def my_open(buf, page, parent=None):
dircache = {}
vbaiter = None
docdata = ""
docdataiter = None
tbliter = None
dirflag = 0
ftype = ""
try:
if parent is None:
gsffilename = page.fname
else:
# need to save tmp file to pass to gsf
gsffilename = "tmp%s" % time.time()
f = open(gsffilename, "wb")
f.write(buf)
f.close()
gsfout = subprocess.check_output(["gsf", "list", gsffilename])
print gsfout
print "-----------------"
for i in gsfout.split("\n")[1:-1]:
if i[0] == "f":
# gsf sometimes lists date even for files. Or, rather, it
# seems that it misrepresents empty dirs as (empty) files.
# I have observed this with 'Objects' in many .pub files.
if i[5] != ' ':
fullname = " ".join(i.split()[4:])
else:
fullname = " ".join(i.split()[2:])
if not len(fullname):
fullname = " ".join(i.split()[2:])
if "/" in fullname:
fns = fullname.split("/")
cdir = "/".join(fns[:-1])
fn = fns[-1]
else:
fn = fullname
cdir = ""
if len(fn) and ord(fn[0]) < 32:
fn = fn[1:]
pn = None
if cdir:
cdir_to_treeiter(page, parent, cdir, dircache)
pn = dircache["/" + cdir]
data = subprocess.check_output(
["gsf", "cat", gsffilename, fullname])
iter1 = add_pgiter(page, fn, "ole", fn, data, pn)
if fn == "DesignerDoc":
ftype = "dsf"
page.model.set_value(iter1, 1, ("dsf", dirflag))
dsf.open(page, data, iter1)
if (fn == "EscherStm"
or fn == "EscherDelayStm"): # and infchild.size()>0:
ftype = "escher"
page.model.set_value(iter1, 1, ("escher", dirflag))
escher.parse(
page.model, data, iter1, "pub"
) # currently I don't parse it automagically for MSDOC
if fn == "MagicTab":
ftype = "wls"
page.model.set_value(iter1, 1, ("wls", dirflag))
wls.parse(page, data, iter1)
if fn == "CONTENTS":
if data[6:11] == "WT602":
ftype = "wt602"
page.model.set_value(iter1, 1, ("wt602", dirflag))
wt602.parse(page, data, iter1)
elif fullname.split('/')[0] == "OleObjects":
# Nested OLE objects (or images) in WT602
wt602.parse_object(page, data, iter1)
else:
ftype = "quill"
page.model.set_value(iter1, 1, ("quill", dirflag))
quill.parse(page, data, iter1)
if fn == "Contents":
if data[:2] == "\xe8\xac": # take signature into account
ftype = "pub"
page.model.set_value(iter1, 1, ("pub", dirflag))
pub.parse(page, data, iter1)
if fn == "VisioDocument":
ftype = "vsd"
page.model.set_value(iter1, 1,
("vsd", dirflag)) # level = 1?
vsd.parse(page, data, iter1)
if fn == "PageMaker":
ftype = "pm"
page.model.set_value(iter1, 1, ("pm", dirflag))
pm6.open(page, data, iter1)
if fn == "WordDocument":
ftype = "doc"
page.model.set_value(iter1, 1,
("doc", dirflag)) #level = 1
doc.parse(page, data, iter1)
if fn == "1Table" or fn == "0Table":
page.wtable = iter1
if fn == "Data" and page.type == "DOC":
page.wdata = iter1
if fn == "Book" or fn == "Workbook":
page.model.set_value(iter1, 1, ("xls", dirflag))
ftype = xls.parse(page, data, iter1)
if fn == "PowerPoint Document" or fn == "Pictures":
ftype = "ppt"
page.model.set_value(iter1, 1, ("ppt", dirflag))
ppt.parse(page, data, iter1)
if fn == "NativeContent_MAIN":
ftype = "qpw"
page.model.set_value(iter1, 1, ("qpw", dirflag))
qpw.parse(page, data, iter1)
if fn == "Signature" and data[:4] == '\x60\x67\x01\x00':
ftype = "ppp" #PagePlus OLE version (9.x?)
