| 1 | # PyKota |
|---|
| 2 | # -*- coding: ISO-8859-15 -*- |
|---|
| 3 | # |
|---|
| 4 | # PyKota - Print Quotas for CUPS and LPRng |
|---|
| 5 | # |
|---|
| 6 | # (c) 2003-2004 Jerome Alet <alet@librelogiciel.com> |
|---|
| 7 | # This program is free software; you can redistribute it and/or modify |
|---|
| 8 | # it under the terms of the GNU General Public License as published by |
|---|
| 9 | # the Free Software Foundation; either version 2 of the License, or |
|---|
| 10 | # (at your option) any later version. |
|---|
| 11 | # |
|---|
| 12 | # This program is distributed in the hope that it will be useful, |
|---|
| 13 | # but WITHOUT ANY WARRANTY; without even the implied warranty of |
|---|
| 14 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|---|
| 15 | # GNU General Public License for more details. |
|---|
| 16 | # |
|---|
| 17 | # You should have received a copy of the GNU General Public License |
|---|
| 18 | # along with this program; if not, write to the Free Software |
|---|
| 19 | # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307, USA. |
|---|
| 20 | # |
|---|
| 21 | # $Id$ |
|---|
| 22 | # |
|---|
| 23 | # $Log$ |
|---|
| 24 | # Revision 1.36 2004/09/01 22:31:49 jalet |
|---|
| 25 | # Some more work on ESC/P2 analyzer to avoid missing \r\n sequences. Not |
|---|
| 26 | # exactly optimal though... |
|---|
| 27 | # |
|---|
| 28 | # Revision 1.35 2004/08/30 23:10:24 jalet |
|---|
| 29 | # Improved the ESC/P2 analyzer so that more GhostScript devices are supported |
|---|
| 30 | # |
|---|
| 31 | # Revision 1.34 2004/08/27 09:08:22 jalet |
|---|
| 32 | # Improvement in PostScript parser to avoid being fooled by clever "students" |
|---|
| 33 | # |
|---|
| 34 | # Revision 1.33 2004/08/27 09:02:34 jalet |
|---|
| 35 | # Forgot to remove some special debugging code... |
|---|
| 36 | # |
|---|
| 37 | # Revision 1.32 2004/08/27 08:58:50 jalet |
|---|
| 38 | # Relax checks for PCL5 header to accomodate strange printer drivers |
|---|
| 39 | # |
|---|
| 40 | # Revision 1.31 2004/08/22 08:25:33 jalet |
|---|
| 41 | # Improved ESC/P2 miniparser thanks to Paulo Silva |
|---|
| 42 | # |
|---|
| 43 | # Revision 1.30 2004/08/21 23:16:57 jalet |
|---|
| 44 | # First draft of ESC/P2 (mini-)parser. |
|---|
| 45 | # |
|---|
| 46 | # Revision 1.29 2004/08/11 16:25:38 jalet |
|---|
| 47 | # Fixed index problem in PCLXL parser when retrieving number of copies for |
|---|
| 48 | # each page |
|---|
| 49 | # |
|---|
| 50 | # Revision 1.28 2004/08/10 23:01:49 jalet |
|---|
| 51 | # Fixed number of copies in PCL5 parser |
|---|
| 52 | # |
|---|
| 53 | # Revision 1.27 2004/08/09 18:14:22 jalet |
|---|
| 54 | # Added workaround for number of copies and some PostScript drivers |
|---|
| 55 | # |
|---|
| 56 | # Revision 1.26 2004/07/22 13:49:51 jalet |
|---|
| 57 | # Added support for binary PostScript through GhostScript if native DSC |
|---|
| 58 | # compliant PostScript analyzer doesn't find any page. This is much |
|---|
| 59 | # slower though, so native analyzer is tried first. |
|---|
| 60 | # |
|---|
| 61 | # Revision 1.25 2004/07/10 14:06:36 jalet |
|---|
| 62 | # Fix for Python2.1 incompatibilities |
|---|
| 63 | # |
|---|
| 64 | # Revision 1.24 2004/07/05 21:00:39 jalet |
|---|
| 65 | # Fix for number of copies for each page in PCLXL parser |
|---|
| 66 | # |
|---|
| 67 | # Revision 1.23 2004/07/03 08:21:59 jalet |
|---|
| 68 | # Testsuite for PDL Analyzer added |
|---|
| 69 | # |
|---|
| 70 | # Revision 1.22 2004/06/29 14:21:41 jalet |
|---|
| 71 | # Smallish optimization |
|---|
| 72 | # |
|---|
| 73 | # Revision 1.21 2004/06/28 23:11:26 jalet |
|---|
| 74 | # Code de-factorization in PCLXL parser |
|---|
| 75 | # |
|---|
| 76 | # Revision 1.20 2004/06/28 22:38:41 jalet |
|---|
| 77 | # Increased speed by a factor of 2 in PCLXL parser |
|---|
| 78 | # |
|---|
| 79 | # Revision 1.19 2004/06/28 21:20:30 jalet |
|---|
| 80 | # PCLXL support now works ! |
|---|
| 81 | # |
|---|
| 82 | # Revision 1.18 2004/06/27 22:59:37 jalet |
|---|
| 83 | # More work on PCLXL parser |
|---|
| 84 | # |
|---|
| 85 | # Revision 1.17 2004/06/26 23:20:01 jalet |
|---|
| 86 | # Additionnal speedup for GhostScript generated PCL5 files |
|---|
| 87 | # |
|---|
| 88 | # Revision 1.16 2004/06/26 15:31:00 jalet |
|---|
| 89 | # mmap reintroduced in PCL5 parser |
|---|
| 90 | # |
|---|
| 91 | # Revision 1.15 2004/06/26 14:14:31 jalet |
