feat(version): port rpm-style version comparison from ZETA

2026-08-08 17:22:06 -04:00
parent 348f0e3bac
commit c545879115
3 changed files with 421 additions and 0 deletions
@@ -0,0 +1,35 @@
+     Warning 
+     Warning ## Warning for package tofu ##
+     Warning 
+     Warning The following compiler flags have been specified in the package description
+     Warning file. They are handled by DUB and direct use in packages is discouraged.
+     Warning Alternatively, you can set the DFLAGS environment variable to pass custom flags
+     Warning to the compiler, or use one of the suggestions below:
+     Warning 
+     Warning warningsAsErrors: Use "buildRequirements" to control the warning level
+     Warning 
+             Generating test runner configuration 'tofu-test-application' for 'application' (executable).
+     Warning Excluding main source file src/main.d from test.
+    Starting Performing "unittest" build using /usr/bin/dmd for x86_64.
+  Up-to-date toml 1.0.0: target for configuration [library] is up to date.
+    Building tofu ~main: building configuration [tofu-test-application]
+     Linking tofu-test-application
+    Finished To force a rebuild of up-to-date targets, run again with --force
+     Running tofu-test-application 
+Warning: malformed TOML config at /tmp/tofu-test-config-bad-178345.toml: Invalid table key declaration (2:0)
+Warning: invalid TOFU_DEFAULT_JOBS 'not-a-number', using default 1
+4 modules passed unittests
+     Warning 
+     Warning ## Warning for package tofu ##
+     Warning 
+     Warning The following compiler flags have been specified in the package description
+     Warning file. They are handled by DUB and direct use in packages is discouraged.
+     Warning Alternatively, you can set the DFLAGS environment variable to pass custom flags
+     Warning to the compiler, or use one of the suggestions below:
+     Warning 
+     Warning warningsAsErrors: Use "buildRequirements" to control the warning level
+     Warning 
+    Starting Performing "debug" build using /usr/bin/dmd for x86_64.
+  Up-to-date toml 1.0.0: target for configuration [library] is up to date.
+  Up-to-date tofu ~main: target for configuration [application] is up to date.
+    Finished To force a rebuild of up-to-date targets, run again with --force
@@ -6,6 +6,46 @@ _Auto-scaffolded by /start-work. Append new entries below - never overwrite._

 ---

+## Task 5 — tofu.vercmp (RPM-style version comparison)
+
+### Module naming
+- `version` is a D reserved keyword (conditional compilation). Module is named
+  `tofu.vercmp`, file `src/tofu/vercmp.d`.  The `DepConstraint` struct in
+  `types.d` uses field `ver` (not `version`) for the same reason.
+
+### Algorithm (exact port from ZETA/lib/vercmp.lua)
+- `nextSegment`: skips any non-digit/non-letter chars (separators); then reads
+  a contiguous run of same-type chars.  Returns `(segment, nextIndex)`.
+- `cmpNumeric`: strips leading zeros from both segments (`"007"` → `"7"`,
+  `"000"` → `"0"`).  Longer stripped string wins; then lexicographic.
+- `compare`: strips all whitespace, then loops over alternating segments.
+  If both segments are digit-typed → `cmpNumeric`, else lexical.  When one
+  version exhausts segments: the exhausted one is *older* (shorter = older).
+  This matches RPM semantics: `compare("1.0.0", "1.0") > 0`.
+- `satisfies`: uses `final switch` on `DepOp` enum.  Maps to `compare` result
+  as expected (ge→c>=0, le→c<=0, eq→c==0, ne→c!=0, gt→c>0, lt→c<0).
+- `parseDep`: delegates to `DepConstraint.parse` (already ported in types.d by
+  task 4).  Both implementations are kept consistent.
+
+### whitespace stripping
+- Could not use `filter`+`array` because Phobos auto-decodes `string` to
+  `dchar` range elements, returning `dchar[]` not `string`.  Implemented a
+  manual `removeWhitespace` helper using raw code-unit indexing, with a
+  `@trusted` cast since the freshly-allocated `char[]` has no aliasing.
+
+### Test porting
+- All 10 ZETA test categories ported as separate `@safe unittest` blocks.
+- 4 modules pass unittests (types, vercmp, plus pre-existing ones).
+- `dub build` passes with warnings-as-errors enabled.
