[Libreoffice-commits] .: 3 commits - src/docrecord.py src/docstream.py test/doc

[Miklos Vajna]

 src/docrecord.py      |    7 +++----
 src/docstream.py      |   26 ++++++++++++++++++++++++++
 test/doc/header.doc   |binary
 test/doc/header.rtf   |    4 ++++
 test/doc/headerlo.doc |binary
 test/doc/test.py      |   12 ++++++++++++
 6 files changed, 45 insertions(+), 4 deletions(-)

New commits:
commit 0ccbaa56a706ec398e4cea1a656ced09829f1f8f
Author: Miklos Vajna <vmiklos at suse.cz>
Date:   Sat May 4 16:02:55 2013 +0200

    PlcfSed: use retrieveCPs

diff --git a/src/docrecord.py b/src/docrecord.py
index bcd156b..20c63b4 100644
--- a/src/docrecord.py
+++ b/src/docrecord.py
@@ -300,7 +300,6 @@ class PlcfSed(DOCDirStream, PLC):
 
     def dump(self):
         print '<plcfSed type="PlcfSed" offset="%d" size="%d bytes">' % (self.pos, self.size)
-        offset = self.mainStream.fcMin
         pos = self.pos
         for i in range(self.getElements()):
             # aCp
@@ -313,7 +312,7 @@ class PlcfSed(DOCDirStream, PLC):
             aSed = Sed(self, self.getOffset(self.pos, i))
             aSed.dump()
 
-            print '<transformed value="%s"/>' % self.quoteAttr(self.mainStream.retrieveText(offset + start, offset + end, logicalLength = True))
+            print '<transformed value="%s"/>' % self.quoteAttr(self.mainStream.retrieveCPs(start, end))
             print '</aCP>'
         print '</plcfSed>'
 
commit d8165133205a3c3533211b8e4f0da14de450bdaa
Author: Miklos Vajna <vmiklos at suse.cz>
Date:   Sat May 4 15:57:43 2013 +0200

    doc: test header dumping of both MSO and LO-generated doc

diff --git a/test/doc/header.doc b/test/doc/header.doc
new file mode 100755
index 0000000..a351d72
Binary files /dev/null and b/test/doc/header.doc differ
diff --git a/test/doc/header.rtf b/test/doc/header.rtf
new file mode 100644
index 0000000..b3fd8b4
--- /dev/null
+++ b/test/doc/header.rtf
@@ -0,0 +1,4 @@
+{\rtf1
+{\header This is a header.}
+Hello world!\par
+}
diff --git a/test/doc/headerlo.doc b/test/doc/headerlo.doc
new file mode 100644
index 0000000..73ec689
Binary files /dev/null and b/test/doc/headerlo.doc differ
diff --git a/test/doc/test.py b/test/doc/test.py
index d02f680..e50e447 100755
--- a/test/doc/test.py
+++ b/test/doc/test.py
@@ -170,6 +170,18 @@ class Test(unittest.TestCase):
         levels = self.root.findall('stream[@name="WordDocument"]/fib/fibRgFcLcbBlob/lcbPlfLst/plfLst/lvl')
         self.assertEqual("•", levels[0].findall('xst/rgtchar')[0].attrib['value'])
 
+    def test_header(self):
+        self.dump('header')
+
+        firstHeader = self.root.findall('stream[@name="WordDocument"]/fib/fibRgFcLcbBlob/lcbPlcfHdd/plcfHdd/aCP[@index="7"]')
+        self.assertEqual("This is a header.\\x0D\\x0D", firstHeader[0].findall('transformed')[0].attrib['value'])
+
+    def test_headerlo(self):
+        self.dump('headerlo')
+
+        firstHeader = self.root.findall('stream[@name="WordDocument"]/fib/fibRgFcLcbBlob/lcbPlcfHdd/plcfHdd/aCP[@index="7"]')
+        self.assertEqual("This is a header.\\x0D\\x0D", firstHeader[0].findall('transformed')[0].attrib['value'])
+
 if __name__ == '__main__':
     unittest.main()
 
commit 89cb847a68c1db6a1177d9a0673a68702de48da2
Author: Miklos Vajna <vmiklos at suse.cz>
Date:   Sat May 4 15:50:25 2013 +0200

    doc: fix dumping header/footer text in LO-produced files

diff --git a/src/docrecord.py b/src/docrecord.py
index 62ca85f..bcd156b 100644
--- a/src/docrecord.py
+++ b/src/docrecord.py
@@ -816,7 +816,7 @@ class PlcfHdd(DOCDirStream, PLC):
 
     def dump(self):
         print '<plcfHdd type="PlcfHdd" offset="%d" size="%d bytes">' % (self.pos, self.size)
-        offset = self.mainStream.fcMin + self.mainStream.ccpText
+        offset = self.mainStream.ccpText + self.mainStream.ccpFtn
         pos = self.pos
         for i in range(self.getElements() - 1):
             start = self.getuInt32(pos = pos)
@@ -837,7 +837,7 @@ class PlcfHdd(DOCDirStream, PLC):
                     11: "First page footer",
                     }
             print '<aCP index="%d" contents="%s" start="%d" end="%d">' % (i, contentsMap[i], start, end)
-            print '<transformed value="%s"/>' % self.quoteAttr(self.mainStream.retrieveText(offset + start, offset + end))
+            print '<transformed value="%s"/>' % self.quoteAttr(self.mainStream.retrieveCPs(offset + start, offset + end))
             pos += 4
             print '</aCP>'
         print '</plcfHdd>'
diff --git a/src/docstream.py b/src/docstream.py
index 7f3cc51..dc36baf 100644
--- a/src/docstream.py
+++ b/src/docstream.py
@@ -9,6 +9,7 @@ import ole
 import struct
 from docdirstream import DOCDirStream
 import docrecord
+import globals
 
 class DOCFile:
     """Represents the whole word file - feed will all bytes."""
@@ -674,8 +675,33 @@ class WordDocumentStream(DOCDirStream):
         return index
 
     def retrieveText(self, start, end, logicalLength = False):
+        """Deprecated, use retrieveCPs instead."""
         plcPcd = self.clx.pcdt.plcPcd
         idx = self.__findText(plcPcd, start)
         return plcPcd.aPcd[idx].fc.getTransformedValue(start, end, logicalPositions = False, logicalLength = logicalLength)
 
+    def retrieveCP(self, cp):
+        """Implements 2.4.1 Retrieving Text."""
+        plcPcd = self.clx.pcdt.plcPcd
+        for i in range(len(plcPcd.aCp)):
+            if plcPcd.aCp[i] <= cp:
+                index = i
+                break
+        aPcd = plcPcd.aPcd[index]
+        fcCompressed = aPcd.fc
+        if fcCompressed.fCompressed == 1:
+            return globals.encodeName(self.bytes[(fcCompressed.fc/2) + (cp - plcPcd.aCp[i])])
+        else:
+            pos = fcCompressed.fc + 2 * (cp - plcPcd.aCp[i])
+            return globals.encodeName(self.bytes[pos:pos+2].decode('utf-16'), lowOnly = True)
+
+    def retrieveCPs(self, start, end):
+        """Retrieves a range of characters."""
+        ret = []
+        i = start
+        while i < end:
+            ret.append(self.retrieveCP(i))
+            i += 1
+        return "".join(ret)
+
 # vim:set filetype=python shiftwidth=4 softtabstop=4 expandtab: