Added regexp/no-obscure-range rule (#122)

* Added `regexp/no-obscure-range` rule

* Added docs

* Added `regexp/no-obscure-range` rule

* Added docs

* Updated recommended

* Updated docs

* Support setting

* Added test

* Renamed setting

* Fixed false negative

* Updated docs

Author: RunDevelopment

README.md

@@ -97,6 +97,7 @@ The rules with the following star :star: are included in the `plugin:regexp/reco
 | [regexp/no-invisible-character](https://ota-meshi.github.io/eslint-plugin-regexp/rules/no-invisible-character.html) | disallow invisible raw character | :star::wrench: |
 | [regexp/no-lazy-ends](https://ota-meshi.github.io/eslint-plugin-regexp/rules/no-lazy-ends.html) | disallow lazy quantifiers at the end of an expression |  |
 | [regexp/no-legacy-features](https://ota-meshi.github.io/eslint-plugin-regexp/rules/no-legacy-features.html) | disallow legacy RegExp features |  |
+| [regexp/no-obscure-range](https://ota-meshi.github.io/eslint-plugin-regexp/rules/no-obscure-range.html) | disallow obscure character ranges |  |
 | [regexp/no-octal](https://ota-meshi.github.io/eslint-plugin-regexp/rules/no-octal.html) | disallow octal escape sequence | :star: |
 | [regexp/no-potentially-useless-backreference](https://ota-meshi.github.io/eslint-plugin-regexp/rules/no-potentially-useless-backreference.html) | disallow backreferences that reference a group that might not be matched |  |
 | [regexp/no-trivially-nested-assertion](https://ota-meshi.github.io/eslint-plugin-regexp/rules/no-trivially-nested-assertion.html) | disallow trivially nested assertions | :wrench: |

docs/.vuepress/config.js

@@ -44,6 +44,7 @@ module.exports = {
             { text: "Introduction", link: "/" },
             { text: "User Guide", link: "/user-guide/" },
             { text: "Rules", link: "/rules/" },
+            { text: "Settings", link: "/settings/" },
             { text: "Playground", link: "/playground/" },
         ],
 
@@ -93,7 +94,7 @@ module.exports = {
                       ]
                     : []),
             ],
-            "/": ["/", "/user-guide/", "/rules/", "/playground/"],
+            "/": ["/", "/user-guide/", "/rules/", "/settings/", "/playground/"],
         },
     },
 }

docs/README.md

@@ -30,6 +30,10 @@ See [User Guide](./user-guide/README.md).
 
 See [Available Rules](./rules/README.md).
 
+## :gear: Settings
+
+See [Settings](./settings/README.md).
+
 ## :lock: License
 
 See the [LICENSE](LICENSE) file for license rights and limitations (MIT).

docs/rules/README.md

@@ -25,6 +25,7 @@ The rules with the following star :star: are included in the `plugin:regexp/reco
 | [regexp/no-invisible-character](./no-invisible-character.md) | disallow invisible raw character | :star::wrench: |
 | [regexp/no-lazy-ends](./no-lazy-ends.md) | disallow lazy quantifiers at the end of an expression |  |
 | [regexp/no-legacy-features](./no-legacy-features.md) | disallow legacy RegExp features |  |
+| [regexp/no-obscure-range](./no-obscure-range.md) | disallow obscure character ranges |  |
 | [regexp/no-octal](./no-octal.md) | disallow octal escape sequence | :star: |
 | [regexp/no-potentially-useless-backreference](./no-potentially-useless-backreference.md) | disallow backreferences that reference a group that might not be matched |  |
 | [regexp/no-trivially-nested-assertion](./no-trivially-nested-assertion.md) | disallow trivially nested assertions | :wrench: |

docs/rules/no-obscure-range.md

@@ -0,0 +1,126 @@
+---
+pageClass: "rule-details"
+sidebarDepth: 0
+title: "regexp/no-obscure-range"
+description: "disallow obscure character ranges"
+---
+# regexp/no-obscure-range
+
+> disallow obscure character ranges
+
+- :exclamation: <badge text="This rule has not been released yet." vertical="middle" type="error"> ***This rule has not been released yet.*** </badge>
+
+## :book: Rule Details
+
+The character range operator (the `-` inside character classes) can easily be misused (mostly unintentionally) to construct non-obvious character class. This rule will disallow all non-obvious uses of the character range operator.
+
+<eslint-code-block>
+
+```js
+/* eslint regexp/no-obscure-range: "error" */
+
+/* ✓ GOOD */
+var foo = /[a-z]/;
+var foo = /[J-O]/;
+var foo = /[1-9]/;
+var foo = /[\x00-\x40]/;
+var foo = /[\0-\uFFFF]/;
+var foo = /[\0-\u{10FFFF}]/u;
+var foo = /[\1-\5]/;
+var foo = /[\cA-\cZ]/;
+
+/* ✗ BAD */
+var foo = /[A-\x43]/;
+var foo = /[\41-\x45]/;
+var foo = /[!-$]/;
+var foo = /[😀-😄]/u;
+```
+
+</eslint-code-block>
+
+## :wrench: Options
+
+
+```json5
+{
+  "regexp/no-obscure-range": ["error",
+    {
+      "allowed": "alphanumeric" // or "all" or [...]
+    }
+  ]
+}
+```
+
+This option can be used to override the [allowedCharacterRanges] setting.
+
+It allows all values that the [allowedCharacterRanges] setting allows.
+
+[allowedCharacterRanges]: ../settings/README.md#allowedCharacterRanges
+
+### `"allowed": "alphanumeric"`
+
+<eslint-code-block fix>
+
+```js
+/* eslint regexp/no-obscure-range: ["error", { "allowed": "alphanumeric" }] */
+
+/* ✓ GOOD */
+var foo = /[a-z]/;
+var foo = /[J-O]/;
+var foo = /[1-9]/;
+
+/* ✗ BAD */
+var foo = /[A-\x43]/;
+var foo = /[\41-\x45]/;
+var foo = /[!-$]/;
+var foo = /[😀-😄]/u;
+```
+
+</eslint-code-block>
+
+### `"allowed": "all"`
+
+<eslint-code-block fix>
+
+```js
+/* eslint regexp/no-obscure-range: ["error", { "allowed": "all" }] */
+
+/* ✓ GOOD */
+var foo = /[a-z]/;
+var foo = /[J-O]/;
+var foo = /[1-9]/;
+var foo = /[!-$]/;
+var foo = /[😀-😄]/u;
+
+/* ✗ BAD */
+var foo = /[A-\x43]/;
+var foo = /[\41-\x45]/;
+```
+
+</eslint-code-block>
+
+### `"allowed": [ "alphanumeric", "😀-😏" ]`
+
+<eslint-code-block fix>
+
+```js
+/* eslint regexp/no-obscure-range: ["error", { "allowed": [ "alphanumeric", "😀-😏" ] }] */
+
+/* ✓ GOOD */
+var foo = /[a-z]/;
+var foo = /[J-O]/;
+var foo = /[1-9]/;
+var foo = /[😀-😄]/u;
+
+/* ✗ BAD */
+var foo = /[A-\x43]/;
+var foo = /[\41-\x45]/;
+var foo = /[!-$]/;
+```
+
+</eslint-code-block>
+
+## :mag: Implementation
+
+- [Rule source](https://github.com/ota-meshi/eslint-plugin-regexp/blob/master/lib/rules/no-obscure-range.ts)
+- [Test source](https://github.com/ota-meshi/eslint-plugin-regexp/blob/master/tests/lib/rules/no-obscure-range.ts)

