Merge pull request #2912 from chaoss/main

Updating Dev Branch

CITATION.cff

@@ -0,0 +1,14 @@
+# This CITATION.cff reference content was generated from Zotero.
+cff-version: 1.2.0
+message: "If you use this software, please cite it as below."
+authors:
+    - family-names: Goggins
+      given-names: Sean
+    - family-names: Lumbard
+      given-names: Kevin
+    - family-names: Germonprez
+      given-names: Matt
+title: "Open Source Community Health: Analytical Metrics and Their Corresponding Narratives"
+doi: 10.1109/SoHeal52568.2021.00010
+date-released: 2021
+url: https://www.seangoggins.net/wp-content/plugins/zotpress/lib/request/request.dl.php?api_user_id=655145&dlkey=HNG22ZSU&content_type=application/pdf

README.md

@@ -1,4 +1,4 @@
-# Augur NEW Release v0.76.1
+# Augur NEW Release v0.76.2
 
 Augur is primarily a data engineering tool that makes it possible for data scientists to gather open source software community data. Less data carpentry for everyone else! 
 The primary way of looking at Augur data is through [8Knot](https://github.com/oss-aspen/8knot) ... A public instance of 8Knot is available at https://metrix.chaoss.io ... That is tied to a public instance of Augur at https://ai.chaoss.io 
@@ -10,7 +10,7 @@ The primary way of looking at Augur data is through [8Knot](https://github.com/o
 ## NEW RELEASE ALERT!
 ### [If you want to jump right in, updated docker build/compose and bare metal installation instructions are available here](docs/new-install.md)
 
-Augur is now releasing a dramatically improved new version to the main branch. It is also available here: https://github.com/chaoss/augur/releases/tag/v0.76.1  
+Augur is now releasing a dramatically improved new version to the main branch. It is also available here: https://github.com/chaoss/augur/releases/tag/v0.76.2 
 
 - The `main` branch is a stable version of our new architecture, which features:
   - Dramatic improvement in the speed of large scale data collection (100,000+ repos). All data is obtained for 100k+ repos within 2 weeks.

augur/api/metrics/deps.py

@@ -77,4 +77,162 @@ def deps(repo_group_id, repo_id=None, period='day', begin_date=None, end_date=No
     return results
 