|---|
| 92 | # Now uses Psyco if it is available |
|---|
| 93 | # |
|---|
| 94 | # Revision 1.14 2004/06/25 09:50:28 jalet |
|---|
| 95 | # More debug info in PCLXL parser |
|---|
| 96 | # |
|---|
| 97 | # Revision 1.13 2004/06/25 08:10:08 jalet |
|---|
| 98 | # Another fix for PCL5 parser |
|---|
| 99 | # |
|---|
| 100 | # Revision 1.12 2004/06/24 23:09:53 jalet |
|---|
| 101 | # Fix for number of copies in PCL5 parser |
|---|
| 102 | # |
|---|
| 103 | # Revision 1.11 2004/06/23 22:07:50 jalet |
|---|
| 104 | # Fixed PCL5 parser according to the sources of rastertohp |
|---|
| 105 | # |
|---|
| 106 | # Revision 1.10 2004/06/18 22:24:03 jalet |
|---|
| 107 | # Removed old comments |
|---|
| 108 | # |
|---|
| 109 | # Revision 1.9 2004/06/18 22:21:27 jalet |
|---|
| 110 | # Native PDF parser greatly improved. |
|---|
| 111 | # GhostScript based PDF parser completely removed because native code |
|---|
| 112 | # is now portable across Python versions. |
|---|
| 113 | # |
|---|
| 114 | # Revision 1.8 2004/06/18 20:49:46 jalet |
|---|
| 115 | # "ERROR:" prefix added |
|---|
| 116 | # |
|---|
| 117 | # Revision 1.7 2004/06/18 17:48:04 jalet |
|---|
| 118 | # Added native fast PDF parsing method |
|---|
| 119 | # |
|---|
| 120 | # Revision 1.6 2004/06/18 14:00:16 jalet |
|---|
| 121 | # Added PDF support in smart PDL analyzer (through GhostScript for now) |
|---|
| 122 | # |
|---|
| 123 | # Revision 1.5 2004/06/18 10:09:05 jalet |
|---|
| 124 | # Resets file pointer to start of file in all cases |
|---|
| 125 | # |
|---|
| 126 | # Revision 1.4 2004/06/18 06:16:14 jalet |
|---|
| 127 | # Fixes PostScript detection code for incorrect drivers |
|---|
| 128 | # |
|---|
| 129 | # Revision 1.3 2004/05/21 20:40:08 jalet |
|---|
| 130 | # All the code for pkpgcounter is now in pdlanalyzer.py |
|---|
| 131 | # |
|---|
| 132 | # Revision 1.2 2004/05/19 19:09:36 jalet |
|---|
| 133 | # Speed improvement |
|---|
| 134 | # |
|---|
| 135 | # Revision 1.1 2004/05/18 09:59:54 jalet |
|---|
| 136 | # pkpgcounter is now just a wrapper around the PDLAnalyzer class |
|---|
| 137 | # |
|---|
| 138 | # |
|---|
| 139 | # |
|---|
| 140 | |
|---|
| 141 | import sys |
|---|
| 142 | import os |
|---|
| 143 | import re |
|---|
| 144 | from struct import unpack |
|---|
| 145 | import tempfile |
|---|
| 146 | import mmap |
|---|
| 147 | import popen2 |
|---|
| 148 | |
|---|
| 149 | KILOBYTE = 1024 |
|---|
| 150 | MEGABYTE = 1024 * KILOBYTE |
|---|
| 151 | |
|---|
| 152 | class PDLAnalyzerError(Exception): |
|---|
| 153 | """An exception for PDL Analyzer related stuff.""" |
|---|
| 154 | def __init__(self, message = ""): |
|---|
| 155 | self.message = message |
|---|
| 156 | Exception.__init__(self, message) |
|---|
| 157 | def __repr__(self): |
|---|
| 158 | return self.message |
|---|
| 159 | __str__ = __repr__ |
|---|
| 160 | |
|---|
| 161 | class PostScriptAnalyzer : |
|---|
| 162 | def __init__(self, infile) : |
|---|
| 163 | """Initialize PostScript Analyzer.""" |
|---|
| 164 | self.infile = infile |
|---|
| 165 | self.copies = 1 |
|---|
| 166 | |
|---|
| 167 | def throughGhostScript(self) : |
|---|
| 168 | """Get the count through GhostScript, useful for non-DSC compliant PS files.""" |
|---|
| 169 | self.infile.seek(0) |
|---|
| 170 | command = 'gs -sDEVICE=bbox -dNOPAUSE -dBATCH -dQUIET - 2>&1 | grep -c "%%HiResBoundingBox:" 2>/dev/null' |
|---|
| 171 | child = popen2.Popen4(command) |
|---|
| 172 | try : |
|---|
| 173 | data = self.infile.read(MEGABYTE) |
|---|
| 174 | while data : |
|---|
| 175 | child.tochild.write(data) |
|---|
| 176 | data = self.infile.read(MEGABYTE) |
|---|
| 177 | child.tochild.flush() |
|---|
| 178 | child.tochild.close() |
|---|
| 179 | except (IOError, OSError), msg : |
|---|
| 180 | raise PDLAnalyzerError, "Problem during analysis of Binary PostScript document." |
|---|
| 181 | |
|---|
| 182 | pagecount = 0 |
|---|
| 183 | try : |
|---|
| 184 | pagecount = int(child.fromchild.readline().strip()) |
|---|
| 185 | except (IOError, OSError, AttributeError, ValueError) : |
|---|
| 186 | raise PDLAnalyzerError, "Problem during analysis of Binary PostScript document." |
|---|
| 187 | child.fromchild.close() |
|---|
| 188 | |
|---|
| 189 | try : |
|---|
| 190 | retcode = child.wait() |
|---|
| 191 | except OSError, msg : |
|---|
| 192 | raise PDLAnalyzerError, "Problem during analysis of Binary PostScript document." |
