From a3e170e9558c85be2ed539bf50a86458f4c812cd Mon Sep 17 00:00:00 2001
From: Thomas Cram <tcram@ucar.edu>
Date: Wed, 8 Oct 2025 11:42:13 -0600
Subject: [PATCH 1/3] add unidecode to requirements.txt

---
 requirements.txt | 1 +
 1 file changed, 1 insertion(+)

diff --git a/requirements.txt b/requirements.txt
index f038d89..a1482d1 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -4,3 +4,4 @@ pluggy==1.5.0
 psycopg2-binary==2.9.10
 pytest==8.3.5
 rda-python-globus
+unidecode

From e7a35626b3760a13752b8816db47efbed17d2f42 Mon Sep 17 00:00:00 2001
From: Thomas Cram <tcram@ucar.edu>
Date: Wed, 8 Oct 2025 12:03:02 -0600
Subject: [PATCH 2/3] refactor: improve character conversion to use unidecode
 for better ASCII representation

---
 src/rda_python_common/PgLOG.py | 32 +++++++++-----------------------
 1 file changed, 9 insertions(+), 23 deletions(-)

diff --git a/src/rda_python_common/PgLOG.py b/src/rda_python_common/PgLOG.py
index 54ce0e2..d6f8c0d 100644
--- a/src/rda_python_common/PgLOG.py
+++ b/src/rda_python_common/PgLOG.py
@@ -28,6 +28,7 @@
 import socket
 import shutil
 import traceback
+from unidecode import unidecode
 
 # define some constants for logging actions
 MSGLOG = (0x00001)   # logging message
@@ -1579,33 +1580,18 @@ def check_process_host(hosts, chost = None, mflag = None, pinfo = None, logact =
    return ret
 
 #
-# convert special characters
+# convert special foreign characters into ascii characters
 #
 def convert_chars(name, default = 'X'):
 
    if not name: return default
-   if re.match(r'^[a-zA-Z0-9]+$', name): return name  # no need convert
-
-   z = ord('z')
-   newchrs = ochrs = ''
-   for i in range(len(name)):
-      ch = name[i]
-      if re.match(r'^[a-zA-Z0-9]$', ch):
-         newchrs += ch
-      elif (ch == ' ' or ch == '_') and newchrs:
-         newchrs += '_'
-      elif ord(ch) > z and ochrs != None:
-         if not ochrs:
-            ochrs = None
-            with open(PGLOG['DSSHOME'] + "/lib/ExtChrs.txt", "r") as CHR:
-               ochrs = CHR.readline()
-               nchrs = CHR.readline()
-            if ochrs is None: continue
-         idx = ochrs.find(ch)
-         if idx >= 0: newchrs += nchrs[idx]
-
-   if newchrs:
-      return newchrs
+   if re.match(r'^[a-zA-Z0-9]+$', name): return name  # conversion not needed
+
+   decoded_name = unidecode(name).strip()
+   cleaned_name = re.sub(r'[^a-zA-Z0-9]', '', decoded_name)
+
+   if cleaned_name:
+      return cleaned_name
    else:
       return default
 

From 2ba2ee082c7d4bda8594f4dcaf0fbf60d82e5bc7 Mon Sep 17 00:00:00 2001
From: Thomas Cram <tcram@ucar.edu>
Date: Wed, 8 Oct 2025 12:10:57 -0600
Subject: [PATCH 3/3] refactor: enhance character conversion to allow
 underscores in cleaned names

---
 src/rda_python_common/PgLOG.py | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/src/rda_python_common/PgLOG.py b/src/rda_python_common/PgLOG.py
index d6f8c0d..c08b3e3 100644
--- a/src/rda_python_common/PgLOG.py
+++ b/src/rda_python_common/PgLOG.py
@@ -1583,13 +1583,11 @@ def check_process_host(hosts, chost = None, mflag = None, pinfo = None, logact =
 # convert special foreign characters into ascii characters
 #
 def convert_chars(name, default = 'X'):
-
    if not name: return default
    if re.match(r'^[a-zA-Z0-9]+$', name): return name  # conversion not needed
-
    decoded_name = unidecode(name).strip()
-   cleaned_name = re.sub(r'[^a-zA-Z0-9]', '', decoded_name)
-
+   # remove any non-alphanumeric and non-underscore characters
+   cleaned_name = re.sub(r'[^a-zA-Z0-9_]', '', decoded_name)
    if cleaned_name:
       return cleaned_name
    else: