documentaiton of troggle & dup filename detection utility

scripts/detect-filename-clashes.py

@@ -0,0 +1,57 @@
+#!/usr/bin/env python
+"""
+Recursively checks that no filenames from the current working directory 
+down are the same, even if they are in different subdirectories. Exits 
+with status 0 if no matching filenames are found, or 1 if a match is 
+found.
+
+2020-03-25 PMS Copied originally from https://gist.github.com/seanh/229455 
+Modified to lowercase everything before checking, to detect clashes caused
+by capitalisation, and to capture full path for all clashing files
+"""
+import sys,os
+print "Checking for filename clashes...",
+
+def sortSecond(val): 
+    return val[1] 
+
+allfilenames = []
+detect = []
+clashes = []
+report = []
+for root, dirs, files in os.walk('.'):
+    if root.startswith('./.git'):
+        # Ignore  .git directory.
+        continue
+    if root.startswith('./.hg'):
+        # Ignore  mercurial directory.
+        continue
+    for filename in files:
+        if filename.endswith('~'):
+            # Ignore automatically created backup files.
+            continue
+        if os.path.islink(os.path.join(root,filename)):
+            # Don't count symlinks as filename clashes.
+            continue
+        allfilenames.append((root,filename))
+        if filename.lower() in detect:
+            clashes.append(filename.lower())
+        else:
+            detect.append(filename.lower())
+
+print len(allfilenames), 'files found...',
+if clashes == []:
+    print 'OK'
+    sys.exit(0)
+else:
+    print len(clashes)," clash(es) found."
+    for r,f in allfilenames:
+        if f in clashes:
+            report.append((r,f))
+    # sorts the array in descending according to 
+    # second element: the clashing filename
+    report.sort(key = sortSecond) 
+    for r,f in report:
+        print os.path.join(r,f)
+    print len(clashes)," clash(es) found."
+    sys.exit(1)