bpt/guile.git
14 years agoImplementation for the R6RS (rnrs programs) library.
Julian Graham [Sun, 21 Mar 2010 21:03:35 +0000 (17:03 -0400)]
Implementation for the R6RS (rnrs programs) library.

* module/Makefile.am: Add rnrs/6/programs.scm to RNRS_SOURCES.
* module/rnrs/6/programs.scm: New file.

14 years agoImplementation and test cases for the R6RS (rnrs unicode) library.
Julian Graham [Sun, 21 Mar 2010 20:19:06 +0000 (16:19 -0400)]
Implementation and test cases for the R6RS (rnrs unicode) library.

* module/Makefile.am: Add rnrs/6/unicode.scm to RNRS_SOURCES.
* module/rnrs/6/unicode.scm: New file.
* test-suite/Makefile.am: Add tests/r6rs-unicode.test to SCM_TESTS.
* test-suite/tests/r6rs-unicode.test

14 years agoFixes and test cases for R6RS (rnrs hashtables) library.
Julian Graham [Sun, 21 Mar 2010 00:51:37 +0000 (20:51 -0400)]
Fixes and test cases for R6RS (rnrs hashtables) library.

* module/rnrs/6/hashtables.scm: Assorted bugfixes, esp. for wrapping
  single-argument hash functions.
* test-suite/Makefile.am: Add tests/r6rs-hashtables.test to SCM_TESTS.
* test-suite/tests/r6rs-hashtables.test: New file.

14 years agoAdd test cases for record constructor protocols and parent protocol
Julian Graham [Sat, 20 Mar 2010 19:14:46 +0000 (15:14 -0400)]
Add test cases for record constructor protocols and parent protocol
delegation.

* test-suite/tests/r6rs-records-procedural.test ("simple protocol",
  "protocol delegates to parent with protocol"): New tests.

14 years agoImplementation for the R6RS (rnrs hashtables) library;
Julian Graham [Sat, 20 Mar 2010 19:10:11 +0000 (15:10 -0400)]
Implementation for the R6RS (rnrs hashtables) library;
Implementation and test cases for the R6RS (rnrs record syntactic) library.

* module/Makefile.am: Add rnrs/6/hashtables.scm to RNRS_SOURCES.
* module/rnrs/6/hashtables.scm: New file.
* module/rnrs/records/6/inspection.scm: (record-type-generative?) Record
  types are generative iff they have no uid, not vice-versa.
* module/rnrs/records/6/syntactic.scm: Finish `define-record-type'
  implementation; add `record-type-descriptor' and
  `record-constructor-descriptor' forms.
* test-suite/Makefile.am: Add tests/r6rs-records-syntactic.test to
  SCM_TESTS.
* test-suite/tests/r6rs-records-inspection.test: Update tests for
  `record-type-generative?' to reflect corrected behavior.
* test-suite/tests/r6rs-records-syntactic.test: New file.

14 years ago(rnrs conditions) should not depend on (rnrs records syntactic).
Julian Graham [Sat, 20 Mar 2010 18:57:49 +0000 (14:57 -0400)]
(rnrs conditions) should not depend on (rnrs records syntactic).

* module/rnrs/6/conditions.scm: (define-condition-type) Re-implement
  `define-condition-type' in terms of (rnrs records procedural).

14 years agoFix missing export of string->symbol in (rnrs base).
Julian Graham [Sat, 20 Mar 2010 12:36:17 +0000 (08:36 -0400)]
Fix missing export of string->symbol in (rnrs base).

* module/rnrs/6/base.scm: Add string->symbol to library exports.

14 years agoImplementation for the R6RS (rnrs lists) library.
Julian Graham [Sat, 20 Mar 2010 00:03:46 +0000 (20:03 -0400)]
Implementation for the R6RS (rnrs lists) library.

* module/Makefile.am: Add module/rnrs/6/lists.scm to RNRS_SOURCES.
* module/rnrs/6/lists.scm: New file.

14 years agoImplementation and test cases for the R6RS (rnrs records inspection)
Julian Graham [Wed, 10 Mar 2010 06:36:15 +0000 (01:36 -0500)]
Implementation and test cases for the R6RS (rnrs records inspection)
library.

* module/Makefile.am: Add module/rnrs/records/6/inspection.scm to RNRS_SOURCES.
* module/rnrs/records/6/inspection.scm: New file.
* module/rnrs/records/6/procedural.scm: Assorted refactoring:
    Create index constants for record, rtd, and rcd field indexes;
    record-type-vtable, record-constructor-vtable: More informative display
    names;
    (make-record-type-descriptor): fold left, not right when creating vtable;
      store field names as vector, not list;
      detect opaque parents
* test-suite/Makefile.am: Add test-suite/tests/r6rs-records-inspection.test to
  SCM_TESTS.
* test-suite/tests/r6rs-records-inspection.test: New file.

14 years agoRemove needless import of (rnrs io simple).
Julian Graham [Wed, 10 Mar 2010 06:26:12 +0000 (01:26 -0500)]
Remove needless import of (rnrs io simple).

* module/rnrs/6/conditions.scm: Remove (rnrs io simple (6)) from imports.

14 years agoImplementation and test cases for the R6RS (rnrs records procedural) library,
Julian Graham [Mon, 8 Mar 2010 14:00:42 +0000 (09:00 -0500)]
Implementation and test cases for the R6RS (rnrs records procedural) library,
along with its dependencies.

* module/Makefile.am: Add new R6RS libraries below to RNRS_SOURCES.
* module/rnrs/6/conditions.scm, exceptions.scm, syntax-case.scm: New files.
* module/rnrs/io/6/simple.scm: New file.
* module/rnrs/records/6/procedural.scm, syntactic.scm: New files.
* test-suite/Makefile.am: Add tests/r6rs-records-procedural.test to SCM_TESTS.
* test-suite/tests/r6rs-records-procedural.test: New file.

14 years agoFix missing imports for `(rnrs arithmetic bitwise)'.
Julian Graham [Sat, 6 Mar 2010 07:04:25 +0000 (02:04 -0500)]
Fix missing imports for `(rnrs arithmetic bitwise)'.

