Pepakura .PBO to Metasequoia .MQO Conversion Script

[gwlogan]

I've got a script written for Python 2.6 that supposedly converts Pepakura (.PBO) models to Metasequoia (.MQO) models. The problem is I can't get the script to run without getting a bunch of syntax errors! I did not write this script and I have no python programing knowledge.
I hope someone who knows Python scripting can fix the errors.
I'd love to see how well this script works. Any and all help would be much appreciated!

Code: Select all

# pepakura3 (.PDO) to metasequoia (.MQO) converter
# by MrPotatoHead using python 2.6

# You must specify the imput file with the full directory
# You must also specify the output directory AND filename (no extension)
# Also remember to change the scale

# I couldn't be bothered to invert the face orientation
# Just select all the faces in your cad program and invert them

import struct
import zlib
import codecs
from array import array
import sys
#saveout = sys.stdout
#outfile = open('C:/output.txt', 'w')
#sys.stdout = outfile

h1=codecs.decode('424d','hex')
h2=codecs.decode('000000003600000028000000','hex')
h3=codecs.decode('010018000000000000000000120B0000120B00000000000000000000','hex')
h4=codecs.decode('0000','hex')

#***** Change the Scale here *****
scale = 1
#*********************************


f = open("C:/testfile.pdo", "rb")
outputname = "testfile"
outputdirectory = "C:/temp/"


metafile = outputdirectory+outputname
of = open(metafile+".mqo", 'w')

of.write("Metasequoia Document\n")
of.write("Format Text Ver 1.0\n")
of.write("\n")
of.write("Scene {\n")
of.write("pos -30.5374 -301.0406 878.7299\n")
of.write("lookat 0.0000 0.0000 0.0000\n")
of.write("head -6.8400\n")
of.write("pich 0.1908\n")
of.write("ortho 0\n")
of.write("zoom2 10.0094\n")
of.write("amb 0.490 0.490 0.490\n")
of.write("}\n")



try:
   dump = f.read(148)
   print ("dump")
   nobj = f.read(4)
   numobj = struct.unpack("i", nobj)
   kk1 = numobj[0]
   print ("number of objects =" ,kk1)
   if kk1 > 10000: kk1=1
   #loop number of objects
   for i in range(0,kk1):
      dump = f.read(4)
      namelen = struct.unpack("i", dump)
      dump = f.read(namelen[0])
      dump = f.read(1)
      dump = f.read(4)
      numvert = struct.unpack("i", dump)
      kk2 = numvert[0]
      if kk2 > 10000: kk2 = 0
      #loop number of vertices
      for j in range(0, kk2):
         x = f.read(8)
         y = f.read(8)

         z = f.read(8)
      dump = f.read(4)
      numfaces = struct.unpack("i", dump)
      kk3 = numfaces[0]
      if kk3 > 10000: kk3 = 0
      #loop number of faces
      for j in range(0, kk3):
         dump = f.read(40)
         nvif = f.read(4)
         numvif = struct.unpack("i", nvif)
         kk4 = numvif[0]
         if kk4 > 10: kk4 = 0
         #loop number of verticies in face
         for k in range(0, kk4):
            vnum = f.read(4)
            vnumber = struct.unpack("i", vnum)
            dump = f.read(8)
            dump = f.read(8)
            u1 = f.read(8)
            uu = struct.unpack("d", u1)
            v1 = f.read(8)
            vv = struct.unpack("d", v1)
            dump = f.read(49)
      dump = f.read(4)
      numlines = struct.unpack("i", dump)
      kk5 = numlines[0]
      print ("number of lines", kk5)
      if kk5 > 20000: kk5 = 0
      for l in range(0, kk5):
         dump = f.read(22)
      position = f.tell();