if (fn == "contents"
or fn == "SCFFPreview") and ftype == "ppp":
ppp.parse(page, data, iter1, fn)
# I've no idea if this is really the signature, but it is
# present in all files I've seen so far
if fn == "Header" and data[0xc:0xf] == 'xV4':
ftype = 'zmf'
zmf.zmf2_open(page, data, iter1, fn)
if fn[-4:] == '.zmf':
ftype = 'zmf'
zmf.zmf2_open(page, data, iter1, fn)
if fn == "VBA":
page.type = "vba"
ftype = "vba"
#if (ftype == "vba" and fn == "dir") or
if "vba/dir" in fullname.lower():
page.model.set_value(iter1, 1, ("vba", dirflag))
vbaiter = iter1
vbadata = data
if "SummaryInformation" in fn:
page.model.set_value(iter1, 1, ("ole", "propset"))
if parent is None:
try:
os.remove(gsffilename)
except:
pass
else:
if i.split()[2] == "VBA":
page.type = "vba"
ftype = "vba"
if vbaiter != None:
vba.parse(page, vbadata, vbaiter)
except subprocess.CalledProcessError:
print "Failed to run gsf. Please install libgsf."
return
ftype = "TEST"
return ftype
def my_open (buf,page,parent=None):
dircache = {}
vbaiter = None
docdata = ""
docdataiter = None
tbliter = None
dirflag=0
ftype = ""
tmpfile = None
try:
if parent is None:
gsffilename = page.fname
else:
# need to save tmp file to pass to gsf
(tmpfd, tmpfile) = mkstemp()
gsffilename = tmpfile
os.write(tmpfd, buf)
os.close(tmpfd)
gsfout = subprocess.check_output(["gsf", "list", gsffilename])
print gsfout
print "-----------------"
for i in gsfout.split("\n")[1:-1]:
if i[0] == "f":
# gsf sometimes lists date even for files. Or, rather, it
# seems that it misrepresents empty dirs as (empty) files.
# I have observed this with 'Objects' in many .pub files.
if i[5] != ' ':
fullname = " ".join(i.split()[4:])
else:
fullname = " ".join(i.split()[2:])
if not len(fullname):
fullname = " ".join(i.split()[2:])
if "/" in fullname:
fns = fullname.split("/")
cdir = "/".join(fns[:-1])
fn = fns[-1]
else:
fn = fullname
cdir = ""
if len(fn) and ord(fn[0]) < 32:
fn = fn[1:]
if cdir:
cdir_to_treeiter(page,parent,cdir,dircache)
pn = dircache["/"+cdir]
else:
pn = parent
data = subprocess.check_output(["gsf", "cat", gsffilename, fullname])
iter1 = add_pgiter(page,fn,"ole",fn,data,pn)
if fn == "DesignerDoc":
ftype = "dsf"
page.model.set_value(iter1,1,("dsf",dirflag))
dsf.open (page, data, iter1)
if (fn == "EscherStm" or fn == "EscherDelayStm"): # and infchild.size()>0:
ftype = "escher"
page.model.set_value(iter1,1,("escher",dirflag))
escher.parse (page.model,data,iter1,"pub") # currently I don't parse it automagically for MSDOC
if fn == "MagicTab":
ftype = "wls"
page.model.set_value(iter1,1,("wls",dirflag))
wls.parse (page,data,iter1)
if fn == "CONTENTS":
if data[6:11] == "WT602":
ftype = "wt602"
page.model.set_value(iter1,1,("wt602",dirflag))
wt602.parse (page,data,iter1)
elif fullname.split('/')[0] == "OleObjects":
# Nested OLE objects (or images) in WT602
wt602.parse_object(page, data, iter1)
else:
ftype = "quill"
page.model.set_value(iter1,1,("quill",dirflag))
quill.parse (page,data,iter1)
if fn == "Contents":
if data[:2] == "\xe8\xac": # take signature into account
ftype = "pub"
page.model.set_value(iter1,1,("pub",dirflag))
page.appcontentdoc=pub.PublisherContentDoc(page,iter1)
page.appcontentdoc.parse(data)
if fn == "VisioDocument":
ftype = "vsd"
page.model.set_value(iter1,1,("vsd",dirflag)) # level = 1?
vsd.parse (page, data, iter1)
if fn == "PageMaker":
ftype = "pm"
page.model.set_value(iter1,1,("pm",dirflag))
pm6.open (page, data, iter1)
if fn == "WordDocument":
ftype = "doc"
page.model.set_value(iter1,1,("doc",dirflag)) #level = 1
doc.parse (page, data, iter1)
if fn == "1Table" or fn == "0Table":
page.wtable = iter1
if fn == "Data" and page.type == "DOC":
page.wdata = iter1
if fn == "Book" or fn == "Workbook":
page.model.set_value(iter1,1,("xls",dirflag))
ftype = xls.parse (page, data, iter1)
if fn == "PowerPoint Document" or fn == "Pictures":
ftype = "ppt"
page.model.set_value(iter1,1,("ppt",dirflag))
ppt.parse (page, data, iter1)
if fn == "NativeContent_MAIN":
ftype = "qpw"
page.model.set_value(iter1,1,("qpw",dirflag))
qpw.parse (page, data, iter1)
if fn == "Signature" and data[:4] == '\x60\x67\x01\x00':
ftype = "ppp" #PagePlus OLE version (9.x?)