* module/Makefile.am: Add module/rnrs/arithmetic/6/bitwise.scm to
  RNRS_SOURCES.
* module/rnrs/arithmetic/6/bitwise.scm: Import `(rnrs control)' and `modulo'
  from `(guile)'.
* test-suite/Makefile.am: Add r6rs-arithmetic-bitwise.test to SCM_TESTS.

14 years agoImplementation and test cases for the R6RS (rnrs control) library.
Julian Graham [Sat, 6 Mar 2010 06:28:46 +0000 (01:28 -0500)]
Implementation and test cases for the R6RS (rnrs control) library.

* module/Makefile.am: Add rnrs/6/base.scm and rnrs/6/control.scm to
  RNRS_SOURCES.
* module/rnrs/6/base.scm, control.scm: New files.
* test-suite/Makefile.am: Add tests/r6rs-control.test to SCM_TESTS.
* test-suite/tests/r6rs-control.test: New file.

14 years agoImplementation and test cases for the R6RS (rnrs arithmetic bitwise)
Julian Graham [Mon, 8 Feb 2010 14:45:54 +0000 (09:45 -0500)]
Implementation and test cases for the R6RS (rnrs arithmetic bitwise)
library.

* module/rnrs/arithmetic/6/bitwise.scm: New file.
* test-suite/tests/r6rs-arithmetic-bitwise.test: New file.

14 years agoDocumentation for the R6RS `library' and `import' forms.
Julian Graham [Sat, 8 May 2010 21:15:52 +0000 (17:15 -0400)]
Documentation for the R6RS `library' and `import' forms.

* doc/ref/api-modules.texi (R6RS Libraries): New section, documents
  `library' and `import'.

14 years agoeval-when expand for use-modules and export
Andy Wingo [Thu, 20 May 2010 15:52:44 +0000 (17:52 +0200)]
eval-when expand for use-modules and export

* module/ice-9/boot-9.scm (use-modules, use-syntax)
  (export, re-export): Add `expand' to the eval-when list, as we did for
  R6RS.

14 years agopure r6rs modules
Andy Wingo [Thu, 20 May 2010 11:59:34 +0000 (13:59 +0200)]
pure r6rs modules

* module/ice-9/r6rs-libraries.scm (library): Make pure modules.
  (import): Also process imports at expand time, for the `eval' case.

14 years agomacroexpanded pretty-printer
Andy Wingo [Thu, 20 May 2010 11:47:49 +0000 (13:47 +0200)]
macroexpanded pretty-printer

* module/language/tree-il.scm (print-tree-il):
  (borrow-core-vtables, <tree-il>): When tree-il loads, override the
  printers for macroexpanded structures.

14 years agolambda* usage in psyntax.scm
Andy Wingo [Thu, 20 May 2010 11:28:25 +0000 (13:28 +0200)]
lambda* usage in psyntax.scm

* module/ice-9/psyntax.scm (macroexpand): Use lambda*.
* module/ice-9/psyntax-pp.scm: Regenerated.

14 years agotree-il->scheme for lambda* et al
Andy Wingo [Thu, 20 May 2010 11:27:49 +0000 (13:27 +0200)]
tree-il->scheme for lambda* et al

* module/language/tree-il.scm (tree-il->scheme): Fix lambda* /
  case-lambda / case-lambda* transformation.

14 years agoremove *mode* fluid from psyntax
Andy Wingo [Thu, 20 May 2010 10:54:47 +0000 (12:54 +0200)]
remove *mode* fluid from psyntax

* module/ice-9/psyntax.scm: Remove the *mode* fluid. Whee :)
  (macroexpand): No need to bind aa *mode* fluid.

* module/ice-9/psyntax-pp.scm: Regenerated.

14 years agoiron out inconsistency between eval and compile expansion
Andy Wingo [Thu, 20 May 2010 10:53:21 +0000 (12:53 +0200)]
iron out inconsistency between eval and compile expansion

* libguile/expand.c (expand_lambda_star_case): Harmonize with tree-il,
  expanding keywords to (aok? (kw name gensym) ...), not
  (aok? (kw . index) ...).

* libguile/memoize.c (memoize): Process the (kw name gensym) format into
  (kw . index).

* module/ice-9/psyntax.scm (build-lambda-case): Remove a
  compile-versus-eval special case.

* module/ice-9/psyntax-pp.scm: Regenerate.

14 years agoprimitive-eval takes expanded, not memoized, source
Andy Wingo [Thu, 20 May 2010 10:25:52 +0000 (12:25 +0200)]
primitive-eval takes expanded, not memoized, source

* module/ice-9/eval.scm (primitive-eval):
* libguile/eval.c (scm_c_primitive_eval): Don't expect a memoized
  expression -- expect either raw source or an *expanded* expression. We
  handle memoization ourselves.

* libguile/expand.c (scm_macroexpand): Settle down into its proper name,
  "macroexpand", even as we comment that it is but a fleeting boot
  expander.
  (scm_macroexpanded_p): New predicate for expanded code.

* libguile/expand.h: Add scm_macroexpanded_p.

* libguile/memoize.c (scm_memoize_expression): Require that the
  expression be expanded.
  (scm_init_memoize): Don't alias memoize-expression to macroexpand.

* module/ice-9/psyntax-pp.scm:
* module/ice-9/psyntax.scm: Always produce macroexpanded expressions,
  and hand them to primitive-eval. No more calls to memoize-expression
  here.

* test-suite/tests/optargs.test: Remove some tests, as unfortunately we
  have no way to invoke the boot expander after boot.

14 years ago%null-pointer properly aligned
Andy Wingo [Wed, 19 May 2010 23:00:37 +0000 (01:00 +0200)]
%null-pointer properly aligned

* libguile/foreign.c: Allocate %null-pointer on the heap, to assure
  proper alignment.

14 years agopsyntax produces "expanded" structures directly
Andy Wingo [Wed, 19 May 2010 21:56:00 +0000 (23:56 +0200)]
psyntax produces "expanded" structures directly

* module/ice-9/psyntax.scm: Remove a number of eval-versus-compile
  cases, instead producing "expanded" structures directly, which happen
  to coincide with tree-il.

* module/ice-9/psyntax-pp.scm: Regenerate.

14 years agoexpanded type constructors in psyntax
Andy Wingo [Wed, 19 May 2010 21:37:27 +0000 (23:37 +0200)]
expanded type constructors in psyntax

* module/ice-9/psyntax.scm (define-expansion-constructors): Define
  constructors for expanded types.

* module/ice-9/psyntax-pp.scm: Regenerated.

14 years agotree-il on expanded-vtables
Andy Wingo [Wed, 19 May 2010 21:27:14 +0000 (23:27 +0200)]
tree-il on expanded-vtables

* module/language/tree-il.scm: In a somewhat daring move, inherit
  the "core" tree-il constructs from %expanded-vtables.

14 years agofix a couple expander naming bugs
Andy Wingo [Wed, 19 May 2010 21:25:50 +0000 (23:25 +0200)]
fix a couple expander naming bugs

* libguile/expand.h:
* libguile/memoize.c: Fix a couple bugs: it's application-args, not
  application-exps, and public?, not public.

14 years ago%expanded-vtables is a vector
Andy Wingo [Wed, 19 May 2010 20:57:52 +0000 (22:57 +0200)]
%expanded-vtables is a vector

* libguile/expand.c (scm_init_expand): %expanded-vtables is a vector.

14 years agoseparate boot expansion from memoization
Andy Wingo [Wed, 19 May 2010 20:51:31 +0000 (22:51 +0200)]
separate boot expansion from memoization

* libguile/Makefile.am:
* libguile/init.c:
* libguile/expand.c:
* libguile/expand.h: Add new expander. The idea is that macroexpansion
  is one thing, and whether to compile or interpret the result of that
  is another thing.

* libguile/memoize.c: Adapt to expand as necessary, and then memoize
  expanded source without worrying about syntax errors.

* module/ice-9/eval.scm (make-general-closure): Allow alt clauses to not
  possess the full make-general-closure arity.

14 years agoremove the new scm_memoize_begin, etc
Andy Wingo [Tue, 18 May 2010 20:30:58 +0000 (22:30 +0200)]
remove the new scm_memoize_begin, etc

* libguile/memoize.h:
* libguile/memoize.c: Remove the recently-added scm_memoize_begin,
  scm_memoize_lambda, et al. We'll take a different tack.

14 years agoscm_memoize_lambda update
Andy Wingo [Sun, 16 May 2010 10:21:27 +0000 (12:21 +0200)]
scm_memoize_lambda update

* libguile/memoize.h:
* libguile/memoize.c (scm_memoize_lambda): Update to memoize
  "full-arity" lambdas.

14 years agopsyntax builds case-lambda expressions
Andy Wingo [Thu, 13 May 2010 21:27:00 +0000 (23:27 +0200)]
psyntax builds case-lambda expressions

* module/ice-9/psyntax.scm (build-case-lambda, build-lambda-case):
  Actually build case-lambda expressions, knowing that the memoizer will
  do the right thing.

* module/ice-9/psyntax-pp.scm: Regenerated.

14 years agoevaluator support for case-lambda
Andy Wingo [Thu, 13 May 2010 19:43:35 +0000 (21:43 +0200)]
evaluator support for case-lambda

* libguile/memoize.c (patch_case_lambda, scm_m_case_lambda)
  (scm_m_case_lambda_star): Add memoizers for case-lambda and
  case-lambda*.
  (unmemoize): Unmemoize lambdas with multiple arities.

* libguile/eval.c (prepare_boot_closure_env_for_apply):
  (prepare_boot_closure_env_for_eval): Adapt to return both body and
  env, so that case-lambda clauses can be selected appropriately.
  (eval, boot_closure_apply): Adapt callers.

* module/ice-9/eval.scm (make-general-closure): Support multiple
  arities.

14 years agoprimitive support for lambda*
Andy Wingo [Thu, 13 May 2010 15:15:10 +0000 (17:15 +0200)]
primitive support for lambda*

* libguile/memoize.c (scm_m_lambda_star): Define lambda* in the
  pre-psyntax env, and make it memoize lambda* expressions.

* libguile/eval.c (BOOT_CLOSURE_PARSE_FULL): New helper.
  (error_invalid_keyword, error_unrecognized_keyword): New helpers.
  (prepare_boot_closure_env_for_apply): Flesh out application of boot
  closures with "full" arity.
  (prepare_boot_closure_env_for_eval): Punt to
  prepare_boot_closure_env_for_eval for the full-arity case.

* module/ice-9/eval.scm (make-fixed-closure): Rename from `closure', and
  just handle fixed arities, where there is no rest argument..
  (make-general-closure): New helper, a procedure, that returns a
  closure that can take rest, optional, and keyword arguments.
  (eval): Adapt to call make-fixed-closure or make-general-closure as
  appropriate.

* test-suite/tests/optargs.test ("lambda* inits"): Test the memoizer as
  well.

14 years agotweak to lambda* memoization format
Andy Wingo [Thu, 13 May 2010 14:56:31 +0000 (16:56 +0200)]
tweak to lambda* memoization format

* libguile/memoize.c (FULL_ARITY): Change the form to more closely
  approximate <lambda-case>.
  (unmemoize): Unmemoize lambda* expressions.

14 years agochange to format of memoization lambda expressions
Andy Wingo [Wed, 12 May 2010 10:17:18 +0000 (12:17 +0200)]
change to format of memoization lambda expressions

* libguile/memoize.c (scm_m_lambda, memoize_named_let)
  (scm_memoize_lambda, unmemoize)
* libguile/eval.c (eval, prepare_boot_closure_env_for_apply)
  (prepare_boot_closure_env_for_eval, boot_closure_apply):
* module/ice-9/eval.scm (primitive-eval): Change the format for memoized
  lambda expressions, so as to allow, in the future, case-lambda* to be
  supported by the evaluator.

