* module/ice-9/boot-9.scm (primitive-load): Define a version in Scheme
that uses read-syntax. This allows the expander to be able to access
source locations.
It was never documented, never used, and uses the gross assoc interface
to doubly-weak tables. Good-bye!
* module/ice-9/hcons.scm: Remove.
* am/bootstrap.am: Remove hcons from build.
* module/ice-9/poe.scm: Extract the one procedure from hcons that poe
uses, which notably doesn't have anything to do with hcons.
* module/language/tree-il.scm (tree-il-src): Always a vector now;
tree-il-srcv is gone. An incompatible change but we are in the
compiler.
(location): For parse-tree-il, make vector locations instead of alists.
* module/language/tree-il/analyze.scm:
* module/language/tree-il/compile-bytecode.scm:
* module/language/tree-il/compile-cps.scm:
* module/language/tree-il/debug.scm:
* module/language/tree-il/letrectify.scm:
* module/language/tree-il/peval.scm:
* module/system/vm/assembler.scm: Update all uses to expect vectors
instead of alists and to use tree-il-src instead of tree-il-srcv.
* module/language/elisp/compile-tree-il.scm (location): Create vectors,
not alists.
* test-suite/tests/compiler.test ("psyntax"): Update syntax-source
expectation.
* module/ice-9/psyntax.scm (source-annotation): Only get source info
from syntax objects.
(strip): Don't attach source info.
(macroexpand): Don't proxy source info in that isn't in a syntax object.
(datum->syntax): Don't proxy source info from source-properties.
* test-suite/tests/compiler.test ("psyntax"):
* test-suite/tests/coverage.test (code):
* test-suite/tests/eval-string.test ("basic"):
* test-suite/tests/syntax.test ("expressions"):
* test-suite/tests/tree-il.test ("warnings"): Update tests that attach
source properties to use read-and-compile, or read-syntax.
This will allow for weak tables to be implemented partly in Scheme.
* module/ice-9/object-properties.scm: New file.
* am/bootstrap.am (SOURCES): Add new file.
* module/ice-9/boot-9.scm:
* module/ice-9/deprecated.scm (make-object-property*): Deprecate
make-object-property in default env and add a shim.
* module/ice-9/buffered-input.scm (ice-9):
* module/language/elisp/boot.el (plist-function):
* module/scripts/frisk.scm (scripts):
* module/web/http.scm (web): Adapt users to import the new module.
* libguile/threads.h: Add join data directly on the thread instead of
using a Scheme-side weak table. It's less complicated and it will let
the weak table implementation use locks in Scheme; otherwise you would
have threads depending on weak tables and vice versa.
* libguile/threads.c (scm_trace_thread, guilify_self_1): Init and mark
the new members.
(thread_join_cond, thread_join_lock, thread_join_results)
(thread_init_joinable_x, thread_set_join_results_x): New accessors.
* module/ice-9/threads.scm (call-with-new-thread, join-thread): Use the
new accessors.
If that's what you want, you need to bring your own weak hash table on a
normal observer.
* module/ice-9/boot-9.scm (module): Remove weak-observers field.
(make-module, make-autoload-interface): Don't pass weak table to
constructor.
(module-observe-weak): Remove.
(module-unobserve, module-call-observers): Remove weak case.
* module/ice-9/deprecated.scm (module-observe-weak): Dispatch to
module-observe.
* test-suite/tests/modules.test ("observers"): Adapt.
* module/ice-9/weak-vector.scm: New implementation, same interface.
* doc/ref/api-memory.texi (Weak vectors): Default weak vector value was
documented as empty list when it was actually unspecified, but #f is
most useful, so we change documentation and code to match.
* libguile/Makefile.am (libguile_@GUILE_EFFECTIVE_VERSION@_la_SOURCES):
(DOT_X_FILES):
(DOT_DOC_FILES):
(noinst_HEADERS):
(modinclude_HEADERS):
* libguile.h:
* libguile/deprecated.c:
* libguile/deprecated.h:
* libguile/init.c:
* libguile/weak-vector.c:
* libguile/weak-vector.h: Remove C weak vector implementation, replaced
with deprecation stubs that call out to Scheme.
