[Libreoffice-commits] .: src/docrecord.py src/docstream.py test/doc

[Miklos Vajna]

 src/docrecord.py   |   61 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 src/docstream.py   |    5 +++-
 test/doc/field.doc |binary
 test/doc/field.rtf |    8 ++++++
 test/doc/test.py   |    9 +++++++
 5 files changed, 82 insertions(+), 1 deletion(-)

New commits:
commit 134ed95ebe314473ad0333394f0ee0cf1dbad49e
Author: Miklos Vajna <vmiklos at suse.cz>
Date:   Wed Jan 9 11:38:36 2013 +0100

    dump PlcFld

diff --git a/src/docrecord.py b/src/docrecord.py
index 0aa271d..9e30066 100644
--- a/src/docrecord.py
+++ b/src/docrecord.py
@@ -142,6 +142,67 @@ class PlcfBkf(DOCDirStream, PLC):
             print '</aCP>'
         print '</plcfBkf>'
 
+class Fldch(DOCDirStream):
+    """The fldch structure determines the type of the field character."""
+    def __init__(self, parent):
+        DOCDirStream.__init__(self, parent.bytes)
+        self.pos = parent.pos
+        self.parent = parent
+
+    def dump(self):
+        print '<fldch type="fldch" offset="%d" size="1 byte">' % self.pos
+        buf = self.readuInt8()
+        self.printAndSet("ch", buf & 0x1f) # 1..5th bits
+        self.printAndSet("reserved", (buf & 0xe0) >> 5) # 6..8th bits
+        print '</fldch>'
+        self.parent.pos = self.pos
+
+class Fld(DOCDirStream):
+    """The Fld structure specifies a field character."""
+    def __init__(self, parent, offset):
+        DOCDirStream.__init__(self, parent.bytes)
+        self.pos = offset
+
+    def dump(self):
+        print '<fld type="FLD" offset="%d" size="2 bytes">' % self.pos
+        self.fldch = Fldch(self)
+        self.fldch.dump()
+        self.printAndSet("grffld", self.readuInt8()) # TODO parse flt and grffldEnd
+        print '</fld>'
+
+class PlcFld(DOCDirStream, PLC):
+    """The Plcfld structure specifies the location of fields in the document."""
+    def __init__(self, mainStream):
+        DOCDirStream.__init__(self, mainStream.doc.getDirectoryStreamByName("1Table").bytes, mainStream = mainStream)
+        PLC.__init__(self, mainStream.lcbPlcfFldMom, 2) # 2 is defined by 2.8.25
+        self.pos = mainStream.fcPlcfFldMom
+        self.size = mainStream.lcbPlcfFldMom
+
+    def dump(self):
+        print '<plcFld type="PlcFld" offset="%d" size="%d bytes">' % (self.pos, self.size)
+        offset = self.mainStream.fcMin # 2.8.25: CPs relative to the start of that document part.
+        pos = self.pos
+        aFlds = []
+        for i in range(self.getElements()):
+            # aCp
+            value = self.getuInt32(pos = pos)
+            print '<aCP index="%d" value="%d">' % (i, value)
+            pos += 4
+
+            # aFld
+            aFld = Fld(self, self.getOffset(self.pos, i))
+            aFld.dump()
+
+            # This is a separator and the previous was a start: display the field instructions.
+            if aFld.fldch.ch == 0x14 and aFlds[-1][1].fldch.ch == 0x13:
+                print '<transformed value="%s"/>' % self.quoteAttr(self.mainStream.retrieveText(offset + aFlds[-1][0] + 1, offset + value))
+            # This is an end and the previous was a separator: display the field result.
+            elif aFld.fldch.ch == 0x15 and aFlds[-1][1].fldch.ch == 0x14:
+                print '<transformed value="%s"/>' % self.quoteAttr(self.mainStream.retrieveText(offset + aFlds[-1][0] + 1, offset + value))
+            aFlds.append((value, aFld))
+            print '</aCP>'
+        print '</plcFld>'
+
 class PlcfBkl(DOCDirStream, PLC):
     """The Plcfbkl structure is a PLC that contains only CPs and no additional data."""
     def __init__(self, mainStream, offset, size):
diff --git a/src/docstream.py b/src/docstream.py
index 167fb9c..a0aba5e 100644
--- a/src/docstream.py
+++ b/src/docstream.py
@@ -236,7 +236,7 @@ class WordDocumentStream(DOCDirStream):
             ["fcSttbfFfn"],
             ["lcbSttbfFfn", self.handleLcbSttbfFfn],
             ["fcPlcfFldMom"],
-            ["lcbPlcfFldMom"],
+            ["lcbPlcfFldMom", self.handleLcbPlcfFldMom],
             ["fcPlcfFldHdr"],
             ["lcbPlcfFldHdr"],
             ["fcPlcfFldFtn"],
@@ -509,6 +509,9 @@ class WordDocumentStream(DOCDirStream):
     def handleLcbDggInfo(self):
         docrecord.OfficeArtContent(self).dump()
 
+    def handleLcbPlcfFldMom(self):
+        docrecord.PlcFld(self).dump()
+
     def dumpFibRgFcLcb97(self, name):
         print '<%s type="FibRgFcLcb97" size="744 bytes">' % name
         self.__dumpFibRgFcLcb97()
diff --git a/test/doc/field.doc b/test/doc/field.doc
new file mode 100644
index 0000000..247f024
Binary files /dev/null and b/test/doc/field.doc differ
diff --git a/test/doc/field.rtf b/test/doc/field.rtf
new file mode 100644
index 0000000..48e39d1
--- /dev/null
+++ b/test/doc/field.rtf
@@ -0,0 +1,8 @@
+{\rtf1
+Page number:
+{\field
+{\*\fldinst  PAGE }
+{\fldrslt 1}
+}
+\par
+}
diff --git a/test/doc/test.py b/test/doc/test.py
index a97d0f5..c2b955c 100755
--- a/test/doc/test.py
+++ b/test/doc/test.py
@@ -137,6 +137,15 @@ class Test(unittest.TestCase):
         # This first caused unhandled exceptions, then later invalid XML output.
         self.dump('escape')
 
+    def test_field(self):
+        self.dump('field')
+
+        instruction = self.root.findall('stream[@name="WordDocument"]/fib/fibRgFcLcbBlob/lcbPlcfFldMom/plcFld/aCP[@index="1"]/transformed')
+        self.assertEqual(' PAGE ', instruction[0].attrib['value'])
+
+        result = self.root.findall('stream[@name="WordDocument"]/fib/fibRgFcLcbBlob/lcbPlcfFldMom/plcFld/aCP[@index="2"]/transformed')
+        self.assertEqual('1', result[0].attrib['value'])
+
 if __name__ == '__main__':
     unittest.main()