14 years agoscm_badargsp is deprecated
Andy Wingo [Wed, 12 May 2010 10:02:09 +0000 (12:02 +0200)]
scm_badargsp is deprecated

* libguile/eval.h:
* libguile/eval.c:
* libguile/deprecated.h
* libguile/deprecated.c (scm_badargsp): Deprecate scm_badargsp.

14 years agoHave `sxml->xml' handle `*TOP*' nodes (bug #29260).
Ludovic Courtès [Tue, 11 May 2010 20:54:14 +0000 (22:54 +0200)]
Have `sxml->xml' handle `*TOP*' nodes (bug #29260).

* module/sxml/simple.scm (universal-sxslt-rules): Add handler for
  `*TOP*'.  Suggested by Stefan Israelsson Tampe at
  <https://savannah.gnu.org/bugs/index.php?29260>.

* test-suite/Makefile.am (SCM_TESTS): Add `sxml.simple.test'.

* test-suite/tests/sxml.simple.test: New file.

* THANKS: Add Stefan.

14 years agoadd memoize-begin, memoize-let, etc functions
Andy Wingo [Fri, 7 May 2010 21:12:34 +0000 (23:12 +0200)]
add memoize-begin, memoize-let, etc functions

* libguile/memoize.h:
* libguile/memoize.c (scm_memoize_begin, scm_memoize_if,
  scm_memoize_lambda, scm_memoize_let, scm_memoize_quote,
  scm_memoize_define, scm_memoize_with_fluids, scm_memoize_call,
  scm_memoize_lexical_ref, scm_memoize_lexical_set,
  scm_memoize_toplevel_ref, scm_memoize_toplevel_set,
  scm_memoize_module_ref, scm_memoize_module_set): New functions, public
  to Scheme but private to C. For use by psyntax, in the future.

14 years agoadd "memoizer" data type
Andy Wingo [Fri, 7 May 2010 20:19:36 +0000 (22:19 +0200)]
add "memoizer" data type

* libguile/memoize.h:
* libguile/memoize.c (MAKMEMO_APPLY): Take the proc and args separately.
  (scm_tc16_memoizer, SCM_MEMOIZER_P, SCM_MEMOIZER): New smob type, for
  "syntax" like @apply that have normal evaluation order.
  (memoize_env_ref_macro, memoize): Refactor to handle "memoizers"
  generically.

  Reorganize the list of SCM_SYNTAX forms.

  Move @apply, @dynamic-wind, @call-with-current-continuation,
  @call-with-values, @prompt handling down to be generic "memoizer" data
  types.

  (scm_memoizer_p, scm_memoizer): New functions, exposed to Scheme. Will
  be used by psyntax.

14 years agomacro expanders return memoized expressions
Andy Wingo [Fri, 7 May 2010 08:51:28 +0000 (10:51 +0200)]
macro expanders return memoized expressions

* libguile/eval.c (scm_c_primitive_eval):
* module/ice-9/eval.scm (primitive-eval): Rely on the expander to
  produce a memoized expression. If the expression is already memoized,
  just pass it through (the equivalent of the old "noexpand" hack).

* libguile/memoize.c (scm_init_memoize): Initialize `memoize-expression'
  as the initial binding of `macroexpand'.

* libguile/modules.c (scm_module_transformer): Before modules are
  booted, look for `macroexpand', not `%pre-modules-transformer'.

* module/ice-9/boot-9.scm: No more %pre-modules-transformer. Loading
  psyntax extends `macroexpand'.
  (make-module): `macroexpand' is the default transformer.

* module/ice-9/psyntax.scm: No more `noexpand'.
  (top-level-eval-hook, local-eval-hook): Instead of annotating with
  noexpand, memoize the expression before handing it to primitive-eval.
  (macroexpand): No more noexpand hack -- in its place we have another
  hack, to memoize the result when running in eval mode.
* module/ice-9/psyntax-pp.scm: Regenerated.

14 years agoAttempt to guarantee 8-byte alignment of on-stack frame objects.
Ludovic Courtès [Fri, 7 May 2010 13:59:50 +0000 (15:59 +0200)]
Attempt to guarantee 8-byte alignment of on-stack frame objects.

* libguile/vm.c (vm_dispatch_hook): Make FRAME a union with a `double'
  member.

14 years agoAdd (system vm coverage).
Ludovic Courtès [Sun, 2 May 2010 12:17:41 +0000 (14:17 +0200)]
Add (system vm coverage).

* module/system/vm/coverage.scm: New file.

* module/Makefile.am (SYSTEM_SOURCES): Add `system/vm/coverage.scm'.

* test-suite/guile-test (main): Use (system vm coverage).  Handle
  `--coverage' and `-c'.

* test-suite/tests/coverage.test: New file.

* test-suite/Makefile.am (SCM_TESTS): Add `tests/coverage.test'.

* doc/ref/Makefile.am (guile_TEXINFOS): Add `api-coverage.texi'.

* doc/ref/api-coverage.texi: New file.

* doc/ref/guile.texi (API Reference): Include it.

14 years agoAllocate frame objects on the stack when invoking VM hooks.
Ludovic Courtès [Tue, 20 Apr 2010 23:05:42 +0000 (01:05 +0200)]
Allocate frame objects on the stack when invoking VM hooks.

* libguile/vm.c (vm_dispatch_hook): Don't call `scm_c_make_frame ()'.
  Instead, allocate FRAME on the stack.

14 years agoAdd `program-free-variables' to `(system vm program)'.
Ludovic Courtès [Wed, 21 Apr 2010 21:30:48 +0000 (23:30 +0200)]
Add `program-free-variables' to `(system vm program)'.

* module/system/vm/program.scm (program-free-variables): New procedure.

* module/language/objcode/spec.scm (program-free-variables): Remove.

14 years agoFix the type tag produced by `SCM_STATIC_PROGRAM'.
Ludovic Courtès [Sun, 18 Apr 2010 22:27:47 +0000 (00:27 +0200)]
Fix the type tag produced by `SCM_STATIC_PROGRAM'.

* libguile/snarf.h (SCM_STATIC_PROGRAM): Fix typo.

14 years agoWhen printing a VM, show the type of engine being used.
Ludovic Courtès [Wed, 14 Apr 2010 21:46:41 +0000 (23:46 +0200)]
When printing a VM, show the type of engine being used.

* libguile/vm.c (scm_i_vm_print): Print the engine type.

14 years agofix macros that expand into (begin (define-module ...) ...)
Andy Wingo [Thu, 6 May 2010 20:34:36 +0000 (22:34 +0200)]
fix macros that expand into (begin (define-module ...) ...)

* module/ice-9/boot-9.scm (define-module): Add `expand' to the
  situations, so that we ensure the current module changes at expand
  time.

* module/ice-9/r6rs-libraries.scm (library): Expand the body within @@
  expressions.

14 years agopsyntax no longer attempts to track (current-module)
Andy Wingo [Thu, 6 May 2010 20:32:50 +0000 (22:32 +0200)]
psyntax no longer attempts to track (current-module)

* module/ice-9/psyntax.scm: Revert tracking of the current module while
  expanding toplevel sequences. Better to preserve lexical scoping
  (relative to the module of the original expression) and allow for @@
  to override it.

* module/ice-9/psyntax-pp.scm: Regenerated.

* test-suite/tests/syncase.test: Test module-changing with @@.

14 years ago@@ generalized to arbitrary expressions.
Andy Wingo [Thu, 6 May 2010 19:12:09 +0000 (21:12 +0200)]
@@ generalized to arbitrary expressions.

* module/ice-9/psyntax.scm (@@): The RHS of an @@ is no longer
  restricted to be an identifier, it can be an arbitrary expression.

* module/ice-9/psyntax-pp.scm: Regenerated.

14 years agochange form of @ and @@ psyntax handlers
Andy Wingo [Wed, 5 May 2010 22:15:06 +0000 (00:15 +0200)]
change form of @ and @@ psyntax handlers

* module/ice-9/psyntax.scm (@, @@): Take r and w args, and also return
  r, w, and s. Adapt callers.

14 years agoadd `expand' eval-when situation
Andy Wingo [Wed, 5 May 2010 20:59:32 +0000 (22:59 +0200)]
add `expand' eval-when situation

* module/ice-9/psyntax.scm (chi-when-list, chi-top): Add an undocumented
  `expand' situation to eval-when, which will trigger at expand-time for
  toplevel expressions. Useful when you are expanding for eval, but need
  to do some side effect to affect the expander -- like defining a
  module.

* module/ice-9/psyntax-pp.scm: Regenerate.

14 years agoadd explanatory comment to chi-macro
Andy Wingo [Tue, 4 May 2010 21:47:22 +0000 (23:47 +0200)]
add explanatory comment to chi-macro

* module/ice-9/psyntax.scm (chi-macro): Add comment.

14 years agocheck layout when making vtables
Andy Wingo [Tue, 4 May 2010 21:20:12 +0000 (23:20 +0200)]
check layout when making vtables

* libguile/struct.c (scm_is_valid_vtable_layout): New private layout
  validating function.
  (scm_i_struct_inherit_vtable_magic): Do a more proper layout
  validation, and don't abort.

14 years agoadd rnrs libraries test suite
Andy Wingo [Mon, 3 May 2010 17:09:35 +0000 (19:09 +0200)]
add rnrs libraries test suite

* test-suite/Makefile.am:
* test-suite/tests/rnrs-libraries.test:
* test-suite/tests/rnrs-test-a.scm: Add rnrs libraries test suite.

14 years agoadd support for r6rs libraries
Julian Graham [Sat, 6 Feb 2010 17:33:20 +0000 (12:33 -0500)]
add support for r6rs libraries

* module/ice-9/boot-9.scm: Include ice-9/r6rs-libraries.scm.
* module/ice-9/r6rs-libraries.scm: New file.
* module/Makefile.am (NOCOMP_SOURCES): Add r6rs-libraries.scm.

14 years agoonly ensure local vars for definitions during expansion when compiling
Andy Wingo [Mon, 3 May 2010 20:06:44 +0000 (22:06 +0200)]
only ensure local vars for definitions during expansion when compiling

* module/ice-9/psyntax.scm (chi-top): Only modify the compile-time
  environment to indicate that there is a local binding if we are
  running in compile mode. Otherwise some eval-when things might not
  have gotten executed, and our current-module could be wrong.

* module/ice-9/psyntax-pp.scm: Regenerate.

14 years agoexport, re-export etc as syntax-rules macros
Andy Wingo [Mon, 3 May 2010 14:34:12 +0000 (16:34 +0200)]
export, re-export etc as syntax-rules macros

* module/ice-9/boot-9.scm (export, re-export, export-syntax)
  (re-export-syntax): Rewrite as syntax-rules macros.

14 years agorewrite use-modules and use-syntax using syntax-case
Andy Wingo [Mon, 3 May 2010 14:31:32 +0000 (16:31 +0200)]
rewrite use-modules and use-syntax using syntax-case

* module/ice-9/boot-9.scm (use-modules): Rewrite as a syntax-case macro.
  (use-syntax): Likewise.
  (compile-interface-spec): Remove unused function

14 years agorewrite define-module as a syntax-case macro
Andy Wingo [Mon, 3 May 2010 13:38:29 +0000 (15:38 +0200)]
rewrite define-module as a syntax-case macro

* module/ice-9/boot-9.scm (define-module): Rewrite as a syntax-case
  macro, so that the expansion has proper module hygiene. Otherwise
  process-define-module isn't properly resolved against the root module
  -- a bytecode file that starts with a define-module would just try to
  look up process-define-module from the current module.
  (compile-define-module-args): Remove. Internal, and no one else used
  it.

