Should use clearText instead of rawData in InflaterInputBuffer.ReadLeByte()?

[uscq]

In InflaterInputBuffer, when specifies the CryptoTransform by call zip.cryptoTransform = new MyCustomTransform();, the raw data (rawData) may has been transformed to clearText.
Why still use the clearText to read the data in InflaterInputBuffer.ReadLeByte() instead use rawData.

code at

SharpZipLib/src/ICSharpCode.SharpZipLib/Zip/Compression/Streams/InflaterInputStream.cs

Line 239 in cdd0931

byte result = rawData[rawLength - available];

[piksel]

I think what you are asking is about why the crypto transform is not performed on the data read from ReadLeByte() et al.
The reason is basically that InflaterInputStream is used for two purposes, ZIP and GZip, and only the latter of them contains a header in the "deflated" data. The ReadLe*() methods are used to read said header, which should not be encrypted. That being said, GZip doesn't support encryption at all, so it shouldn't matter if the transform is used or not.