definiteIymaybe · December 2, 2022 03:43 · Jan 28, 2019 · Jan 28, 2019 · Jan 28, 2019 · Jan 28, 2019
diff --git a/compare-files.js b/compare-files.js
@@ -68,7 +68,7 @@ export function diff (files, hashes, base = '/') {
 
   // Check sizes
   for (const file of files) {
-    if (statSync(resolve(base, file)) !== hashes[file][0]) {
+    if (statSync(resolve(base, file)).size !== hashes[file][0]) {
       return { type: 'changed', file }
     }
   }

diff --git a/compare-files.js b/compare-files.js
@@ -37,7 +37,7 @@ export function profile (files, base = '/') {
   const map = {}
   for (const [ file, hash ] of hashFileList(files, base)) {
     map[file] = [
-      statsync(resolve(base, file)).size,
+      statSync(resolve(base, file)).size,
       hash
     ]
   }

diff --git a/readme.md b/readme.md
@@ -44,7 +44,7 @@ diff(
 )
 ```
 
-The `diff()` returns either `null` if no changes have occurred, or an object with a `type` (which is either `added`, `removed` or `changed`) and a `file` property pointing to the according file.
+The `diff()` returns either `null` if no changes have occurred, or an object with a `type` (which is either `added`, `removed` or `changed`) and a `file` property pointing to the (first) according file.
 
 For example:
 ```js

diff --git a/readme.md b/readme.md
@@ -1,6 +1,6 @@
 # Compare File Lists in Node.js
 
-I encouter a certain situation pretty frequently: I have a known set of files on my file system which I need to scan to compare them against the same (or a similar) set of files at a later point in time.
+I encouter a certain situation pretty frequently: I have a known set of files on my file system which I need to scan to compare them against the same (or a similar) set of files at a later point in time (i.e. "Did any of the files in this list change?" or "Did any of the files matching these criteria change?").
 
 A quick example:
 

diff --git a/readme.md b/readme.md
@@ -1,6 +1,6 @@
 # Compare File Lists in Node.js
 
-I encouter a certain situation pretty frequently: I have a known set of files on my file system which I need to scan to compare them against another (or possibly the same) set of files at a later point in time.
+I encouter a certain situation pretty frequently: I have a known set of files on my file system which I need to scan to compare them against the same (or a similar) set of files at a later point in time.
 
 A quick example:
 

diff --git a/compare-files.js b/compare-files.js
@@ -52,14 +52,14 @@ export function profile (files, base = '/') {
  * @return {object}
  */
 export function diff (files, hashes, base = '/') {
-  // Check new files
+  // Detect new files
   for (const file of files) {
     if (!(file in hashes)) {
       return { type: 'added', file }
     }
   }
 
-  // Check missing files
+  // Detect missing files
   for (const file in hashes) {
     if (!files.includes(file)) {
       return { type: 'removed', file }

diff --git a/readme.md b/readme.md
@@ -31,7 +31,7 @@ import { profile } from './compare-files.js'
 profile(glob.sync('/home/images/**/*.jpg'))
 ```
 
-`profile()` returns a plain JavaScript object. You may serialize and store it somewhere for later access.
+Calling `profile()` returns a plain JavaScript object. You may serialize and store it somewhere for later access.
 
 ## Check for Changes
 Comparing is the other way around: Read the stored profile and use `diff()`:

diff --git a/readme.md b/readme.md
@@ -16,7 +16,7 @@ const images_later = glob.sync('/home/images/**/*.jpg')
 // Do the files in images_now and images_later differ in any way?
 ```
 
-The script below tackles this problem in a way as read-performant as possible (i.e. one scan to potentially many comparisons). It checks if there are any differences between an earlier file list and the current file system and returns the first difference it finds.
+The script below tackles this problem in a way as read-performant as possible (i.e. scan once, compare potentially many times). It checks if there are any differences between an earlier file list and the current file system and returns the first difference it finds.
 
 > **Note:** The script uses ES modules (transpiling most likely needed) and needs the [`rev-hash`](https://www.npmjs.com/package/rev-hash) package to be installed.
 

diff --git a/readme.md b/readme.md
@@ -49,7 +49,9 @@ The `diff()` returns either `null` if no changes have occurred, or an object wit
 For example:
 ```js
 { type: 'added', file: 'new.jpg' }
+// or
 { type: 'removed', file: 'no-longer-exists.jpg' }
+// or
 { type: 'changed', file: 'photoshopped.jpg' }
 ```
 

diff --git a/readme.md b/readme.md
@@ -47,7 +47,7 @@ diff(
 The `diff()` returns either `null` if no changes have occurred, or an object with a `type` (which is either `added`, `removed` or `changed`) and a `file` property pointing to the according file.
 
 For example:
-```json
+```js
 { type: 'added', file: 'new.jpg' }
 { type: 'removed', file: 'no-longer-exists.jpg' }
 { type: 'changed', file: 'photoshopped.jpg' }

diff --git a/readme.md b/readme.md
@@ -47,7 +47,7 @@ diff(
 The `diff()` returns either `null` if no changes have occurred, or an object with a `type` (which is either `added`, `removed` or `changed`) and a `file` property pointing to the according file.
 
 For example:
-```
+```json
 { type: 'added', file: 'new.jpg' }
 { type: 'removed', file: 'no-longer-exists.jpg' }
 { type: 'changed', file: 'photoshopped.jpg' }

diff --git a/readme.md b/readme.md
@@ -46,4 +46,11 @@ diff(
 
 The `diff()` returns either `null` if no changes have occurred, or an object with a `type` (which is either `added`, `removed` or `changed`) and a `file` property pointing to the according file.
 