if (fn == "contents" or fn == "SCFFPreview") and ftype == "ppp":
ppp.parse(page,data,iter1,fn)
# I've no idea if this is really the signature, but it is
# present in all files I've seen so far
if fn == "Header" and data[0xc:0xf] == 'xV4':
ftype = 'zmf'
zmf.zmf2_open(page, data, iter1, fn)
if fn[-4:] == '.zmf':
ftype = 'zmf'
zmf.zmf2_open(page, data, iter1, fn)
if fn[-4:] == '.BMI' and fullname.split('/')[0] == 'Bitmaps':
ftype = 'bmi'
bmi.open(data, page, iter1)
if fn == "VBA":
page.type = "vba"
ftype = "vba"
#if (ftype == "vba" and fn == "dir") or
if "vba/dir" in fullname.lower():
page.model.set_value(iter1,1,("vba",dirflag))
vbaiter = iter1
vbadata = data
if "SummaryInformation" in fn:
page.model.set_value(iter1,1,("ole","propset"))
else:
if i.split()[2] == "VBA":
page.type = "vba"
ftype = "vba"
if vbaiter != None:
vba.parse (page, vbadata, vbaiter)
except subprocess.CalledProcessError:
print "Failed to run gsf. Please install libgsf."
if tmpfile:
try: os.remove(tmpfile)
except: pass
return ftype
def gsf_get_children(page, infile, parent, ftype, dirflag=0):
vbaiter = None
docdata = ""
docdataiter = None
tbliter = None
for i in range(infile.num_children()):
infchild = infile.child_by_index(i)
infname = infile.name_by_index(i)
chsize = infchild.size()
# print "Name ", infname, dirflag
if ord(infname[0]) < 32:
infname = infname[1:]
if infname == "dir":
infuncomp = infchild.uncompress()
data = infuncomp.read(infuncomp.size())
else:
data = infchild.read(chsize)
iter1 = add_pgiter(page, infname, "ole", dirflag, data)
if (infname == "EscherStm"
or infname == "EscherDelayStm") and chsize > 0:
ftype = "escher"
page.model.set_value(iter1, 1, ("escher", dirflag))
escher.parse(
page.model, data, iter1,
"pub") # currently I don't parse it automagically for MSDOC
if infname == "CONTENTS":
if data[6:11] == "WT602":
ftype = "wt602"
page.model.set_value(iter1, 1, ("wt602", dirflag))
wt602.parse(page, data, iter1)
else:
ftype = "quill"
page.model.set_value(iter1, 1, ("quill", dirflag))
quill.parse(page, data, iter1)
if infname == "Contents":
if data and data[:2] == "\xe8\xac": # take signature into account
ftype = "pub"
page.model.set_value(iter1, 1, ("pub", dirflag))
pub.parse(page, data, iter1)
if infname == "VisioDocument":
ftype = "vsd"
page.model.set_value(iter1, 1, ("vsd", dirflag)) # level = 1?
# choose vsd or vsd2
vsd.parse(page, data, iter1)
if infname == "PageMaker":
ftype = "pm"
page.model.set_value(iter1, 1, ("pm", dirflag))
pm6.open(page, data, iter1)
if infname == "WordDocument":
ftype = "doc"
page.model.set_value(iter1, 1, ("doc", dirflag)) #level = 1
doc.parse(page, data, iter1)
if infname == "1Table" or infname == "0Table":
page.wtable = iter1
if infname == "Data":
page.wdata = iter1
if infname == "Book" or infname == "Workbook":
page.model.set_value(iter1, 1, ("xls", dirflag))
ftype = xls.parse(page, data, iter1)
if infname == "PowerPoint Document" or infname == "Pictures" and data != None:
ftype = "ppt"
page.model.set_value(iter1, 1, ("ppt", dirflag))
ppt.parse(page, data, iter1)
if infname == "NativeContent_MAIN":
ftype = "qpw"
page.model.set_value(iter1, 1, ("qpw", dirflag))
qpw.parse(page, data, iter1)
if infname == "Signature" and data[:4] == '\x60\x67\x01\x00':
ftype = "ppp" #PagePlus OLE version (9.x?)