docs/rules/prefer-range.md

@@ -43,14 +43,13 @@ var foo = /[a-cd-f]/
 }
 ```
 
-- `target` ... Specify the range of characters you want to check with this rule.
-  - `"alphanumeric"` ... Check only alphanumeric characters (`0-9`,`a-z` and `A-Z`). This is the default.
-  - `"all"` ... Check all characters. Use `"all"`, if you want to focus on regular expression optimization.
-  - `[...]` (Array) ... Specify as an array of character ranges. List the character ranges that your team is familiar with in this option, and replace redundant contiguous characters with ranges.
-    Specify the range as a three-character string in which the from and to characters are connected with a hyphen (`-`) using. e.g. `"!-/"` (U+0021 - U+002F), `"😀-😏"` (U+1F600 - U+1F60F)  
-    You can also use `"alphanumeric"`.
-
-### `"target": "alphanumeric"` (Default)
+This option can be used to override the [allowedCharacterRanges] setting.
+
+It allows all values that the [allowedCharacterRanges] setting allows.
+
+[allowedCharacterRanges]: ../settings/README.md#allowedCharacterRanges
+
+### `"target": "alphanumeric"`
 
 <eslint-code-block fix>
 

docs/settings/README.md

@@ -0,0 +1,75 @@
+# Settings
+
+[Shared settings](https://eslint.org/docs/user-guide/configuring/configuration-files#adding-shared-settings) are a way to configure multiple rules at once.
+
+## :book: Usage
+
+All settings for this plugin use the `regexp` namespace.
+
+Example **.eslintrc.js**:
+
+```js
+module.exports = {
+  ..., // rules, plugins, etc.
+
+  settings: {
+    // all settings for this plugin have to be in the `regexp` namespace
+    regexp: {
+      // define settings here, such as:
+      // allowedCharacterRanges: 'all'
+    }
+  }
+}
+```
+
+## :gear: Available settings
+
+### `allowedCharacterRanges`
+
+Defines a set of allowed character ranges. Rules will only allow, create, and fix character ranges defined here.
+
+#### Values
+
+The following values are allowed:
+
+- `"alphanumeric"`
+
+  This will allow only alphanumeric ranges (`0-9`, `A-Z`, and `a-z`). Only ASCII character are included.
+
+- `"all"`
+
+  This will allow only all ranges (roughly equivalent to `"\x00-\uFFFF"`).
+
+- `"<min>-<max>"`
+
+  A custom range that allows all character from `<min>` to `<max>`. Both `<min>` and `<max>` have to be single Unicode code points.
+
+  E.g. `"A-Z"` (U+0041 - U+005A), `"а-я"` (U+0430 - U+044F), `"😀-😏"` (U+1F600 - U+1F60F).
+
+- A non-empty array of the string values mentioned above. All ranges of the array items will be allowed.
+
+#### Default
+
+If the setting isn't defined, its value defaults to `"alphanumeric"`.
+
+#### Example
+
+```js
+module.exports = {
+  ..., // rules, plugins, etc.
+  settings: {
+    regexp: {
+      // allow alphanumeric and cyrillic ranges
+      allowedCharacterRanges: ['alphanumeric', 'а-я', 'А-Я']
+    }
+  }
+}
+```
+
+#### Affected rules
+
+- [regexp/no-obscure-range]
+- [regexp/prefer-range]
+
+[regexp/no-obscure-range]: ../rules/no-obscure-range.md
+[regexp/prefer-range]: ../rules/prefer-range.md

docs/user-guide/README.md

@@ -37,7 +37,9 @@ module.exports = {
 
 This plugin provides one config:
 
-- `plugin:regexp/recommended` ... This is the recommended configuration for this plugin.  
+- `plugin:regexp/recommended` ... This is the recommended configuration for this plugin.
   See [lib/configs/recommended.ts](https://github.com/ota-meshi/eslint-plugin-regexp/blob/master/lib/configs/recommended.ts) for details.
 
 See [the rule list](../rules/README.md) to get the `rules` that this plugin provides.
+
+Some rules also support [shared settings](../settings/README.md).

lib/rules/no-obscure-range.ts

@@ -0,0 +1,106 @@
+import type { Expression } from "estree"
+import {
+    getAllowedCharRanges,
+    inRange,
+    getAllowedCharValueSchema,
+} from "../utils/char-ranges"
+import type { RegExpVisitor } from "regexpp/visitor"
+import {
+    createRule,
+    defineRegexpVisitor,
+    getRegexpLocation,
+    isControlEscape,
+    isEscapeSequence,
+    isHexadecimalEscape,
+    isOctalEscape,
+} from "../utils"
+
+export default createRule("no-obscure-range", {
+    meta: {
+        docs: {
+            description: "disallow obscure character ranges",
+            // TODO Switch to recommended in the major version.
+            // recommended: true,
+            recommended: false,
+        },
+        schema: [
+            {
+                type: "object",
+                properties: {
+                    allowed: getAllowedCharValueSchema(),
+                },
+                additionalProperties: false,
+            },
+        ],
+        messages: {
+            unexpected:
+                "Unexpected obscure character range. The characters of '{{range}}' ({{unicode}}) are not obvious.",
+        },
+        type: "suggestion", // "problem",
+    },
+    create(context) {
+        const allowedRanges = getAllowedCharRanges(
+            context.options[0]?.allowed,
+            context,
+        )
+        const sourceCode = context.getSourceCode()
+
+        /**
+         * Create visitor
+         * @param node
+         */
+        function createVisitor(node: Expression): RegExpVisitor.Handlers {
+            return {
+                onCharacterClassRangeEnter(rNode) {
+                    const { min, max } = rNode
+
+                    if (min.value === max.value) {
+                        // we don't deal with that
+                        return
+                    }
+
+                    if (isControlEscape(min.raw) && isControlEscape(max.raw)) {
+                        // both min and max are control escapes
+                        return
+                    }
+                    if (isOctalEscape(min.raw) && isOctalEscape(max.raw)) {
+                        // both min and max are either octal
+                        return
+                    }
+                    if (
+                        (isHexadecimalEscape(min.raw) || min.value === 0) &&
+                        isHexadecimalEscape(max.raw)
+                    ) {
+                        // both min and max are hexadecimal (with a small exception for \0)
+                        return
+                    }
+
+                    if (
+                        !isEscapeSequence(min.raw) &&
+                        !isEscapeSequence(max.raw) &&
+                        inRange(allowedRanges, min.value, max.value)
+                    ) {
+                        return
+                    }
+
+                    const uMin = `U+${min.value.toString(16).padStart(4, "0")}`
+                    const uMax = `U+${max.value.toString(16).padStart(4, "0")}`
+
+                    context.report({
+                        node,
+                        loc: getRegexpLocation(sourceCode, node, rNode),
+                        messageId: "unexpected",
+                        data: {
+                            range: rNode.raw,
+                            unicode: `${uMin} - ${uMax}`,
+                        },
+                    })
+                },
+            }
+        }
+
+        return defineRegexpVisitor(context, {
+            createVisitor,
+        })
+    },
+})