[libc-commits] [libc] [libc] Make hdrgen deduce header type lists from function signatures (PR #127251)
Roland McGrath via libc-commits
libc-commits at lists.llvm.org
Fri Feb 14 11:54:16 PST 2025
https://github.com/frobtech created https://github.com/llvm/llvm-project/pull/127251
With this, the `types` list in YAML files should only be used to
list the types that a standard specifies should be in that header
per se. All the types referenced in function signatures will be
collected automatically.
>From 5aae70262eb927a0e6f882ac66c944850f09d447 Mon Sep 17 00:00:00 2001
From: Roland McGrath <mcgrathr at google.com>
Date: Fri, 14 Feb 2025 11:46:23 -0800
Subject: [PATCH] [libc] Make hdrgen deduce header type lists from function
signatures
With this, the `types` list in YAML files should only be used to
list the types that a standard specifies should be in that header
per se. All the types referenced in function signatures will be
collected automatically.
---
libc/utils/hdrgen/function.py | 57 ++++++++++++++++---
libc/utils/hdrgen/header.py | 57 +++++++++++++++----
.../tests/expected_output/test_header.h | 2 +-
.../utils/hdrgen/tests/input/subdir/test.yaml | 3 -
.../utils/hdrgen/tests/input/test_small.h.def | 1 -
libc/utils/hdrgen/type.py | 7 +++
6 files changed, 103 insertions(+), 24 deletions(-)
diff --git a/libc/utils/hdrgen/function.py b/libc/utils/hdrgen/function.py
index 8ae47e574785f..25a7fa80e4501 100644
--- a/libc/utils/hdrgen/function.py
+++ b/libc/utils/hdrgen/function.py
@@ -6,25 +6,68 @@
#
# ==-------------------------------------------------------------------------==#
+import re
+from type import Type
+
+
+# These are the keywords that appear in C type syntax but are not part of the
+# include file name. This is all of the modifiers, qualifiers, and base types,
+# but not "struct".
+KEYWORDS = [
+ "_Atomic",
+ "_Complex",
+ "_Float16",
+ "_Noreturn",
+ "__restrict",
+ "accum",
+ "char",
+ "const",
+ "double",
+ "float",
+ "fract",
+ "int",
+ "long",
+ "short",
+ "signed",
+ "unsigned",
+ "void",
+ "volatile",
+]
+NONIDENTIFIER = re.compile("[^a-zA-Z0-9_]+")
+
class Function:
def __init__(
self, return_type, name, arguments, standards, guard=None, attributes=[]
):
+ assert return_type
self.return_type = return_type
self.name = name
self.arguments = [
arg if isinstance(arg, str) else arg["type"] for arg in arguments
]
+ assert all(self.arguments)
self.standards = standards
self.guard = guard
- self.attributes = attributes or ""
+ self.attributes = attributes or []
+
+ def signature_types(self):
+ def collapse(type_string):
+ assert type_string
+ # Split into words at nonidentifier characters (`*`, `[`, etc.),
+ # filter out keywords and numbers, and then rejoin with "_".
+ return "_".join(
+ word
+ for word in NONIDENTIFIER.split(type_string)
+ if word and not word.isdecimal() and word not in KEYWORDS
+ )
+
+ all_types = [self.return_type] + self.arguments
+ return {
+ Type(string) for string in filter(None, (collapse(t) for t in all_types))
+ }
def __str__(self):
- attributes_str = " ".join(self.attributes)
+ attrs_str = "".join(f"{attr} " for attr in self.attributes)
arguments_str = ", ".join(self.arguments) if self.arguments else "void"
- if attributes_str == "":
- result = f"{self.return_type} {self.name}({arguments_str})"
- else:
- result = f"{attributes_str} {self.return_type} {self.name}({arguments_str})"
- return result
+ return attrs_str + f"{self.return_type} {self.name}({arguments_str})"
diff --git a/libc/utils/hdrgen/header.py b/libc/utils/hdrgen/header.py
index f94cb8e628fa6..c817c02b83b75 100644
--- a/libc/utils/hdrgen/header.py
+++ b/libc/utils/hdrgen/header.py
@@ -6,9 +6,33 @@