14 years agoremove duplicate quasisyntax.scm EXTRA_DIST
Andy Wingo [Sun, 2 May 2010 17:08:26 +0000 (19:08 +0200)]
remove duplicate quasisyntax.scm EXTRA_DIST

* module/Makefile.am (EXTRA_DIST): Remove quasisyntax.scm, as the
  NOCOMP_SOURCES handles that case.

14 years agoTest cases for `version-matches?'.
Julian Graham [Sun, 2 May 2010 18:58:15 +0000 (14:58 -0400)]
Test cases for `version-matches?'.

* test-suite/tests/modules.test ("module versions"): New tests for
  `version-matches?'.

14 years agoadd tour.texi to guile_TEXINFOS
Andy Wingo [Sun, 2 May 2010 12:27:46 +0000 (14:27 +0200)]
add tour.texi to guile_TEXINFOS

* doc/ref/Makefile.am (guile_TEXINFOS): Add tour.texi.

14 years agoUpdate example disassemblies
Andy Wingo [Sun, 2 May 2010 12:18:58 +0000 (14:18 +0200)]
Update example disassemblies

* doc/ref/compiler.texi:
* doc/ref/vm.texi: Update example disassemblies for current compiler,
  which uses assert-nargs-ee/locals.

14 years agoadd assert-nargs-ee/locals instruction
Andy Wingo [Sun, 2 May 2010 11:41:31 +0000 (13:41 +0200)]
add assert-nargs-ee/locals instruction

* libguile/vm-i-system.c (assert-nargs-ee/locals): New instruction, a
  combination of assert-nargs-ee and reserve-locals in the case in which
  nreq and nlocs can both be represented in 8 bits.

* module/language/glil/compile-assembly.scm (glil->assembly): Add
  compiler case.

* doc/ref/vm.texi (Function Prologue Instructions): Update docs.

14 years agoadd underflow check in make-array
Andy Wingo [Sun, 2 May 2010 11:15:22 +0000 (13:15 +0200)]
add underflow check in make-array

* libguile/vm-i-scheme.c (make-array): Check for underflow before making
  the array.

14 years agoadd SCM_UNLIKELY to CHECK_UNDERFLOW
Andy Wingo [Sun, 2 May 2010 11:14:58 +0000 (13:14 +0200)]
add SCM_UNLIKELY to CHECK_UNDERFLOW

* libguile/vm-engine.h (CHECK_OVERFLOW, CHECK_UNDERFLOW): Add
  SCM_UNLIKELY blocks. Fix off-by-one error in CHECK_UNDERFLOW.
  (PRE_CHECK_UNDERFLOW): New macro, for checking underflow before
  accessing the current sp.
  (POP): Use PRE_CHECK_UNDERFLOW.

14 years agoupdate compiler.texi
Andy Wingo [Sun, 2 May 2010 10:46:50 +0000 (12:46 +0200)]
update compiler.texi

* doc/ref/compiler.texi (Compiler Tower): Update for removal of version
  from <language>, and add joiner and make-default-environment fields.
  Update examples.
  (The Scheme Compiler): Update for `macroexpand' instead of
  `sc-expand', and that the environment must be a module.
  (Tree-IL): Update for new Tree-IL, and change from "vars" to
  "gensyms".
  (GLIL): Update for new GLIL, including preludes and prompts.
  (Assembly): Update for current output (which seems quite verbose).
  (Bytecode and Objcode): Update for current output, and some procedure
  name changes.

14 years agorename `vars' field in tree-il binding constructs to `gensyms'
Andy Wingo [Sun, 2 May 2010 09:22:23 +0000 (11:22 +0200)]
rename `vars' field in tree-il binding constructs to `gensyms'

* module/language/tree-il.scm (<tree-il>): Rename `vars' fields of
  <let>, <letrec>, <fix>, and <lambda-case> to `gensyms'. For clarity,
  and to match <lexical-ref>.

* module/language/tree-il.scm:
* module/language/tree-il/analyze.scm:
* module/language/tree-il/compile-glil.scm:
* module/language/tree-il/fix-letrec.scm:
* module/language/tree-il/inline.scm: Update all callers.

14 years agofix a comment about environments and the scheme compiler
Andy Wingo [Sun, 2 May 2010 09:19:46 +0000 (11:19 +0200)]
fix a comment about environments and the scheme compiler

* module/language/scheme/compile-tree-il.scm: Fix a comment about the
  format of environments.

14 years agoremove `version' field from <language>
Andy Wingo [Sun, 2 May 2010 09:19:13 +0000 (11:19 +0200)]
remove `version' field from <language>

* module/system/base/language.scm (<language>): Remove the `version'
  field from languages. It just wasn't useful.

* module/language/assembly/spec.scm:
* module/language/brainfuck/spec.scm:
* module/language/bytecode/spec.scm:
* module/language/ecmascript/spec.scm:
* module/language/elisp/spec.scm:
* module/language/glil/spec.scm:
* module/language/objcode/spec.scm:
* module/language/scheme/spec.scm:
* module/language/tree-il/spec.scm:
* module/language/value/spec.scm: Remove #:version from all language
  definitions. Shorten some language names (e.g. "Guile Scheme" ->
  "Scheme").

14 years agoupdate documentation for vm instructions
Andy Wingo [Sat, 1 May 2010 15:02:39 +0000 (17:02 +0200)]
update documentation for vm instructions

* doc/ref/vm.texi: Update with documentation for new-ish instructions,
  and reorganize the sections a bit.

* doc/ref/compiler.texi (GLIL): Fix up a couple xrefs.

14 years agosimplify make-struct
Andy Wingo [Fri, 30 Apr 2010 22:49:06 +0000 (00:49 +0200)]
simplify make-struct

* libguile/vm-i-scheme.c: Use DROPN idiom.

14 years agooptimize and bugfix make-struct VM opcode
Andy Wingo [Fri, 30 Apr 2010 22:31:18 +0000 (00:31 +0200)]
optimize and bugfix make-struct VM opcode

* libguile/_scm.h (SCM_OBJCODE_MINOR_VERSION): Bump for make-struct
  change.

