From 1bb3858d6177437f2beadd0649fdcefcd2c158cd Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Mon, 20 Jan 2025 13:47:20 +0100
Subject: [PATCH 01/21] Implementation of publish experiment feature

---
 deep_code/constants.py                  |  4 ++
 deep_code/tools/publish.py              | 76 +++++++++++++++++++-
 deep_code/utils/ogc_api_record.py       | 93 +++++++++++++++++++++++++
 deep_code/utils/ogc_record_generator.py | 53 ++++++++++++++
 pyproject.toml                          |  3 +-
 5 files changed, 226 insertions(+), 3 deletions(-)
 create mode 100644 deep_code/utils/ogc_api_record.py
 create mode 100644 deep_code/utils/ogc_record_generator.py

diff --git a/deep_code/constants.py b/deep_code/constants.py
index 68982bc..d40d728 100644
--- a/deep_code/constants.py
+++ b/deep_code/constants.py
@@ -9,3 +9,7 @@
 OSC_REPO_OWNER = "ESA-EarthCODE"
 OSC_REPO_NAME = "open-science-catalog-metadata-testing"
 OSC_BRANCH_NAME = "add-new-collection"
+DEFAULT_THEME_SCHEME = (
+    "https://gcmd.earthdata.nasa.gov/kms/concepts/concept_scheme/sciencekeywords"
+)
+OGC_API_RECORD_SPEC = "http://www.opengis.net/spec/ogcapi-records-1/1.0/req/record-core"
diff --git a/deep_code/tools/publish.py b/deep_code/tools/publish.py
index 26b49f3..27eff98 100644
--- a/deep_code/tools/publish.py
+++ b/deep_code/tools/publish.py
@@ -4,13 +4,16 @@
 # Permissions are hereby granted under the terms of the MIT License:
 # https://opensource.org/licenses/MIT.
 
-import fsspec
 import logging
+
+import fsspec
 import yaml
 
-from deep_code.constants import OSC_REPO_OWNER, OSC_REPO_NAME, OSC_BRANCH_NAME
+from deep_code.constants import OSC_BRANCH_NAME, OSC_REPO_NAME, OSC_REPO_OWNER
 from deep_code.utils.dataset_stac_generator import OSCProductSTACGenerator
 from deep_code.utils.github_automation import GitHubAutomation
+from deep_code.utils.ogc_record_generator import OSCWorkflowOGCApiRecordGenerator
+from utils.ogc_api_record import OgcRecord
 
 logger = logging.getLogger(__name__)
 logging.basicConfig(level=logging.INFO)
@@ -99,3 +102,72 @@ def publish_dataset(self, dataset_config_path: str):
 
