fix: exclude non-root README.md/LICENSE files (#173)

BREAKING CHANGE: The files array can now be used to exclude non-root readme, license, licence, and copying files.

Co-authored-by: rahulio96 <[email protected]>

Author: AaronHamilton965

README.md

@@ -30,9 +30,11 @@ arborist.loadActual().then((tree) => {
 This uses the following rules:
 
 1. If a `package.json` file is found, and it has a `files` list,
-   then ignore everything that isn't in `files`.  Always include the
+   then ignore everything that isn't in `files`.  Always include the root
    readme, license, licence and copying files, if they exist, as well
-   as the package.json file itself.
+   as the package.json file itself. Non-root readme, license, licence and
+   copying files are included by default, but can be excluded using the 
+   `files` list e.g. `"!readme"`.
 2. If there's no `package.json` file (or it has no `files` list), and
    there is a `.npmignore` file, then ignore all the files in the
    `.npmignore` file.

lib/index.js

@@ -38,13 +38,22 @@ const defaults = [
 ]
 
 const strictDefaults = [
-  // these are forcibly included at all levels
+  // these are forcibly excluded
+  '/.git',
+]
+
+const allLevels = [
+  // these are included by default but can be excluded by package.json files array
   '!/readme{,.*[^~$]}',
   '!/copying{,.*[^~$]}',
   '!/license{,.*[^~$]}',
   '!/licence{,.*[^~$]}',
-  // these are forcibly excluded
-  '/.git',
+]
+
+const rootOnly = [
+  /^!.*readme/i,
+  /^!.*copying/i,
+  /^!.*licen[sc]e/i,
 ]
 
 const normalizePath = (path) => path.split('\\').join('/')
@@ -132,6 +141,7 @@ class PackWalker extends IgnoreWalker {
       // known required files for this directory
       this.injectRules(strictRules, [
         ...strictDefaults,
+        ...allLevels,
         ...this.requiredFiles.map((file) => `!${file}`),
       ])
     }
@@ -284,6 +294,7 @@ class PackWalker extends IgnoreWalker {
     const ignores = []
     const strict = [
       ...strictDefaults,
+      ...allLevels,
       '!/package.json',
       '/.git',
       '/node_modules',
@@ -304,6 +315,9 @@ class PackWalker extends IgnoreWalker {
           file = file.slice(0, -2)
         }
         const inverse = `!${file}`
+
+        this.excludeNonRoot(file)
+
         try {
           // if an entry in the files array is a specific file, then we need to include it as a
           // strict requirement for this package. if it's a directory or a pattern, it's a default
@@ -352,6 +366,20 @@ class PackWalker extends IgnoreWalker {
     this.injectRules(strictRules, strict, callback)
   }
 
+  // excludes non root files by checking if elements from the files array in
+  // package.json contain an ! and readme/license/licence/copying, and then
+  // removing readme/license/licence/copying accordingly from strict defaults
+  excludeNonRoot (file) {
+    // Find the pattern
+    const matchingPattern = rootOnly.find(regex => regex.test(file))
+
+    if (matchingPattern) {
+      // Find which index matches the pattern and remove it from allLevels
+      const indexToRemove = allLevels.findIndex(element => matchingPattern.test(element))
+      allLevels.splice(indexToRemove, 1)
+    }
+  }
+
   // custom method: after we've finished gathering the files for the root package, we call this
   // before emitting the 'done' event in order to gather all of the files for bundled deps
   async gatherBundles () {

test/package-json-negate-non-root.js

@@ -0,0 +1,69 @@
+// exclude readme, license, and licnce files if package.json
+// files array includes !readme, !license, or !licence
+'use strict'
+
+const Arborist = require('@npmcli/arborist')
+const t = require('tap')
+const packlist = require('../')
+
+const pkg = t.testdir({
+  'package.json': JSON.stringify({
+    files: [
+      '**/*.js',
+      '!readme.md',
+      '!licence',
+      '!license',
+      '!copying',
+    ],
+
+  }),
+  'readme.md': 'one',
+  licence: 'two',
+  license: 'tre',
+  copying: 'for',
+  lib: {
+    'readme.md': 'one',
+    licence: 'two',
+    license: 'tre',
+    copying: 'for',
+    a: {
+      'readme.md': 'one',
+      licence: 'two',
+      license: 'tre',
+      copying: 'for',
+      b: {
+        'readme.md': 'one',
+        licence: 'two',
+        license: 'tre',
+        copying: 'for',
+        c: {
+          'readme.md': 'one',
+          licence: 'two',
+          license: 'tre',
+          copying: 'for',
+          'file.txt': 'one',
+          'c.js': 'two',
+        },
+        'file.txt': 'one',
+        'b.js': 'two',
+      },
+      'file.txt': 'one',
+      'a.js': 'two',
+    },
+  } })
+
+t.test('package with negated readme, licence and license files', async (t) => {
+  const arborist = new Arborist({ path: pkg })
+  const tree = await arborist.loadActual()
+  const files = await packlist(tree)
+  t.same(files, [
+    'copying',
+    'licence',
+    'license',
+    'lib/a/a.js',
+    'lib/a/b/b.js',
+    'lib/a/b/c/c.js',
+    'package.json',
+    'readme.md',
+  ])
+})