* libguile/struct.c (scm_i_alloc_struct): Use scm_words instead of
  scm_gc_malloc to simplify the code and inline the call to GC_MALLOC.

* module/language/tree-il/compile-glil.scm (*primcall-ops*): Compile
  make-struct/no-tail to make-struct.

* module/language/tree-il/primitives.scm (define-primitive-expander):
  Allow a conditional branch of #f to aboirt inlining.
  (make-struct): Expand into make-struct/no-tail in the case that
  tail-size is 0.

* libguile/vm-i-scheme.c (make-struct): Adapt to always assume tail-size
  is 0. Inline allocation if possible. Don't decrement the SP past live
  objects on the stack, which could cause GC to miss references. Use the
  NULLSTACK macro.

14 years agofix bug in compile-glil.scm for return opcode-hack
Andy Wingo [Tue, 27 Apr 2010 21:04:39 +0000 (23:04 +0200)]
fix bug in compile-glil.scm for return opcode-hack

* module/language/tree-il/compile-glil.scm (*primcall-ops*): Fix entry
  for `return'. Thanks to Jon Herron for the report.

14 years agoadd module-export-all!
Andy Wingo [Tue, 27 Apr 2010 20:57:54 +0000 (22:57 +0200)]
add module-export-all!

* module/ice-9/boot-9.scm (module-export-all!): New function, exports
  all current and future local variables from a module.

14 years agoMerge branch 'wip-module-namespaces'
Andy Wingo [Tue, 27 Apr 2010 20:04:33 +0000 (22:04 +0200)]
Merge branch 'wip-module-namespaces'

14 years agocomment some global variables in modules.c
Andy Wingo [Tue, 27 Apr 2010 20:04:24 +0000 (22:04 +0200)]
comment some global variables in modules.c

* libguile/modules.c: Comment some global variables.

14 years agoFix inconsistent sectioning, causing make to fail
Neil Jerram [Mon, 26 Apr 2010 21:09:20 +0000 (22:09 +0100)]
Fix inconsistent sectioning, causing make to fail

* doc/ref/guile.texi (Top): Remove @raisesections and @lowersections
  around scheme-ideas.texi.

* doc/ref/scheme-ideas.texi (About Data, ...): Instead remove one
  "sub" from every node here apart from the chapter node.

* doc/ref/scheme-ideas.texi (Evaluating, Eval Variable, Eval
  Procedure, Eval Special, Lexical Scope, Scoping Example): Turn
  subheadings back into subsubsections, and reinstate corresponding
  menus.  (These had to become headings, rather than (sub)*sections, when
  the top level item in scheme-ideas.texi was a section.  Now it's a
  chapter again, they can go back to being (sub)*sections.)

14 years agoBring the manual organisation section a little closer to reality
Neil Jerram [Mon, 26 Apr 2010 04:24:24 +0000 (05:24 +0100)]
Bring the manual organisation section a little closer to reality

FWIW I'm not sure I like this bullety presentation - but for now this
just aligns the descriptions a bit closer to the actual chapters and
chapter contents.

* doc/ref/intro.texi (Organisation of this Manual): Cover `Hello
  Guile!' and `Hello Scheme!' chapters.  Thin out text for
  `Programming in Scheme'.  Renumber other chapters.

14 years agoAssorted minor manual changes
Neil Jerram [Mon, 26 Apr 2010 03:27:01 +0000 (04:27 +0100)]
Assorted minor manual changes

* doc/ref/guile.texi (Programming in Scheme): Remove "awful" and add
  "as opposed to having to dive into C code", to explain what we
  really mean here.  Remove incorrect para about the `Hello Scheme!'
  material.

* doc/ref/intro.texi (Interactive Programming): Change "allows" to "is
  designed for" - sounds less accidental!

* doc/ref/scheme-ideas.texi (Hello Scheme!): Remove unnecessary
  "reference", and inappropriate "randomness".

14 years agoMore `Guile and the GNU Project' text
Neil Jerram [Mon, 26 Apr 2010 01:43:54 +0000 (02:43 +0100)]
More `Guile and the GNU Project' text

* doc/ref/intro.texi (Guile and the GNU Project): More text about
  extensibility and software freedom.

14 years agoMove description of manual organisation from Preface to Introduction
Neil Jerram [Sun, 25 Apr 2010 17:53:40 +0000 (18:53 +0100)]
Move description of manual organisation from Preface to Introduction

This material isn't "meta-" enough to be in the Preface, and it
supports (by its structure) some of the points that the Introduction
makes about Guile, so I think it belongs here.

* doc/ref/intro.texi (Introduction): Add `Organisation of this Manual'
  menu item and material to here...

* doc/ref/preface.texi (Preface): ...from here.

14 years agoRemove weird instruction
Neil Jerram [Sun, 25 Apr 2010 17:28:19 +0000 (18:28 +0100)]
Remove weird instruction

* doc/ref/intro.texi: Remove `Do not confuse the two.'  (Or else
  what?)

14 years agoFix typo
Neil Jerram [Sun, 25 Apr 2010 17:12:01 +0000 (18:12 +0100)]
Fix typo

* doc/ref/compiler.texi (Bytecode and Objcode): Remove extraneous
  comma.

14 years agoMove `Reporting Bugs' section out of Introduction
Neil Jerram [Sun, 25 Apr 2010 16:56:56 +0000 (17:56 +0100)]
Move `Reporting Bugs' section out of Introduction

I think it's too detailed, when compared with the other Introduction
sections.  It probably also refers to concepts that haven't been
mentioned yet, but are mentioned in the following quick tour chapter;
so let's move it there.

* doc/ref/guile.texi: Move @raisesections and @lowersections for
  tour.texi material from here to inside tour.texi.

* doc/ref/intro.texi (Introduction): Don't mention reporting problems
  in first para.  Move menu item and material for `Reporting Bugs'
  from here...

* doc/ref/tour.texi (Hello Guile!): ...to here.  Also mention
  reporting problems in the introductory para.

