Support base64 padding in federated tokens
Fernet tokens can have base64 padding, which gets urlencoded as one or more "%3D" strings at the end of the token. This change expands the parsing of tokens to support the % character. Change-Id: I62ff7bd94b475cddbc3cfcc7782ce4ef3b65b268
This commit is contained in:
parent
5a07dd96b1
commit
df6084573b
@ -72,7 +72,7 @@ if [ "$?" != "0" ]; then
|
||||
echo "Could not obtain IdP token, did you forget to import your openrc file? See token.json and error.log for details."
|
||||
exit 1
|
||||
fi
|
||||
IDP_TOKEN=`grep X-Subject-Token token.txt | grep -Po ': .*' | grep -Po '[a-zA-Z0-9-_]*'`
|
||||
IDP_TOKEN=`grep X-Subject-Token token.txt | grep -Po ': .*' | grep -Po '[a-zA-Z0-9-_%]*'`
|
||||
echo - Obtained IdP token.
|
||||
|
||||
# obtain the service provider URLs
|
||||
@ -106,7 +106,7 @@ if [ "$?" != "0" ] || ! grep -q X-Subject-Token unscoped.txt; then
|
||||
echo Could not obtain unscoped token from service provider. See unscoped.txt and error.log for details.
|
||||
exit 1
|
||||
fi
|
||||
UNSCOPED_TOKEN=`grep X-Subject-Token unscoped.txt | grep -Po ': .*' | grep -Po '[a-zA-Z0-9-_]*'`
|
||||
UNSCOPED_TOKEN=`grep X-Subject-Token unscoped.txt | grep -Po ': .*' | grep -Po '[a-zA-Z0-9-_%]*'`
|
||||
echo - Obtained unscoped token from SP: $UNSCOPED_TOKEN
|
||||
|
||||
echo '- Domains available at sp: '
|
||||
|
Loading…
Reference in New Issue
Block a user