   #now get materials
   print ("position")
   dump = f.read(4)
   numtx = struct.unpack("i", dump)

   numtex = numtx[0]
   tstr1 = ("Material"+str(numtex)+" {\n")
   of.write(tstr1)
   for m in range(0, numtex):
      print ("processing material", m+1, "of", numtex)
      tstr1 = "\"mat"+str(m+1)+"\" shader(3) col("
      of.write(tstr1)
      dump = f.read(4)
      nameln = struct.unpack("i", dump)
      namelen = nameln[0]
      dump = f.read(namelen)
      dump = f.read(4)
      rrrr = struct.unpack("f", dump)
      dump = f.read(4)
      gggg = struct.unpack("f", dump)
      dump = f.read(4)
      bbbb = struct.unpack("f", dump)
      dump = f.read(4)
      aaaa = struct.unpack("f", dump)
      tstr1 = ("%.3f" % rrrr)+" "+("%.3f" % gggg)+" "+("%.3f" % bbbb)+" "+("%.3f" % aaaa)
      tstr1 += ") dif(0.800) amb(0.600) emi(0.160) spc(0.000) power(5.00)"
      of.write(tstr1)
      dump = f.read(64)
      dump = f.read(1)
      d2 = struct.unpack("B", dump[0])
      print ("d2[0]")
      if d2[0] == 1:
         tstr1 = "text(\""+outputname+str(m)+"tex.bmp\")"
         of.write(tstr1)
         print ("texture found")
         dump = f.read(4)
         bmpw = struct.unpack("i", dump)
         bmpwidth = bmpw[0]
         dump = f.read(4)
         bmph = struct.unpack("i", dump)


         bmpheight = bmph[0]
         dump = f.read(4)
         zlibl = struct.unpack("i", dump)
         zliblen = zlib[0]
         numpixels = bmpwidth*bmpheight
         print ("numpixels")
         #decompress zlib data
         dump = f.read(zliblen)
      of.write("\n")



except EOFError:
   print ("oops")

finally:
   f.seek(0,0);


#just to make sure
f.seek(0,0);
of.write("}\n")
out_str = ""
try:
   dump = f.read(148)
   nobj - f.read(4)
   numobj = struct.unpack("i", nobj)
   #print numobj[0]
   kk1 = numobj[0]
   if kk1 > 10000: kk1 = 1
   for i in range(0, kk1):
      dump = f.read(4)
      namelen = struct.unpack("i", dump)
      dump = f.read(namelen[0])
      dump = f.read(1)
      print ("processing object", i+1)
      name = "obj"+str(i+1)
      of.write("Object \""+name+"\" {\n")

      of.write(" depth 0\n")
      of.write(" folding 0\n")
      of.write(" scale 1.000000 1.000000 1.000000\n")
      of.write(" rotation 0.000000 0.000000 0.000000\n")
      of.write(" translation 0.000000 0.000000 0.000000\n")
      of.write(" visible 15\n")
      of.write(" locking 0\n")
      of.write(" shading 1\n")
      of.write(" facet 59.5\n")
      of.write(" color 0.000 0.000 0.000\n")
      of.write(" color_type 0\n")
      of.write(" verex ")
      dump = f.read(4)
      numvert = struct.unpack("i", dump)
      of.write(str(numvert[0])),
      of.write("{\n")
      kk2 = numvert[0]
      if kk2 > 10000: kk2 = 0
      for j in range(0, kk2):
         x = f.read(8)
         y = f.read(8)
         z = f.read(8)
         outx = struct.unpack("d",x)
         outy = struct.unpack("d",y)
         outz = struct.unpack("d",z)
         ox = outx[0] * scale
         oy = outy[0] * scale
         oz = outz[0] * scale
         tmpstr = "\t"+("%.4f" % ox)+" "+("%.4f" % oy)+" "+("%.4f" % oz)
         of.write(tmpstr+"\n")
      of.write(" }\n")
      dump = f.read(4)
      numfaces = struct.unpack("i", dump)
      tmpstr = "\t face"+" "+str(numfaces[0])+" {\n"
      of.write(tmpstr)
      kk3 = numfaces[0]

