anubis/lib/policy/config/expressionorlist_test.go
Jason Cameron e0781e4560
feat: add robots2policy CLI to convert robots.txt to Anubis CEL (#657)
* feat: add robots2policy CLI utility to convert robots.txt to Anubis challenge policies

* feat: add documentation for robots2policy CLI tool

* feat: implement crawl delay handling as weight adjustment in Anubis rules

* feat: add various robots.txt and YAML configurations for user agent handling and crawl delays

* test: add comprehensive tests for robots2policy conversion and parsing

* fix: update example URL in usage instructions for robots2policy CLI

* Update metadata

check-spelling run (pull_request) for json/robots2policycli

Signed-off-by: check-spelling-bot <check-spelling-bot@users.noreply.github.com>
on-behalf-of: @check-spelling <check-spelling-bot@check-spelling.dev>

* docs: add crawl delay weight adjustment and deny user agents option to robots2policy CLI

* Update cmd/robots2policy/main.go

Co-authored-by: Xe Iaso <me@xeiaso.net>
Signed-off-by: Jason Cameron <jasoncameron.all@gmail.com>

* Update cmd/robots2policy/main.go

Co-authored-by: Xe Iaso <me@xeiaso.net>
Signed-off-by: Jason Cameron <jasoncameron.all@gmail.com>

* fix(robots2policy): use sigs.k8s.io/yaml

Signed-off-by: Xe Iaso <me@xeiaso.net>

* feat(config): properly marshal bot policy rules

Signed-off-by: Xe Iaso <me@xeiaso.net>

* chore(yeetfile): expose robots2policy in libexec

Signed-off-by: Xe Iaso <me@xeiaso.net>

* fix(yeetfile): put robots2policy in $PATH

Signed-off-by: Xe Iaso <me@xeiaso.net>

* Update metadata

check-spelling run (pull_request) for json/robots2policycli

Signed-off-by: check-spelling-bot <check-spelling-bot@users.noreply.github.com>
on-behalf-of: @check-spelling <check-spelling-bot@check-spelling.dev>

* style: reorder imports

* refactor: use preexisting structs in config

* fix: correct flag check in main function

* fix: reorder fields in AnubisRule struct for better alignment

* style: improve alignment of struct fields in AnubisRule and OGTagCache

* Update metadata

check-spelling run (pull_request) for json/robots2policycli

Signed-off-by: check-spelling-bot <check-spelling-bot@users.noreply.github.com>
on-behalf-of: @check-spelling <check-spelling-bot@check-spelling.dev>

* fix: add validation for generated Anubis rules from robots.txt

* feat: add batch processing for robots.txt files to generate Anubis CEL policies

* fix: improve usage message and error handling for input file requirement

* refactor: update AnubisRule structure to use ExpressionOrList for improved expression handling

* refactor: reorganize policy definitions in YAML files for consistency and clarity

* fix: correct indentation in blacklist and complex YAML files for consistency

* test: enhance output comparison in robots2policy tests for YAML and JSON formats

* Revert "fix: improve usage message and error handling for input file requirement"

This reverts commit ddcde1f2a326545d3ef2ec32e5e03f55f4f931a8.

* fix: improve usage message and error handling in robots2policy

Signed-off-by: Jason Cameron <git@jasoncameron.dev>

---------

Signed-off-by: check-spelling-bot <check-spelling-bot@users.noreply.github.com>
Signed-off-by: Jason Cameron <jasoncameron.all@gmail.com>
Signed-off-by: Xe Iaso <me@xeiaso.net>
Signed-off-by: Jason Cameron <git@jasoncameron.dev>
Co-authored-by: Xe Iaso <me@xeiaso.net>
2025-06-14 23:41:00 -04:00

216 lines
4.1 KiB
Go

package config
import (
"bytes"
"encoding/json"
"errors"
"testing"
yaml "sigs.k8s.io/yaml/goyaml.v3"
)
func TestExpressionOrListMarshalJSON(t *testing.T) {
for _, tt := range []struct {
name string
input *ExpressionOrList
output []byte
err error
}{
{
name: "single expression",
input: &ExpressionOrList{
Expression: "true",
},
output: []byte(`"true"`),
err: nil,
},
{
name: "all",
input: &ExpressionOrList{
All: []string{"true", "true"},
},
output: []byte(`{"all":["true","true"]}`),
err: nil,
},
{
name: "all one",
input: &ExpressionOrList{
All: []string{"true"},
},
output: []byte(`"true"`),
err: nil,
},
{
name: "any",
input: &ExpressionOrList{
Any: []string{"true", "false"},
},
output: []byte(`{"any":["true","false"]}`),
err: nil,
},
{
name: "any one",
input: &ExpressionOrList{
Any: []string{"true"},
},
output: []byte(`"true"`),
err: nil,
},
} {
t.Run(tt.name, func(t *testing.T) {
result, err := json.Marshal(tt.input)
if !errors.Is(err, tt.err) {
t.Errorf("wanted marshal error: %v but got: %v", tt.err, err)
}
if !bytes.Equal(result, tt.output) {
t.Logf("wanted: %s", string(tt.output))
t.Logf("got: %s", string(result))
t.Error("mismatched output")
}
})
}
}
func TestExpressionOrListMarshalYAML(t *testing.T) {
for _, tt := range []struct {
name string
input *ExpressionOrList
output []byte
err error
}{
{
name: "single expression",
input: &ExpressionOrList{
Expression: "true",
},
output: []byte(`"true"`),
err: nil,
},
{
name: "all",
input: &ExpressionOrList{
All: []string{"true", "true"},
},
output: []byte(`all:
- "true"
- "true"`),
err: nil,
},
{
name: "all one",
input: &ExpressionOrList{
All: []string{"true"},
},
output: []byte(`"true"`),
err: nil,
},
{
name: "any",
input: &ExpressionOrList{
Any: []string{"true", "false"},
},
output: []byte(`any:
- "true"
- "false"`),
err: nil,
},
{
name: "any one",
input: &ExpressionOrList{
Any: []string{"true"},
},
output: []byte(`"true"`),
err: nil,
},
} {
t.Run(tt.name, func(t *testing.T) {
result, err := yaml.Marshal(tt.input)
if !errors.Is(err, tt.err) {
t.Errorf("wanted marshal error: %v but got: %v", tt.err, err)
}
result = bytes.TrimSpace(result)
if !bytes.Equal(result, tt.output) {
t.Logf("wanted: %q", string(tt.output))
t.Logf("got: %q", string(result))
t.Error("mismatched output")
}
})
}
}
func TestExpressionOrListUnmarshalJSON(t *testing.T) {
for _, tt := range []struct {
err error
validErr error
result *ExpressionOrList
name string
inp string
}{
{
name: "simple",
inp: `"\"User-Agent\" in headers"`,
result: &ExpressionOrList{
Expression: `"User-Agent" in headers`,
},
},
{
name: "object-and",
inp: `{
"all": ["\"User-Agent\" in headers"]
}`,
result: &ExpressionOrList{
All: []string{
`"User-Agent" in headers`,
},
},
},
{
name: "object-or",
inp: `{
"any": ["\"User-Agent\" in headers"]
}`,
result: &ExpressionOrList{
Any: []string{
`"User-Agent" in headers`,
},
},
},
{
name: "both-or-and",
inp: `{
"all": ["\"User-Agent\" in headers"],
"any": ["\"User-Agent\" in headers"]
}`,
validErr: ErrExpressionCantHaveBoth,
},
{
name: "expression-empty",
inp: `{
"any": []
}`,
validErr: ErrExpressionEmpty,
},
} {
t.Run(tt.name, func(t *testing.T) {
var eol ExpressionOrList
if err := json.Unmarshal([]byte(tt.inp), &eol); !errors.Is(err, tt.err) {
t.Errorf("wanted unmarshal error: %v but got: %v", tt.err, err)
}
if tt.result != nil && !eol.Equal(tt.result) {
t.Logf("wanted: %#v", tt.result)
t.Logf("got: %#v", &eol)
t.Fatal("parsed expression is not what was expected")
}
if err := eol.Valid(); !errors.Is(err, tt.validErr) {
t.Errorf("wanted validation error: %v but got: %v", tt.err, err)
}
})
}
}