From d6a45e5c9f868515c2ac515466f3cdc232aaa6d7 Mon Sep 17 00:00:00 2001
From: Andreas Gohr <andi@splitbrain.org>
Date: Wed, 13 Mar 2024 12:09:34 +0100
Subject: [PATCH] ignore typical temporary files when listing directories

Operating systems love to cluter the file system with all kinds of
cruft. This adds a gitignore like config to skip those files when
listing files.
---
 Crawler.php                        | 37 +++++++++++++
 _test/filelistdata/~$ignoreme.docx |  0
 conf/ignore.txt                    | 85 ++++++++++++++++++++++++++++++
 3 files changed, 122 insertions(+)
 create mode 100644 _test/filelistdata/~$ignoreme.docx
 create mode 100644 conf/ignore.txt

diff --git a/Crawler.php b/Crawler.php
index e8d4ccb..09e5806 100644
--- a/Crawler.php
+++ b/Crawler.php
@@ -13,6 +13,9 @@ class Crawler
     /** @var bool */
     protected $sortreverse = false;
 
+    /** @var string[] patterns to ignore */
+    protected $ignore = [];
+
     /**
      * Initializes the crawler
      *
@@ -24,6 +27,8 @@ public function __construct($extensions)
         $this->ext = array_map('trim', $this->ext);
         $this->ext = array_map('preg_quote_cb', $this->ext);
         $this->ext = implode('|', $this->ext);
+
+        $this->ignore = $this->loadIgnores();
     }
 
     public function setSortBy($sortby)
@@ -67,6 +72,9 @@ public function crawl($root, $local, $pattern, $recursive, $titlefile)
                 if (!is_dir($filepath) && !$this->isExtensionAllowed($file)) {
                     continue;
                 }
+                if ($this->isFileIgnored($file)) {
+                    continue;
+                }
 
                 // get title file
                 $filename = $file;
@@ -141,6 +149,35 @@ protected function isExtensionAllowed($file)
         return preg_match('/(' . $this->ext . ')$/i', $file);
     }
 
+    /**
+     * Check if a file is ignored by the ignore patterns
+     *
+     * @param string $file
+     * @return bool
+     */
+    protected function isFileIgnored($file)
+    {
+        foreach ($this->ignore as $pattern) {
+            if ($this->fnmatch($pattern, $file)) return true;
+        }
+        return false;
+    }
+
+    /**
+     * Load the ignore patterns from the ignore.txt file
+     *
+     * @return string[]
+     */
+    protected function loadIgnores()
+    {
+        $file = __DIR__ . '/conf/ignore.txt';
+        $ignore = file($file, FILE_IGNORE_NEW_LINES | FILE_SKIP_EMPTY_LINES);
+        $ignore = array_map(function ($line) {
+            return trim(preg_replace('/\s*#.*$/', '', $line));
+        }, $ignore);
+        $ignore = array_filter($ignore);
+        return $ignore;
+    }
 
     /**
      * Replacement for fnmatch() for windows systems.
diff --git a/_test/filelistdata/~$ignoreme.docx b/_test/filelistdata/~$ignoreme.docx
new file mode 100644
index 0000000..e69de29
diff --git a/conf/ignore.txt b/conf/ignore.txt
new file mode 100644
index 0000000..869b332
--- /dev/null
+++ b/conf/ignore.txt
@@ -0,0 +1,85 @@
+# This is a gitignore style file to ignore typical temporary files and directories
+
+### Linux ###
+*~
+
+# temporary files which can be created if a process still has a handle open of a deleted file
+.fuse_hidden*
+
+# KDE directory preferences
+.directory
+
+# Linux trash folder which might appear on any partition or disk
+.Trash-*
+
+# .nfs files are created when an open file is removed but is still being accessed
+.nfs*
+
+### macOS ###
+# General
+.DS_Store
+.AppleDouble
+.LSOverride
+
+
+# Thumbnails
+._*
+
+# Files that might appear in the root of a volume
+.DocumentRevisions-V100
+.fseventsd
+.Spotlight-V100
+.TemporaryItems
+.Trashes
+.VolumeIcon.icns
+.com.apple.timemachine.donotpresent
+
+# Directories potentially created on remote AFP share
+.AppleDB
+.AppleDesktop
+Network Trash Folder
+Temporary Items
+.apdisk
+
+# iCloud generated files
+*.icloud
+
+### MicrosoftOffice ###
+*.tmp
+
+# Word temporary
+~$*.doc*
+
+# Word Auto Backup File
+Backup of *.doc*
+
+# Excel temporary
+~$*.xls*
+
+# Excel Backup File
+*.xlk
+
+# PowerPoint temporary
+~$*.ppt*
+
+# Visio autosave temporary files
+*.~vsd*
+
+### Windows ###
+# Windows thumbnail cache files
+Thumbs.db
+Thumbs.db:encryptable
+ehthumbs.db
+ehthumbs_vista.db
+
+# Dump file
+*.stackdump
+
+# Folder config file
+[Dd]esktop.ini
+
+# Recycle Bin used on file shares
+$RECYCLE.BIN
+
+# Windows shortcuts
+*.lnk