|---|
| 193 | return pagecount * self.copies |
|---|
| 194 | |
|---|
| 195 | def natively(self) : |
|---|
| 196 | """Count pages in a DSC compliant PostScript document.""" |
|---|
| 197 | self.infile.seek(0) |
|---|
| 198 | pagecount = 0 |
|---|
| 199 | for line in self.infile.xreadlines() : |
|---|
| 200 | if line.startswith("%%Page: ") : |
|---|
| 201 | pagecount += 1 |
|---|
| 202 | elif line.startswith("%%BeginNonPPDFeature: NumCopies ") : |
|---|
| 203 | # handle # of copies set by some Windows printer driver |
|---|
| 204 | try : |
|---|
| 205 | number = int(line.strip().split()[2]) |
|---|
| 206 | except : |
|---|
| 207 | pass |
|---|
| 208 | else : |
|---|
| 209 | if number > self.copies : |
|---|
| 210 | self.copies = number |
|---|
| 211 | elif line.startswith("1 dict dup /NumCopies ") : |
|---|
| 212 | # handle # of copies set by mozilla/kprinter |
|---|
| 213 | try : |
|---|
| 214 | number = int(line.strip().split()[4]) |
|---|
| 215 | except : |
|---|
| 216 | pass |
|---|
| 217 | else : |
|---|
| 218 | if number > self.copies : |
|---|
| 219 | self.copies = number |
|---|
| 220 | return pagecount * self.copies |
|---|
| 221 | |
|---|
| 222 | def getJobSize(self) : |
|---|
| 223 | """Count pages in PostScript document.""" |
|---|
| 224 | return self.natively() or self.throughGhostScript() |
|---|
| 225 | |
|---|
| 226 | class PDFAnalyzer : |
|---|
| 227 | def __init__(self, infile) : |
|---|
| 228 | """Initialize PDF Analyzer.""" |
|---|
| 229 | self.infile = infile |
|---|
| 230 | |
|---|
| 231 | def getJobSize(self) : |
|---|
| 232 | """Counts pages in a PDF document.""" |
|---|
| 233 | regexp = re.compile(r"(/Type) ?(/Page)[/ \t\r\n]") |
|---|
| 234 | pagecount = 0 |
|---|
| 235 | for line in self.infile.xreadlines() : |
|---|
| 236 | pagecount += len(regexp.findall(line)) |
|---|
| 237 | return pagecount |
|---|
| 238 | |
|---|
| 239 | class ESCP2Analyzer : |
|---|
| 240 | def __init__(self, infile) : |
|---|
| 241 | """Initialize ESC/P2 Analyzer.""" |
|---|
| 242 | self.infile = infile |
|---|
| 243 | |
|---|
| 244 | def getJobSize(self) : |
|---|
| 245 | """Counts pages in an ESC/P2 document.""" |
|---|
| 246 | # with Gimpprint, at least, for each page there |
|---|
| 247 | # are two Reset Printer sequences (ESC + @) |
|---|
| 248 | marker1 = "\033@" |
|---|
| 249 | |
|---|
| 250 | # with other software or printer driver, we |
|---|
| 251 | # may prefer to search for "\r\n\fESCAPE" |
|---|
| 252 | # or "\r\fESCAPE" |
|---|
| 253 | marker2r = "\r\f\033" |
|---|
| 254 | marker2rn = "\r\n\f\033" |
|---|
| 255 | |
|---|
| 256 | # and ghostscript's stcolor for example seems to |
|---|
| 257 | # output ESC + @ + \f for each page plus one |
|---|
| 258 | marker3 = "\033@\f" |
|---|
| 259 | |
|---|
| 260 | # while ghostscript's escp driver outputs instead |
|---|
| 261 | # \f + ESC + @ |
|---|
| 262 | marker4 = "\f\033@" |
|---|
| 263 | |
|---|
| 264 | data = self.infile.read() |
|---|
| 265 | pagecount1 = data.count(marker1) |
|---|
| 266 | pagecount2 = max(data.count(marker2r), data.count(marker2rn)) |
|---|
| 267 | pagecount3 = data.count(marker3) |
|---|
| 268 | pagecount4 = data.count(marker4) |
|---|
| 269 | |
|---|
| 270 | if pagecount2 : |
|---|
| 271 | return pagecount2 |
|---|
| 272 | elif pagecount3 > 1 : |
|---|
| 273 | return pagecount3 - 1 |
|---|
| 274 | elif pagecount4 : |
|---|
| 275 | return pagecount4 |
|---|
| 276 | else : |
|---|
| 277 | return int(pagecount1 / 2) |
|---|
| 278 | |
|---|
| 279 | class PCLAnalyzer : |
|---|
| 280 | def __init__(self, infile) : |
|---|
| 281 | """Initialize PCL Analyzer.""" |
|---|
| 282 | self.infile = infile |
|---|
| 283 | |
|---|
| 284 | def getJobSize(self) : |
|---|
| 285 | """Count pages in a PCL5 document. |
|---|
| 286 | |
|---|
| 287 | Should also work for PCL3 and PCL4 documents. |
|---|
| 288 | |
|---|
| 289 | Algorithm from pclcount |
|---|
| 290 | (c) 2003, by Eduardo Gielamo Oliveira & Rodolfo Broco Manin |
|---|
| 291 | published under the terms of the GNU General Public Licence v2. |
|---|
| 292 | |
|---|
| 293 | Backported from C to Python by Jerome Alet, then enhanced |
|---|
| 294 | with more PCL tags detected. I think all the necessary PCL tags |
|---|
| 295 | are recognized to correctly handle PCL5 files wrt their number |
|---|
| 296 | of pages. The documentation used for this was : |
