add internal/cpe

3 files changed, 264 insertions, 0 deletions

diff --git a/internal/cpe/cpe.go b/internal/cpe/cpe.go
new file mode 100644
index 0000000..73930e8
--- /dev/null
+++ b/internal/cpe/cpe.go
@@ -0,0 +1,94 @@
+// CPE 2.3 formatted string parser.
+//
+// Source: NISTIR 7605, figure 6-3:
+// https://nvlpubs.nist.gov/nistpubs/Legacy/IR/nistir7695.pdf
+package cpe
+
+import (
+  "fmt"
+)
+
+//go:generate stringer -linecomment -type=tokenType
+
+// token type
+type tokenType byte
+
+const (
+  anyToken tokenType = iota // any
+  naToken // na
+  valToken // val
+)
+
+// token
+type token struct {
+  Type tokenType // token type
+  Val string // token value
+}
+
+// parse buffer into token.
+func newToken(val []byte) token {
+  if len(val) > 0 {
+    switch val[0] {
+    case '*':
+      return token { Type: anyToken }
+    case '-':
+      return token { Type: naToken }
+    default:
+      return token { Type: valToken, Val: string(val) }
+    }
+  } else {
+    // empty value
+    return token { Type: valToken, Val: "" }
+  }
+}
+
+// Parse buffer into slice of tokens.
+func tokenize(buf []byte) ([]token, error) {
+  // build result
+  var r []token
+
+  // current token and escape state
+  var curr []byte
+  esc := false
+
+  // build result
+  for _, b := range(buf) {
+    if esc {
+      switch b {
+      // valid escaped characters
+      case '\\', '*', '-', '!', '"', '#', '$', '%', '&', '\'', '(', ')',
+           '+', ',', '/', ':', ';', '<', '=', '>', '@', '[', ']', '^', '`',
+           '{', '|', '}', '~':
+        curr = append(curr, b)
+        esc = false
+      default:
+        return r, fmt.Errorf("invalid escape byte: 0x%02x", b)
+      }
+    } else {
+      switch b {
+      case '\\':
+        esc = true
+      case ':':
+        // push token, clear buffer
+        r = append(r, newToken(curr))
+        curr = nil
+      case 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l',
+           'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x',
+           'y', 'z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9',
+           '-', '.', '_', '*', '?':
+        curr = append(curr, b)
+      default:
+        return r, fmt.Errorf("invalid byte: 0x%02x", b)
+      }
+    }
+  }
+
+  if len(curr) > 0 {
+    // push token, clear buffer
+    r = append(r, newToken(curr))
+    curr = nil
+  }
+
+  // return success
+  return r, nil
+}
diff --git a/internal/cpe/cpe_test.go b/internal/cpe/cpe_test.go
new file mode 100644
index 0000000..e842c75
--- /dev/null
+++ b/internal/cpe/cpe_test.go
@@ -0,0 +1,145 @@
+package cpe
+
+import (
+  "reflect"
+  "testing"
+)
+
+func TestTokenTypeString(t *testing.T) {
+  tests := []struct {
+    val tokenType
+    exp string
+  } {
+    { anyToken, "any" },
+    { naToken, "na" },
+    { valToken, "val" },
+    { tokenType(255), "tokenType(255)" },
+  }
+
+  for _, test := range(tests) {
+    t.Run(test.exp, func(t *testing.T) {
+      got := test.val.String()
+      if got != test.exp {
+        t.Errorf("got \"%s\", exp \"%s\"", got, test.exp)
+      }
+    })
+  }
+}
+
+func TestNewToken(t *testing.T) {
+  passTests := []struct {
+    name string
+    val string
+    exp token
+  } {
+    { "any", "*", token { Type: anyToken } },
+    { "na", "-", token { Type: naToken } },
+    { "empty", "", token { Type: valToken } },
+    { "foo", "foo", token { Type: valToken, Val: "foo" } },
+  }
+
+  for _, test := range(passTests) {
+    t.Run(test.name, func(t *testing.T) {
+      got := newToken([]byte(test.val))
+      if got.Type != test.exp.Type {
+        t.Errorf("token: got %s, exp %s", got.Type, test.exp.Type)
+      } else if got.Type == valToken && got.Val != test.exp.Val {
+        t.Errorf("value: got \"%s\", exp \"%s\"", got.Val, test.exp.Val)
+      }
+    })
+  }
+}
+
+func TestTokenize(t *testing.T) {
+  passTests := []struct {
+    val string
+    exp []token
+  } {{
+    val: "foo",
+    exp: []token { token { Type: valToken, Val: "foo" } },
+  }, {
+    val: "foo:bar",
+    exp: []token {
+      token { Type: valToken, Val: "foo" },
+      token { Type: valToken, Val: "bar" },
+    },
+  }, {
+    val: "*",
+    exp: []token { token { Type: anyToken } },
+  }, {
+    val: "-",
+    exp: []token { token { Type: naToken } },
+  }, {
+    val: "*:bar",
+    exp: []token {
+      token { Type: anyToken },
+      token { Type: valToken, Val: "bar" },
+    },
+  }, {
+    val: "foo:*",
+    exp: []token {
+      token { Type: valToken, Val: "foo" },
+      token { Type: anyToken },
+    },
+  }, {
+    val: "-:bar",
+    exp: []token {
+      token { Type: naToken },
+      token { Type: valToken, Val: "bar" },
+    },
+  }, {
+    val: "foo:-",
+    exp: []token {
+      token { Type: valToken, Val: "foo" },
+      token { Type: naToken },
+    },
+  }, {
+    val: "foo\\*:-",
+    exp: []token {
+      token { Type: valToken, Val: "foo*" },
+      token { Type: naToken },
+    },
+  }}
+
+  for _, test := range(passTests) {
+    t.Run(test.val, func(t *testing.T) {
+      // tokenize, check for error
+      got, err := tokenize([]byte(test.val))
+      if err != nil {
+        t.Error(err)
+        return
+      }
+
+      if !reflect.DeepEqual(got, test.exp) {
+        t.Errorf("token: got %v, exp %v", got, test.exp)
+        return
+      }
+    })
+  }
+
+  failTests := []struct {
+    id  string
+    val string
+    exp string
+  } {{
+    id:  "invalid escape",
+    val: "foo\\.",
+    exp: "invalid escape byte: 0x2e",
+  }, {
+    id:  "invalid byte",
+    val: "\n",
+    exp: "invalid byte: 0x0a",
+  }}
+
+  for _, test := range(failTests) {
+    t.Run(test.id, func(t *testing.T) {
+      // tokenize, check for error
+      got, err := tokenize([]byte(test.val))
+      if err == nil {
+        t.Errorf("got %v, exp error", got)
+      } else if err.Error() != test.exp {
+        t.Errorf("got \"%s\", exp \"%s\"", err.Error(), test.exp)
+      }
+    })
+  }
+}
diff --git a/internal/cpe/tokentype_string.go b/internal/cpe/tokentype_string.go
new file mode 100644
index 0000000..7b53758
--- /dev/null
+++ b/internal/cpe/tokentype_string.go
@@ -0,0 +1,25 @@
+// Code generated by "stringer -linecomment -type=tokenType"; DO NOT EDIT.
+
+package cpe
+
+import "strconv"
+
+func _() {
+	// An "invalid array index" compiler error signifies that the constant values have changed.
+	// Re-run the stringer command to generate them again.
+	var x [1]struct{}
+	_ = x[anyToken-0]
+	_ = x[naToken-1]
+	_ = x[valToken-2]
+}
+
+const _tokenType_name = "anynaval"
+
+var _tokenType_index = [...]uint8{0, 3, 5, 8}
+
+func (i tokenType) String() string {
+	if i >= tokenType(len(_tokenType_index)-1) {
+		return "tokenType(" + strconv.FormatInt(int64(i), 10) + ")"
+	}
+	return _tokenType_name[_tokenType_index[i]:_tokenType_index[i+1]]
+}