process-cve-records: improve description parsing

1 files changed, 11 insertions, 2 deletions

diff --git a/bin/process-cve-records b/bin/process-cve-records
index 7cf9d2605c..a5d074655d 100755
--- a/bin/process-cve-records
+++ b/bin/process-cve-records
@@ -67,10 +67,19 @@ def parse_record(record, cve):
                 if desc['lang'].startswith('en')]
         if desc:
             desc = desc[0]
-        if desc and len(desc) > 70:
+
             # for some reason descriptions may contain new lines
             desc = desc.replace('\n', ' ')
-            desc = desc[:70] + ' ...'
+
+            # and even non-printable characters such as \xa0 ( )
+            desc = "".join([ c for c in desc if c.isprintable() ])
+
+            # and some contain leading spaces
+            desc = desc.strip()
+
+            if len(desc) > 70:
+                desc = desc[:70] + ' ...'
+
         cve.header.description = f"({desc})"
 
     if not is_reserved(record) and not is_rejected(record) \