#
# ==-------------------------------------------------------------------------==#
+from functools import reduce
from pathlib import PurePosixPath
+STDINT_SIZES = [
+ "16",
+ "32",
+ "64",
+ "8",
+ "least16",
+ "least32",
+ "least64",
+ "least8",
+ "max",
+ "ptr",
+]
+
+COMPILER_HEADER_TYPES = (
+ {
+ "bool": "<stdbool.h>",
+ "va_list": "<stdarg.h>",
+ }
+ | {f"int{size}_t": "<stdint.h>" for size in STDINT_SIZES}
+ | {f"uint{size}_t": "<stdint.h>" for size in STDINT_SIZES}
+)
+
+
class HeaderFile:
def __init__(self, name):
self.template_file = None
@@ -34,19 +58,25 @@ def add_object(self, object):
def add_function(self, function):
self.functions.append(function)
- def includes(self):
- return sorted(
- {
- PurePosixPath("llvm-libc-macros") / macro.header
- for macro in self.macros
- if macro.header is not None
- }
- | {
- PurePosixPath("llvm-libc-types") / f"{typ.type_name}.h"
- for typ in self.types
- }
+ def all_types(self):
+ return reduce(
+ lambda a, b: a | b,
+ [f.signature_types() for f in self.functions],
+ set(self.types),
)
+ def includes(self):
+ return {
+ PurePosixPath("llvm-libc-macros") / macro.header
+ for macro in self.macros
+ if macro.header is not None
+ } | {
+ COMPILER_HEADER_TYPES.get(
+ typ.type_name, PurePosixPath("llvm-libc-types") / f"{typ.type_name}.h"
+ )
+ for typ in self.all_types()
+ }
+
def public_api(self):
# Python 3.12 has .relative_to(dir, walk_up=True) for this.
path_prefix = PurePosixPath("../" * (len(PurePosixPath(self.name).parents) - 1))
@@ -56,7 +86,10 @@ def relpath(file):
content = [
f"#include {file}"
- for file in sorted(f'"{relpath(file)!s}"' for file in self.includes())
+ for file in sorted(
+ file if isinstance(file, str) else f'"{relpath(file)!s}"'
+ for file in self.includes()
+ )
]
for macro in self.macros:
diff --git a/libc/utils/hdrgen/tests/expected_output/test_header.h b/libc/utils/hdrgen/tests/expected_output/test_header.h
index 838b6d7728150..748c09808c128 100644
--- a/libc/utils/hdrgen/tests/expected_output/test_header.h
+++ b/libc/utils/hdrgen/tests/expected_output/test_header.h
@@ -11,10 +11,10 @@
#include "__llvm-libc-common.h"
#include "llvm-libc-macros/float16-macros.h"
-#include "llvm-libc-types/float128.h"
#include "llvm-libc-macros/test_more-macros.h"
#include "llvm-libc-macros/test_small-macros.h"
+#include "llvm-libc-types/float128.h"
#include "llvm-libc-types/type_a.h"
#include "llvm-libc-types/type_b.h"
diff --git a/libc/utils/hdrgen/tests/input/subdir/test.yaml b/libc/utils/hdrgen/tests/input/subdir/test.yaml
index a414b6f387ccb..e68af00849b0e 100644
--- a/libc/utils/hdrgen/tests/input/subdir/test.yaml
+++ b/libc/utils/hdrgen/tests/input/subdir/test.yaml
@@ -1,8 +1,5 @@
header: subdir/test.h
header_template: test.h.def
-types:
- - type_name: type_a
- - type_name: type_b
functions:
- name: func
return_type: type_a
diff --git a/libc/utils/hdrgen/tests/input/test_small.h.def b/libc/utils/hdrgen/tests/input/test_small.h.def
index 587b163b68d96..1f484a8a2bdb3 100644
--- a/libc/utils/hdrgen/tests/input/test_small.h.def
+++ b/libc/utils/hdrgen/tests/input/test_small.h.def
@@ -11,7 +11,6 @@
#include "__llvm-libc-common.h"
#include "llvm-libc-macros/float16-macros.h"
-#include "llvm-libc-types/float128.h"
%%public_api()
diff --git a/libc/utils/hdrgen/type.py b/libc/utils/hdrgen/type.py
index b90480267d8d5..0dbd8a5837d15 100644
--- a/libc/utils/hdrgen/type.py
+++ b/libc/utils/hdrgen/type.py
@@ -9,4 +9,11 @@
class Type:
def __init__(self, type_name):
+ assert type_name
self.type_name = type_name
+
+ def __eq__(self, other):
+ return self.type_name == other.type_name
+
+ def __hash__(self):
+ return self.type_name.__hash__()
More information about the libc-commits
mailing list