* libguile/weak-set.c:
* libguile/weak-table.c:
* libguile/weak-list.h: Remove unused internal header.
* libguile/eq.c:
* libguile/evalext.c:
* libguile/goops.c:
* libguile/hash.c:
* libguile/scm.h:
* module/system/base/types.scm:
* module/system/base/types/internal.scm:
* module/system/vm/assembler.scm: Remove wvect tc7.
* module/ice-9/boot-9.scm(symbol-property, set-symbol-property!)
(symbol-property-remove!): Remove.
* module/ice-9/boot-9.scm (make-record-type): Name must be symbol.
(record-constructor): Alias record-type-constructor.
(make-module): Require size to be zero. Should fix this with keyword
args :/
(try-load-module): Inline definition of try-module-autoload. Remove
try-module-autoload binding.
(make-soft-port): Add deprecation warning, so we can remove it
eventually.
* module/ice-9/save-stack.scm: Remove deprecation comment.
* module/ice-9/top-repl.scm:
* module/ice-9/threads.scm: Export instead of replace bindings.
* module/language/bytecode.scm: Remove instruction-arity et al.
* module/language/tree-il/analyze.scm: Remove deprecated
unbound-variable-analysis and macro-use-before-definition-analysis.
* module/rnrs.scm: Fix syntax-case export now that module and value
namespaces are separate.
* module/system/base/language.scm (invalidate-compilation-cache!):
Remove.
* module/system/base/language.scm (*current-language*): Remove.
* module/ice-9/atomic.scm: Don't add-interesting-primitive! here.
* module/language/tree-il/primitives.scm
(*interesting-primitive-names*): Instead, import atomics here and
register the primitives as usual. This will let atomics be used earlier
in the boot graph.
Headed towards a 4.0. The binary format will change slightly but
incompatibly (e.g. renumbering of intrinsics). Having a new GC is
enough of a change to warrant a minor version bump, and it's been 5
years so why not go major.
* GUILE-VERSION (GUILE_EFFECTIVE_VERSION): Bump to 4.0.
(GUILE_MICRO_VERSION, GUILE_MINOR_VERSION): Bump to 3.9.0.
(LIBGUILE_INTERFACE_CURRENT, LIBGUILE_INTERFACE_AGE): Reset.
* meta/guile-4.0.pc.in:
* meta/guile-4.0-uninstalled.pc.in: Rename from guile-3.0.
* .gitignore: Ignore pkg-config files with 4.0 effective version.
* README: Update slightly.
* doc/ref/api-evaluation.texi (Load Paths):
* doc/ref/api-options.texi (Build Config):
* doc/ref/guile-invoke.texi (Environment Variables):
* doc/ref/history.texi (A Timeline of Selected Guile Releases):
* doc/ref/libguile-parallel.texi (Parallel Installations):
* doc/ref/srfi-modules.texi (SRFI-0): Update references to stable
version in paths.
* module/system/vm/assembler.scm (*bytecode-minor-version*):
* libguile/loader.h (SCM_OBJCODE_MAJOR_VERSION):
(SCM_OBJCODE_MINIMUM_MINOR_VERSION):
(SCM_OBJCODE_MINOR_VERSION): Use the same major objcode version for now
but bump the minor to 99.
* meta/guile.m4 (GUILE_PKG, GUILE_PROGS): Check for Guile 4.0.
* module/ice-9/boot-9.scm (%cond-expand-features): Add new cond-expand
features.
* libguile/scm.h (scm_tc7_thread): Give threads their own tc7.
* libguile/threads.h (struct scm_thread): Add a tag, so that struct
thread can be a SCM directly. Add a struct gc_mutator* member.
(scm_thread_handle): New inline function.
