Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

This reverts commit 332d809c, reversing
changes made to 0c72bc88.

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Use a more common type of conflict in the solver basic memory usage test.

The test previously caused the solver to backtrack because of a missing package.
Missing packages are handled as a special case in the solver, so the test didn't
exercise as much code as it could have.  This commit causes the solver to
backtrack with a version conflict, which is a more common type of failure.
Other tests continue to test the missing package case.

Don't put VQuiet in VFlags; that's reserved for user-toggles.

Since VQuiet is set programmatically, it's inappropriate
for it to be propagated to the command line.  But unfortunately,
if it was set, we were accidentally using the fancy flag format.
This patch moves VQuiet to its own field in Verbosity so we
don't get confused.

Fixes #4393.

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Filter out internal deps when forming package ElaboratedConfiguredPackage.

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Prior to this patch, we were unconditionally sticking all dependencies
in a "whole package' ElaboratedConfiguredPackage, which resulted in
cycles when there were internal dependencies on the library in question,
and broken deps with convenience libraries.  Happily, the fix is simple,
although I did have to add ConfiguredId to the "extra" dependency fields
to make it easier to tell if a dependency was external or internal.

I added an extra test to make sure that we give a good error message
if a package requires per-component builds (apparently, internal
libraries require per-component builds!)

Fixes #4388.

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

ProjectPlanning refactor

* toConfiguredComponents and friends are now monadic.  This means
  we can report a user-friendly error when we fail to find a
  dependency in the dependency map.  I had to rejigger a bit
  of the logic in ProjectPlanning since we were knot-tying through
  this function, but it all worked out.  This means
  that unbuildable_external_lib_deps is no more.

* cc_internal_build_tools is no more; instead it's cc_exe_deps,
  which tracks ALL dependencies.  It also comes with a PackageId
  so we can build ConfiguredId cabal-install side.  This change
  propagates all the way to 'ReadyComponent'

* ProjectPlanning: Instead of recomputing dependencies from scratch,
  we instead use the ElaboratedConfiguredPackages we just finished
  making to build the ComponentDeps.

* ProjectPlanning now constructs a skeletal setupComponent.  This
  is used to setup the above with correct setup dependencies.
  In principle this component might also be used for building, but
  lots of functionality isn't written in yet.

* filterExeMapDep is no more; it's all handled by Cabal now.

* The ConfiguredComponentMap now handles both libraries and
  executables in one data structure.  This is nice.

* compSetupDependencies is no more, because elaborated components
  never have custom setup.

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

There's still more refactoring to do, but here is a good
first step.

* New 'foldPlanPackage' for taking care of that annoying
  thing where you really wanted an OR-pattern on Configured/Installed

* Big Note [SolverId to ConfiguredId]

* elaborate(Lib|Exe)SolverId(') has been pared down to just
  elaborateLibSolverId and elaborateExeSolverId; call sites
  are responsible for projecting out the components they're actually
  interested in.  That gives us a new 'planPackageCacheFile' and
  'planPackageExePath' helper functions (which maybe should be
  lifted out.)

* external_lib_dep_sids/etc bindings have been moved so they're
  closer together; easier to see.

* Pile of new helper functions to help shorten code.

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Rework build target handling

Distinguish between true package names, and munged package names

A few things I forgot relating to build tool dependencies + slight test organization

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

There should be little-to-no functional changes with this commit

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Fixes bug where cabal exec with foreign libraries didn't
actually work on Windows (because it was selecting
the wrong library directory).

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Needed to stop logging dir change to stdout
to make test output deterministic.

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Windows behavior for argument translation always quotes, which
is not so great for deterministic output.  And it wasn't really
necessary either, so just don't do it at all.

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

The problem was that if we kill GHCi sufficiently early,
C^C will *actually* kill it, and then we'll have some unbuffered
input that can't make it there.

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>

Signed-off-by: Edward Z. Yang <ezyang@cs.stanford.edu>