1 | # |
---|
2 | # pkpgcounter : a generic Page Description Language parser |
---|
3 | # |
---|
4 | # (c) 2003, 2004, 2005 Jerome Alet <alet@librelogiciel.com> |
---|
5 | # This program is free software; you can redistribute it and/or modify |
---|
6 | # it under the terms of the GNU General Public License as published by |
---|
7 | # the Free Software Foundation; either version 2 of the License, or |
---|
8 | # (at your option) any later version. |
---|
9 | # |
---|
10 | # This program is distributed in the hope that it will be useful, |
---|
11 | # but WITHOUT ANY WARRANTY; without even the implied warranty of |
---|
12 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
---|
13 | # GNU General Public License for more details. |
---|
14 | # |
---|
15 | # You should have received a copy of the GNU General Public License |
---|
16 | # along with this program; if not, write to the Free Software |
---|
17 | # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. |
---|
18 | # |
---|
19 | # $Id$ |
---|
20 | # |
---|
21 | |
---|
22 | import sys |
---|
23 | import tempfile |
---|
24 | |
---|
25 | from pdlanalyzer import version, pdlparser, postscript, pdf, pcl345, pclxl, escp2, dvi, tiff |
---|
26 | |
---|
27 | class PDLAnalyzer : |
---|
28 | """Class for PDL autodetection.""" |
---|
29 | def __init__(self, filename, debug=0) : |
---|
30 | """Initializes the PDL analyzer. |
---|
31 | |
---|
32 | filename is the name of the file or '-' for stdin. |
---|
33 | filename can also be a file-like object which |
---|
34 | supports read() and seek(). |
---|
35 | """ |
---|
36 | self.debug = debug |
---|
37 | self.filename = filename |
---|
38 | |
---|
39 | def getJobSize(self) : |
---|
40 | """Returns the job's size.""" |
---|
41 | self.openFile() |
---|
42 | try : |
---|
43 | pdlhandler = self.detectPDLHandler() |
---|
44 | except pdlparser.PDLParserError, msg : |
---|
45 | self.closeFile() |
---|
46 | raise pdlparser.PDLParserError, "Unknown file format for %s (%s)" % (self.filename, msg) |
---|
47 | else : |
---|
48 | try : |
---|
49 | size = pdlhandler.getJobSize() |
---|
50 | finally : |
---|
51 | self.closeFile() |
---|
52 | return size |
---|
53 | |
---|
54 | def openFile(self) : |
---|
55 | """Opens the job's data stream for reading.""" |
---|
56 | self.mustclose = 0 # by default we don't want to close the file when finished |
---|
57 | if hasattr(self.filename, "read") and hasattr(self.filename, "seek") : |
---|
58 | # filename is in fact a file-like object |
---|
59 | infile = self.filename |
---|
60 | elif self.filename == "-" : |
---|
61 | # we must read from stdin |
---|
62 | infile = sys.stdin |
---|
63 | else : |
---|
64 | # normal file |
---|
65 | self.infile = open(self.filename, "rb") |
---|
66 | self.mustclose = 1 |
---|
67 | return |
---|
68 | |
---|
69 | # Use a temporary file, always seekable contrary to standard input. |
---|
70 | self.infile = tempfile.TemporaryFile(mode="w+b") |
---|
71 | while 1 : |
---|
72 | data = infile.read(pdlparser.MEGABYTE) |
---|
73 | if not data : |
---|
74 | break |
---|
75 | self.infile.write(data) |
---|
76 | self.infile.flush() |
---|
77 | self.infile.seek(0) |
---|
78 | |
---|
79 | def closeFile(self) : |
---|
80 | """Closes the job's data stream if we can close it.""" |
---|
81 | if self.mustclose : |
---|
82 | self.infile.close() |
---|
83 | else : |
---|
84 | # if we don't have to close the file, then |
---|
85 | # ensure the file pointer is reset to the |
---|
86 | # start of the file in case the process wants |
---|
87 | # to read the file again. |
---|
88 | try : |
---|
89 | self.infile.seek(0) |
---|
90 | except : |
---|
91 | pass # probably stdin, which is not seekable |
---|
92 | |
---|
93 | def detectPDLHandler(self) : |
---|
94 | """Tries to autodetect the document format. |
---|
95 | |
---|
96 | Returns the correct PDL handler class or None if format is unknown |
---|
97 | """ |
---|
98 | # Try to detect file type by reading first and last blocks of datas |
---|
99 | # Each parser can read them automatically, but here we do this only once. |
---|
100 | self.infile.seek(0) |
---|
101 | firstblock = self.infile.read(pdlparser.FIRSTBLOCKSIZE) |
---|
102 | try : |
---|
103 | self.infile.seek(-pdlparser.LASTBLOCKSIZE, 2) |
---|
104 | lastblock = self.infile.read(pdlparser.LASTBLOCKSIZE) |
---|
105 | except IOError : |
---|
106 | lastblock = "" |
---|
107 | self.infile.seek(0) |
---|
108 | if not firstblock : |
---|
109 | raise pdlparser.PDLParserError, "input file %s is empty !" % str(self.filename) |
---|
110 | else : |
---|
111 | for module in (postscript, \ |
---|
112 | pclxl, \ |
---|
113 | pdf, \ |
---|
114 | pcl345, \ |
---|
115 | escp2, \ |
---|
116 | dvi, \ |
---|
117 | tiff) : |
---|
118 | try : |
---|
119 | return getattr(module, "Parser")(self.infile, self.debug, firstblock, lastblock) |
---|
120 | except pdlparser.PDLParserError : |
---|
121 | pass # try next parser |
---|
122 | raise pdlparser.PDLParserError, "Analysis of first data block failed." |
---|
123 | |
---|
124 | def main() : |
---|
125 | """Entry point for PDL Analyzer.""" |
---|
126 | if (len(sys.argv) < 2) or ((not sys.stdin.isatty()) and ("-" not in sys.argv[1:])) : |
---|
127 | sys.argv.append("-") |
---|
128 | |
---|
129 | if ("-h" in sys.argv[1:]) or ("--help" in sys.argv[1:]) : |
---|
130 | print "usage : pkpgcounter file1 file2 ... fileN" |
---|
131 | elif ("-v" in sys.argv[1:]) or ("--version" in sys.argv[1:]) : |
---|
132 | print "%s" % version.__version__ |
---|
133 | else : |
---|
134 | totalsize = 0 |
---|
135 | debug = 0 |
---|
136 | minindex = 1 |
---|
137 | if sys.argv[1] in ("-d", "--debug") : |
---|
138 | minindex = 2 |
---|
139 | debug = 1 |
---|
140 | for arg in sys.argv[minindex:] : |
---|
141 | try : |
---|
142 | parser = PDLAnalyzer(arg, debug) |
---|
143 | totalsize += parser.getJobSize() |
---|
144 | except pdlparser.PDLParserError, msg : |
---|
145 | sys.stderr.write("ERROR: %s\n" % msg) |
---|
146 | sys.stderr.flush() |
---|
147 | print "%s" % totalsize |
---|
148 | |
---|
149 | if __name__ == "__main__" : |
---|
150 | main() |
---|