Merge pull request #19108 from mvdbeek/access_public_history_in_job_c…

…ache

Access public history in job cache / job search

client/src/api/schema/schema.ts

@@ -8753,6 +8753,11 @@ export interface components {
              * @description The email of the user that owns this job. Only the owner of the job and administrators can see this value.
              */
             user_email?: string | null;
+            /**
+             * User Id
+             * @description User ID of user that ran this job
+             */
+            user_id?: string | null;
         };
         /** EncodedJobParameterHistoryItem */
         EncodedJobParameterHistoryItem: {
@@ -16177,6 +16182,11 @@ export interface components {
              * @description The email of the user that owns this job. Only the owner of the job and administrators can see this value.
              */
             user_email?: string | null;
+            /**
+             * User Id
+             * @description User ID of user that ran this job
+             */
+            user_id?: string | null;
         };
         /**
          * Src

lib/galaxy/managers/jobs.py

@@ -493,11 +493,18 @@ def _build_job_subquery(
         self, tool_id: str, user_id: int, tool_version: Optional[str], job_state, wildcard_param_dump
     ):
         """Build subquery that selects a job with correct job parameters."""
-        stmt = select(model.Job.id).where(
-            and_(
-                model.Job.tool_id == tool_id,
-                model.Job.user_id == user_id,
-                model.Job.copied_from_job_id.is_(None),  # Always pick original job
+        stmt = (
+            select(model.Job.id)
+            .join(model.History, model.Job.history_id == model.History.id)
+            .where(
+                and_(
+                    model.Job.tool_id == tool_id,
+                    or_(
+                        model.Job.user_id == user_id,
+                        model.History.published == true(),
+                    ),
+                    model.Job.copied_from_job_id.is_(None),  # Always pick original job
+                )
             )
         )
         if tool_version:

lib/galaxy/model/__init__.py

@@ -1549,6 +1549,7 @@ class Job(Base, JobLike, UsesCreateAndUpdateTime, Dictifiable, Serializable):
         "galaxy_version",
         "command_version",
         "copied_from_job_id",
+        "user_id",
     ]
 
     _numeric_metric = JobMetricNumeric

lib/galaxy/schema/jobs.py

@@ -191,6 +191,7 @@ class EncodedJobDetails(JobSummary):
         title="Output collections",
         description="",
     )
+    user_id: Optional[EncodedDatabaseIdField] = Field(default=None, description="User ID of user that ran this job")
 
 
 class JobDestinationParams(Model):

lib/galaxy_test/api/test_tools.py

@@ -8,7 +8,9 @@
     Any,
     Dict,
     List,
+    Optional,
 )
+from uuid import uuid4
 
 import pytest
 from requests import (
@@ -974,15 +976,19 @@ def test_run_cat1(self):
             output1_content = self.dataset_populator.get_history_dataset_content(history_id, dataset=output1)
             assert output1_content.strip() == "Cat1Test"
 
+    def _get_cat1_inputs(self, history_id):
+        new_dataset = self.dataset_populator.new_dataset(history_id, content="Cat1Test")
+        inputs = dict(
+            input1=dataset_to_param(new_dataset),
+        )
+        return inputs
+
     @skip_without_tool("cat1")
     @requires_new_history
     def test_run_cat1_use_cached_job(self):
         with self.dataset_populator.test_history_for(self.test_run_cat1_use_cached_job) as history_id:
             # Run simple non-upload tool with an input data parameter.
-            new_dataset = self.dataset_populator.new_dataset(history_id, content="Cat1Test")
-            inputs = dict(
-                input1=dataset_to_param(new_dataset),
-            )
+            inputs = self._get_cat1_inputs(history_id)
             outputs_one = self._run_cat1(history_id, inputs=inputs, assert_ok=True, wait_for_job=True)
             outputs_two = self._run_cat1(
                 history_id, inputs=inputs, use_cached_job=False, assert_ok=True, wait_for_job=True
@@ -998,6 +1004,31 @@ def test_run_cat1_use_cached_job(self):
             assert len(filenames) == 3, filenames
             assert len(set(filenames)) <= 2, filenames
 
+    @skip_without_tool("cat1")
+    @requires_new_history
+    def test_run_cat1_use_cached_job_from_public_history(self):
+        with self.dataset_populator.test_history_for(self.test_run_cat1_use_cached_job) as history_id:
+            # Run simple non-upload tool with an input data parameter.
+            inputs = self._get_cat1_inputs(history_id)
+            original_output = self._run_cat1(history_id, inputs=inputs, assert_ok=True, wait_for_job=True)
+            original_job = self.dataset_populator.get_job_details(original_output["jobs"][0]["id"], full=True).json()
+
+            def run_again(user_email):
+                with self._different_user_and_history(user_email=user_email) as different_history_id:
+                    cached_output = self._run_cat1(
+                        different_history_id, inputs=inputs, use_cached_job=True, assert_ok=True, wait_for_job=True
+                    )
+                    return self.dataset_populator.get_job_details(cached_output["jobs"][0]["id"], full=True).json()
+
+            job = run_again(f"{uuid4()}@test.com")
+            assert job["user_id"] != original_job["user_id"]
+            assert not job["copied_from_job_id"]
+            # publish history, now we can use cached job
+            self.dataset_populator.make_public(history_id=history_id)
+            cached_job = run_again(f"{uuid4()}@test.com")
+            assert cached_job["user_id"] != original_job["user_id"]
+            assert cached_job["copied_from_job_id"] == original_output["jobs"][0]["id"]
+
     @skip_without_tool("cat1")
     def test_run_cat1_listified_param(self):
         with self.dataset_populator.test_history_for(self.test_run_cat1_listified_param) as history_id:
@@ -2809,8 +2840,8 @@ def _assert_dataset_permission_denied_response(self, response):
         # assert "User does not have permission to use a dataset" in err_message, err_message
 
     @contextlib.contextmanager
-    def _different_user_and_history(self):
-        with self._different_user():
+    def _different_user_and_history(self, user_email: Optional[str] = None):
+        with self._different_user(email=user_email):
             with self.dataset_populator.test_history() as other_history_id:
                 yield other_history_id
 

lib/galaxy_test/base/api.py

@@ -147,7 +147,7 @@ def _setup_user_get_key(self, email, password=None):
         return user, self._post(f"users/{user['id']}/api_key", admin=True).json()
 
     @contextmanager
-    def _different_user(self, email=OTHER_USER, anon=False):
+    def _different_user(self, email: Optional[str] = None, anon=False):
         """Use in test cases to switch get/post operations to act as new user
 
         ..code-block:: python
@@ -164,6 +164,7 @@ def _different_user(self, email=OTHER_USER, anon=False):
             self.galaxy_interactor.cookies = cookies
             new_key = None
         else:
+            email = OTHER_USER if email is None else email
             _, new_key = self._setup_user_get_key(email)
         try:
             self.user_api_key = new_key