+
+### Future considerations
+- When task 12 (constraint resolution) imports `tofu.vercmp`, the `compare`
+  and `satisfies` functions are ready.  The `parseDep` integration with
+  `DepConstraint.parse` ensures consistency between ZETA and tofu's dep
+  constraint semantics.
+
+---
+
 ## Task 2: Config Loading (src/tofu/config.d)

 ### toml package API (v1.0.0, Kripth/toml)
@@ -0,0 +1,346 @@
+/// tofu.vercmp — RPM-style version comparison and dependency constraint checking.
+///
+/// Direct port of ZETA's `vercmp.lua` (references/ZETA/lib/vercmp.lua).
+/// Splits version strings into alternating digit and letter segments;
+/// digit segments compare numerically (leading zeros ignored),
+/// letter segments compare lexically.
+/// Shorter versions (fewer segments) are considered older.
+///
+/// Note: named `vercmp` rather than `version` because `version` is a
+/// reserved keyword in D.
+module tofu.vercmp;
+
+import tofu.types;
+
+// ─────────────────────────────────────────────────────────────────────────────
+//  Helpers
+// ─────────────────────────────────────────────────────────────────────────────
+
+@safe @nogc pure nothrow
+bool isDigit(char c) { return c >= '0' && c <= '9'; }
+
+@safe @nogc pure nothrow
+bool isLetter(char c) {
+    return (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z');
+}
+
+// ─────────────────────────────────────────────────────────────────────────────
+//  Segment extraction
+// ─────────────────────────────────────────────────────────────────────────────
+
+/// Reads the next digit or letter segment from `s` starting at index `i`,
+/// skipping separator characters (anything not digit/letter).
+/// Sets `seg` and `ni` (next index). Returns `true` if a segment was found.
+@safe pure
+bool nextSegment(scope const string s, scope size_t i,
+                 out string seg, out size_t ni) {
+    while (i < s.length) {
+        if (isDigit(s[i]) || isLetter(s[i]))
+            break;
+        i++;
+    }
+    if (i >= s.length) {
+        seg = null;
+        ni = i;
+        return false;
+    }
+    bool digit = isDigit(s[i]);
+    size_t j = i + 1;
+    while (j < s.length) {
+        if (digit && !isDigit(s[j]))
+            break;
+        if (!digit && !isLetter(s[j]))
+            break;
+        j++;
+    }
+    seg = s[i .. j];
+    ni = j;
+    return true;
+}
+
+// ─────────────────────────────────────────────────────────────────────────────
+//  Numeric segment comparison
+// ─────────────────────────────────────────────────────────────────────────────
+
+/// Compares two numeric segments.
+/// Leading zeros are stripped before comparison.
+/// Longer stripped string wins; equal-length strings compare lexicographically.
+@safe pure
+int cmpNumeric(scope const string a, scope const string b) {
+    // Strip leading zeros (Lua: gsub("^0+", ""), fallback to "0")
+    size_t ai = 0;
+    while (ai < a.length && a[ai] == '0')
+        ai++;
+    string x = ai < a.length ? a[ai .. $] : "0";
+
+    size_t bi = 0;
+    while (bi < b.length && b[bi] == '0')
+        bi++;
+    string y = bi < b.length ? b[bi .. $] : "0";
+
+    if (x.length > y.length) return 1;
+    if (x.length < y.length) return -1;
+    if (x > y) return 1;
+    if (x < y) return -1;
+    return 0;
+}
+
+// ─────────────────────────────────────────────────────────────────────────────
+//  Whitespace helper
+// ─────────────────────────────────────────────────────────────────────────────
+
+/// Remove all whitespace from a string.  Iterates raw code units so
+/// it only strips ASCII whitespace (space, tab, newline, carriage return).
+/// This matches Lua's `gsub("%s+", "")` for version strings.
+@safe
+string removeWhitespace(string s) {
+    size_t newLen = 0;
+    for (size_t i = 0; i < s.length; i++) {
+        char c = s[i];
+        if (c != ' ' && c != '\t' && c != '\n' && c != '\r')
+            newLen++;
+    }
+    if (newLen == s.length)
+        return s; // no whitespace — return original
+    char[] buf = new char[newLen];
+    size_t pos = 0;
+    for (size_t i = 0; i < s.length; i++) {
+        char c = s[i];
+        if (c != ' ' && c != '\t' && c != '\n' && c != '\r')
+            buf[pos++] = c;
+    }
+    // Safe: buf is freshly allocated, no other references
+    return (() @trusted => cast(string) buf)();
+}
+
+// ─────────────────────────────────────────────────────────────────────────────
+//  Version comparison
+// ─────────────────────────────────────────────────────────────────────────────
+
+/// Compares two version strings using RPM-style semantics.
