Unknown host when host doesn't exist as fallback mechanism.

Author: chanchiem

aws_xray_sdk/ext/util.py

@@ -14,6 +14,7 @@
 
 first_cap_re = re.compile('(.)([A-Z][a-z]+)')
 all_cap_re = re.compile('([a-z0-9])([A-Z])')
+UNKNOWN_HOSTNAME = "UNKNOWN HOST"
 
 
 def inject_trace_header(headers, entity):
@@ -126,9 +127,11 @@ def strip_url(url):
 
 def get_hostname(url):
     if url is None:
-        return None
+        return UNKNOWN_HOSTNAME
     url_parse = urlparse(url)
     hostname = url_parse.hostname
+    if hostname is None:
+        return UNKNOWN_HOSTNAME
     return hostname if hostname else url  # If hostname is none, we return the regular URL; indication of malformed url
 
 

tests/ext/requests/test_requests.py

@@ -84,7 +84,7 @@ def test_fault():
     assert http_meta['response']['status'] == status_code
 
 
-def test_invalid_url():
+def test_nonexistent_domain():
     try:
         requests.get('http://doesnt.exist')
     except Exception:
@@ -97,6 +97,24 @@ def test_invalid_url():
     assert exception.type == 'ConnectionError'
 
 
+def test_invalid_url():
+    url = 'KLSDFJKLSDFJKLSDJF'
+    try:
+        requests.get(url)
+    except Exception:
+        # prevent uncatch exception from breaking test run
+        pass
+    subsegment = xray_recorder.current_segment().subsegments[0]
+    assert subsegment.name == get_hostname(url)
+    assert subsegment.fault
+
+    http_meta = subsegment.http
+    assert http_meta['request']['url'] == strip_url(url)
+
+    exception = subsegment.cause['exceptions'][0]
+    assert exception.type == 'MissingSchema'
+
+
 def test_name_uses_hostname():
     url1 = 'http://{}/fakepath/stuff/koo/lai/ahh'.format(BASE_URL)
     requests.get(url1)

tests/test_utils.py

@@ -1,5 +1,7 @@
 from aws_xray_sdk.ext.util import to_snake_case, get_hostname, strip_url
 
+UNKNOWN_HOST = "UNKNOWN HOST"
+
 
 def test_to_snake_case():
     s1 = to_snake_case('Bucket')
@@ -29,13 +31,13 @@ def test_get_hostname():
     assert s4 == "amazon.com"
 
     s5 = get_hostname("INVALID_URL")
-    assert s5 == "INVALID_URL"
+    assert s5 == UNKNOWN_HOST
 
     s6 = get_hostname("")
-    assert s6 == ""
+    assert s6 == UNKNOWN_HOST
 
     s7 = get_hostname(None)
-    assert not s7
+    assert s7 == UNKNOWN_HOST
 
 
 def test_strip_url():