anubis/lib/policy/config/expressionorlist.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

103 lines
2.3 KiB
Go

package config
import (
"encoding/json"
"errors"
"slices"
)
var (
ErrExpressionOrListMustBeStringOrObject = errors.New("config: this must be a string or an object")
ErrExpressionEmpty = errors.New("config: this expression is empty")
ErrExpressionCantHaveBoth = errors.New("config: expression block can't contain multiple expression types")
)
type ExpressionOrList struct {
Expression string `json:"-" yaml:"-"`
All []string `json:"all,omitempty" yaml:"all,omitempty"`
Any []string `json:"any,omitempty" yaml:"any,omitempty"`
}
func (eol ExpressionOrList) Equal(rhs *ExpressionOrList) bool {
if eol.Expression != rhs.Expression {
return false
}
if !slices.Equal(eol.All, rhs.All) {
return false
}
if !slices.Equal(eol.Any, rhs.Any) {
return false
}
return true
}
func (eol *ExpressionOrList) MarshalYAML() (any, error) {
switch {
case len(eol.All) == 1 && len(eol.Any) == 0:
eol.Expression = eol.All[0]
eol.All = nil
case len(eol.Any) == 1 && len(eol.All) == 0:
eol.Expression = eol.Any[0]
eol.Any = nil
}
if eol.Expression != "" {
return eol.Expression, nil
}
type RawExpressionOrList ExpressionOrList
return RawExpressionOrList(*eol), nil
}
func (eol *ExpressionOrList) MarshalJSON() ([]byte, error) {
switch {
case len(eol.All) == 1 && len(eol.Any) == 0:
eol.Expression = eol.All[0]
eol.All = nil
case len(eol.Any) == 1 && len(eol.All) == 0:
eol.Expression = eol.Any[0]
eol.Any = nil
}
if eol.Expression != "" {
return json.Marshal(string(eol.Expression))
}
type RawExpressionOrList ExpressionOrList
val := RawExpressionOrList(*eol)
return json.Marshal(val)
}
func (eol *ExpressionOrList) UnmarshalJSON(data []byte) error {
switch string(data[0]) {
case `"`: // string
return json.Unmarshal(data, &eol.Expression)
case "{": // object
type RawExpressionOrList ExpressionOrList
var val RawExpressionOrList
if err := json.Unmarshal(data, &val); err != nil {
return err
}
eol.All = val.All
eol.Any = val.Any
return nil
}
return ErrExpressionOrListMustBeStringOrObject
}
func (eol *ExpressionOrList) Valid() error {
if eol.Expression == "" && len(eol.All) == 0 && len(eol.Any) == 0 {
return ErrExpressionEmpty
}
if len(eol.All) != 0 && len(eol.Any) != 0 {
return ErrExpressionCantHaveBoth
}
return nil
}