      if kk3 > 10000: kk3 = 0
      for j in range(0, kk3):
         mtid = f.read(4)
         matid = struct.unpack("i", mtid)
         matid2 = matid[0]
         matstr = ""
         if matid2 != -1:
            matstr = "M("+str(matid2)+")"
         dump = f.read(36)
         nvif = f.read(4)
         numvif = struct.unpack("i", nvif)
         tmpstr = str(numvif[0])+" V("
         of.write(tmpstr)
         kk4 = numvif[0]
         if kk4 > 10: kk4 = 0
         #redo this to reverse the faces: either 3 or 4
         for k in range(0, kk4):
            vnum = f.read(4)
            vnumber = struct.unpack("i", vnum)
            of.write(str(vnumber[0])+" "),
            dump = f.read(8)
            dump = f.read(8)
            u1 = f.read(8)
            uu = struct.unpack("d", u1)
            out_str += ("%.5f" % uu[0])
            out_str += " "
            v1 = f.read(8)
            vv = struct.unpack("d", v1)
            out_str += ("%.5f" % vv[0])
            out_str += " "
            dump = f.read(49)
         of.write(")"+matstr+" UV( "+out_str+" )\n")
         out_str = ""
      of.write("}\n")
      of.write("}\n")
      dump = f.read(4)

      numlines = struct.unpack("i", dump)
      kk5 = numlines[0]
      if kk5 > 20000: kk5 = 0
      for l in range(0, kk5):
         dump = f.read(22)
      position = f.tell();
   of.write("Eof\n")
   of.close()
   #read number of materials
   dump = f.read(4)
   numtx = struct.unpack("i", dump)
   numtex = numtx[0]
   #for k = 1 to numtex do
   for m in range(0, numtex):
      print ("processing materials",m)
      dump = f.read(4)
      nameln = struct.unpack("i", dump)
      namelen = nameln[0]
      dump = f.read(nameln)
      dump = f.read(80)
      dump = f.read(1)
      d2 = struct.unpack("B", dump[0])
      print ("d2[0]")
      if d2[0] == 1:
         print ("doing bitmap")
         dump = f.read(4)
         bmpw = struct.unpack("i", dump)
         bmpwidth = bmpw[0]
         dump = f.read(4)
         bmph = struct.unpack("i", dump)
         bmpheight = bmph[0]
         dump = f.read(4)
         zlibl = struct.unpack("i", dump)
         zliblen = zlib[0]
         numpixels = bmpwidth*bmpheight
         print ("numpixels")

         #decompress zlib data
         dump = f.read(zliblen)
         out = zlib.decompress(dump)
         out = out[::-1]
         out2 = ""
         for n in range(0, bmpheight):
            pstart = int(n*bmpwidth*3)
            pend = int((pstart)+(bmpwidth*3))
            tout = out[pstart:pend]
            outk = ""
            for knn in range(0, bmpwidth):
               myoutk = ""
               kstart = (knn*3)
               kend = (knn*3)+3
               myoutk = tout[kstart:kend]
               outk += myoutk[::-1]
            out2 += outk[::-1]
         print ("n")
         #write bitmap image to file
         of2name = metafile+str(m)+"tex.bmp"
         of2 = open(of2name, 'wb')
         of2.write(h1)
         bmpfilesize = "%08x" % ((bmpwidth*bmpheight*3)+56)
         print ("bmpfilesize, bmpwidth, bmpheight")
         bmpfs = codecs.decode(bmpfilesize,'hex')
         bmpfs = bmpfs[::-1]
         of2.write(bmpfs)
         of2.write(h2)
         bmpws = "%08x" % (bmpwidth)
         bmpws2 = codecs.decode(bmpws,'hex')
         bmpws2 = bmpws2[::-1]
         of2.write(str(bmpws2))
         bmphs = "%08x" % (bmpheight)
         bmphs2 = codecs.decode(bmphs,'hex')
         bmphs2 = bmphs2[::-1]
         of2.write(str(bmphs2))

         of2.write(h3)
         of2.write(out2)
         of2.write(h4)
         of2.close()

except EOFError:
   print ("oops")

finally:
   f.close()

print ("finished")

[finale00]

Upload some samples to test against.