|---|
| 297 | |
|---|
| 298 | HP PCL/PJL Reference Set |
|---|
| 299 | PCL5 Printer Language Technical Quick Reference Guide |
|---|
| 300 | http://h20000.www2.hp.com/bc/docs/support/SupportManual/bpl13205/bpl13205.pdf |
|---|
| 301 | """ |
|---|
| 302 | infileno = self.infile.fileno() |
|---|
| 303 | minfile = mmap.mmap(infileno, os.fstat(infileno)[6], prot=mmap.PROT_READ, flags=mmap.MAP_SHARED) |
|---|
| 304 | tagsends = { "&n" : "W", |
|---|
| 305 | "&b" : "W", |
|---|
| 306 | "*i" : "W", |
|---|
| 307 | "*l" : "W", |
|---|
| 308 | "*m" : "W", |
|---|
| 309 | "*v" : "W", |
|---|
| 310 | "*c" : "W", |
|---|
| 311 | "(f" : "W", |
|---|
| 312 | "(s" : "W", |
|---|
| 313 | ")s" : "W", |
|---|
| 314 | "&p" : "X", |
|---|
| 315 | "&l" : "XH", |
|---|
| 316 | "&a" : "G", |
|---|
| 317 | # "*b" : "VW", # treated specially because it occurs very often |
|---|
| 318 | } |
|---|
| 319 | pagecount = resets = ejects = backsides = 0 |
|---|
| 320 | tag = None |
|---|
| 321 | copies = {} |
|---|
| 322 | pos = 0 |
|---|
| 323 | try : |
|---|
| 324 | while 1 : |
|---|
| 325 | char = minfile[pos] ; pos += 1 |
|---|
| 326 | if char == "\014" : |
|---|
| 327 | pagecount += 1 |
|---|
| 328 | elif char == "\033" : |
|---|
| 329 | # |
|---|
| 330 | # <ESC>*b###W -> Start of a raster data row/block |
|---|
| 331 | # <ESC>*b###V -> Start of a raster data plane |
|---|
| 332 | # <ESC>*c###W -> Start of a user defined pattern |
|---|
| 333 | # <ESC>*i###W -> Start of a viewing illuminant block |
|---|
| 334 | # <ESC>*l###W -> Start of a color lookup table |
|---|
| 335 | # <ESC>*m###W -> Start of a download dither matrix block |
|---|
| 336 | # <ESC>*v###W -> Start of a configure image data block |
|---|
| 337 | # <ESC>(s###W -> Start of a characters description block |
|---|
| 338 | # <ESC>)s###W -> Start of a fonts description block |
|---|
| 339 | # <ESC>(f###W -> Start of a symbol set block |
|---|
| 340 | # <ESC>&b###W -> Start of configuration data block |
|---|
| 341 | # <ESC>&l###X -> Number of copies for current page |
|---|
| 342 | # <ESC>&n###W -> Starts an alphanumeric string ID block |
|---|
| 343 | # <ESC>&p###X -> Start of a non printable characters block |
|---|
| 344 | # <ESC>&a2G -> Back side when duplex mode as generated by rastertohp |
|---|
| 345 | # <ESC>&l0H -> Eject if NumPlanes > 1, as generated by rastertohp |
|---|
| 346 | # |
|---|
| 347 | tagstart = minfile[pos] ; pos += 1 |
|---|
| 348 | if tagstart in "E9=YZ" : # one byte PCL tag |
|---|
| 349 | if tagstart == "E" : |
|---|
| 350 | resets += 1 |
|---|
| 351 | continue # skip to next tag |
|---|
| 352 | tag = tagstart + minfile[pos] ; pos += 1 |
|---|
| 353 | if tag == "*b" : |
|---|
| 354 | tagend = "VW" |
|---|
| 355 | else : |
|---|
| 356 | try : |
|---|
| 357 | tagend = tagsends[tag] |
|---|
| 358 | except KeyError : |
|---|
| 359 | continue # Unsupported PCL tag |
|---|
| 360 | # Now read the numeric argument |
|---|
| 361 | size = 0 |
|---|
| 362 | while 1 : |
|---|
| 363 | char = minfile[pos] ; pos += 1 |
|---|
| 364 | if not char.isdigit() : |
|---|
| 365 | break |
|---|
| 366 | size = (size * 10) + int(char) |
|---|
| 367 | if char in tagend : |
|---|
| 368 | if (tag == "&l") and (char == "X") : # copies for current page |
|---|
| 369 | copies[pagecount] = size |
|---|
| 370 | elif (tag == "&l") and (char == "H") and (size == 0) : |
|---|
| 371 | ejects += 1 # Eject |
|---|
| 372 | elif (tag == "&a") and (size == 2) : |
|---|
| 373 | backsides += 1 # Back side in duplex mode |
|---|
| 374 | else : |
|---|
| 375 | # we just ignore the block. |
|---|
| 376 | if tag == "&n" : |
|---|
| 377 | # we have to take care of the operation id byte |
|---|
| 378 | # which is before the string itself |
|---|
| 379 | size += 1 |
|---|
| 380 | pos += size |
|---|
| 381 | except IndexError : # EOF ? |
|---|
| 382 | minfile.close() # reached EOF |
|---|
| 383 | |
|---|
| 384 | # if pagecount is still 0, we will use the number |
|---|
| 385 | # of resets instead of the number of form feed characters. |
|---|
| 386 | # but the number of resets is always at least 2 with a valid |
|---|
| 387 | # pcl file : one at the very start and one at the very end |
|---|
| 388 | # of the job's data. So we substract 2 from the number of |
|---|
| 389 | # resets. And since on our test data we needed to substract |
|---|
| 390 | # 1 more, we finally substract 3, and will test several |
