Merge pull request #27 from xsuite/release/v0.4.3

Release 0.4.3

Author: freddieknets

README.md

@@ -114,8 +114,9 @@ When the jobs are completed, the Xboinc server will store the results in your al
 ```python
 import xboinc as xb
 
-for job_name, result_particles in xb.JobRetriever.iterate("mycernshortname", "a_relevant_study_name", dev_server=True):
+for job_name, job_metadata, result_particles in xb.JobRetriever.iterate("mycernshortname", "a_relevant_study_name", dev_server=True):
     print(f"Job {job_name} completed with particles: {result_particles.to_dict()}")
+    print(f"Job metadata: {job_metadata}")
 
 ```
 

examples/results.py

@@ -8,6 +8,7 @@
 list_of_succeeded_jobs = []
 user='sixtadm'
 study_name='example_study'
-for jobname, new_particles in xb.JobRetriever(user=user, study_name=study_name, dev_server=True):
+for jobname, metadata, new_particles in xb.JobRetriever(user=user, study_name=study_name, dev_server=True):
     print(jobname)
+    print(metadata)
     print(f"Particles: {new_particles.at_turn}")

pyproject.toml

@@ -5,7 +5,7 @@
 
 [tool.poetry]
 name = "xboinc"
-version = "0.4.2"
+version = "0.4.3"
 description = "Xsuite BOINC interface"
 homepage = "https://github.com/xsuite/xboinc"
 repository = "https://github.com/xsuite/xboinc"

tests/test_03_submission_and_retrieval.py

@@ -358,7 +358,7 @@ def test_retrieval(registered_user):
         shutil.copy(tar_file, output_dir)
 
     # Iterate through jobs and validate results
-    for _, result_particles in xb.JobRetriever.iterate(
+    for _, _, result_particles in xb.JobRetriever.iterate(
         "testuser", "example_study_fourth", dev_server=True
     ):
         assert len(result_particles.x) == 100

tests/test_version.py

@@ -12,7 +12,7 @@
 
 
 def test_version():
-    assert __version__ == '0.4.2'
+    assert __version__ == '0.4.3'
 
 
 def test_xb_ver():

xboinc/df_wu.py

@@ -94,23 +94,23 @@ def query_registered_work_units(
     return all_df
 
 
-def query_subscribed_users() -> list[str]:
+def query_subscribed_users() -> list[tuple[str, str]]:
     """
-    Get a list of all users subscribed to the work unit database.
+    Get a list of all users subscribed to the work unit database with their status.
 
     Returns
     -------
-    list[str]
-        List of usernames subscribed to the work unit database.
+    list[tuple[str, str]]
+        List of tuples containing (username, status) for users subscribed to the work unit database.
     """
     with _get_read_only_user_db_connection() as conn:
         cursor = conn.cursor()
-        cursor.execute("SELECT user FROM users")
-        users = [row[0] for row in cursor.fetchall()]
+        cursor.execute("SELECT user, status FROM users")
+        users = [(row[0], row[1]) for row in cursor.fetchall()]
     return users
 
 
-def check_user_subscription(user: str) -> bool:
+def check_user_subscription(user: str) -> str:
     """
     Check if a user is subscribed to the work unit database.
 
@@ -121,7 +121,12 @@ def check_user_subscription(user: str) -> bool:
 
     Returns
     -------
-    bool
-        True if the user is subscribed, False otherwise.
+    str
+        The status of the user if subscribed, "not_subscribed" otherwise,
+        "broken" if the user's directory is invalid.
     """
-    return user in query_subscribed_users()
+    users = query_subscribed_users()
+    for u, status in users:
+        if u == user:
+            return status
+    return "not_subscribed"

xboinc/general.py

@@ -12,7 +12,7 @@
 # Do not change
 # ==============================================================================
 
-__version__ = '0.4.2'
+__version__ = '0.4.3'
 
 # These are the xsuite modules that are used by boinc and the versions they are
 # tied to. This will be automatically updated from the active environment when

xboinc/retrieve.py

@@ -104,13 +104,22 @@ def _index_results(self, path: FsPath, silent: bool = False) -> pd.DataFrame:
                 job_name = parts[2]
                 wu_name = bin_file.name.replace(".bin", "")
                 # append to the DataFrame
-                new_row = pd.DataFrame([{
-                    "user": user,
-                    "study_name": study_name,
-                    "job_name": job_name,
-                    "wu_name": wu_name,
-                    "bin_file": bin_file,
-                }])
+                new_row = pd.DataFrame(
+                    [
+                        {
+                            "user": user,
+                            "study_name": study_name,
+                            "job_name": job_name,
+                            "wu_name": wu_name,
+                            "bin_file": bin_file,
+                            "json_file": bin_file.with_name(
+                                bin_file.name.replace(
+                                    "__file_xboinc_state_out.bin", ".json"
+                                )
+                            ),
+                        }
+                    ]
+                )
                 df = pd.concat([df, new_row], ignore_index=True)
         return df
 
@@ -299,8 +308,8 @@ def iterate_results(self, study_name):
 
         Yields
         ------
-        tuple of (str, xpart.Particles)
-            Job name and corresponding particles object for each result
+        tuple of (str, dict, xpart.Particles)
+            Job name, corresponding metadata, and particles object for each result
 
         Raises
         ------
@@ -324,6 +333,7 @@ def iterate_results(self, study_name):
         for row in self._df[self._df["study_name"] == study_name].itertuples():
             job_name = row.job_name
             bin_file = row.bin_file
+            json_file = row.json_file
             result = XbState.from_binary(bin_file, raise_version_error=False)
             if result is None:
                 warn(
@@ -332,7 +342,22 @@ def iterate_results(self, study_name):
                     UserWarning,
                 )
                 continue
-            yield job_name, result.particles
+            try:
+                with open(json_file, "r") as f:
+                    metadata = json.load(f)
+                # is metadata an empty dict?
+                if not metadata:
+                    warn(
+                        f"Warning: The JSON file {json_file} is empty.",
+                        UserWarning,
+                    )
+            except FileNotFoundError:
+                warn(
+                    f"Warning: The JSON file {json_file} was not found.",
+                    UserWarning,
+                )
+                metadata = {}
+            yield job_name, metadata, result.particles
 
     def clean(self, study_name):
         """
@@ -362,6 +387,9 @@ def clean(self, study_name):
             bin_file = row.bin_file
             if bin_file.exists():
                 bin_file.unlink()
+            json_file = row.json_file
+            if json_file.exists():
+                json_file.unlink()
         # Remove empty directories
         for folder in self._directory.glob("*/"):
             if not any(folder.iterdir()):
@@ -487,4 +515,4 @@ def study_list(cls, user, dev_server=False, silent=False):
         >>> print(studies)
         """
         instance = cls(user, dev_server=dev_server, silent=silent)
-        return instance.get_study_list()
+        return instance.get_study_list()