thoughtsync/android/tools/check-symbols.py

#!/usr/bin/env python3
"""Flag capitalised identifiers that are neither imported nor declared locally.

This exists because ktlint and detekt are both structurally blind to it: they
parse Kotlin without resolving symbols, so a *missing import* is invisible to
them and both pass a file that cannot compile. The first sync-screen push failed
in CI on exactly that (`Unresolved reference 'Build'` — `android.os.Build` was
lost in a file split), after a clean local analyzer run.

Not a type checker and not trying to be. `compileDebugKotlin` in CI is the real
one; this is a cheap pre-push filter for the single mistake that survives every
other local gate. It errs toward false positives — anything it cannot account
for is reported rather than assumed fine.

It also checks MEMBERS of this package's own `object` declarations — `Foo.bar()`
where `Foo` is an object declared here. That case was added after moving a
function between two objects and forgetting to paste it into the second: the
call site read `Other.thing()`, resolved fine as far as the leading token, and
failed in CI (785ebdb).

Still NOT caught, so a clean run is not over-read: members of anything declared
outside this package, members reached through a variable rather than a type
name, and every question about types. Those are what `compileDebugKotlin` is for.

    python3 android/tools/check-symbols.py [source-root]

Exits non-zero when something is unaccounted for.
"""

import collections
import os
import re
import sys

DEFAULT_ROOT = os.path.join(
    os.path.dirname(os.path.abspath(__file__)), "..", "app", "src", "main", "java"
)

# Available without an import: kotlin.* and kotlin.collections.*, plus
# java.lang.* which Kotlin/JVM also imports by default.
IMPLICIT = set(
    """
String Int Long Short Byte Boolean Char Float Double Unit Any Nothing Number
UInt ULong UShort UByte Array List Set Map MutableList MutableSet MutableMap
Collection Iterable Iterator Sequence Pair Triple Comparable Comparator
Throwable Exception RuntimeException IllegalArgumentException IllegalStateException
Error Result Regex StringBuilder CharSequence Enum Annotation Function
Deprecated Suppress OptIn JvmStatic JvmField JvmName JvmOverloads Volatile
Synchronized Throws Target Retention Repeatable MustBeDocumented
System Math Object Class Thread Runnable Void Integer Character
StringBuffer
""".split()
)

# `R` is generated at build time and never imported from the app's own package.
GENERATED = {"R"}

DECL = re.compile(
    r"^\s*(?:@\w+\s+)*(?:public |private |internal |protected )?"
    r"(?:expect |actual |external |abstract |final |open |sealed |data |value |"
    r"inline |enum |annotation |fun |companion |const |lateinit )*"
    r"(?:class|interface|object|typealias|fun|val|var)\s+"
    r"(?:<[^>]*>\s*)?([A-Za-z_]\w*)",
    re.M,
)


def strip(src: str) -> str:
    """Blank out comments and string literals.

    Order matters: raw strings before block comments, and line comments must NOT
    use DOTALL — `//.*` with re.S eats from the first comment to end of file,
    which silently empties the input and makes the whole check pass vacuously.
    """
    src = re.sub(r'"""(?:.|\n)*?"""', '""', src)
    src = re.sub(r"/\*(?:.|\n)*?\*/", " ", src)
    src = re.sub(r"//[^\n]*", " ", src)
    src = re.sub(r'"(?:\\.|[^"\\\n])*"', '""', src)
    return src


def object_members(src: str) -> dict:
    """Map each `object Foo` declared here to the names declared directly in it.

    Brace-counted rather than regex-matched: an object body contains nested
    braces (lambdas, apply blocks, companions) and no regex closes correctly over
    them. Only top-level members count — anything nested deeper is not reachable
    as `Foo.member` anyway.
    """
    members = {}
    for match in re.finditer(r"^(?:internal |private )?object (\w+)\s*\{", src, re.M):
        name = match.group(1)
        depth = 0
        body_start = match.end() - 1
        for i in range(body_start, len(src)):
            if src[i] == "{":
                depth += 1
            elif src[i] == "}":
                depth -= 1
                if depth == 0:
                    break
        body = src[body_start + 1 : i]
        own = set()
        depth = 0
        for line in body.splitlines():
            if depth == 0:
                # Nested TYPES count as members too: `Foo.Bar` where Bar is a
                # data class inside object Foo is an ordinary reference, and
                # leaving them out made the checker report four false positives
                # the first time an object held one.
                decl = re.match(
                    r"\s*(?:@\w+\s+)*(?:public |private |internal |protected )?"
                    r"(?:const |lateinit |inline |suspend |data |sealed |enum |value |abstract |open )*"
                    r"(?:fun|val|var|class|object|interface)\s+"
                    r"(?:<[^>]*>\s*)?(\w+)",
                    line,
                )
                if decl:
                    own.add(decl.group(1))
            depth += line.count("{") - line.count("}")
        members[name] = own
    return members


def main() -> int:
    root = sys.argv[1] if len(sys.argv) > 1 else DEFAULT_ROOT
    files = [
        os.path.join(d, n)
        for d, _, names in os.walk(root)
        for n in names
        if n.endswith(".kt")
    ]

    declared = collections.defaultdict(set)
    objects = {}
    parsed = {}
    for path in files:
        with open(path, encoding="utf-8") as fh:
            raw = fh.read()
        package = re.search(r"^package\s+([\w.]+)", raw, re.M).group(1)
        src = strip(raw)
        parsed[path] = (package, src, raw)
        objects.update(object_members(src))
        for match in DECL.finditer(src):
            declared[package].add(match.group(1))
        # Enum entries are declarations too; DECL only sees the class itself.
        for match in re.finditer(r"enum class \w+[^{]*\{([^};]*)", src):
            for entry in match.group(1).split(","):
                name = entry.strip().split("(")[0].strip()
                if re.fullmatch(r"[A-Z]\w*", name):
                    declared[package].add(name)

    problems = 0
    for path in sorted(files):
        package, src, raw = parsed[path]
        imported = set()
        for match in re.finditer(r"^import\s+([\w.]+)(?:\s+as\s+(\w+))?", raw, re.M):
            imported.add(match.group(2) or match.group(1).split(".")[-1])
        # Type parameters are declared inline at their use site.
        type_params = set()
        for match in re.finditer(r"(?:fun|class|interface)\s*<([^>]*)>", src):
            type_params |= set(
                re.findall(r"\b([A-Z]\w*)\b(?=\s*(?::|,|$))", match.group(1))
            )
        known = imported | declared[package] | IMPLICIT | GENERATED | type_params

        # Capitalised tokens NOT preceded by a dot: `Icons.Filled` resolves
        # through `Icons`, so only the leading segment needs to be accounted for.
        for match in re.finditer(r"(?<![\w.])@?([A-Z][A-Za-z0-9_]*)\b", src):
            name = match.group(1)
            if name in known:
                continue
            line = src[: match.start()].count("\n") + 1
            print(f"{os.path.relpath(path, root)}:{line}: unresolved '{name}'")
            problems += 1

        # Members of objects declared in this package.
        for match in re.finditer(r"(?<![\w.])([A-Z][A-Za-z0-9_]*)\.(\w+)", src):
            owner, member = match.group(1), match.group(2)
            if owner not in objects or member in objects[owner]:
                continue
            line = src[: match.start()].count("\n") + 1
            print(
                f"{os.path.relpath(path, root)}:{line}: "
                f"'{owner}' has no member '{member}'"
            )
            problems += 1

    print(f"\n{len(files)} files, {problems} unresolved")
    return 1 if problems else 0


if __name__ == "__main__":
    sys.exit(main())