| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
| |
* prefix to-be-separated modules with js
* remove dynstreams dependency
* untangle from EmptyPromise
* move typeptr into tojs
|
| |
|
|
|
|
|
| |
Equivalent to curl --insecure.
Note: unfortunately this does not help if the server is using unsafe
legacy renegotiation, you have to allow that in the OpenSSL config.
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
| |
std's version is known to be broken on versions we still support, and it
makes no sense to use different decoders anyway.
(This does introduce a bit of a dependency hell, because js/base64
depends on js/javascript which tries to bring in the entire QuickJS
runtime. So we move that out into twtstr, and manually convert a
Result[string, string] to DOMException in js/base64.)
|
| |
|
|
|
|
|
|
|
| |
glibc apparently calls fstat from fread, and we didn't allow it in
seccomp. So:
* allow fstat in the sandbox; no reason not to, and it seems too big of
a footgun to assume we never call fread
* use read(2) in http; no need for buffered i/o here
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
pledge is a bit more fine-grained than Capsicum's capability mode,
so the buffer & http ("network") sandboxes are now split up into
two parts.
I applied the same hack as in FreeBSD for overriding the buffer
selector kqueue, because a) I didn't want to request sysctl promise
b) I'm not sure if it would even work and c) if it breaks on OpenBSD,
then it's broken on FreeBSD too, so there's a greater chance of
discovering the bug.
|
| |
|
|
| |
isCommand was completely broken
|
| |
|
|
| |
causes problems with header parsing
|
| |
|
|
|
|
|
|
|
|
|
| |
The empty string comparison here was in fact pointless; in cw-out.c,
libcurl only calls cwb (which is curlWriteHeader in this case) if blen
is not 0, so the string will never be empty. (Instead, it is expected
to be \r\n; I've added \n too since CGI can already parse headers like
that.)
Normally it still worked because we just passed through the line to
cgi.nim. However, it choked horribly on HTTP/2 early hints.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
It's the sandboxing system of FreeBSD. Quite pleasant to work with.
(Just trying to figure out the basics with this one before tackling the
abomination that is seccomp.)
Indeed, the only non-trivial part was getting newSelector to work with
Capsicum. Long story short it doesn't, so we use an ugly pointer cast +
assignment. But even that is stdlib's "fault", not Capsicum's.
This also gets rid of that ugly SocketPath global.
|
| |
|
|
| |
upstream now calls it unicode
|
| |
|
|
|
|
|
| |
it still sucks, but it is at least slightly more usable.
this also fixes a bug in dirlist where sort would mess up item name
association
|
| |
|
|
|
|
| |
it is best to base our interpretation of ^H_ on the current formatting,
because a) it works predictably well on man output and b) a formatting
mismatch results in manRe malfunctioning
|
| |
|
|
|
|
| |
The "redirect stderr to stdout" scheme broke with groff/man-db, as
it was spitting out warnings during execution. So now we handle stderr
and stdout separately.
|
| |
|
|
|
|
|
| |
* do not skip first 5 chars (this is legacy from when we used query
strings)
* allow practically anything but control chars (so we can use
parameters)
|
| |
|
|
|
|
|
| |
* run processBackspace on the first line, because groff likes to print
formatting there too
* check man references like SAMEPAGE(1) with isCommand because it's
commonly found in footers
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is horrible.
-s means completely different things on various systems. -l does not
exist on various systems. Nothing is standardized, except that man
should take at least one parameter and that -k should perform a search.
(Seriously, that's all.)
So what we do is:
* add a separate env var for overriding apropos
* for man:, never use -s to specify sections
* for man-k:, fall back to man, EXCEPT on FreeBSD which does not have a
working section specifier on man -k (neither -S nor MANSECT does
anything)
* for man-l:, just pass the path wholesale to man and hope it does
something useful.
Also, we now set MANCOLOR to 1 so FreeBSD man gives us formatting as
well.
|
| |
|
|
|
|
|
| |
* do not use query string for arguments
* accept symlinks as man binaries
* improve error message reporting
* run all regexes on the original line
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Depending on Perl just for this is silly.
Now we use libregexp for filtering basically the same things as
w3mman2html did. This required another patch to QuickJS to avoid
pulling in the entire JS engine, but in return, we can now run regexes
without a dummy JS context global variable.
Also, man.nim now tries to find a man command on the system even if it's
not in /usr/bin/man.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Originally we had several loader processes so that the loader did not
need asynchronity for loading several buffers at once. Since then, the
scope of what loader does has been reduced significantly, and with
that loader has become mostly asynchronous.
This patch finishes the above work as follows:
* We only fork a single loader process for the browser. It is a waste of
resources to do otherwise, and would have made future work on a
download manager very difficult.
* loader becomes (almost) fully async. Now the only sync part is a)
processing commands and b) waiting for clients to consume responses.
b) is a bit more problematic than a), but should not cause problems
unless some other horrible bug exists in a client. (TODO: make it
fully async.)
This gives us a noticable improvement in CSS loading speed, since all
resources can now be queried at once (even before the previous ones
are connected).
* Buffers now only get processes when the *connection* is finished. So
headers, status code, etc. are handled by the client, and the buffer
is forked when the loader starts streaming the response body.
As a result, mailcap entries can simply dup2 the first UNIX domain
socket connection as their stdin. This allows us to remove the ugly
(and slow) `canredir' hack, which required us to send file handles on
a tour accross the entire codebase.
* The "cache" has been reworked somewhat:
- Since canredir is gone, buffer-level requests usually start
in a suspended state, and are explicitly resumed only after
the client could decide whether it wants to cache the response.
- Instead of a flag on Request and the URL as the cache key,
we now use a global counter and the special `cache:' scheme.
* misc fixes: referer_from is now actually respected by buffers (not
just the pager), load info display should work slightly better, etc.
|
| |
|
|
| |
or it breaks on BSD
|
| |
|
|
|
|
|
| |
buffering output kind of defeats the purpose of the entire loader select
machinery
(we don't buffer streams either for the same reason)
|
| | |
|
| | |
|
| |
|
|
|
|
| |
extract_hostname is no more, hooray.
+ add standard error reporting
|
| |
|
|
|
|
| |
* move out half width <-> full width converters
* snake_case -> camelCase
* improve toScreamingSnakeCase slicing
|
| | |
|
| |
|
|
|
|
|
|
|
|
| |
derived from w3mman2html.cgi, there are only a few minor differences:
* different man page opener command
* use man:, man-k:, man-l: instead of query string to specify action
* no form input (C-lC-uman:pageC-m is faster anyway)
TODO rewrite in Nim so we don't have to depend on Perl...
|
| | |
|
| | |
|
| |
|
|
| |
why not
|
| |
|
|
| |
It was never reached anyway.
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
also in ftp: clean up resources before exit
|
| |
|
|
| |
fixes error on reloading stdin
|
| |
|
|
|
|
|
| |
* Makefile: fix parallel build, add new binaries to install target
* twtstr: split out libunicode-related stuff to luwrap
* config: quote default gopher2html URL env var for unquote
* adapter/: get rid of types/url dependency, use CURL url in all cases
|
| | |
|
| |
|
|
|
| |
Avoid computing e.g. charwidth data for http which does not need it
at all.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Now it is (technically) no longer mandatory to link to libcurl.
Also, Chawan is at last completely protocol and network backend
agnostic :)
* Implement multipart requests in local CGI
* Implement simultaneous download of CGI data
* Add REQUEST_HEADERS env var with all headers
* cssparser: add a missing check in consumeEscape
|
| |
|
|
| |
Also, move default urimethodmap config to res.
|
| | |
|
| | |
|
| |
|