Good morning Ralph.
Ralph Corderoy <[email protected]> wrote:
|> + soll = (i > 0 && cbuf[i - 1] == '/');
...
|> + }else
|> + res = str_concat_csvl(&s, cbuf, (soll ? "" : "/"), &res[1], \
|> NULL)->s;
|
|It occurs to me that the two assignments to res are nearly identical.
|Perhaps
...
Honestly spoken i don't care as long as this mess works. ^.^ Only
the latter, the "else" _must_ remain in v15: it is a property of
the protocol and as such covered by an URL object, which is then
passed all along the call chain without any adjustment, and
whether the protocol-specific code performs adjustments or not is
up to the protocol. All this is
URL *urlp = URL::create_from_string(...);
if(!urlp->concat(&res[1]))
panic("Invalid URL: cannot append %s to %s\n",
urlp->to_string(), &res[1]);
VFS *vfsp = urlp->vfs_factory();
if(vfsp == NULL)
panic("We don't support protocol %s, sorry!\n",
urlp->protocol());
or something very similar. I will never get this C-style
programming style that over and over and over again performs
string manipulations and checks right, like which_protocol() and
protbase() above, and looses more and more properties of what is
being worked on, just to finally being unable to know just about
anything of what was intented, and then guessing.
So no, it is nothing but me not being able to open my mind to this
code. Humiliatingly the test in the announcement of v14.8.11:
? Fi+/
"imap://user1@localhost/Mail": 20 messages [Read only]
? Fi@/
"imap://user1@localhost/INBOX": 9 messages [Read only]
should have rung the bell, but... Unfortunately not. It seems
i was too happy to get this out, after not being able to get to
this at all for over two weeks, once Stephen reported the issue.
So, _not_ getting to this half of the brain at all seems also not
to be the way... (But happy not to be a mathematician
nonetheless. Thanks to the rightist teacher who fucked up this
hopeful leftwing with a lot of isolation and pressure!)
--steffen
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
__________________________________
[email protected]
