54 | | |
55 | | class PDLAnalyzer : |
56 | | """Class for PDL autodetection.""" |
57 | | def __init__(self, filename, debug=0) : |
58 | | """Initializes the PDL analyzer. |
59 | | |
60 | | filename is the name of the file or '-' for stdin. |
61 | | filename can also be a file-like object which |
62 | | supports read() and seek(). |
63 | | """ |
64 | | self.debug = debug |
65 | | self.filename = filename |
66 | | try : |
67 | | import psyco |
68 | | except ImportError : |
69 | | sys.stderr.write("pkpgcounter : you should install psyco if possible, this would greatly speedup parsing.\n") |
70 | | pass # Psyco is not installed |
71 | | else : |
72 | | # Psyco is installed, tell it to compile |
73 | | # the CPU intensive methods : PCL and PCLXL |
74 | | # parsing will greatly benefit from this, |
75 | | # for PostScript and PDF the difference is |
76 | | # barely noticeable since they are already |
77 | | # almost optimal, and much more speedy anyway. |
78 | | psyco.bind(PostScriptParser.getJobSize) |
79 | | psyco.bind(PDFParser.getJobSize) |
80 | | psyco.bind(ESCP2Parser.getJobSize) |
81 | | psyco.bind(PCLParser.getJobSize) |
82 | | psyco.bind(PCLXLParser.getJobSize) |
83 | | |
84 | | def getJobSize(self) : |
85 | | """Returns the job's size.""" |
86 | | self.openFile() |
87 | | try : |
88 | | pdlhandler = self.detectPDLHandler() |
89 | | except PDLParserError, msg : |
90 | | self.closeFile() |
91 | | raise PDLParserError, "ERROR : Unknown file format for %s (%s)" % (self.filename, msg) |
92 | | else : |
93 | | try : |
94 | | size = pdlhandler(self.infile, self.debug).getJobSize() |
95 | | finally : |
96 | | self.closeFile() |
97 | | return size |
98 | | |
99 | | def openFile(self) : |
100 | | """Opens the job's data stream for reading.""" |
101 | | self.mustclose = 0 # by default we don't want to close the file when finished |
102 | | if hasattr(self.filename, "read") and hasattr(self.filename, "seek") : |
103 | | # filename is in fact a file-like object |
104 | | infile = self.filename |
105 | | elif self.filename == "-" : |
106 | | # we must read from stdin |
107 | | infile = sys.stdin |
108 | | else : |
109 | | # normal file |
110 | | self.infile = open(self.filename, "rb") |
111 | | self.mustclose = 1 |
112 | | return |
113 | | |
114 | | # Use a temporary file, always seekable contrary to standard input. |
115 | | self.infile = tempfile.TemporaryFile(mode="w+b") |
116 | | while 1 : |
117 | | data = infile.read(MEGABYTE) |
118 | | if not data : |
119 | | break |
120 | | self.infile.write(data) |
121 | | self.infile.flush() |
122 | | self.infile.seek(0) |
123 | | |
124 | | def closeFile(self) : |
125 | | """Closes the job's data stream if we can close it.""" |
126 | | if self.mustclose : |
127 | | self.infile.close() |
128 | | else : |
129 | | # if we don't have to close the file, then |
130 | | # ensure the file pointer is reset to the |
131 | | # start of the file in case the process wants |
132 | | # to read the file again. |
133 | | try : |
134 | | self.infile.seek(0) |
135 | | except : |
136 | | pass # probably stdin, which is not seekable |
137 | | |
138 | | def isPostScript(self, sdata, edata) : |
139 | | """Returns 1 if data is PostScript, else 0.""" |
140 | | if sdata.startswith("%!") or \ |
141 | | sdata.startswith("\004%!") or \ |
142 | | sdata.startswith("\033%-12345X%!PS") or \ |
143 | | ((sdata[:128].find("\033%-12345X") != -1) and \ |
144 | | ((sdata.find("LANGUAGE=POSTSCRIPT") != -1) or \ |
145 | | (sdata.find("LANGUAGE = POSTSCRIPT") != -1) or \ |
146 | | (sdata.find("LANGUAGE = Postscript") != -1))) or \ |
147 | | (sdata.find("%!PS-Adobe") != -1) : |
148 | | if self.debug : |
149 | | sys.stderr.write("%s is a PostScript file\n" % str(self.filename)) |
150 | | return 1 |
151 | | else : |
152 | | return 0 |
153 | | |
154 | | def isPDF(self, sdata, edata) : |
155 | | """Returns 1 if data is PDF, else 0.""" |
156 | | if sdata.startswith("%PDF-") or \ |
157 | | sdata.startswith("\033%-12345X%PDF-") or \ |
158 | | ((sdata[:128].find("\033%-12345X") != -1) and (sdata.upper().find("LANGUAGE=PDF") != -1)) or \ |
159 | | (sdata.find("%PDF-") != -1) : |
160 | | if self.debug : |
161 | | sys.stderr.write("%s is a PDF file\n" % str(self.filename)) |
162 | | return 1 |
163 | | else : |
164 | | return 0 |
165 | | |
166 | | def isPCL(self, sdata, edata) : |
167 | | """Returns 1 if data is PCL, else 0.""" |
168 | | if sdata.startswith("\033E\033") or \ |
169 | | (sdata.startswith("\033*rbC") and (not edata[-3:] == "\f\033@")) or \ |
170 | | sdata.startswith("\033%8\033") or \ |
171 | | (sdata.find("\033%-12345X") != -1) : |
172 | | if self.debug : |
173 | | sys.stderr.write("%s is a PCL3/4/5 file\n" % str(self.filename)) |
174 | | return 1 |
175 | | else : |
176 | | return 0 |
177 | | |
178 | | def isPCLXL(self, sdata, edata) : |
179 | | """Returns 1 if data is PCLXL aka PCL6, else 0.""" |
180 | | if ((sdata[:128].find("\033%-12345X") != -1) and \ |
181 | | (sdata.find(" HP-PCL XL;") != -1) and \ |
182 | | ((sdata.find("LANGUAGE=PCLXL") != -1) or \ |
183 | | (sdata.find("LANGUAGE = PCLXL") != -1))) : |
184 | | if self.debug : |
185 | | sys.stderr.write("%s is a PCLXL (aka PCL6) file\n" % str(self.filename)) |
186 | | return 1 |
187 | | else : |
188 | | return 0 |
189 | | |
190 | | def isESCP2(self, sdata, edata) : |
191 | | """Returns 1 if data is ESC/P2, else 0.""" |
192 | | if sdata.startswith("\033@") or \ |
193 | | sdata.startswith("\033*") or \ |
194 | | sdata.startswith("\n\033@") or \ |
195 | | sdata.startswith("\0\0\0\033\1@EJL") : # ESC/P Raster ??? Seen on Stylus Photo 1284 |
196 | | if self.debug : |
197 | | sys.stderr.write("%s is an ESC/P2 file\n" % str(self.filename)) |
198 | | return 1 |
199 | | else : |
200 | | return 0 |
201 | | |
202 | | def detectPDLHandler(self) : |
203 | | """Tries to autodetect the document format. |
204 | | |
205 | | Returns the correct PDL handler class or None if format is unknown |
206 | | """ |
207 | | # Try to detect file type by reading first block of datas |
208 | | self.infile.seek(0) |
209 | | firstblock = self.infile.read(4 * KILOBYTE) |
210 | | try : |
211 | | self.infile.seek(-LASTBLOCKSIZE, 2) |
212 | | lastblock = self.infile.read(LASTBLOCKSIZE) |
213 | | except IOError : |
214 | | lastblock = "" |
215 | | self.infile.seek(0) |
216 | | if self.isPostScript(firstblock, lastblock) : |
217 | | return PostScriptParser |
218 | | elif self.isPCLXL(firstblock, lastblock) : |
219 | | return PCLXLParser |
220 | | elif self.isPDF(firstblock, lastblock) : |
221 | | return PDFParser |
222 | | elif self.isPCL(firstblock, lastblock) : |
223 | | return PCLParser |
224 | | elif self.isESCP2(firstblock, lastblock) : |
225 | | return ESCP2Parser |
226 | | else : |
227 | | raise PDLParserError, "Analysis of first data block failed." |
228 | | |
229 | | def main() : |
230 | | """Entry point for PDL Analyzer.""" |
231 | | if (len(sys.argv) < 2) or ((not sys.stdin.isatty()) and ("-" not in sys.argv[1:])) : |
232 | | sys.argv.append("-") |
233 | | |
234 | | if ("-h" in sys.argv[1:]) or ("--help" in sys.argv[1:]) : |
235 | | print "usage : pkpgcounter file1 file2 ... fileN" |
236 | | elif ("-v" in sys.argv[1:]) or ("--version" in sys.argv[1:]) : |
237 | | print "%s" % version.__version__ |
238 | | else : |
239 | | totalsize = 0 |
240 | | debug = 0 |
241 | | minindex = 1 |
242 | | if sys.argv[1] == "--debug" : |
243 | | minindex = 2 |
244 | | debug = 1 |
245 | | for arg in sys.argv[minindex:] : |
246 | | try : |
247 | | parser = PDLAnalyzer(arg, debug) |
248 | | totalsize += parser.getJobSize() |
249 | | except PDLParserError, msg : |
250 | | sys.stderr.write("ERROR: %s\n" % msg) |
251 | | sys.stderr.flush() |
252 | | print "%s" % totalsize |
253 | | |
254 | | if __name__ == "__main__" : |
255 | | main() |