|---|
| 391 | # PCL files with this. If resets < 2, then the file is |
|---|
| 392 | # probably not a valid PCL file, so we use 0 |
|---|
| 393 | if not pagecount : |
|---|
| 394 | pagecount = (pagecount or ((resets - 3) * (resets > 2))) |
|---|
| 395 | else : |
|---|
| 396 | # here we add counters for other ways new pages may have |
|---|
| 397 | # been printed and ejected by the printer |
|---|
| 398 | pagecount += ejects + backsides |
|---|
| 399 | |
|---|
| 400 | # now handle number of copies for each page (may differ). |
|---|
| 401 | # in duplex mode, number of copies may be sent only once. |
|---|
| 402 | for pnum in range(pagecount) : |
|---|
| 403 | # if no number of copies defined, take the preceding one else the one set before any page else 1. |
|---|
| 404 | nb = copies.get(pnum, copies.get(pnum-1, copies.get(0, 1))) |
|---|
| 405 | pagecount += (nb - 1) |
|---|
| 406 | return pagecount |
|---|
| 407 | |
|---|
| 408 | class PCLXLAnalyzer : |
|---|
| 409 | def __init__(self, infile) : |
|---|
| 410 | """Initialize PCLXL Analyzer.""" |
|---|
| 411 | self.infile = infile |
|---|
| 412 | self.endianness = None |
|---|
| 413 | found = 0 |
|---|
| 414 | while not found : |
|---|
| 415 | line = self.infile.readline() |
|---|
| 416 | if not line : |
|---|
| 417 | break |
|---|
| 418 | if line[1:12] == " HP-PCL XL;" : |
|---|
| 419 | found = 1 |
|---|
| 420 | endian = ord(line[0]) |
|---|
| 421 | if endian == 0x29 : |
|---|
| 422 | self.littleEndian() |
|---|
| 423 | elif endian == 0x28 : |
|---|
| 424 | self.bigEndian() |
|---|
| 425 | # elif endian == 0x27 : TODO : What can we do here ? |
|---|
| 426 | # |
|---|
| 427 | else : |
|---|
| 428 | raise PDLAnalyzerError, "Unknown endianness marker 0x%02x at start !" % endian |
|---|
| 429 | if not found : |
|---|
| 430 | raise PDLAnalyzerError, "This file doesn't seem to be PCLXL (aka PCL6)" |
|---|
| 431 | else : |
|---|
| 432 | # Initialize table of tags |
|---|
| 433 | self.tags = [ 0 ] * 256 |
|---|
| 434 | |
|---|
| 435 | # GhostScript's sources tell us that HP printers |
|---|
| 436 | # only accept little endianness, but we can handle both. |
|---|
| 437 | self.tags[0x28] = self.bigEndian # BigEndian |
|---|
| 438 | self.tags[0x29] = self.littleEndian # LittleEndian |
|---|
| 439 | |
|---|
| 440 | self.tags[0x43] = self.beginPage # BeginPage |
|---|
| 441 | self.tags[0x44] = self.endPage # EndPage |
|---|
| 442 | |
|---|
| 443 | self.tags[0xc0] = 1 # ubyte |
|---|
| 444 | self.tags[0xc1] = 2 # uint16 |
|---|
| 445 | self.tags[0xc2] = 4 # uint32 |
|---|
| 446 | self.tags[0xc3] = 2 # sint16 |
|---|
| 447 | self.tags[0xc4] = 4 # sint32 |
|---|
| 448 | self.tags[0xc5] = 4 # real32 |
|---|
| 449 | |
|---|
| 450 | self.tags[0xc8] = self.array_8 # ubyte_array |
|---|
| 451 | self.tags[0xc9] = self.array_16 # uint16_array |
|---|
| 452 | self.tags[0xca] = self.array_32 # uint32_array |
|---|
| 453 | self.tags[0xcb] = self.array_16 # sint16_array |
|---|
| 454 | self.tags[0xcc] = self.array_32 # sint32_array |
|---|
| 455 | self.tags[0xcd] = self.array_32 # real32_array |
|---|
| 456 | |
|---|
| 457 | self.tags[0xd0] = 2 # ubyte_xy |
|---|
| 458 | self.tags[0xd1] = 4 # uint16_xy |
|---|
| 459 | self.tags[0xd2] = 8 # uint32_xy |
|---|
| 460 | self.tags[0xd3] = 4 # sint16_xy |
|---|
| 461 | self.tags[0xd4] = 8 # sint32_xy |
|---|
| 462 | self.tags[0xd5] = 8 # real32_xy |
|---|
| 463 | |
|---|
| 464 | self.tags[0xe0] = 4 # ubyte_box |
|---|
| 465 | self.tags[0xe1] = 8 # uint16_box |
|---|
| 466 | self.tags[0xe2] = 16 # uint32_box |
|---|
| 467 | self.tags[0xe3] = 8 # sint16_box |
|---|
| 468 | self.tags[0xe4] = 16 # sint32_box |
|---|
| 469 | self.tags[0xe5] = 16 # real32_box |
|---|
| 470 | |
|---|
| 471 | self.tags[0xf8] = 1 # attr_ubyte |
|---|
| 472 | self.tags[0xf9] = 2 # attr_uint16 |
|---|
| 473 | |
|---|
| 474 | self.tags[0xfa] = self.embeddedData # dataLength |
|---|
| 475 | self.tags[0xfb] = self.embeddedDataSmall # dataLengthByte |
|---|
| 476 | |
|---|
| 477 | def beginPage(self) : |
|---|
| 478 | """Indicates the beginning of a new page.""" |
|---|
| 479 | self.pagecount += 1 |
|---|
| 480 | return 0 |
|---|
| 481 | |
|---|
| 482 | def endPage(self) : |
|---|
| 483 | """Indicates the end of a page.""" |
|---|
| 484 | pos = self.pos |
|---|
| 485 | minfile = self.minfile |
|---|
