private static void ReadModule(DocumentInputStream dis, String name, ModuleMap modules)
{
Module module = modules.Get(name);
// TODO Refactor this to fetch dir then do the rest
if (module == null)
{
// no DIR stream with offsets yet, so store the compressed bytes for later
module = new Module();
modules.Put(name, module);
module.Read(dis);
}
else
{
if (module.offset == null)
{
//This should not happen. bug 59858
throw new IOException("Module offset for '" + name + "' was never Read.");
}
// we know the offset already, so decompress immediately on-the-fly
long skippedBytes = dis.Skip(module.offset.Value);
if (skippedBytes != module.offset)
{
throw new IOException("tried to skip " + module.offset + " bytes, but actually skipped " + skippedBytes + " bytes");
}
InputStream stream = new RLEDecompressingInputStream(dis);
module.Read(stream);
stream.Close();
}
}
public void Decompress()
{
byte[] compressed =
{
0x01, 0x03, (byte)0xB0, 0x02, 0x61, 0x45, 0x00
};
byte[] expanded = RLEDecompressingInputStream.Decompress(compressed);
byte[] expected = Encoding.UTF8.GetBytes("aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa");
CollectionAssert.AreEqual(expected, expanded);
}
/**
* Reads module from DIR node in input stream and Adds it to the modules map for decompression later
* on the second pass through this function, the module will be decompressed
*
* Side-effects: Adds a new module to the module map or Sets the buf field on the module
* to the decompressed stream contents (the VBA code for one module)
*
* @param in the Run-length encoded input stream to read from
* @param streamName the stream name of the module
* @param modules a map to store the modules
* @throws IOException
*/
private static void ReadModule(RLEDecompressingInputStream in1, String streamName, ModuleMap modules)
{
int moduleOffset = in1.ReadInt();
Module module = modules.Get(streamName);
if (module == null)
{
// First time we've seen the module. Add it to the ModuleMap and decompress it later
module = new Module();
module.offset = moduleOffset;
modules.Put(streamName, module);
// Would Adding module.Read(in1) here be correct?
}
else
{
// Decompress a previously found module and store the decompressed result into module.buf
InputStream stream = new RLEDecompressingInputStream(
new MemoryStream(module.buf, moduleOffset, module.buf.Length - moduleOffset)
);
module.Read(stream);
stream.Close();
}
}
private static void CheckRLEDecompression(String expected, byte[] RunLengthEncodedData)
{
Stream compressedStream = new MemoryStream(RunLengthEncodedData);
MemoryStream out1 = new MemoryStream();
try
{
InputStream stream = new RLEDecompressingInputStream(compressedStream);
try
{
IOUtils.Copy(stream, out1);
}
finally
{
out1.Close();
stream.Close();
}
}
catch (IOException e)
{
//throw new Exception(e);
throw e;
}
String expanded;
try
{
//expanded = out1.ToString(StringUtil.UTF8.Name());
expanded = Encoding.UTF8.GetString(out1.ToArray());
}
catch (EncoderFallbackException e)
{
//throw new Exception(e);
throw e;
}
Assert.AreEqual(expected, expanded);
}
/**
* Reads VBA Project modules from a VBA Project directory located at
* <tt>macroDir</tt> into <tt>modules</tt>.
*
* @since 3.15-beta2
*/
protected void ReadMacros(DirectoryNode macroDir, ModuleMap modules)
{
foreach (Entry entry in macroDir)
{
if (!(entry is DocumentNode))
{
continue;
}
String name = entry.Name;
DocumentNode document = (DocumentNode)entry;
DocumentInputStream dis = new DocumentInputStream(document);
try
{
if ("dir".Equals(name, StringComparison.OrdinalIgnoreCase))
{
// process DIR
RLEDecompressingInputStream in1 = new RLEDecompressingInputStream(dis);
String streamName = null;
int recordId = 0;
try
{
while (true)
{
recordId = in1.ReadShort();
if (EOF == recordId ||
VERSION_INDEPENDENT_TERMINATOR == recordId)
{
break;
}
int recordLength = in1.ReadInt();
switch (recordId)
{
case PROJECTVERSION:
TrySkip(in1, 6);
break;
case PROJECTCODEPAGE:
int codepage = in1.ReadShort();
ModuleMap.charset = Encoding.GetEncoding(codepage); //Charset.ForName("Cp" + codepage);
break;
case STREAMNAME:
streamName = ReadString(in1, recordLength, ModuleMap.charset);
break;
case MODULEOFFSET:
ReadModule(in1, streamName, modules);
break;
default:
TrySkip(in1, recordLength);
break;
}
}
}
catch (IOException e)
{
throw new IOException(
"Error occurred while Reading macros at section id "
+ recordId + " (" + HexDump.ShortToHex(recordId) + ")", e);
}
finally
{
in1.Close();
}
}
else if (!name.StartsWith("__SRP", StringComparison.OrdinalIgnoreCase) &&
!name.StartsWith("_VBA_PROJECT", StringComparison.OrdinalIgnoreCase))
{
// process module, skip __SRP and _VBA_PROJECT since these do not contain macros
ReadModule(dis, name, modules);
}
}
finally
{
dis.Close();
}
}
}