EDIT: I found this http://www.tamasoft.co.jp/pepakura-en/d ... index.html
Guess these are samples.

Though, I can't follow his code lol

[gwlogan]

Here is a place to download some free models:
http://www.swpm.tym.sk/index.files/alliance.htm
The web is full of free Pepakura models so finding some to test with shouldn't be too hard.

[porimac]

I tried running script.
error messaged "Invalid syntax @ line 158".
but i have no idea sorry

[finale00]

I don't think you're supposed to run it through metaseq script editor.
Just run it through command-line, seeing how it's writing out mqo files.

Anyways I ran it through command-line and it was running into a bunch of odd errors that shouldn't occur in tested code

1: strings not ended properly (missing quotes, extra quotes)

Code: Select all

  for j in range(0, kk2):
         x = f.read(8)
         y = f.read(8)
         z = f.read(8)
         outx = struct.unpack("d", x")
         outy = struct.unpack("d", y")
         outz = struct.unpack("d", z")
         ox = outx[0] * scale
         oy = outy[0] * scale
         oz = outz[0] * scale

2: hardcoded input file (some people would probably have no idea what's wrong)

After making some attempt to fix it, it still doesn't work for the models I downloaded from the official site.
It tells me 0 objects were in the file and so everything fails.

Here's what I've made out of the version 3 binary format:

Code: Select all

#pepakura PDO version 3 format

Char_10 idstring (Version 3\n)
dword unk1
dword null
dword unk2

dword charLen
charLen keyboard
dword charLen
charLen encoding

dword null
dword charLen
charLen hash

dword_8 ???
dword numMesh

numMesh mesh {
   dword charLen
   charLen meshName
   
   byte unkByte
   dword unkCount
   #skip unkCount * 24 bytes
   dword numVerts
   
   numVerts vertex {
       299 bytes or 384 bytes
   }
   dword numFaces
}

From what I've written here, the code doesn't even look like it's going to get the right values.

[alon]

Me too.
"Invalid syntax."

[gwlogan]

Well after some further research I found this:
http://uppit.com/psx3tn8va1b8/Pdo2mqo.zip
This is the original Python code. How it works is you must open the pdo2mqo.pdo file in Pepakura Viewer (NOT Pepakura Designer), and view the model's texture to read the code. You then have to manually type out the lines of code text yourself into Python and save the script. What a headache! From what I've seen of the original code it looks to be meant to work with 3D Studio Max 7. Thanks for all the continued help here guys!

[finale00]

We might as well just figure out the format, post it up, and then let people write the respective importers
I looked around and it seemed some other members on other forums had some interest in the format for version 3, but couldn't figure it out.

I'm also have trouble just getting the faces section right (I can't find a solid pattern) and have no idea how the vertex is stored besides having 299 or 384 bytes depending on some unknown reason.

I believe the designer (or maybe the viewer) has an option to export to OBJ though.

[Szkaradek123]

The contents of this post was deleted because of possible forum rules violation.

[finale00]

It looks like the ones from the official site are different; they didn't have the "pepakura designer 3" in their header. That makes sense then.

[gwlogan]

The contents of this post was deleted because of possible forum rules violation.

[finale00]

Post up the ones you are having trouble with.

[gwlogan]

The contents of this post was deleted because of possible forum rules violation.

[Ares722]

Szkaradek123 , you are great!!!

[youngmark]

Please help me.
This pdo file could not be converted
http://www.mediafire.com/?dkdas8shmhckk26
need someone to fix error on python script.