if (infname == "contents"
or infname == "SCFFPreview") and ftype == "ppp":
ppp.parse(page, data, iter1, infname)
# I've no idea if this is really the signature, but it is
# present in all files I've seen so far
if infname == "Header" and data[0xc:0xf] == 'xV4':
ftype = 'zmf'
zmf.zmf2_open(page, data, iter1, infname)
if infname[-4:] == '.zmf':
ftype = 'zmf'
zmf.zmf2_open(page, data, iter1, infname)
if infname == "VBA":
page.type = ftype
ftype = "vba"
if ftype == "vba" and infname == "dir":
page.model.set_value(iter1, 1, ("vba", dirflag))
vbaiter = iter1
vbadata = data
if (infile.num_children() > 0):
page.model.set_value(iter1, 1, (ftype, 1))
gsf_get_children(page, infchild, iter1, ftype, 0)
if "SummaryInformation" in infname:
page.model.set_value(iter1, 1, ("ole", "propset"))
if vbaiter != None:
vba.parse(page, vbadata, vbaiter)
return ftype
def gsf_get_children(page,infile,parent,ftype,dirflag=0):
vbaiter = None
docdata = ""
docdataiter = None
tbliter = None
for i in range(infile.num_children()):
infchild = infile.child_by_index(i)
infname = infile.name_by_index(i)
chsize = infchild.size()
# print "Name ", infname, dirflag
if ord(infname[0]) < 32:
infname = infname[1:]
if infname == "dir":
infuncomp = infchild.uncompress()
data = infuncomp.read(infuncomp.size())
else:
data = infchild.read(chsize)
iter1 = add_pgiter (page, infname, "ole", dirflag, data)
if (infname == "EscherStm" or infname == "EscherDelayStm") and chsize>0:
ftype = "escher"
page.model.set_value(iter1,1,("escher",dirflag))
escher.parse (page.model,data,iter1,"pub") # currently I don't parse it automagically for MSDOC
if infname == "CONTENTS":
if data[6:11] == "WT602":
ftype = "wt602"
page.model.set_value(iter1,1,("wt602",dirflag))
wt602.parse (page,data,iter1)
else:
ftype = "quill"
page.model.set_value(iter1,1,("quill",dirflag))
quill.parse (page,data,iter1)
if infname == "Contents":
if data and data[:2] == "\xe8\xac": # take signature into account
ftype = "pub"
page.model.set_value(iter1,1,("pub",dirflag))
pub.parse (page,data,iter1)
if infname == "VisioDocument":
ftype = "vsd"
page.model.set_value(iter1,1,("vsd",dirflag)) # level = 1?
# choose vsd or vsd2
vsd.parse (page, data, iter1)
if infname == "PageMaker":
ftype = "pm"
page.model.set_value(iter1,1,("pm",dirflag))
pm6.open (page, data, iter1)
if infname == "WordDocument":
ftype = "doc"
page.model.set_value(iter1,1,("doc",dirflag)) #level = 1
doc.parse (page, data, iter1)
if infname == "1Table" or infname == "0Table":
page.wtable = iter1
if infname == "Data":
page.wdata = iter1
if infname == "Book" or infname == "Workbook":
page.model.set_value(iter1,1,("xls",dirflag))
ftype = xls.parse (page, data, iter1)
if infname == "PowerPoint Document" or infname == "Pictures" and data != None:
ftype = "ppt"
page.model.set_value(iter1,1,("ppt",dirflag))
ppt.parse (page, data, iter1)
if infname == "NativeContent_MAIN":
ftype = "qpw"
page.model.set_value(iter1,1,("qpw",dirflag))
qpw.parse (page, data, iter1)
if infname == "Signature" and data[:4] == '\x60\x67\x01\x00':
ftype = "ppp" #PagePlus OLE version (9.x?)
if (infname == "contents" or infname == "SCFFPreview") and ftype == "ppp":
ppp.parse(page,data,iter1,infname)
# I've no idea if this is really the signature, but it is
# present in all files I've seen so far
if infname == "Header" and data[0xc:0xf] == 'xV4':
ftype = 'zmf'
zmf.zmf2_open(page, data, iter1, infname)
if infname[-4:] == '.zmf':
ftype = 'zmf'
zmf.zmf2_open(page, data, iter1, infname)
if infname == "VBA":
page.type = ftype
ftype = "vba"
if ftype == "vba" and infname == "dir":
page.model.set_value(iter1,1,("vba",dirflag))
vbaiter = iter1
vbadata = data
if (infile.num_children()>0):
page.model.set_value(iter1,1,(ftype,1))
gsf_get_children(page,infchild,iter1,ftype,0)
if "SummaryInformation" in infname:
page.model.set_value(iter1,1,("ole","propset"))
if vbaiter != None:
vba.parse (page, vbadata, vbaiter)
return ftype