14 years agoMove whirlwind tour material to its own chapter
Neil Jerram [Sun, 25 Apr 2010 16:01:01 +0000 (17:01 +0100)]
Move whirlwind tour material to its own chapter

* doc/ref/guile.texi (Top): New top level `Hello Guile!' menu item;
  include new file tour.texi.

* doc/ref/intro.texi (Introduction): Remove whirlwind tour menu item
  and material from here...

* doc/ref/tour.texi: ...to new file here.

14 years agoMove Scheme introduction (Guile-independent) to its own chapter
Neil Jerram [Sun, 25 Apr 2010 12:20:17 +0000 (13:20 +0100)]
Move Scheme introduction (Guile-independent) to its own chapter

This magically fixes all the internal wordings that say "chapter"
rather than "section", and separates Guile-independent and skippable
material from material that is Guile-specific and
probably-not-skippable - which I think is helpful.

* doc/ref/guile.texi (Top): New `Hello Scheme!' menu item, and removed
  `Basic Ideas' item.  Move include of scheme-ideas.texi before
  `Programming in Scheme' chapter, using @raisesections and
  @lowersections.  Also move scheme-reading.texi to same place.

* doc/ref/scheme-ideas.texi: Change name from `Basic Ideas' (which is
  a bit boring) to `Hello Scheme!'.  Move `Further Reading' menu item
  here.

14 years agomodule-public-interface is a field in the module record
Andy Wingo [Fri, 23 Apr 2010 15:16:56 +0000 (17:16 +0200)]
module-public-interface is a field in the module record

* module/ice-9/boot-9.scm (module-public-interface)
  (set-module-public-interface!): Instead of using
  '%module-public-interface, use a field in the module instead.
  (make-module, make-autoload-interface): Adapt.

* module/ice-9/deprecated.scm (module-public-interface):
  (set-module-public-interface!): Add shims so that manually munging
  %module-public-interface should continue to work.

14 years agofix some uses of %module-public-interface
Andy Wingo [Fri, 23 Apr 2010 15:22:00 +0000 (17:22 +0200)]
fix some uses of %module-public-interface

* module/ice-9/r5rs.scm:
* module/ice-9/safe-r5rs.scm:
* module/oop/goops/save.scm:
* module/oop/goops/simple.scm:
* module/oop/goops/stklos.scm: Fix some uses of
  %module-public-interface.

14 years agomodule-public-interface in Scheme
Andy Wingo [Fri, 23 Apr 2010 15:03:34 +0000 (17:03 +0200)]
module-public-interface in Scheme

* libguile/modules.c: Consolidate all variables to the top of the file.
  (scm_module_public_interface): Dispatch to Scheme.
  (scm_post_boot_init_modules): Resolve module-public-interface.

* module/ice-9/boot-9.scm (module-public-interface): Implement in
  Scheme.

14 years agoadd submodule binders
Andy Wingo [Fri, 23 Apr 2010 14:31:09 +0000 (16:31 +0200)]
add submodule binders

* module/ice-9/boot-9.scm (module-submodule-binder)
  (set-module-submodule-binder!): New field on modules.
  (make-module, make-autoload-interface): Adapt.
  (module-ref-submodule): If we miss the submodules table, call the
  submodules binder, if any.

* module/ice-9/deprecated.scm (module-ref-submodule): Check the
  submodule binder before the deprecated look into the obarray.

14 years agomodule-{ref,define}-submodule use the submodules table
Andy Wingo [Fri, 23 Apr 2010 14:13:06 +0000 (16:13 +0200)]
module-{ref,define}-submodule use the submodules table

* module/ice-9/boot-9.scm (module-ref-submodule)
  (module-define-submodule!): Implement in terms of the
  module-submodules table, instead of looking for bindings in the value
  namespace.

* module/ice-9/deprecated.scm (module-ref-submodule):
  (module-define-submodule!): Define deprecated versions of these that
  duplicate the submodules table in the normal values namespace, for
  back compatibility.

14 years agoadd submodules field to modules
Andy Wingo [Thu, 22 Apr 2010 12:12:05 +0000 (14:12 +0200)]
add submodules field to modules

* module/ice-9/boot-9.scm (module-type, module-constructor): Add a field
  to modules, a table that will hold submodules.
  (make-module, make-autoload-interface): Adapt.

14 years agouse nested-ref-module in (system xref)
Andy Wingo [Fri, 23 Apr 2010 14:08:01 +0000 (16:08 +0200)]
use nested-ref-module in (system xref)

* module/system/xref.scm (program-callee-rev-vars): Use
  nested-ref-module instead of nested-ref.

14 years agonested module fixen to (ice-9 ls)
Andy Wingo [Fri, 23 Apr 2010 14:03:23 +0000 (16:03 +0200)]
nested module fixen to (ice-9 ls)

* module/ice-9/ls.scm (local-definitions-in, definitions-in): Use nested
  module procedures, as appropriate.
  (recursive-local-define): Fix attempt to treat null as false. Whoops..

14 years agodeprecated %app shims use module-define-submodule!
Andy Wingo [Fri, 23 Apr 2010 13:58:08 +0000 (15:58 +0200)]
deprecated %app shims use module-define-submodule!

* module/ice-9/deprecated.scm (%app, app, modules): Don't use the module
  interface to provide %app shims, use module-define-submodule! directly
  to side-effect the root module.

14 years agoresolve-module uses -module variants
Andy Wingo [Fri, 23 Apr 2010 13:55:32 +0000 (15:55 +0200)]
resolve-module uses -module variants

* module/ice-9/boot-9.scm (resolve-module): Use module-define-submodule!
  when binding the root, and nested-ref-module for the hot lookup, which
  is guaranteed to return a module or #f.

14 years agouse the define-module variants in module-name and make-modules-in
Andy Wingo [Fri, 23 Apr 2010 13:44:31 +0000 (15:44 +0200)]
use the define-module variants in module-name and make-modules-in

* module/ice-9/boot-9.scm (module-name): Use module-define-submodule!
  instead of nested-define!.
  (make-modules-in): Rewrite in terms of nested-define-module!.