... by removing export.m and options_file.m from that set.
compiler/export.m:
compiler/options_file.m:
Replace bespoke error reporting code with invocations of standardized
error reporting predicates.
compiler/file_util.m:
Add versions of existing error reporting predicates and functions that
work without access to a globals structure, for use by options_file.m.
... by removing compiler_util.m from that set.
compiler/tupling.m:
Replace the only two calls to report_warning with code to return
error_specs.
compiler/mercury_compile_middle_passes.m:
Accumulate the resulting error_specs.
compiler/write_error_spec.m:
Inline the only call to record_warning_opt_table.
compiler/compiler_util.m:
Delete report_warning and record_warning_opt_table, which now have
no callers, and record_warning, which had no callers to begin with.
samples/c_interface/short_example.m:
Do not output two trailing newlines.
samples/concurrency/midimon/concurrent_stream.m:
Module qualify some calls.
samples/java_interface/short_example.m:
s/JAva/Java/
samples/muz/muz.m:
Fix some spelling errors.
samples/calculator2.m:
Do not output error messages that begin with "unexpected unexpected".
samples/sort.m:
Fix inadvertently escaped newline in usage message.
samples/ultra_sub.m:
Fix inverted arguments in usage message.
C99 introduced additional conversion specifiers for use with the printf and
scanf family of functions. Specifically, the 'z' size specifier for size_t
values and the 't' size specifier for ptrdiff_t values. Historically, we have
not used them because the C99 support in a certain C compiler was lacking.
Since this is no longer the case, use them. This removes the need of a bunch
of casts, some of which were incorrect on 64-bit Windows.
runtime/mercury_accurate_gc.c:
runtime/mercury_memory_zones.c:
runtime/mercury_stacks.c:
runtime/mercury_wrapper.c:
As above.
The date/0 type is misnamed. Values of the type have both a date and a time
component. The common name for combined date and time values is a "date_time",
for which we have had a type synonym since 2014. This change makes date_time
the proper name for type and make date into the type synonym.
Deprecate the date/0 name and note that we will change its meaning in a future
release. (It will eventually be used for data values that do not have a time
component.)
Rename predicates and functions accordingly and mark the existing versions as
obsolete.
library/calendar.m:
Make the above renamings.
library/hard_coded/stream.string_writer.m:
Replace a call to a now obsolete function.
NEWS.md:
Add entry describing the above.
tests/hard_coded/calendar_init_date.{m,exp}:
tests/hard_coded/calendar_test.m:
tests/hard_coded/fold_days.m:
tests/hard_coded/stream_string_writer_types.m:
Conform to the above changes.
Add a predicate to the calendar module for testing if a year is a leap year.
Use a more efficient method for determining this than the existing code
in the implementation of this module used (and replace that code with
a call to the new predicate).
Add the function days_to_month/2, which is a strongly typed wrapper
for the implementation function max_day_in_month_for/2.
Add a new test case covering basic operations in the calendar module,
together with the newly added operations.
library/calendar.m:
As above.
NEWS.md:
Announce the new additions.
tests/hard_coded/Mmakefile:
tests/hard_coded/calendar_basics.{m,exp}:
Add the new test case.
tests/debugger/Mmakefile:
Delete the bespoke make rules for test cases that do not require
their own special handling.
tests/debugger/all_solutions.exp:
tests/debugger/all_solutions.exp2:
tests/debugger/all_solutions.exp4:
tests/debugger/browse_packed.exp:
tests/debugger/browse_pretty.exp:
tests/debugger/class_decl.exp:
tests/debugger/cmd_quote.exp:
tests/debugger/debugger_regs.exp:
tests/debugger/field_names.exp:
tests/debugger/implied_instance.exp:
tests/debugger/label_layout.exp:
tests/debugger/list_cmd.exp:
tests/debugger/lval_desc_array.exp:
tests/debugger/multi_parameter.exp:
tests/debugger/multi_parameter.exp2:
tests/debugger/mutrec.exp:
tests/debugger/no_inline_builtins.exp:
tests/debugger/queens_rep.exp:
tests/debugger/shallow.exp:
tests/debugger/shallow.exp2:
Expect standardized event numbers and call sequence numbers,
since that is what the implicit rule specifies.
tests/debugger/all_solutions.m:
Fill in missing info about which .expN file is for what situations.
library/float.m:
Remove out-of-date references to ANSI C, POSIX, C9X (C99), and
platform-specific behaviour on Linux, Digital Unix and Solaris.
