 1204 | .PP |
| 1205 | vary in size. The alignment constraint of the structure can be greater than |
| 1206 | any of its elements. [And if you think that this doesn't affect anything |
| 1207 | common, dismember the next cellphone that you see. Many have \s-1ARM\s0 cores, and |
| 1208 | the \s-1ARM\s0 structure rules make \f(CW\*(C`sizeof (foo_t)\*(C'\fR == 4] |
| 1209 | .Sh "Pointers for How to Use Them" |
| 1210 | .IX Subsection "Pointers for How to Use Them" |
| 1211 | The title of this section indicates the second problem you may run into |
| 1212 | sooner or later when you pack C structures. If the function you intend |
| 1213 | to call expects a, say, \f(CW\*(C`void *\*(C'\fR value, you \fIcannot\fR simply take |
| 1214 | a reference to a Perl variable. (Although that value certainly is a |
| 1215 | memory address, it's not the address where the variable's contents are |
| 1216 | stored.) |
| 1217 | .PP |
| 1218 | Template code \f(CW\*(C`P\*(C'\fR promises to pack a \*(L"pointer to a fixed length string\*(R". |
| 1219 | Isn't this what we want? Let's try: |
| 1220 | .PP |
| 1221 | .Vb 3 |
| 1222 | \& # allocate some storage and pack a pointer to it |
| 1223 | \& my $memory = "\ex00" x $size; |
| 1224 | \& my $memptr = pack( 'P', $memory ); |
| 1225 | .Ve |
| 1226 | .PP |
| 1227 | But wait: doesn't \f(CW\*(C`pack\*(C'\fR just return a sequence of bytes? How can we pass this |
| 1228 | string of bytes to some C code expecting a pointer which is, after all, |
| 1229 | nothing but a number? The answer is simple: We have to obtain the numeric |
| 1230 | address from the bytes returned by \f(CW\*(C`pack\*(C'\fR. |
| 1231 | .PP |
| 1232 | .Vb 1 |
| 1233 | \& my $ptr = unpack( 'L!', $memptr ); |
| 1234 | .Ve |
| 1235 | .PP |
| 1236 | Obviously this assumes that it is possible to typecast a pointer |
| 1237 | to an unsigned long and vice versa, which frequently works but should not |
| 1238 | be taken as a universal law. \- Now that we have this pointer the next question |
| 1239 | is: How can we put it to good use? We need a call to some C function |
| 1240 | where a pointer is expected. The \fIread\fR\|(2) system call comes to mind: |
| 1241 | .PP |
| 1242 | .Vb 1 |
| 1243 | \& ssize_t read(int fd, void *buf, size_t count); |
| 1244 | .Ve |
| 1245 | .PP |
| 1246 | After reading perlfunc explaining how to use \f(CW\*(C`syscall\*(C'\fR we can write |
| 1247 | this Perl function copying a file to standard output: |
| 1248 | .PP |
| 1249 | .Vb 12 |
| 1250 | \& require 'syscall.ph'; |
| 1251 | \& sub cat($){ |
| 1252 | \& my $path = shift(); |
| 1253 | \& my $size = -s $path; |
| 1254 | \& my $memory = "\ex00" x $size; # allocate some memory |
| 1255 | \& my $ptr = unpack( 'L', pack( 'P', $memory ) ); |
| 1256 | \& open( F, $path ) || die( "$path: cannot open ($!)\en" ); |
| 1257 | \& my $fd = fileno(F); |
| 1258 | \& my $res = syscall( &SYS_read, fileno(F), $ptr, $size ); |
| 1259 | \& print $memory; |
| 1260 | \& close( F ); |
| 1261 | \& } |
| 1262 | .Ve |
| 1263 | .PP |
| 1264 | This is neither a specimen of simplicity nor a paragon of portability but |
| 1265 | it illustrates the point: We are able to sneak behind the scenes and |
| 1266 | access Perl's otherwise well-guarded memory! (Important note: Perl's |
| 1267 | \&\f(CW\*(C`syscall\*(C'\fR does \fInot\fR require you to construct pointers in this roundabout |
| 1268 | way. You simply pass a string variable, and Perl forwards the address.) |
| 1269 | .PP |
| 1270 | How does \f(CW\*(C`unpack\*(C'\fR with \f(CW\*(C`P\*(C'\fR work? Imagine some pointer in the buffer |
| 1271 | about to be unpacked: If it isn't the null pointer (which will smartly |
| 1272 | produce the \f(CW\*(C`undef\*(C'\fR value) we have a start address \- but then what? |
| 1273 | Perl has no way of knowing how long this \*(L"fixed length string\*(R" is, so |
| 1274 | it's up to you to specify the actual size as an explicit length after \f(CW\*(C`P\*(C'\fR. |
| 1275 | .PP |
| 1276 | .Vb 2 |
| 1277 | \& my $mem = "abcdefghijklmn"; |
| 1278 | \& print unpack( 'P5', pack( 'P', $mem ) ); # prints "abcde" |
| 1279 | .Ve |
| 1280 | .PP |
| 1281 | As a consequence, \f(CW\*(C`pack\*(C'\fR ignores any number or \f(CW\*(C`*\*(C'\fR after \f(CW\*(C`P\*(C'\fR. |
| 1282 | .PP |
| 1283 | Now that we have seen \f(CW\*(C`P\*(C'\fR at work, we might as well give \f(CW\*(C`p\*(C'\fR a whirl. |