 
+@register_metric()
+def libyear(repo_group_id, repo_id=None, period='day', begin_date=None, end_date=None):
+    """
+    Returns a list of all the dependencies in a project/repo/repo_group.
+
+    :param repo_id: The repository's id
+    :param repo_group_id: The repository's group id
+    :param period: To set the periodicity to 'day', 'week', 'month' or 'year', defaults to 'day'
+    :param begin_date: Specifies the begin date, defaults to '1970-1-1 00:00:00'
+    :param end_date: Specifies the end date, defaults to datetime.now()
+    :return: DataFrame of persons/period
+    """
+
+    if not begin_date:
+        begin_date = '1970-1-1 00:00:01'
+    if not end_date:
+        end_date = datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S')
+
+    if repo_id:
+
+        libyearSQL = s.sql.text("""
+            SELECT
+                rg_name,
+                repo_group_id,
+                repo_name,
+                d.repo_id,
+                repo_git,
+                forked_from,
+                repo_archived,
+                c.name, 
+                c.libyear,
+                MAX ( C.data_collection_date ) AS most_recent_collection
+            FROM
+                (
+                SELECT A.rg_name AS rg_name,
+                    A.repo_group_id AS repo_group_id,
+                    b.repo_name AS repo_name,
+                    b.repo_id AS repo_id,
+                    b.repo_git AS repo_git,
+                    b.forked_from AS forked_from,
+                    b.repo_archived AS repo_archived 
+                FROM
+                    repo_groups A,
+                    repo b 
+                WHERE
+                    A.repo_group_id = b.repo_group_id 
+                ORDER BY
+                    rg_name,
+                    repo_name 
+                ) d,
+                (
+                SELECT DISTINCT
+                    f.repo_id,
+                    f.NAME,
+                    f.libyear, 
+                    f.data_collection_date
+                FROM
+                    ( SELECT repo_id, NAME, MAX ( data_collection_date ) AS data_collection_date FROM augur_data.repo_deps_libyear WHERE repo_id = :repo_id GROUP BY repo_id, NAME ORDER BY NAME ) e,
+                    augur_data.repo_deps_libyear f 
+                WHERE
+                    e.data_collection_date = f.data_collection_date and 
+                    e.repo_id = f.repo_id 
+                ORDER BY
+                NAME 
+                ) C 
+            WHERE
+                d.repo_id = C.repo_id 
+                AND C.repo_id = :repo_id
+            GROUP BY
+                rg_name,
+                repo_git,
+                repo_group_id,
+                repo_name,
+                d.repo_id,
+                forked_from,
+                repo_archived, 
+                c.name, 
+                c.libyear
+            ORDER BY
+                repo_id;
+            """)
+
+        with current_app.engine.connect() as conn:
+            results = pd.read_sql(libyearSQL, conn, params={'repo_id': repo_id})    	
+
+    else:
+
+        libyearSQL = s.sql.text("""
+            Select w.* from 
+            (
+            SELECT
+                rg_name,
+                repo_group_id,
+                repo_name,
+                d.repo_id,
+                repo_git,
+                forked_from,
+                repo_archived,
+                c.name, 
+                c.libyear,
+                MAX ( C.data_collection_date ) AS most_recent_collection
+            FROM
+                (
+                SELECT A.rg_name AS rg_name,
+                    A.repo_group_id AS repo_group_id,
+                    b.repo_name AS repo_name,
+                    b.repo_id AS repo_id,
+                    b.repo_git AS repo_git,
+                    b.forked_from AS forked_from,
+                    b.repo_archived AS repo_archived 
+                FROM
+                    repo_groups A,
+                    repo b 
+                WHERE
+                    A.repo_group_id = b.repo_group_id 
+                ORDER BY
+                    rg_name,
+                    repo_name 
+                ) d,
+                (
+                SELECT DISTINCT
+                    f.repo_id,
+                    f.NAME,
+                    f.libyear, 
+                    f.data_collection_date
+                FROM
+                    ( SELECT repo_id, NAME, MAX ( data_collection_date ) AS data_collection_date FROM augur_data.repo_deps_libyear GROUP BY repo_id, NAME ORDER BY NAME ) e,
+                    augur_data.repo_deps_libyear f 
+                WHERE
+                    e.data_collection_date = f.data_collection_date and 
+                    e.repo_id = f.repo_id 
+                ORDER BY
+                NAME 
+                ) C 
+            WHERE
+                d.repo_id = C.repo_id 
+            GROUP BY
+                rg_name,
+                repo_git,
+                repo_group_id,
+                repo_name,
+                d.repo_id,
+                forked_from,
+                repo_archived, 
+                c.name, 
+                c.libyear
+            ORDER BY
+                repo_id) w, 
+                repo_groups y, 
+                repo z 
+            where w.repo_id=z.repo_id and 
+            y.repo_group_id=z.repo_group_id
+            and z.repo_group_id = :repo_group_id
+            """)
+
+        with current_app.engine.connect() as conn:
+            results = pd.read_sql(libyearSQL, conn, params={'repo_group_id': repo_group_id})
+    return results
 

augur/application/db/lib.py

@@ -182,6 +182,22 @@ def get_working_commits_by_repo_id(repo_id):
 
     return working_commits
 
+def get_missing_commit_message_hashes(repo_id):
+
+    fetch_missing_hashes_sql = s.sql.text("""
+    SELECT DISTINCT cmt_commit_hash FROM commits
+    WHERE repo_id=:repo_id 
+    AND cmt_commit_hash NOT IN 
+    (SELECT DISTINCT cmt_hash FROM commit_messages WHERE repo_id=:repo_id);
+    """).bindparams(repo_id=repo_id)
+
+    try:
+        missing_commit_hashes = fetchall_data_from_sql_text(fetch_missing_hashes_sql)
+    except:
+        missing_commit_hashes = []
+
+    return missing_commit_hashes
+
 def get_worker_oauth_keys(platform: str):
 
     with get_session() as session:

augur/tasks/git/facade_tasks.py

@@ -1,9 +1,11 @@
 #SPDX-License-Identifier: MIT
 
 import logging
+import datetime
 from celery import group, chain
 
-from augur.application.db.lib import get_session, get_repo_by_repo_git, get_repo_by_repo_id, remove_working_commits_by_repo_id_and_hashes, get_working_commits_by_repo_id, facade_bulk_insert_commits, bulk_insert_dicts
+from subprocess import check_output
+from augur.application.db.lib import get_session, get_repo_by_repo_git, get_repo_by_repo_id, remove_working_commits_by_repo_id_and_hashes, get_working_commits_by_repo_id, facade_bulk_insert_commits, bulk_insert_dicts, get_missing_commit_message_hashes
 
 from augur.tasks.git.util.facade_worker.facade_worker.utilitymethods import trim_commits
 from augur.tasks.git.util.facade_worker.facade_worker.utilitymethods import get_absolute_repo_path, get_parent_commits_set, get_existing_commits_set
@@ -157,6 +159,50 @@ def facade_start_contrib_analysis_task():
     facade_helper.update_status('Updating Contributors')
     facade_helper.log_activity('Info', 'Updating Contributors with commits')
 
+@celery.task(base=AugurFacadeRepoCollectionTask)
+def facade_fetch_missing_commit_messages(repo_git):
+    logger = logging.getLogger(facade_fetch_missing_commit_messages.__name__)
+    facade_helper = FacadeHelper(logger)
+
+    repo = get_repo_by_repo_git(repo_git)
+
+    logger.debug(f"Fetching missing commit message records for repo {repo_git}")
+
+    missing_message_hashes = get_missing_commit_message_hashes(repo.repo_id)
+
+    to_insert = []
+
+    for hash in missing_message_hashes:
+        #Get the huge list of commits to process.
+        absolute_path = get_absolute_repo_path(facade_helper.repo_base_directory, repo.repo_id, repo.repo_path, repo.repo_name)
+        repo_loc = (f"{absolute_path}/.git")
+
+        try: 
+            commit_message = check_output(
+                f"git --git-dir {repo_loc} log --format=%B -n 1 {hash}".split()
+            ).decode('utf-8').strip()
+
+            msg_record = {
+                'repo_id' : repo.repo_id,
+                'cmt_msg' : commit_message,
+                'cmt_hash' : hash,
+                'tool_source' : 'Facade',
+                'tool_version' : '0.78?',
+                'data_source' : 'git',
+                'data_collection_date' : datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S')
+            }
+
+            if len(to_insert) >= 1000:
+                bulk_insert_dicts(logger,to_insert, CommitMessage, ["repo_id","cmt_hash"])
+                to_insert = []
+
+            to_insert.append(msg_record)
+        except Exception as e: 
+            logger.info(f'The exception is : {e}.')
+
+    if to_insert:
+        bulk_insert_dicts(logger, to_insert, CommitMessage, ["repo_id","cmt_hash"])
+
 
 #enable celery multithreading
 @celery.task(base=AugurFacadeRepoCollectionTask)
@@ -354,6 +400,7 @@ def generate_analysis_sequence(logger,repo_git, facade_helper):
 
     analysis_sequence.append(trim_commits_post_analysis_facade_task.si(repo_git))
 
+    analysis_sequence.append(facade_fetch_missing_commit_messages.si(repo_git))
 
     analysis_sequence.append(facade_analysis_end_facade_task.si())
 

docs/new-install.md

@@ -76,8 +76,9 @@ alter database augur owner to augur;
 
 **If you're using PostgreSQL 15 or later**, default database permissions will prevent Augur's installer from configuring the database. Add one last line after the above to fix this:
 ```sql
-set search_path=augur; 
-grant all privileges on schema public to augur with grant option; 
+\connect augur
+GRANT ALL PRIVILEGES ON ALL TABLES IN SCHEMA public TO augur;
+GRANT ALL PRIVILEGES ON SCHEMA public TO augur;
 ```
 
 After that, return to your user by exiting `psql`

docs/rabbitmq-troubleshooting.md

@@ -0,0 +1,6 @@
+#Troubleshooting RabbitMQ Issues: 
+
+set rabbitmq consumer timeout to 200 hours
+```
+sudo rabbitmqctl eval 'application:set_env(rabbit, consumer_timeout, 720000000).'
+```

metadata.py

@@ -5,8 +5,8 @@
 
 __short_description__ = "Python 3 package for free/libre and open-source software community metrics, models & data collection"
 
-__version__ = "0.76.1"
-__release__ = "v0.76.1 (For the Good of the Data)"
+__version__ = "0.76.2"
+__release__ = "v0.76.2 (Pumpkin Space)"
 
 __license__ = "MIT"
 __copyright__ = "University of Missouri, University of Nebraska-Omaha, CHAOSS, Brian Warner & Augurlabs 2024"