(SCM_I_IS_THREAD, SCM_I_THREAD_DATA, SCM_VALIDATE_THREAD): Update to use
tc7 instead of SMOB tags.
* libguile/continuations.c (scm_i_with_continuation_barrier)
* libguile/finalizers.c (queue_finalizer_async)
* libguile/jit.c (compile_current_thread)
* libguile/threads.c (block_self, guilify_self_2)
(lock_mutex, unlock_mutex, timed_wait scm_current_thread)
(scm_all_threads)
* libguile/vm-engine.c (current-thread): Use scm_thread_handle instead
of thread->handle.
* libguile/evalext.c (scm_self_evaluating_p):
* libguile/goops.c (class_thread, scm_class_of, scm_sys_goops_early_init)
* libguile/print.c (iprin1)
* module/language/cps/compile-bytecode.scm (compile-function)
* module/oop/goops.scm (<thread>)
* module/system/base/types.scm (cell->object)
* module/system/base/types/internal.scm (heap-tags)
* module/system/vm/assembler.scm: (emit-thread?): Adapt to
scm_tc7_thread.
* libguile/gc-internal.h: Move init functions that take "struct
gc_stack_addr" here, so that internal Whippet uses don't cause Whippet
to be added to public headers.
* libguile/gc.c (scm_storage_prehistory): Take struct gc_stack_addr as
arg, and pass to gc_init. Return a mutator pointer.
* libguile/init.c (scm_i_init_guile): Pass mutator and stack base to GC
and thread init routines.
* libguile/threads.c (scm_trace_dynstack, scm_trace_thread)
(scm_trace_thread_mutator_roots): New infra for marking threads in terms
of Whippet API.
* libguile/threads.c (guilify_self_1): Since we don't use a separate GC
kind for threads any more, and thread marking is keyed off
gc_mutator_set_roots, we can avoid some of the gnarly synchronization.
(on_thread_exit): Arrange to gc_finish_for_thread.
(scm_i_init_thread_for_guile): Use gc_init_for_thread.
(init_main_thread, with_guile, scm_i_with_guile): Use Whippet API.
(scm_threads_prehistory): Take main-thread mutator and the stack base as
arguments.
* libguile/vm.c (scm_trace_vm): Rework in terms of Whippet API.
* libguile/whippet-embedder.h (gc_trace_mutator_roots): Arrange to trace
the current mutator's SCM thread object.
* libguile/trace.h: New file, to declare implementations of trace
routines.
* libguile/Makefile.am (noinst_HEADERS): Add trace.h.
The new non-libltdl foreign library loading algorithm from 3.0.6
fails to cover common cases regarding how libtool names and installs
DLL files. Notably, it fails to recognize when libtool has added the
major version number into the filename itself, such as libfoo-1.dll
Also, it does not search in binary directories and the PATH for DLL
files, where libtool is likely to install DLLs.
This adds the option to search for dlls with major version numbers
in the filename, and modifies the search strategy for DLL-using
OSs to check bindir and PATH.
For MSYS, libraries are installed with the 'msys-' prefix. So this
modifies load-foreign-library to handle that prefix as well.
It changes the #:rename-on-cygwin? option to #:host-type-rename? to
better reflect that is works on both Cygwin and MSYS.
Partially based on a patch by Hannes Müller.
* NEWS: updated
* doc/ref/api-foreign.texi: document updates to load-foreign-library
and system-dll-path
* module/system/foreign-library.scm (is-integer-string?): new utility function
(dll-name-match?): new utility function
(find-best-dll-from-matches): new utility function
(dll-exists-with-version): new function that implements new dll search logic
(file-exists-with-extension): add flag argument to allow new dll search
(file-exists-in-path-with-extension): add flag argument to all new dll search
(system-dll-path): new parameter
(lib->msys): new helper function
(load-foreign-library): add new optarg flag #:allow-dll-version-suffix?
Pass new flag to library search functions.
Implement new search strategy for #:search-system-paths? on DLL systems'
replace #:rename-on-cygwin? with #:host-type-rename?