| 1284 | Why do we need a second template code for packing pointers at all? The |
| 1285 | answer lies behind the simple fact that an \f(CW\*(C`unpack\*(C'\fR with \f(CW\*(C`p\*(C'\fR promises |
| 1286 | a null-terminated string starting at the address taken from the buffer, |
| 1287 | and that implies a length for the data item to be returned: |
| 1288 | .PP |
| 1289 | .Vb 2 |
| 1290 | \& my $buf = pack( 'p', "abc\ex00efhijklmn" ); |
| 1291 | \& print unpack( 'p', $buf ); # prints "abc" |
| 1292 | .Ve |
| 1293 | .PP |
| 1294 | Albeit this is apt to be confusing: As a consequence of the length being |
| 1295 | implied by the string's length, a number after pack code \f(CW\*(C`p\*(C'\fR is a repeat |
| 1296 | count, not a length as after \f(CW\*(C`P\*(C'\fR. |
| 1297 | .PP |
| 1298 | Using \f(CW\*(C`pack(..., $x)\*(C'\fR with \f(CW\*(C`P\*(C'\fR or \f(CW\*(C`p\*(C'\fR to get the address where \f(CW$x\fR is |
| 1299 | actually stored must be used with circumspection. Perl's internal machinery |
| 1300 | considers the relation between a variable and that address as its very own |
| 1301 | private matter and doesn't really care that we have obtained a copy. Therefore: |
| 1302 | .IP "\(bu" 4 |
| 1303 | Do not use \f(CW\*(C`pack\*(C'\fR with \f(CW\*(C`p\*(C'\fR or \f(CW\*(C`P\*(C'\fR to obtain the address of variable |
| 1304 | that's bound to go out of scope (and thereby freeing its memory) before you |
| 1305 | are done with using the memory at that address. |
| 1306 | .IP "\(bu" 4 |
| 1307 | Be very careful with Perl operations that change the value of the |
| 1308 | variable. Appending something to the variable, for instance, might require |
| 1309 | reallocation of its storage, leaving you with a pointer into no\-man's land. |
| 1310 | .IP "\(bu" 4 |
| 1311 | Don't think that you can get the address of a Perl variable |
| 1312 | when it is stored as an integer or double number! \f(CW\*(C`pack('P', $x)\*(C'\fR will |
| 1313 | force the variable's internal representation to string, just as if you |
| 1314 | had written something like \f(CW\*(C`$x .= ''\*(C'\fR. |
| 1315 | .PP |
| 1316 | It's safe, however, to P\- or p\-pack a string literal, because Perl simply |
| 1317 | allocates an anonymous variable. |
| 1318 | .SH "Pack Recipes" |
| 1319 | .IX Header "Pack Recipes" |
| 1320 | Here are a collection of (possibly) useful canned recipes for \f(CW\*(C`pack\*(C'\fR |
| 1321 | and \f(CW\*(C`unpack\*(C'\fR: |
| 1322 | .PP |
| 1323 | .Vb 2 |
| 1324 | \& # Convert IP address for socket functions |
| 1325 | \& pack( "C4", split /\e./, "123.4.5.6" ); |
| 1326 | .Ve |
| 1327 | .PP |
| 1328 | .Vb 2 |
| 1329 | \& # Count the bits in a chunk of memory (e.g. a select vector) |
| 1330 | \& unpack( '%32b*', $mask ); |
| 1331 | .Ve |
| 1332 | .PP |
| 1333 | .Vb 3 |
| 1334 | \& # Determine the endianness of your system |
| 1335 | \& $is_little_endian = unpack( 'c', pack( 's', 1 ) ); |
| 1336 | \& $is_big_endian = unpack( 'xc', pack( 's', 1 ) ); |
| 1337 | .Ve |
| 1338 | .PP |
| 1339 | .Vb 2 |
| 1340 | \& # Determine the number of bits in a native integer |
| 1341 | \& $bits = unpack( '%32I!', ~0 ); |
| 1342 | .Ve |
| 1343 | .PP |
| 1344 | .Vb 2 |
| 1345 | \& # Prepare argument for the nanosleep system call |
| 1346 | \& my $timespec = pack( 'L!L!', $secs, $nanosecs ); |
| 1347 | .Ve |
| 1348 | .PP |
| 1349 | For a simple memory dump we unpack some bytes into just as |
| 1350 | many pairs of hex digits, and use \f(CW\*(C`map\*(C'\fR to handle the traditional |
| 1351 | spacing \- 16 bytes to a line: |
| 1352 | .PP |
| 1353 | .Vb 4 |
| 1354 | \& my $i; |
| 1355 | \& print map( ++$i % 16 ? "$_ " : "$_\en", |
| 1356 | \& unpack( 'H2' x length( $mem ), $mem ) ), |
| 1357 | \& length( $mem ) % 16 ? "\en" : ''; |
| 1358 | .Ve |
| 1359 | .SH "Funnies Section" |
| 1360 | .IX Header "Funnies Section" |
| 1361 | .Vb 5 |
| 1362 | \& # Pulling digits out of nowhere... |
| 1363 | \& print unpack( 'C', pack( 'x' ) ), |
| 1364 | \& unpack( '%B*', pack( 'A' ) ), |
| 1365 | \& unpack( 'H', pack( 'A' ) ), |
| 1366 | \& unpack( 'A', unpack( 'C', pack( 'A' ) ) ), "\en"; |
| 1367 | .Ve |
| 1368 | .PP |
| 1369 | .Vb 2 |
| 1370 | \& # One for the road ;-) |
| 1371 | \& my $advice = pack( 'all u can in a van' ); |
| 1372 | .Ve |
| 1373 | .SH "Authors" |
| 1374 | .IX Header "Authors" |
| 1375 | Simon Cozens and Wolfgang Laun. |