| 486 | if (ord(minfile[pos-3]) == 0xf8) and (ord(minfile[pos-2]) == 0x31) : |
|---|
| 487 | # The EndPage operator is preceded by a PageCopies attribute |
|---|
| 488 | # So set number of copies for current page. |
|---|
| 489 | # From what I read in PCLXL documentation, the number |
|---|
| 490 | # of copies is an unsigned 16 bits integer |
|---|
| 491 | self.copies[self.pagecount] = unpack(self.endianness + "H", minfile[pos-5:pos-3])[0] |
|---|
| 492 | return 0 |
|---|
| 493 | |
|---|
| 494 | def array_8(self) : |
|---|
| 495 | """Handles byte arrays.""" |
|---|
| 496 | pos = self.pos |
|---|
| 497 | datatype = self.minfile[pos] |
|---|
| 498 | pos += 1 |
|---|
| 499 | length = self.tags[ord(datatype)] |
|---|
| 500 | if callable(length) : |
|---|
| 501 | self.pos = pos |
|---|
| 502 | length = length() |
|---|
| 503 | pos = self.pos |
|---|
| 504 | posl = pos + length |
|---|
| 505 | self.pos = posl |
|---|
| 506 | if length == 1 : |
|---|
| 507 | return unpack("B", self.minfile[pos:posl])[0] |
|---|
| 508 | elif length == 2 : |
|---|
| 509 | return unpack(self.endianness + "H", self.minfile[pos:posl])[0] |
|---|
| 510 | elif length == 4 : |
|---|
| 511 | return unpack(self.endianness + "I", self.minfile[pos:posl])[0] |
|---|
| 512 | else : |
|---|
| 513 | raise PDLAnalyzerError, "Error on array size at %s" % self.pos |
|---|
| 514 | |
|---|
| 515 | def array_16(self) : |
|---|
| 516 | """Handles byte arrays.""" |
|---|
| 517 | pos = self.pos |
|---|
| 518 | datatype = self.minfile[pos] |
|---|
| 519 | pos += 1 |
|---|
| 520 | length = self.tags[ord(datatype)] |
|---|
| 521 | if callable(length) : |
|---|
| 522 | self.pos = pos |
|---|
| 523 | length = length() |
|---|
| 524 | pos = self.pos |
|---|
| 525 | posl = pos + length |
|---|
| 526 | self.pos = posl |
|---|
| 527 | if length == 1 : |
|---|
| 528 | return 2 * unpack("B", self.minfile[pos:posl])[0] |
|---|
| 529 | elif length == 2 : |
|---|
| 530 | return 2 * unpack(self.endianness + "H", self.minfile[pos:posl])[0] |
|---|
| 531 | elif length == 4 : |
|---|
| 532 | return 2 * unpack(self.endianness + "I", self.minfile[pos:posl])[0] |
|---|
| 533 | else : |
|---|
| 534 | raise PDLAnalyzerError, "Error on array size at %s" % self.pos |
|---|
| 535 | |
|---|
| 536 | def array_32(self) : |
|---|
| 537 | """Handles byte arrays.""" |
|---|
| 538 | pos = self.pos |
|---|
| 539 | datatype = self.minfile[pos] |
|---|
| 540 | pos += 1 |
|---|
| 541 | length = self.tags[ord(datatype)] |
|---|
| 542 | if callable(length) : |
|---|
| 543 | self.pos = pos |
|---|
| 544 | length = length() |
|---|
| 545 | pos = self.pos |
|---|
| 546 | posl = pos + length |
|---|
| 547 | self.pos = posl |
|---|
| 548 | if length == 1 : |
|---|
| 549 | return 4 * unpack("B", self.minfile[pos:posl])[0] |
|---|
| 550 | elif length == 2 : |
|---|
| 551 | return 4 * unpack(self.endianness + "H", self.minfile[pos:posl])[0] |
|---|
| 552 | elif length == 4 : |
|---|
| 553 | return 4 * unpack(self.endianness + "I", self.minfile[pos:posl])[0] |
|---|
| 554 | else : |
|---|
| 555 | raise PDLAnalyzerError, "Error on array size at %s" % self.pos |
|---|
| 556 | |
|---|
| 557 | def embeddedDataSmall(self) : |
|---|
| 558 | """Handle small amounts of data.""" |
|---|
| 559 | pos = self.pos |
|---|
| 560 | length = ord(self.minfile[pos]) |
|---|
| 561 | self.pos = pos + 1 |
|---|
| 562 | return length |
|---|
| 563 | |
|---|
| 564 | def embeddedData(self) : |
|---|
| 565 | """Handle normal amounts of data.""" |
|---|
| 566 | pos = self.pos |
|---|
| 567 | pos4 = pos + 4 |
|---|
| 568 | self.pos = pos4 |
|---|
| 569 | return unpack(self.endianness + "I", self.minfile[pos:pos4])[0] |
|---|
| 570 | |
|---|
| 571 | def littleEndian(self) : |
|---|
| 572 | """Toggles to little endianness.""" |
|---|
| 573 | self.endianness = "<" # little endian |
|---|
| 574 | return 0 |
|---|
| 575 | |
|---|
| 576 | def bigEndian(self) : |
|---|
| 577 | """Toggles to big endianness.""" |
|---|
| 578 | self.endianness = ">" # big endian |
|---|
| 579 | return 0 |
|---|
| 580 | |
|---|
| 581 | def getJobSize(self) : |
|---|
| 582 | """Counts pages in a PCLXL (PCL6) document. |
|---|
| 583 | |
|---|
| 584 | Algorithm by Jerome Alet. |
|---|
| 585 | |
|---|
| 586 | The documentation used for this was : |
|---|
| 587 | |
|---|
| 588 | HP PCL XL Feature Reference |
|---|
| 589 | Protocol Class 2.0 |
|---|
| 590 | http://www.hpdevelopersolutions.com/downloads/64/358/xl_ref20r22.pdf |
|---|
| 591 | """ |
|---|
| 592 | infileno = self.infile.fileno() |
|---|
| 593 | self.copies = {} |
|---|
| 594 | self.minfile = minfile = mmap.mmap(infileno, os.fstat(infileno)[6], prot=mmap.PROT_READ, flags=mmap.MAP_SHARED) |
|---|
| 595 | tags = self.tags |
|---|
| 596 | self.pagecount = 0 |
|---|
| 597 | self.pos = pos = self.infile.tell() |
|---|
| 598 | try : |
|---|
| 599 | while 1 : |
|---|
| 600 | char = minfile[pos] |
|---|
| 601 | pos += 1 |
|---|
| 602 | length = tags[ord(char)] |
|---|
| 603 | if not length : |
|---|
| 604 | continue |
|---|
| 605 | if callable(length) : |
|---|
| 606 | self.pos = pos |
|---|
| 607 | length = length() |
|---|
| 608 | pos = self.pos |
|---|
| 609 | pos += length |
|---|
| 610 | except IndexError : # EOF ? |
|---|
| 611 | self.minfile.close() # reached EOF |
|---|
| 612 | |
|---|
| 613 | # now handle number of copies for each page (may differ). |
|---|
| 614 | for pnum in range(1, self.pagecount + 1) : |
|---|
| 615 | # if no number of copies defined, take 1, as explained |
|---|
| 616 | # in PCLXL documentation. |
|---|
| 617 | # NB : is number of copies is 0, the page won't be output |
|---|
| 618 | # but the formula below is still correct : we want |
|---|
| 619 | # to decrease the total number of pages in this case. |
|---|
| 620 | self.pagecount += (self.copies.get(pnum, 1) - 1) |
|---|
| 621 | |
|---|
| 622 | return self.pagecount |
|---|
| 623 | |
|---|
| 624 | class PDLAnalyzer : |
|---|
| 625 | """Generic PDL Analyzer class.""" |
|---|
| 626 | def __init__(self, filename) : |
|---|
| 627 | """Initializes the PDL analyzer. |
|---|
| 628 | |
|---|
| 629 | filename is the name of the file or '-' for stdin. |
|---|
| 630 | filename can also be a file-like object which |
|---|
| 631 | supports read() and seek(). |
|---|
| 632 | """ |
|---|
| 633 | self.filename = filename |
|---|
| 634 | try : |
|---|
| 635 | import psyco |
|---|
| 636 | except ImportError : |
|---|
| 637 | pass # Psyco is not installed |
|---|
| 638 | else : |
|---|
| 639 | # Psyco is installed, tell it to compile |
|---|
| 640 | # the CPU intensive methods : PCL and PCLXL |
|---|
| 641 | # parsing will greatly benefit from this, |
|---|
| 642 | # for PostScript and PDF the difference is |
|---|
| 643 | # barely noticeable since they are already |
|---|
| 644 | # almost optimal, and much more speedy anyway. |
|---|
| 645 | psyco.bind(PostScriptAnalyzer.getJobSize) |
|---|
| 646 | psyco.bind(PDFAnalyzer.getJobSize) |
|---|
| 647 | psyco.bind(ESCP2Analyzer.getJobSize) |
|---|
| 648 | psyco.bind(PCLAnalyzer.getJobSize) |
|---|
| 649 | psyco.bind(PCLXLAnalyzer.getJobSize) |
|---|
| 650 | |
|---|
| 651 | def getJobSize(self) : |
|---|
| 652 | """Returns the job's size.""" |
|---|
| 653 | self.openFile() |
|---|
| 654 | try : |
|---|
| 655 | pdlhandler = self.detectPDLHandler() |
|---|
| 656 | except PDLAnalyzerError, msg : |
|---|
| 657 | self.closeFile() |
|---|
| 658 | raise PDLAnalyzerError, "ERROR : Unknown file format for %s (%s)" % (self.filename, msg) |
|---|
| 659 | else : |
|---|
| 660 | try : |
|---|
| 661 | size = pdlhandler(self.infile).getJobSize() |
|---|
| 662 | finally : |
|---|
| 663 | self.closeFile() |
|---|
| 664 | return size |
|---|
| 665 | |
|---|
| 666 | def openFile(self) : |
|---|
| 667 | """Opens the job's data stream for reading.""" |
|---|
| 668 | self.mustclose = 0 # by default we don't want to close the file when finished |
|---|
| 669 | if hasattr(self.filename, "read") and hasattr(self.filename, "seek") : |
|---|
| 670 | # filename is in fact a file-like object |
|---|
| 671 | infile = self.filename |
|---|
| 672 | elif self.filename == "-" : |
|---|
| 673 | # we must read from stdin |
|---|
| 674 | infile = sys.stdin |
|---|
| 675 | else : |
|---|
| 676 | # normal file |
|---|
| 677 | self.infile = open(self.filename, "rb") |
|---|
| 678 | self.mustclose = 1 |
|---|
| 679 | return |
|---|
| 680 | |
|---|
| 681 | # Use a temporary file, always seekable contrary to standard input. |
|---|
| 682 | self.infile = tempfile.TemporaryFile(mode="w+b") |
|---|
| 683 | while 1 : |
|---|
| 684 | data = infile.read(MEGABYTE) |
|---|
| 685 | if not data : |
|---|
| 686 | break |
|---|
| 687 | self.infile.write(data) |
|---|
| 688 | self.infile.flush() |
|---|
| 689 | self.infile.seek(0) |
|---|
| 690 | |
|---|
| 691 | def closeFile(self) : |
