Parse OpenScanHub URL correctly from stdout (#2477)

Parse OpenScanHub URL correctly from stdout

Stdout may include lot of info e.g. about Kerberos ticket processing before the actual dict response from the osh-client, which should be in the last line of the output.

Reviewed-by: Siteshwar Vashisht
Reviewed-by: Laura Barcziová
Reviewed-by: Maja Massarini
Reviewed-by: Nikola Forró

packit_service/worker/handlers/copr.py

@@ -3,6 +3,7 @@
 
 import json
 import logging
+import re
 import tempfile
 from datetime import datetime, timezone
 from os import getenv
@@ -564,17 +565,20 @@ def handle_scan(self):
                 logger.debug("Something went wrong, skipping the reporting.")
                 return
 
-            logger.info(f"Scan submitted successfully, output: {output}")
+            logger.info("Scan submitted successfully.")
 
-            response = json.loads(output)
-            if not (url := response.get("url")):
+            response_dict = self.parse_dict_from_output(output)
+
+            logger.debug(f"Parsed dict from output: {response_dict} ")
+
+            if not (url := response_dict.get("url")):
                 logger.debug("It was not possible to get the URL from the response.")
                 return
 
             self.copr_build_helper._report(
                 state=BaseCommitStatus.success,
                 description=(
-                    "Scan in OpenScanHub submitted succesfully. Check the URL for more details."
+                    "Scan in OpenScanHub submitted successfully. Check the URL for more details."
                 ),
                 url=url,
                 check_names=["osh-diff-scan:fedora-rawhide-x86_64"],
@@ -587,6 +591,17 @@ def handle_scan(self):
                 ),
             )
 
+    @staticmethod
+    def parse_dict_from_output(output: str) -> dict:
+        json_pattern = r"\{.*?\}"
+        matches = re.findall(json_pattern, output, re.DOTALL)
+
+        if not matches:
+            return {}
+
+        json_str = matches[-1]
+        return json.loads(json_str)
+
     def find_base_build_job(self) -> Optional[JobConfig]:
         """
         Find the job in the config that can provide the base build for the scan

tests/unit/test_scan.py

@@ -34,7 +34,7 @@ def test_handle_scan():
     ).and_return([base_copr_model])
 
     flexmock(PackitAPI).should_receive("run_osh_build").once().and_return(
-        '{"url": "scan-url"}'
+        'some\nmultiline\noutput\n{"id": 123}\nand\nmore\n{"url": "scan-url"}\n'
     )
 
     flexmock(CoprBuildJobHelper).should_receive("_report")