+///
+/// Returns -1 if a < b, 0 if a == b, 1 if a > b.
+///
+/// Whitespace is stripped from both inputs.  Versions are split into
+/// alternating digit and letter segments (separators like `.`, `-`, `_`,
+/// `~` are skipped).  Digit segments compare numerically (leading zeros
+/// ignored), letter segments compare lexically.  When one version runs
+/// out of segments, the shorter version is considered older.
+@safe
+int compare(string a, string b) {
+    a = removeWhitespace(a);
+    b = removeWhitespace(b);
+
+    size_t ia = 0;
+    size_t ib = 0;
+
+    while (true) {
+        string sa;
+        string sb;
+        size_t na;
+        size_t nb;
+        bool hasA = nextSegment(a, ia, sa, na);
+        bool hasB = nextSegment(b, ib, sb, nb);
+
+        if (!hasA || !hasB) {
+            if (!hasA && !hasB) return 0; // both exhausted → equal
+            if (!hasA) return -1;         // a exhausted first → a older
+            return 1;                     // b exhausted first → a newer
+        }
+
+        ia = na;
+        ib = nb;
+
+        int c;
+        if (isDigit(sa[0]) && isDigit(sb[0]))
+            c = cmpNumeric(sa, sb);
+        else {
+            if (sa < sb)      c = -1;
+            else if (sa > sb) c = 1;
+            else              c = 0;
+        }
+
+        if (c != 0) return c;
+    }
+}
+
+// ─────────────────────────────────────────────────────────────────────────────
+//  Dependency parsing
+// ─────────────────────────────────────────────────────────────────────────────
+
+/// Parses a dependency specification string into a `DepConstraint`.
+///
+/// Examples:
+///   "libffi"       → DepConstraint("libffi", DepOp.none, "")
+///   "pcre2>=10.42" → DepConstraint("pcre2", DepOp.ge, "10.42")
+///   "x=1.0"        → DepConstraint("x", DepOp.eq, "1.0")   ("=" normalized)
+///
+/// Throws `TypesException` or `Exception` on malformed input.
+///
+/// Delegates to `DepConstraint.parse` which provides the canonical
+/// implementation (already ported from ZETA in types.d by task 4).
+@safe
+DepConstraint parseDep(scope const string spec) {
+    return DepConstraint.parse(spec);
+}
+
+// ─────────────────────────────────────────────────────────────────────────────
+//  Constraint satisfaction
+// ─────────────────────────────────────────────────────────────────────────────
+
+/// Checks whether `installed` version satisfies the given `constraint`.
+///
+/// When `constraint.op` is `DepOp.none` (unconstrained), always returns
+/// `true`.  Otherwise compares `installed` against `constraint.ver` using
+/// the operator semantics:
+///
+///   DepOp.ge → compare >= 0    DepOp.le → compare <= 0
+///   DepOp.eq → compare == 0    DepOp.ne → compare != 0
+///   DepOp.gt → compare > 0     DepOp.lt → compare < 0
+@safe
+bool satisfies(scope const string installed, scope const DepConstraint constraint) {
+    int c;
+    final switch (constraint.op) {
+        case DepOp.none:
+            return true;
+        case DepOp.ge:
+            c = compare(installed, constraint.ver);
+            return c >= 0;
+        case DepOp.le:
+            c = compare(installed, constraint.ver);
+            return c <= 0;
+        case DepOp.eq:
+            c = compare(installed, constraint.ver);
+            return c == 0;
+        case DepOp.ne:
+            c = compare(installed, constraint.ver);
+            return c != 0;
+        case DepOp.gt:
+            c = compare(installed, constraint.ver);
+            return c > 0;
+        case DepOp.lt:
+            c = compare(installed, constraint.ver);
+            return c < 0;
+    }
+}
+