|---|
| 692 | """Closes the job's data stream if we can close it.""" |
|---|
| 693 | if self.mustclose : |
|---|
| 694 | self.infile.close() |
|---|
| 695 | else : |
|---|
| 696 | # if we don't have to close the file, then |
|---|
| 697 | # ensure the file pointer is reset to the |
|---|
| 698 | # start of the file in case the process wants |
|---|
| 699 | # to read the file again. |
|---|
| 700 | try : |
|---|
| 701 | self.infile.seek(0) |
|---|
| 702 | except : |
|---|
| 703 | pass # probably stdin, which is not seekable |
|---|
| 704 | |
|---|
| 705 | def isPostScript(self, data) : |
|---|
| 706 | """Returns 1 if data is PostScript, else 0.""" |
|---|
| 707 | if data.startswith("%!") or \ |
|---|
| 708 | data.startswith("\004%!") or \ |
|---|
| 709 | data.startswith("\033%-12345X%!PS") or \ |
|---|
| 710 | ((data[:128].find("\033%-12345X") != -1) and \ |
|---|
| 711 | ((data.find("LANGUAGE=POSTSCRIPT") != -1) or \ |
|---|
| 712 | (data.find("LANGUAGE = POSTSCRIPT") != -1) or \ |
|---|
| 713 | (data.find("LANGUAGE = Postscript") != -1))) or \ |
|---|
| 714 | (data.find("%!PS-Adobe") != -1) : |
|---|
| 715 | return 1 |
|---|
| 716 | else : |
|---|
| 717 | return 0 |
|---|
| 718 | |
|---|
| 719 | def isPDF(self, data) : |
|---|
| 720 | """Returns 1 if data is PDF, else 0.""" |
|---|
| 721 | if data.startswith("%PDF-") or \ |
|---|
| 722 | data.startswith("\033%-12345X%PDF-") or \ |
|---|
| 723 | ((data[:128].find("\033%-12345X") != -1) and (data.upper().find("LANGUAGE=PDF") != -1)) or \ |
|---|
| 724 | (data.find("%PDF-") != -1) : |
|---|
| 725 | return 1 |
|---|
| 726 | else : |
|---|
| 727 | return 0 |
|---|
| 728 | |
|---|
| 729 | def isPCL(self, data) : |
|---|
| 730 | """Returns 1 if data is PCL, else 0.""" |
|---|
| 731 | if data.startswith("\033E\033") or \ |
|---|
| 732 | (data[:128].find("\033%-12345X") != -1) : |
|---|
| 733 | return 1 |
|---|
| 734 | else : |
|---|
| 735 | return 0 |
|---|
| 736 | |
|---|
| 737 | def isPCLXL(self, data) : |
|---|
| 738 | """Returns 1 if data is PCLXL aka PCL6, else 0.""" |
|---|
| 739 | if ((data[:128].find("\033%-12345X") != -1) and \ |
|---|
| 740 | (data.find(" HP-PCL XL;") != -1) and \ |
|---|
| 741 | ((data.find("LANGUAGE=PCLXL") != -1) or \ |
|---|
| 742 | (data.find("LANGUAGE = PCLXL") != -1))) : |
|---|
| 743 | return 1 |
|---|
| 744 | else : |
|---|
| 745 | return 0 |
|---|
| 746 | |
|---|
| 747 | def isESCP2(self, data) : |
|---|
| 748 | """Returns 1 if data is ESC/P2, else 0.""" |
|---|
| 749 | if data.startswith("\033@") or \ |
|---|
| 750 | data.startswith("\033*") or \ |
|---|
| 751 | data.startswith("\n\033@") : |
|---|
| 752 | return 1 |
|---|
| 753 | else : |
|---|
| 754 | return 0 |
|---|
| 755 | |
|---|
| 756 | def detectPDLHandler(self) : |
|---|
| 757 | """Tries to autodetect the document format. |
|---|
| 758 | |
|---|
| 759 | Returns the correct PDL handler class or None if format is unknown |
|---|
| 760 | """ |
|---|
| 761 | # Try to detect file type by reading first block of datas |
|---|
| 762 | self.infile.seek(0) |
|---|
| 763 | firstblock = self.infile.read(KILOBYTE) |
|---|
| 764 | self.infile.seek(0) |
|---|
| 765 | if self.isPostScript(firstblock) : |
|---|
| 766 | return PostScriptAnalyzer |
|---|
| 767 | elif self.isPCLXL(firstblock) : |
|---|
| 768 | return PCLXLAnalyzer |
|---|
| 769 | elif self.isPDF(firstblock) : |
|---|
| 770 | return PDFAnalyzer |
|---|
| 771 | elif self.isPCL(firstblock) : |
|---|
| 772 | return PCLAnalyzer |
|---|
| 773 | elif self.isESCP2(firstblock) : |
|---|
| 774 | return ESCP2Analyzer |
|---|
| 775 | else : |
|---|
| 776 | raise PDLAnalyzerError, "Analysis of first data block failed." |
|---|
| 777 | |
|---|
| 778 | def main() : |
|---|
| 779 | """Entry point for PDL Analyzer.""" |
|---|
| 780 | if (len(sys.argv) < 2) or ((not sys.stdin.isatty()) and ("-" not in sys.argv[1:])) : |
|---|
| 781 | sys.argv.append("-") |
|---|
| 782 | |
|---|
| 783 | totalsize = 0 |
|---|
| 784 | for arg in sys.argv[1:] : |
|---|
| 785 | try : |
|---|
| 786 | parser = PDLAnalyzer(arg) |
|---|
| 787 | totalsize += parser.getJobSize() |
|---|
| 788 | except PDLAnalyzerError, msg : |
|---|
| 789 | sys.stderr.write("ERROR: %s\n" % msg) |
|---|
| 790 | sys.stderr.flush() |
|---|
| 791 | print "%s" % totalsize |
|---|
| 792 | |
|---|
| 793 | if __name__ == "__main__" : |
|---|
| 794 | main() |
|---|
