| Age | Commit message (Collapse) | Author |
|---|
|
|
|
I'm not including the preload file yet.
|
|
|
|
|
|
A site triggers this with a background image where the style is deleted
upon </div>, and Capi_stop_client() finds that a_Cache_client_get_if_unique()
is false, so nothing aborts the connection. And there's time for this to
happen because we're doing TLS handshake.
I don't know whether all of what triggered this is doing the right thing,
but at least when it comes to capi, we can see that there's the idea of
permitting it (with whether we ever actually want that in practice being
yet another question).
In any case, Http_make_query_str() definitely thinks the web is there.
If we really decided that we wanted connections to continue without webs,
we could stuff 1) what sort of thing are we requesting? 2) is this a
third-party request? into the socket data. Making the query earlier is
probably not advisable because we'd want the cookies available at the
time that we send the query and not the cookies that were available
somewhat earlier.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
I used 'hg rename' and expected (at least hoped) that 'hg diff' would
do what I would naturally want, but no.
|
|
And separate http from https for safety while we're at it.
We were checking this where we needed to, but it would be easy to
forget about in the future. Not that very much happens when you
try http://example.com:443 or https://example.com:80, but I'm
being careful nevertheless.
|
|
openbsd tripped over this for me
|
|
titles)
I've noticed how users on forums can be like "Oh, it must be something about
bugs in dillo. But it manages to load the page". This is a degree of
misunderstanding which I wouldn't expect from anyone interested in using
dillo, but there it is, so I should deal with it.
|
|
I tried dillo on openbsd and, unsurprisingly, it seems fine with libressl.
I still would like it if some other TLS library would become the clear
choice for dillo at some point...
|
|
|
|
|
|
|
|
I see that openbsd likes to complain when it's used, and we certainly
don't have a deep need for it in this case.
|
|
Just recently I added some mention of domainrc to the website. I hadn't
made enough of an effort to communicate the fact that it exists, and now
it's time to remedy that.
|
|
Now have a computer with windows keys :( and, at least for me, the
one that looks like a menu is apparently "Menu" in xev. (Although
some FLTK documentation and other stuff on the web suggests that
this used to [or maybe still does for some people] cause Super_R (xev),
which would then turn into FL_Meta_R in FLTK.)
|
|
|
|
Normally I really like to make commits in small pieces that all compile
and make sense in isolation, but with this https work, the effort vs the
reward just wasn't going to make sense.
|
|
|
|
|
|
|
|
|
|
|
|
Not heading/sectioning ones, but P is legal, for example.
|
|
Followed a link to instructables.com and found that they use one of these.
I'm a little surprised to see one of these strings around.
A minute of research shows: Apparently it generally has something to do
with xslt restrictions.
|
|
|
|
|
|
|
|
https://github.com/torvalds/linux/pull/17 has a five-megabyte title
attribute, which is just a bit excessive. Since it has tons of < and
>, dillo couldn't cope with it. Over five minutes to parse as much
of it as it got before the connection broke. With this change, it's
about fifty seconds (on this old computer) to get/show the full 24 megs,
which is an improvement, at least.
|
|
|
|
noticed-by: eocene
|
|
jeremy's valgrind logs have an instance of
==15610== Invalid read of size 4
==15610== at 0x8090B53: Http_socket_reuse (http.c:668)
|
|
|
|
|
