Align job registry implementations: omit process/job_options from get…

…_user_jobs listings

Process graph (and job options) are not to be included in user job listings,
so align `ZkJobRegistry.get_user_jobs` to not produce then like `ElasticJobRegistry`

refs: eu-cdse/openeo-cdse-infra#141, #498

openeogeotrellis/job_registry.py

@@ -215,9 +215,22 @@ def get_running_jobs(self, *, user_limit: Optional[int] = 1000, parse_specificat
                 else:
                     stats["user_id without jobs"] += 1
 
-    def get_job(self, job_id: str, user_id: str, parse_specification: bool = False) -> dict:
+    def get_job(
+        self,
+        job_id: str,
+        user_id: str,
+        *,
+        parse_specification: bool = False,
+        omit_raw_specification: bool = False,
+    ) -> dict:
         """Returns details of a job."""
-        job_info, _ = self._read(job_id, user_id, include_done=True, parse_specification=parse_specification)
+        job_info, _ = self._read(
+            job_id=job_id,
+            user_id=user_id,
+            include_done=True,
+            parse_specification=parse_specification,
+            omit_raw_specification=omit_raw_specification,
+        )
         return job_info
 
     def get_user_jobs(self, user_id: str) -> List[Dict]:
@@ -227,13 +240,19 @@ def get_user_jobs(self, user_id: str) -> List[Dict]:
 
         try:
             done_job_ids = self._zk.get_children(self._done(user_id))
-            jobs.extend([self.get_job(job_id, user_id) for job_id in done_job_ids])
+            jobs.extend(
+                self.get_job(job_id=job_id, user_id=user_id, parse_specification=False, omit_raw_specification=True)
+                for job_id in done_job_ids
+            )
         except NoNodeError:
             pass
 
         try:
             ongoing_job_ids = self._zk.get_children(self._ongoing(user_id))
-            jobs.extend([self.get_job(job_id, user_id) for job_id in ongoing_job_ids])
+            jobs.extend(
+                self.get_job(job_id=job_id, user_id=user_id, parse_specification=False, omit_raw_specification=True)
+                for job_id in ongoing_job_ids
+            )
         except NoNodeError:
             pass
 
@@ -336,11 +355,18 @@ def _create(self, job_info: Dict, done: bool = False) -> None:
         self._zk.create(path, data, makepath=True)
 
     def _read(
-        self, job_id: str, user_id: str, include_done: bool = False, parse_specification: bool = False
+        self,
+        job_id: str,
+        user_id: str,
+        *,
+        include_done: bool = False,
+        parse_specification: bool = False,
+        omit_raw_specification: bool = False,
     ) -> Tuple[Dict, int]:
         """
         :param parse_specification: parse the (JSON encoded) "specification" field
             and inject the process_graph (as `"process": {"process_graph": ...}`) and job_options as additional fields
+        :param omit_raw_specification: remove the original (raw) "specification" field from the result
         """
         assert job_id, "Shouldn't be empty: job_id"
         assert user_id, "Shouldn't be empty: user_id"
@@ -370,6 +396,9 @@ def _read(
             if "job_options" not in job_info:
                 job_info["job_options"] = job_options
 
+        if omit_raw_specification:
+            del job_info["specification"]
+
         return job_info, stat.version
 
     def _update(self, job_info: Dict, version: int) -> None: