ExtractTiff - masahase's memo @ ウィキ

#!/usr/bin/env python
# -*- coding: utf8 -*-
 
import sys, re, struct
from pdflib.pdfparser import PDFDocument, PDFParser, PDFStream, \
     PDFObjRef, PSKeyword, PSLiteral
 
class MakeTiffFile:
    tags={}
    def __init__(self,data=None):
        self.data=data
        self.tags[0x00fe]=[4,1,0]
    def addTag(self,tagid,valuetype,count,value):
        self.tags.add(tagid,(valuetype,count,value))
    def __makeIFD(self):
        if(len(self.tags.keys())<0xffff):
            ret=struct.pack("H",len(self.tags.keys()))
            for tag in self.tags.keys():
                ret+=struct.pack("H",tag)
                ret+=struct.pack("H",self.tags[tag][0])
                ret+=struct.pack("I",self.tags[tag][1])
                ret+=struct.pack("I",self.tags[tag][2])
            ret+=struct.pack("I",0)
            return ret
        else:
            offset=8
            ikeys=self.tags.keys()
            ret=''
            for i in range((len(ikeys)/0xffff)+1):
                if(len(ikeys[0xffff*i:])>0xffff):
                    ret+=struct.pack("H",0xffff)
                    for tag in ikeys[i*0xffff:i+0xffff]:
                        ret+=struct.pack("H",tag)
                        ret+=struct.pack("H",self.tags[tag][0])
                        ret+=struct.pack("I",self.tags[tag][1])
                        ret+=struct.pack("I",self.tags[tag][2])
                    offset+=2+0xffff*12+4
                    ret+=struct.pack("I",offset)
                else:
                    ret+=struct.pack("H",len(ikeys[ioffset:]))
                    for tag in ikeys[i*0xffff:]:
                        ret+=struct.pack("H",tag)
                        ret+=struct.pack("H",self.tags[tag][0])
                        ret+=struct.pack("I",self.tags[tag][1])
                        ret+=struct.pack("I",self.tags[tag][2])
                    ret+=struct.pack("I",0)
            return ret
    def __makeHeader(self):
        if(struct.pack("<I",1000)==struct.pack("=I",1000)):
            ret=struct.pack("2HI",0x4949,42,8)
        else:
            ret=struct.pack("2HI",0x4D4D,42,8)
        ret+=self.__makeIFD()
        return ret
    def setPhotometricInterpretation(self,flag):
        """A bilevel image contains two colors—black and white. TIFF allows an 
        application to write out bilevel data in either a white-is-zero or 
        black-is-zero format. The field that records this information is 
        called PhotometricInterpretation.
 
        0:White is Zero
        1:Black is Zero
        """
        if(flag==0 or flag==1):
            self.addTag(262,3,1,flag)
    def setCompression(self,value):
        """Data can be stored either compressed or uncompressed."""
        self.addTag(259,3,1,value)
    def setImageLength(self,value):
        self.addTag(257,4,1,value)
    def setImageWidth(self,value):
        self.addTag(256,4,1,value)
    def setResolutionUnit(self,value):
        """1= No absolute unit of measurement. Used for images that may have 
        a non-square aspect ratio but no meaningful absolute dimensions.
        2= Inch.
        3= Centimeter.
        Default = 2 (inch).
        """
        self.addTag(296,3,1,value)
    def setXResolution(self,value):
        self.addTag(282,5,1,value)
    def setYResolution(self,value):
        self.addTag(283,5,1,value)
 
class ReadTiffTag:
    def getTags(self,filename):
        tiff=file(filename,"rb")
        tags={}
        buf=tiff.read(2)
        endian="@"
        if(buf=="II"):
            endian="<"
        else:
            endian=">"
        buf=tiff.read(2)
        buf=tiff.read(4)
        while(True):
            tiff.seek(int(struct.unpack(endian+"I",buf)[0]))
            buf=tiff.read(2)
            for i in range(struct.unpack(endian+"H",buf)[0]):
                buf=tiff.read(2)
                tag=struct.unpack(endian+"H",buf)[0]
                tags[tag]=[
                    struct.unpack(endian+"H",tiff.read(2))[0],
                    struct.unpack(endian+"I",tiff.read(4))[0],
                    struct.unpack(endian+"H",tiff.read(2))[0],
                    struct.unpack(endian+"H",tiff.read(2))[0]
                    ]
            buf=tiff.read(4)
            if(struct.unpack(endian+"I",buf)[0]==0):
                break
        return tags