+// ─────────────────────────────────────────────────────────────────────────────
+//  Unittests — ported from references/ZETA/tests/suites/vercmp.lua
+// ─────────────────────────────────────────────────────────────────────────────
+
+@safe unittest {
+    // Equal versions
+    assert(compare("1.0", "1.0") == 0);
+    assert(compare("1.0.0", "1.0.0") == 0);
+    assert(compare("2.3.4", "2.3.4") == 0);
+}
+
+@safe unittest {
+    // Numeric ordering
+    assert(compare("1.10", "1.9") > 0);
+    assert(compare("2.0", "1.99") > 0);
+    assert(compare("1.0", "0.99") > 0);
+    assert(compare("1.1", "1.10") < 0);
+}
+
+@safe unittest {
+    // Extra segments win (rpm-style): longer is newer
+    assert(compare("1.0.0", "1.0") > 0);
+    assert(compare("1.0rc1", "1.0") > 0);
+    assert(compare("1.0a", "1.0") > 0);
+}
+
+@safe unittest {
+    // Letter segments compare lexically
+    assert(compare("1.0beta", "1.0alpha") > 0);
+    assert(compare("1.0alpha", "1.0beta") < 0);
+}
+
+@safe unittest {
+    // Whitespace is ignored
+    assert(compare(" 1.0 ", "1.0") == 0);
+}
+
+@safe unittest {
+    // Leading zeros are ignored in digit segments
+    assert(compare("01.05", "1.5") == 0);
+    assert(compare("1.000", "1.0") >= 0);
+}
+
+@safe unittest {
+    // Empty and single-segment strings
+    assert(compare("", "") == 0);
+    assert(compare("", "1") == -1);
+    assert(compare("1", "") == 1);
+}
+
+@safe unittest {
+    // parseDep: plain name (unconstrained)
+    auto d = parseDep("libffi");
+    assert(d.name == "libffi");
+    assert(d.op == DepOp.none);
+    assert(d.ver == "");
+}
+
+@safe unittest {
+    // parseDep: constrained with operator
+    auto d1 = parseDep("pcre2>=10.42");
+    assert(d1.name == "pcre2");
+    assert(d1.op == DepOp.ge);
+    assert(d1.ver == "10.42");
+
+    // Whitespace around operator
+    auto d2 = parseDep("pcre2 >= 10.42");
+    assert(d2.name == "pcre2");
+    assert(d2.op == DepOp.ge);
+    assert(d2.ver == "10.42");
+
+    auto d3 = parseDep("glib<=2.8");
+    assert(d3.name == "glib");
+    assert(d3.op == DepOp.le);
+    assert(d3.ver == "2.8");
+}
+
+@safe unittest {
+    // Single equals becomes == (Perl-style)
+    auto d = parseDep("x=1.0");
+    assert(d.op == DepOp.eq);
+}
+
+@safe unittest {
+    // parseDep rejects garbage
+    bool threw;
+
+    threw = false;
+    try { parseDep("!!!"); } catch (Exception) { threw = true; }
+    assert(threw, "expected parseDep(\"!!!\") to throw");
+
+    threw = false;
+    try { parseDep("foo bar baz"); } catch (Exception) { threw = true; }
+    assert(threw, "expected parseDep(\"foo bar baz\") to throw");
+
+    threw = false;
+    try { parseDep(""); } catch (Exception) { threw = true; }
+    assert(threw, "expected parseDep(\"\") to throw");
+}
+
+@safe unittest {
+    // satisfies: all operators matched against installed version
+    assert(satisfies("10.42", DepConstraint("", DepOp.ge, "10.42")));
+    assert(satisfies("10.43", DepConstraint("", DepOp.ge, "10.42")));
+    assert(!satisfies("10.2", DepConstraint("", DepOp.ge, "10.42")));
+
+    assert(satisfies("10.41", DepConstraint("", DepOp.le, "10.42")));
+
+    assert(satisfies("10.42", DepConstraint("", DepOp.eq, "10.42")));
+    assert(!satisfies("10.43", DepConstraint("", DepOp.eq, "10.42")));
+
+    assert(satisfies("10.43", DepConstraint("", DepOp.ne, "10.42")));
+
+    assert(satisfies("10.43", DepConstraint("", DepOp.gt, "10.42")));
+
+    assert(satisfies("10.41", DepConstraint("", DepOp.lt, "10.42")));
+
+    // Unconstrained (DepOp.none) — always satisfied
+    assert(satisfies("anything", DepConstraint("", DepOp.none, "")));
+}