In `dm.xmlsec.binding` ("https://pypi.org/project/dm.xmlsec.binding/"), I integrate Aleksey Sanin's XML security library ("http://www.aleksey.com/xmlsec") with `lxml`, both `libxml2` based. Recently, one of its encryption/decryption tests started to fail. This test uses a tree of the form: <root> ... <to_be_encrypted>...</to_be_encrypted> ... </root> The encryption transforms it (directly at the `libxml2` level) to the form <root> ... <encrypted>...</encypted> ... </root> An `lxml` proxy to the new node `encrypted` is returned. The test calls `getroottree` on this proxy to update `lxml`'s view of the modified tree. In older `lxml` versions, this has been successful. However, in `lxml==4.9.1` (maybe already before this version) the view is inconsistent: with `ENC="{http://www.w3.org/2001/04/xmlenc#}EncryptedData` and an element `E`, I get: `E[0].tag == ENC` but `E.find(ENC) is None`. Apparently, `__getitem__` (behind `E[0]`) is aware of the modification but `find` is not. How the `lxml` view of a tree modified at the `libxml2` leval can be reliably updated? For preciseness =============== The XML serialization of the inconsistent view is: <!-- XML Security Library example: Original XML doc file before encryption (encrypt2 example). --> <Envelope xmlns="urn:envelope"> <EncryptedData xmlns="http://www.w3.org/2001/04/xmlenc#" Type="http://www.w3.org/2001/04/xmlenc#Element"> <EncryptionMethod Algorithm="http://www.w3.org/2001/04/xmlenc#tripledes-cbc"/> <KeyInfo xmlns="http://www.w3.org/2000/09/xmldsig#"> <KeyName>deskey.bin</KeyName> </KeyInfo> <CipherData> <CipherValue>FwsuAoPgBzvCjr3K2q+VLcX4U+NzpHZ6orcSzbe3OrTcu3PrxWnHqg==</CipherValue> </CipherData> </EncryptedData> </Envelope> The `E` in the section above is this document's root element (i.e. `{urn:envelope}Envelope`). Its child `0` has been replaced by the encryption. It is this element which is not seen by `find`. -- Dieter