Use that option to rename both MSYS and Cygwin libraries.
(guile-system-extensions-path): prefer bindir to libdir on DLL systems
* test-suite/tests/foreign.test ("dll-name-match?"): new test category
("find-best-dll-from-matches"): new test category
("lib->msys"): new unit tests
* module/ice-9/custom-ports.scm (make-custom-port): Code fails if
(fluid-ref %default-port-encoding) returns #f. In fact this was the
case why readline support on MSYS2 failed for guile 3.0.10, ref.
https://github.com/msys2/MSYS2-packages/issues/5079
But later used canonicalize-encoding is prepared to handle #f for
encoding. So allow encoding to also handle this case.
Co-authored-by: Ludovic Courtès <ludo@gnu.org>
The ~V is supposed to print ISO week number, not a week number. This
commit fixes that.
* module/srfi/srfi-19.scm (date-week-number-iso): New procedure taken
from the reference implementation.
(directives)<#\V>: Use it.
* test-suite/tests/srfi-19.test ("date->string ~V"): Add tests taken
from the reference test suite.
* doc/ref/srfi-modules.texi (SRFI-19 Date to string): Mention ISO-8601
in description for ~V.
Fixes <https://bugs.gnu.org/74841>.
Edited by lloda <lloda@sarc.name>.
* module/language/tree-il/fix-letrec.scm (compute-sccs): Instead of
depending on all previous complex bindings, we can just depend on the
most recent one. Decreases the graph size.
We were using list sets, which when you end up with thousands of
bindings in an SCC reaches the point where we are off the quadratic end
of the curve. Fix to use intsets and intmaps instead.
* module/language/tree-il/fix-letrec.scm (compute-ids): New function.
(compute-referenced-and-assigned): Rename from analyze-lexicals, and
compute intsets.
(make-compute-free-variables): Rename from free-variables, return a
procedure instead of a hash table, and use intsets. Use a global cache
to avoid quadratic behavior with regard to binding depth.
(compute-complex): Compute a global set of "complex" variables, as an
intset.
(compute-sccs): Use intsets and intmaps to compute the free-variable and
ordering edges.
(fix-scc, fix-term): Refactorings.
(reorder-bindings): Avoid a linear search.
(fix-letrec): Refactor.
Looking at the SRFI-19 specification, the argument is called `day', not
`date'. Even the accessor is called `date-day'. So adjust the
documentation to match.
Also adjust the (web http) module, which was using `date' as well.
* doc/ref/srfi-modules.texi (SRFI-19 Date): Use `day' instead of `date'.
* module/web/http.scm (parse-rfc-822-date, parse-rfc-850-date)
(parse-asctime-date): Same.
Signed-off-by: Ludovic Courtès <ludo@gnu.org>
* module/ice-9/peg/string-peg.scm (peg-as-peg): Augment with rules for
hexadecimal digits, “\uXXX” for characters, “\t” for tabs, etc.
Signed-off-by: Ludovic Courtès <ludo@gnu.org>
Modern PEG supports inversed class like `[^a-z]` that would get any
character not in the `a-z` range. This commit adds support for that and
also for a new `not-in-range` PEG pattern for scheme.
* module/ice-9/peg/codegen.scm (cg-not-in-range): New function.
* module/ice-9/peg/string-peg.scm: Add support for `[^...]`
* test-suite/tests/peg.test: Test it.
* doc/ref/api-peg.texi: Document accordingly.
Signed-off-by: Ludovic Courtès <ludo@gnu.org>
This commit adds support for PEG as described in:
<https://bford.info/pub/lang/peg.pdf>
It adds support for the missing features (comments, underscores in
identifiers and escaping) while keeping the extensions (dashes in
identifiers, < and <--).
The naming system tries to be as close as possible to the one proposed
in the paper.
* module/ice-9/peg/string-peg.scm: Rewrite PEG parser.
* test-suite/tests/peg.test: Fix import
Signed-off-by: Ludovic Courtès <ludo@gnu.org>
