Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Rewrite simple glob match + FuzzMatch #20

Merged
merged 9 commits into from
Sep 18, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
79 changes: 79 additions & 0 deletions capability/policy/glob.go
Original file line number Diff line number Diff line change
@@ -0,0 +1,79 @@
package policy

import "fmt"

type glob string

// parseGlob ensures that the pattern conforms to the spec: only '*' and escaped '\*' are allowed.
func parseGlob(pattern string) (glob, error) {
for i := 0; i < len(pattern); i++ {
if pattern[i] == '*' {
continue
}
if pattern[i] == '\\' && i+1 < len(pattern) && pattern[i+1] == '*' {
i++ // skip the escaped '*'
continue
}
if pattern[i] == '\\' && i+1 < len(pattern) {
i++ // skip the escaped character
continue
}
if pattern[i] == '\\' {
return "", fmt.Errorf("invalid escape sequence")
}
}

return glob(pattern), nil
}

func mustParseGlob(pattern string) glob {
g, err := parseGlob(pattern)
if err != nil {
panic(err)
}
return g
}

// Match matches a string against the glob pattern with * wildcards, handling escaped '\*' literals.
func (pattern glob) Match(str string) bool {
// i is the index for the pattern
// j is the index for the string
var i, j int

// starIdx keeps track of the position of the last * in the pattern.
// matchIdx keeps track of the position in the string where the last * matched.
var starIdx, matchIdx int = -1, -1

for j < len(str) {
if i < len(pattern) && (pattern[i] == str[j] || pattern[i] == '\\' && i+1 < len(pattern) && pattern[i+1] == str[j]) {
// characters match or if there's an escaped character that matches
if pattern[i] == '\\' {
// skip the escape character
i++
}
i++
j++
} else if i < len(pattern) && pattern[i] == '*' {
// there's a * wildcard in the pattern
starIdx = i
matchIdx = j
i++
} else if starIdx != -1 {
// there's a previous * wildcard, backtrack
i = starIdx + 1
matchIdx++
j = matchIdx
} else {
// no match found
return false
}
}

// check for remaining characters in the pattern
for i < len(pattern) && pattern[i] == '*' {
i++
}

// the entire pattern is processed, it's a match
return i == len(pattern)
}
73 changes: 73 additions & 0 deletions capability/policy/glob_test.go
Original file line number Diff line number Diff line change
@@ -0,0 +1,73 @@
package policy

import (
"testing"

"github.com/stretchr/testify/require"
)

func TestSimpleGlobMatch(t *testing.T) {
tests := []struct {
pattern string
str string
matches bool
}{
// Basic matching
{"*", "anything", true},
{"a*", "abc", true},
{"*c", "abc", true},
{"a*c", "abc", true},
{"a*c", "abxc", true},
{"a*c", "ac", true},
{"a*c", "a", false},
{"a*c", "ab", false},

// Escaped characters
{"a\\*c", "a*c", true},
{"a\\*c", "abc", false},

// Mixed wildcards and literals
{"a*b*c", "abc", true},
{"a*b*c", "aXbYc", true},
{"a*b*c", "aXbY", false},
{"a*b*c", "abYc", true},
{"a*b*c", "aXbc", true},
{"a*b*c", "aXbYcZ", false},

// Edge cases
{"", "", true},
{"", "a", false},
{"*", "", true},
{"*", "a", true},
{"\\*", "*", true},
{"\\*", "a", false},

