This needs to be a point so that the CONF_VAR() macro can work
correctly.

Some of these helpers will be needed in multiple places in the unit
tests, so we should move them now.

This commit moves VersioningAuthoritativeDirectory,
RecommendedClientVersions, and RecommendedServerVersions.

This way, we can't get out of sync between the two declarations.

I've chosen the "AuthDirMaxServersPerAddr" option here for
simplicity, since it is used literally nowhere else besides the dirauth
module.  Once we have all the infrastructure in place for this, we
can move more options into this structure.

When a subsystem is disabled, there will be no corresponding object
allocated, and no magic numbers on it.

Like "obsolete" variables, these variables produce a warning when
you try to set them, but the warning says that the relevant module
doesn't have support.

The confdecl macros now have a CONF_CONTEXT that you can define to
make all the modules in a given table disabled.

These modules are only built when the selected modules are disabled.
The provide stub implementations of the subsystem blocks.  Later,
other stub implementations could move here.

Having real subsystem blocks here will let us handle disabled
configuration options better.

This change allows other modules to include confdecl.h without
having first to include integer types they might not even use.

Closes ticket 32732.

Except for src/ext, which we may not want to modify.

Closes ticket 32732.

Create a high-level description of the long-term software architecture
goals.  Closes ticket 32206.

eighthave authored 5 years ago and Nick Mathewson committed 5 years ago

There is no /tmp or mkdtemp on Android, there is /data/local/tmp for
root and the shell user. So this fakes mkdtemp.  Also, FYI, tor might
not like the default perms of /data/local/tmp, e.g. 0770.

https://trac.torproject.org/projects/tor/ticket/32172

In cancel_descriptor_fetches(), use connection_list_by_type_purpose() instead of connection_list_by_type_state()

With v3, the "pending_final_cpath" of a circuit is always NULL which means
that for v3, established client rendezvous circuit waiting for the intro point
to ACK, will always end up timing out quickly.

This can increase the delays to which you connect to a service since in order
to succeed, the rendezvous circuit needs to fully established
(CIRCUIT_PURPOSE_C_REND_JOINED) within the cutoff of the introduction circuit
as well which is these days around 2-3 seconds.

Fixes #32021

Signed-off-by: David Goulet <dgoulet@torproject.org>

Part of ticket 30984.

Simplify handle_control_getinfo() by using the new reply lines
abstraction.  Previously, this function explicitly checked for whether
it should generate a MidReplyLine, a DataReplyLine, or an
EndReplyLine.  control_write_reply_lines() now abstracts this check.

Part of #30984.

Simplify handle_control_protocolinfo() by using the new reply line
abstraction.

Part of #30984.

Factor out the parts of handle_control_protocolinfo() that assemble
the AUTHMETHODS and COOKIEFILE strings.

Part of #30984.

In handle_control_getconf(), use the new control reply line
abstraction to simplify output generation.  Previously, this function
explicitly checked for whether it should generate a MidReplyLine or an
EndReplyLine.  control_write_reply_lines() now abstracts this check.

Part of #30984.