From 2c3ca3c0eb564a4adf9259f70a44798a6e0e13b2 Mon Sep 17 00:00:00 2001
From: Michael Hohn <hohn@github.com>
Date: Mon, 6 Dec 2021 11:27:39 -0800
Subject: [PATCH] Fix for KeyError: 'region', caused by `result` without
 `region`

Region / line / column information are present in most messages.  The one that
caused this error refers to the whole file:

    ipdb> p sarif_struct

    {'ruleId': 'com.lgtm/cpp-queries:cpp/missing-header-guard', 'ruleIndex': 12,
    'message': {'text': 'This header file should contain a header guard to prevent
    multiple inclusion.'}, 'locations': [{'physicalLocation': {'artifactLocation':
    {'uri': 'diff/cmpbuf.h', 'uriBaseId': '%SRCROOT%', 'index': 13}}}],
    'partialFingerprints': {'primaryLocationLineHash': 'd04cb834fa64727d:1',
    'primaryLocationStartColumnFingerprint': '0'}}

The goal is fixed-structure output formatting, so whole-file output uses
-1,-1,-1,-1 for line, column information.
---
 bin/sarif-results-summary |  8 ++++++--
 sarif_cli/__init__.py     | 10 +++++++++-
 2 files changed, 15 insertions(+), 3 deletions(-)

diff --git a/bin/sarif-results-summary b/bin/sarif-results-summary
index cfb1269..833459d 100755
--- a/bin/sarif-results-summary
+++ b/bin/sarif-results-summary
@@ -33,13 +33,17 @@ for runi in S.indices(sarif_struct, 'runs'):
             # Non-path problems
             # TODO: just pull out the uri, not the artifact
             message, artifact, region = S.get_location_message_info(result)
-            l1, c1, l2, c2 = S.lineinfo(region)
+            if region == S.WholeFile:
+                l1, c1, l2, c2 = -1, -1, -1, -1
+            else:
+                l1, c1, l2, c2 = S.lineinfo(region)
             filepath = "%s:%d:%d:%d:%d" % (artifact['uri'], l1, c1, l2, c2)
             if args.csv:
                 S.write_csv(cw, "result", artifact['uri'], l1, c1, l2, c2, message)
             else:
                 S.msg("RESULT: %s: %s\n\n" % (filepath, message))
-            if args.list_source:
+
+            if region != S.WholeFile and args.list_source:
                 lines = S.load_lines(args.list_source, artifact['uri'], l1, l2)
                 if args.csv:
                     pass
diff --git a/sarif_cli/__init__.py b/sarif_cli/__init__.py
index ec694ef..913c682 100644
--- a/sarif_cli/__init__.py
+++ b/sarif_cli/__init__.py
@@ -33,14 +33,22 @@ def get_relatedlocation_message_info(related_location):
     region = get(related_location, 'physicalLocation', 'region')
     return message, artifact, region
 
+class WholeFile:
+    pass
+
 def get_location_message_info(result):
     """ Given one of the results, extract message information.
 
     The `result` typically starts from get(sarif_struct, 'runs', run_index, 'results', res_index)
+
+    Returns: (message, artifact, region)
+        For an empty 'region' key, returns (message, artifact, sarif_cli.WholeFile)
+
     """
     message = get(result, 'message', 'text')
     artifact = get(result, 'locations', 0, 'physicalLocation', 'artifactLocation')
-    region = get(result,  'locations', 0, 'physicalLocation', 'region')
+    # If there is no 'region' key, use the whole file
+    region = get(result,  'locations', 0, 'physicalLocation').get('region', WholeFile)
     return (message, artifact, region)
 
 def display_underlined(l1, c1, l2, c2, line, line_num):