+For example:
+```
+{ type: 'added', file: 'new.jpg' }
+{ type: 'removed', file: 'no-longer-exists.jpg' }
+{ type: 'changed', file: 'photoshopped.jpg' }
+```
+
 > **Note:** If a difference is found, don't forget to invalidate the old profile and create a new one if needed.
diff --git a/readme.md b/readme.md
@@ -44,6 +44,6 @@ diff(
 )
 ```
 
-The `diff()` returns either `null` if no changes have occurred or an object with a `type` (which is either `added`, `removed` or `changed`) and a `file` property containing the according file.
+The `diff()` returns either `null` if no changes have occurred, or an object with a `type` (which is either `added`, `removed` or `changed`) and a `file` property pointing to the according file.
 
 > **Note:** If a difference is found, don't forget to invalidate the old profile and create a new one if needed.
diff --git a/readme.md b/readme.md
@@ -29,10 +29,10 @@ We need to create a profile of the files we're interested in:
 import { profile } from './compare-files.js'
 
 profile(glob.sync('/home/images/**/*.jpg'))
-
-// `profile()` returns a plain JavaScript object. You may serialize and store it somewhere for later access.
 ```
 
+`profile()` returns a plain JavaScript object. You may serialize and store it somewhere for later access.
+
 ## Check for Changes
 Comparing is the other way around: Read the stored profile and use `diff()`:
 ```js

diff --git a/compare-files.js b/compare-files.js
@@ -0,0 +1,84 @@
+import revHash from 'rev-hash'
+import { readFileSync, statSync } from 'fs'
+import { resolve } from 'path'
+
+/**
+ * Calculate the rev hash for a file
+ *
+ * @param {string} file The path to the file to hash
+ * @returns {string}
+ */
+function hashFile (file) {
+  return revHash(readFileSync(file))
+}
+
+/**
+ * Calculate the rev hashes for a list of files
+ *
+ * @generator
+ * @param {string[]} files The files to hash
+ * @param {string}   base  A base path prepended to each file (ignored if a file path is absolute)
+ * @yields {string}
+ */
+function* hashFileList (files, base = '/') {
+  for (const file of files) {
+    yield [ file, hashFile(resolve(base, file)) ]
+  }
+}
+
+/**
+ * Profile a list of files to compare against a later state
+ *
+ * @param {string[]} files The files to profile
+ * @param {string}   base  A base path prepended to each file (ignored if a file path is absolute)
+ * @return {object}
+ */
+export function profile (files, base = '/') {
+  const map = {}
+  for (const [ file, hash ] of hashFileList(files, base)) {
+    map[file] = [
+      statsync(resolve(base, file)).size,
+      hash
+    ]
+  }
+  return map
+}
+
+/**
+ * 
+ * @param {string[]} files  The files to profile
+ * @param {hashes}   object A the file list characteristics generated earlier by profile()
+ * @param {string}   base   A base path prepended to each file (ignored if a file path is absolute)
+ * @return {object}
+ */
+export function diff (files, hashes, base = '/') {
+  // Check new files
+  for (const file of files) {
+    if (!(file in hashes)) {
+      return { type: 'added', file }
+    }
+  }
+
+  // Check missing files
+  for (const file in hashes) {
+    if (!files.includes(file)) {
+      return { type: 'removed', file }
+    }
+  }
+
+  // Check sizes
+  for (const file of files) {
+    if (statSync(resolve(base, file)) !== hashes[file][0]) {
+      return { type: 'changed', file }
+    }
+  }
+
+  // Check contents
+  for (const [ file, hash ] of hashFileList(files, base)) {
+    if (hash !== hashes[file][1]) {
+      return { type: 'changed', file }
+    }
+  }
+
+  return null
+}
diff --git a/readme.md b/readme.md
@@ -0,0 +1,49 @@
+# Compare File Lists in Node.js
+
+I encouter a certain situation pretty frequently: I have a known set of files on my file system which I need to scan to compare them against another (or possibly the same) set of files at a later point in time.
+
+A quick example:
+
+```js
+import glob from 'glob' // from the `glob` package
+
+const images_now = glob.sync('/home/images/**/*.jpg')
+
+// Later (maybe even days or weeks later)
+
+const images_later = glob.sync('/home/images/**/*.jpg')
+
+// Do the files in images_now and images_later differ in any way?
+```
+
+The script below tackles this problem in a way as read-performant as possible (i.e. one scan to potentially many comparisons). It checks if there are any differences between an earlier file list and the current file system and returns the first difference it finds.
+
+> **Note:** The script uses ES modules (transpiling most likely needed) and needs the [`rev-hash`](https://www.npmjs.com/package/rev-hash) package to be installed.
+
+It works as follows:
+
+## Scan the Files
+We need to create a profile of the files we're interested in:
+
+```js
+import { profile } from './compare-files.js'
+
+profile(glob.sync('/home/images/**/*.jpg'))
+
+// `profile()` returns a plain JavaScript object. You may serialize and store it somewhere for later access.
+```
+
+## Check for Changes
+Comparing is the other way around: Read the stored profile and use `diff()`:
+```js
+import { diff } from './compare-files.js'
+
+diff(
+  glob.sync('/home/images/**/*.jpg'),
+  referenceData // The object we obtained through profile() earlier
+)
+```
+
+The `diff()` returns either `null` if no changes have occurred or an object with a `type` (which is either `added`, `removed` or `changed`) and a `file` property containing the according file.
+
+> **Note:** If a difference is found, don't forget to invalidate the old profile and create a new one if needed.