Closed schuelermine closed 1 year ago
Browsers don't follow RFC 3986, they do their own thing.
I suggest identifying %2a with *, which currently would only match a literal %2a in the original URI, as a workaround.
I don't follow. Can you show with an example use case what you mean?
I mean removing *=0
when %2a
is requested to be removed.
Supporting %2a
for that seems a little weird unless we support the entire thing percent encoded, but then that might be odd as well.
What if we just allow backslash escaping, the way it is a little more traditionally done? Like: --trim 'query=\*'
?
Actually, a quirky work-around that already works is this: --trim 'query=**'
since it will make a match of all query pairs that begin with *
...
What if we just allow backslash escaping, the way it is a little more traditionally done? Like: --trim 'query=*' ?
this seems like the most intuitive approach -- I don't think I ever would have tried query=**
to solve this.
Query parameters containing the
*
character cannot be effectively targeted as the wildcard matches other parameters as well. I suggest identifying%2a
with*
, which currently would only match a literal%2a
in the original URI, as a workaround.