more robust tripid labelling

core/utils.py

@@ -1,14 +1,15 @@
+import hashlib
 import logging
+import os
 import random
 import resource
+import string
 import subprocess
-import os
 from decimal import getcontext
 from pathlib import Path
 
 getcontext().prec = 2  # use 2 significant figures for decimal calculations
 
-
 import settings
 
 """This file declares TROG a globally visible object for caches.
@@ -30,6 +31,8 @@ thread.
 """
 
 TROG = {"pagecache": {"expedition": {}}, "caves": {"gcavelookup": {}, "gcavecount": {}}}
+alphabet = []
+sha = hashlib.new('sha256')
 
 # This is module-level executable. This is a Bad Thing. Especially when it touches the file system.
 try:
@@ -50,7 +53,27 @@ def chaosmonkey(n):
         return False
     # print("CHAOS strikes !", file=sys.stderr)
     return True
-
+    
+def unique_slug(text, n):
+    """This gives an almost-unique id based on the text, 
+    2 hex digits would seem adequate, but we might get a collision.
+    Not used anywhere.
+    """
+    sha.update(text.encode('utf-8'))
+    return sha.hexdigest()[0:n]
+    
+def alphabet_suffix(n):
+    """This is called repeatedly during initial parsing import, hence the cached list
+    """
+    global alphabet
+    if not alphabet:
+        alphabet = list(string.ascii_lowercase)
+        
+    if n < len(alphabet):
+        suffix = alphabet[n]
+    else:
+        suffix = "_X_" + random.choice(string.ascii_lowercase) + random.choice(string.ascii_lowercase)
+    return suffix        
 
 def only_commit(fname, message):
     """Only used to commit a survex file edited and saved in view/survex.py"""