All current Mercury backends use (and are _required_ to use) IEEE floating
point; delete conditional language about this.
Do not specify halting with runtime error as a resolution for NaNs or
signed zeros; instead just say we should throw an exception.
Generalise the discussion of excess precision.
(XXX ideally we could get rid of this by telling C compilers not to
use extended-precision FP registers on those system that provide
them; most of them do that by default anyway, but I'm not sure if
all of them do.)
Delete parts of an action that are needed only on machines running OSF-1.
Delete parts of an action that have been obsolete since we switched
to '.' as the module qualifier character decades ago.
mul_by_digit(0, Y) produced i(Len, [0, 0, ...]) instead of the
canonical integer.zero = i(0, []). Since is_zero/1 only matched
the canonical form, denormalized zeros were silently treated as
nonzero, causing incorrect results in rational.m (e.g., gcd_2
non-termination, division-by-zero crashes in big_quot_rem).
Three fixes applied:
- Guard mul_by_digit and printbase_mul_by_digit to return
integer.zero when the digit is 0 (root cause fix).
- Make is_zero/1 recognize denormalized all-zero digit lists
as defense in depth.
- Replace structural `= integer.zero` checks in rational.m
with integer.is_zero/1 calls for robustness.
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
library/dir.m:
Add some missing semicolons to some statements in foreign_proc body.
(The code in question compiled because we wrap foreign proc bodies
int { ;}.)
doc/mercury_reference_manual.texi:
Fix inverted characters.
Fix incorrect variable names in examples and the description
of the DCG transformation.
Fix an incorrect Java foreign_proc example.
Fix the example of higher-order impure code; it would not
compile as written.
library/calendar.m:
Reword the module header to mention microsecond resolution.
Reword the description of the leap year rule in the header.
Document the failure conditions of init_date/8, date_from_string/2
and duration_from_string/2. Mention that date_from_string/2 accepts
negative years, years with more than four digits, and fractional
seconds with between one and six digits.
Document that date_to_string/1 can produce negative years in its
output.
Document the microsecond and timezone limitations of
current_local_time/3 and current_utc_time/3.
Expand the documentation of julian_day_number/1 to explain what
a Julian day number is.
Expand the documentation of the duration type to cover sign constraints,
clamping behaviour, the treatment of leap seconds, internal normalisation
rules, and the relationship between init_duration/7 arguments and component
access function return values.
Document the ranges returned by the duration component access functions for
both positive and negative durations.
Document the sign constraint and exception behaviour of init_duration/7.
Expand the documentation of duration_from_string/2 to cover
negative durations, the six-digit fractional seconds limit,
and the normalisation round-trip example.
Rewrite the documentation of duration/2 to describe the
non-invertibility of duration/2 and add_duration/3 more concisely.
Simplify the documentation of foldl_days/5.
Various other minor documentation and formatting changes.
The crash that this diff fixes occurred when giving a command such as
"print Var^1" to mdb, where the first argument of Var is a direct arg.
runtime/mercury_ml_expand_body.h:
When deconstructing a term with a direct arg, return NULL
as the value of expand_info->chosen_arg_word_sized_ptr.
The crash occurred when we returned a non-null pointer,
which violated the expectations of trace/mercury_trace_vars.c
and its callers. (Not surprising, since those that function and
its callers were written long before the direct_arg optimization
was added to the system.)
runtime/mercury_deconstruct.h:
Document the rationale behind the above changes. (The contents of
mercury_ml_expand_body.h are #included in mercury_deconstruct.c.)
trace/mercury_trace_vars.c:
Add the debugging code I used to track down this issue, in disabled form.
Fix missing copyright year.
trace/mercury_trace_browse.c:
Delete obsolete comment.
Fix missing copyright years.
tests/debugger/direct_arg_test.{m,inp,exp}:
A test case for this bug.
tests/debugger/Mmakefile:
Enable the new test case.
compiler/hlds_out_type_table.m:
When dumping out the data constructors in the type table,
if a constructor has names for some of its fields,
put the name and the type of each field on different lines.
In the original test case for this bug, of which direct_arg_test.m
is an extreme simplification, pretty much every line overflows
without this.
Also, factor out some duplicated code, and replace bools with values
of a bespoke type.
Due to a copy-and-paste error introduced in commit 05fd615471,
set_line_number/2 was incorrectly setting the line number of the
current output stream, not the current input stream.
library/io.m:
Fix the above bug.
tests/hard_coded/Mmakefile:
tests/hard_coded/bad_set_line_number.{m,exp}:
Add a regression test.