// Specified test cases
{"Alice\\*, Bob*, Carol.", "Alice*, Bob, Carol.", true},
{"Alice\\*, Bob*, Carol.", "Alice*, Bob, Dan, Erin, Carol.", true},
{"Alice\\*, Bob*, Carol.", "Alice*, Bob , Carol.", true},
{"Alice\\*, Bob*, Carol.", "Alice*, Bob*, Carol.", true},
{"Alice\\*, Bob*, Carol.", "Alice*, Bob, Carol", false},
{"Alice\\*, Bob*, Carol.", "Alice*, Bob*, Carol!", false},
{"Alice\\*, Bob*, Carol.", "Alice, Bob, Carol.", false},
{"Alice\\*, Bob*, Carol.", "Alice Cooper, Bob, Carol.", false},
{"Alice\\*, Bob*, Carol.", " Alice*, Bob, Carol. ", false},
}

for _, tt := range tests {
t.Run(tt.pattern+"_"+tt.str, func(t *testing.T) {
g, err := parseGlob(tt.pattern)
require.NoError(t, err)
require.Equal(t, tt.matches, g.Match(tt.str))
})
}
}

func BenchmarkGlob(b *testing.B) {
b.ReportAllocs()

for i := 0; i < b.N; i++ {
g := mustParseGlob("Alice\\*, Bob*, Carol.")
g.Match("Alice*, Bob*, Carol!")
}
}
2 changes: 1 addition & 1 deletion capability/policy/ipld.go
Original file line number Diff line number Diff line change
Expand Up @@ -232,7 +232,7 @@ func statementToIPLD(statement Statement) (datamodel.Node, error) {
if err != nil {
return nil, err
}
err = listBuilder.AssembleValue().AssignString(statement.pattern)
err = listBuilder.AssembleValue().AssignString(string(statement.pattern))
if err != nil {
return nil, err
}
Expand Down
2 changes: 1 addition & 1 deletion capability/policy/match.go
Original file line number Diff line number Diff line change
Expand Up @@ -112,7 +112,7 @@ func matchStatement(statement Statement, node ipld.Node) bool {
if err != nil {
return false
}
return s.glob.Match(v)
return s.pattern.Match(v)
}
case KindAll:
if s, ok := statement.(quantifier); ok {
Expand Down
49 changes: 49 additions & 0 deletions capability/policy/match_test.go
Original file line number Diff line number Diff line change
Expand Up @@ -490,3 +490,52 @@ func TestPolicyExamples(t *testing.T) {
require.True(t, evaluate(`["any", ".a", ["==", ".b", 2]]`, data))
})
}

func FuzzMatch(f *testing.F) {
// Policy + Data examples
f.Add([]byte(`[["==", ".status", "draft"]]`), []byte(`{"status": "draft"}`))
f.Add([]byte(`[["all", ".reviewer", ["like", ".email", "*@example.com"]]]`), []byte(`{"reviewer": [{"email": "alice@example.com"}, {"email": "bob@example.com"}]}`))
f.Add([]byte(`[["any", ".tags", ["or", [["==", ".", "news"], ["==", ".", "press"]]]]]`), []byte(`{"tags": ["news", "press"]}`))
f.Add([]byte(`[["==", ".name", "Alice"]]`), []byte(`{"name": "Alice"}`))
f.Add([]byte(`[[">", ".age", 30]]`), []byte(`{"age": 31}`))
f.Add([]byte(`[["<=", ".height", 180]]`), []byte(`{"height": 170}`))
f.Add([]byte(`[["not", ["==", ".status", "inactive"]]]`), []byte(`{"status": "active"}`))
f.Add([]byte(`[["and", [["==", ".role", "admin"], [">=", ".experience", 5]]]]`), []byte(`{"role": "admin", "experience": 6}`))
f.Add([]byte(`[["or", [["==", ".department", "HR"], ["==", ".department", "Finance"]]]]`), []byte(`{"department": "HR"}`))
f.Add([]byte(`[["like", ".email", "*@company.com"]]`), []byte(`{"email": "user@company.com"}`))
f.Add([]byte(`[["all", ".projects", [">", ".budget", 10000]]]`), []byte(`{"projects": [{"budget": 15000}, {"budget": 8000}]}`))
f.Add([]byte(`[["any", ".skills", ["==", ".", "Go"]]]`), []byte(`{"skills": ["Go", "Python", "JavaScript"]}`))
f.Add(
[]byte(`[["and", [
["==", ".name", "Bob"],
["or", [[">", ".age", 25],["==", ".status", "active"]]],
["all", ".tasks", ["==", ".completed", true]]
]]]`),
[]byte(`{
"name": "Bob",
"age": 26,
"status": "active",
"tasks": [{"completed": true}, {"completed": true}, {"completed": false}]
}`),
)

f.Fuzz(func(t *testing.T, policyBytes []byte, dataBytes []byte) {
policyNode, err := ipld.Decode(policyBytes, dagjson.Decode)
if err != nil {
t.Skip()
}

dataNode, err := ipld.Decode(dataBytes, dagjson.Decode)
if err != nil {
t.Skip()
}

// policy node -> policy object
policy, err := FromIPLD(policyNode)
if err != nil {
t.Skip()
}

Match(policy, dataNode)
})
}
9 changes: 4 additions & 5 deletions capability/policy/policy.go
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,6 @@ package policy
// https://github.com/ucan-wg/delegation/blob/4094d5878b58f5d35055a3b93fccda0b8329ebae/README.md#policy

import (
"github.com/gobwas/glob"
"github.com/ipld/go-ipld-prime"

"github.com/ucan-wg/go-ucan/capability/policy/selector"
Expand Down Expand Up @@ -90,20 +89,20 @@ func Or(stmts ...Statement) Statement {

type wildcard struct {
selector selector.Selector
pattern string
glob glob.Glob // not serialized
pattern glob
}

func (n wildcard) Kind() string {
return KindLike
}

func Like(selector selector.Selector, pattern string) (Statement, error) {
g, err := glob.Compile(pattern)
g, err := parseGlob(pattern)
if err != nil {
return nil, err
}
return wildcard{selector: selector, pattern: pattern, glob: g}, nil

return wildcard{selector: selector, pattern: g}, nil
}

type quantifier struct {
Expand Down
1 change: 0 additions & 1 deletion go.mod
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,6 @@ go 1.21
toolchain go1.22.1

require (
github.com/gobwas/glob v0.2.3
github.com/ipfs/go-cid v0.4.1
github.com/ipld/go-ipld-prime v0.21.0
github.com/libp2p/go-libp2p v0.36.2
Expand Down
2 changes: 0 additions & 2 deletions go.sum
Original file line number Diff line number Diff line change
Expand Up @@ -5,8 +5,6 @@ github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSs
github.com/frankban/quicktest v1.14.6 h1:7Xjx+VpznH+oBnejlPUj8oUpdxnVs4f8XU8WnHkI4W8=
github.com/frankban/quicktest v1.14.6/go.mod h1:4ptaffx2x8+WTWXmUCuVU6aPUX1/Mz7zb5vbUoiM6w0=
github.com/go-yaml/yaml v2.1.0+incompatible/go.mod h1:w2MrLa16VYP0jy6N7M5kHaCkaLENm+P+Tv+MfurjSw0=
github.com/gobwas/glob v0.2.3 h1:A4xDbljILXROh+kObIiy5kIaPYD8e96x1tgBhUI5J+Y=
github.com/gobwas/glob v0.2.3/go.mod h1:d3Ez4x06l9bZtSvzIay5+Yzi0fmZzPgnTbPcKjJAkT8=
github.com/google/go-cmp v0.5.9 h1:O2Tfq5qg4qc4AmwVlvv0oLiVAGB7enBSJ2x2DqQFi38=
github.com/google/go-cmp v0.5.9/go.mod h1:17dUlkBOakJ0+DkrSSNjCkIjxS6bF9zb3elmeNGIjoY=
github.com/gopherjs/gopherjs v0.0.0-20181017120253-0766667cb4d1 h1:EGx4pi6eqNxGaHF6qqu48+N2wcFQ5qg5FXgOdqsJ5d8=
Expand Down
Loading