         finally:
             self.github_automation.clean_up()
+
+
+class WorkflowPublisher:
+    """Publishes workflow to the OSC GitHub repository.
+
+        Credentials must be provided via a hidden file named `.gitaccess`, located in
+        the root of the repository. This file is expected to contain YAML of the form:
+
+            github-username: "YOUR_GITHUB_USERNAME"
+            github-token: "YOUR_GITHUB_PERSONAL_ACCESS_TOKEN"
+        """
+
+    def __init__(self):
+        with fsspec.open(".gitaccess", "r") as file:
+            git_config = yaml.safe_load(file) or {}
+
+        self.github_username = git_config.get("github-username")
+        self.github_token = git_config.get("github-token")
+
+        if not self.github_username or not self.github_token:
+            raise ValueError("GitHub credentials are missing in the `.gitaccess` file.")
+
+        self.github_automation = GitHubAutomation(
+            self.github_username, self.github_token, OSC_REPO_OWNER, OSC_REPO_NAME
+        )
+
+    def publish_workflow(self, workflow_config_path: str):
+
+        with fsspec.open(workflow_config_path, "r") as file:
+            workflow_config = yaml.safe_load(file)
+
+        try:
+            logger.info("Generating OGC API Record for the workflow...")
+            workflow_id = workflow_config.get("workflow_id")
+            properties_list = workflow_config.get("properties", [])
+
+
+            contacts = workflow_config.get("contact", [])
+            rg = OSCWorkflowOGCApiRecordGenerator()
+            wf_record_properties = rg.build_record_properties(properties_list, contacts)
+
+            links = workflow_config.get("links")
+            ogc_record = OgcRecord(
+                id=workflow_id,
+                type="Feature",
+                time={},
+                properties=wf_record_properties,
+                links=links,
+            )
+
+            file_path = f"workflow/{workflow_id}/collection.json"
+            logger.info("Automating GitHub tasks...")
+            self.github_automation.fork_repository()
+            self.github_automation.clone_repository()
+            OSC_NEW_BRANCH_NAME = OSC_BRANCH_NAME + "-" + workflow_id
+            self.github_automation.create_branch(OSC_NEW_BRANCH_NAME)
+            self.github_automation.add_file(file_path, ogc_record.to_dict())
+            self.github_automation.commit_and_push(
+                OSC_NEW_BRANCH_NAME, f"Add new collection:{workflow_id}"
+            )
+            pr_url = self.github_automation.create_pull_request(
+                OSC_NEW_BRANCH_NAME,
+                f"Add new collection",
+                "This PR adds a new workflow to the OSC repository.",
+            )
+            logger.info(f"Pull request created: {pr_url}")
+
+        finally:
+            self.github_automation.clean_up()
diff --git a/deep_code/utils/ogc_api_record.py b/deep_code/utils/ogc_api_record.py
new file mode 100644
index 0000000..da783e0
--- /dev/null
+++ b/deep_code/utils/ogc_api_record.py
@@ -0,0 +1,93 @@
+from typing import Any, Optional
+
+from xrlint.util.codec import JsonSerializable, MappingConstructible
+
+from deep_code.constants import OGC_API_RECORD_SPEC
+
+
+class Contact(MappingConstructible["Contact"], JsonSerializable):
+    def __init__(
+        self,
+        name: str,
+        organization: str,
+        position: str | None = "",
+        links: list[dict[str, Any]] | None = None,
+        contactInstructions: str | None = "",
+        roles: list[str] = None,
+    ):
+        self.name = name
+        self.organization = organization
+        self.position = position
+        self.links = links or []
+        self.contactInstructions = contactInstructions
+        self.roles = roles or ["principal investigator"]
+
+
+class ThemeConcept(MappingConstructible["ThemeConcept"], JsonSerializable):
+    def __init__(self, id: str):
+        self.id = id
+
+
+class Theme(MappingConstructible["Theme"], JsonSerializable):
+    def __init__(self, concepts: list[ThemeConcept], scheme: str):
+        self.concepts = concepts
+        self.scheme = scheme
+
+
+class JupyterKernelInfo(MappingConstructible["RecordProperties"], JsonSerializable):
+    def __init__(self, name: str, python_version: float, env_file: str):
+        self.name = name
+        self.python_version = python_version
+        self.env_file = env_file
+
+
+class RecordProperties(MappingConstructible["RecordProperties"], JsonSerializable):
+    def __init__(
+        self,
+        created: str,
+        type: str,
+        title: str,
+        description: str,
+        jupyter_kernel_info: JupyterKernelInfo,
+        updated: str = None,
+        contacts: list[Contact] = None,
+        themes: list[Theme] = None,
+        keywords: list[str] | None = None,
+        formats: list[dict] | None = None,
+        license: str = None,
+    ):
+        self.created = created
+        self.updated = updated
+        self.type = type
+        self.title = title
+        self.description = description
+        self.jupyter_kernel_info = jupyter_kernel_info
+        self.keywords = keywords or []
+        self.contacts = contacts
+        self.themes = themes
+        self.formats = formats or []
+        self.license = license
+
+
+class OgcRecord(MappingConstructible["OgcRecord"], JsonSerializable):
+    def __init__(
+        self,
+        id: str,
+        type: str,
+        time: dict,
+        properties: RecordProperties,
+        links: list[dict],
+        linkTemplates: list = [],
+        conformsTo: list[str] = None,
+        geometry: Optional[Any] = None,
+    ):
+        if conformsTo is None:
+            conformsTo = [OGC_API_RECORD_SPEC]
+        self.id = id
+        self.type = type
+        self.conformsTo = conformsTo
+        self.time = time
+        self.geometry = geometry
+        self.properties = properties
+        self.linkTemplates = linkTemplates
+        self.links = links
diff --git a/deep_code/utils/ogc_record_generator.py b/deep_code/utils/ogc_record_generator.py
new file mode 100644
index 0000000..d323a93
--- /dev/null
+++ b/deep_code/utils/ogc_record_generator.py
@@ -0,0 +1,53 @@
+#!/usr/bin/env python3
+
+# Copyright (c) 2025 by Brockmann Consult GmbH
+# Permissions are hereby granted under the terms of the MIT License:
+# https://opensource.org/licenses/MIT.
+
+from datetime import datetime, timezone
+
+from constants import DEFAULT_THEME_SCHEME
+from deep_code.utils.ogc_api_record import (
+    Contact,
+    RecordProperties,
+    Theme,
+    ThemeConcept,
+)
+
+
+class OSCWorkflowOGCApiRecordGenerator:
+    """Generates OGC API record for a workflow
+    """
+    @staticmethod
+    def build_contact_objects(contacts_list: list[dict]) -> list[Contact]:
+        """Build a list of Contact objects from a list of contact dictionaries.
+            Uses the inherited MappingConstructible logic to parse each dict.
+
+            Args:
+                contacts_list: A list of dictionaries, each containing contact information.
+
+            Returns:
+                A list of Contact instances.
+            """
+        return [Contact.from_value(cdict) for cdict in contacts_list]
+
+    @staticmethod
+    def build_theme(osc_themes: list[str]) -> Theme:
+        """Convert each string into a ThemeConcept
+        """
+        concepts = [ThemeConcept(id=theme_str) for theme_str in osc_themes]
+        return Theme(concepts=concepts, scheme=DEFAULT_THEME_SCHEME)
+
+    def build_record_properties(self, properties, contacts) -> RecordProperties:
+        """Build a RecordProperties object from a list of single-key property dicts
+        """
+        now_iso = datetime.now(timezone.utc).isoformat()
+        properties.update({"created": now_iso})
+        properties.update({"updated": now_iso})
+        themes_list = properties.get("themes", [])
+        properties.update({"contacts": self.build_contact_objects(contacts)})
+        if themes_list:
+            theme_obj = self.build_theme(themes_list)
+            properties.update({"themes": [theme_obj]})
+        properties.setdefault("type", "workflow")
+        return RecordProperties.from_value(properties)
diff --git a/pyproject.toml b/pyproject.toml
index 057f7b8..efac413 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -26,7 +26,8 @@ dependencies = [
     "pandas",
     "pystac",
     "pyyaml",
-    "xcube-core"
+    "xcube-core",
+    "xrlint",
 ]
 
 [tool.setuptools.dynamic]

From de3c504341a0eec28f24766ef83987a79cd0f441 Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Mon, 20 Jan 2025 13:47:51 +0100
Subject: [PATCH 02/21] code formatting

---
 deep_code/cli/publish.py                             | 5 +----
 deep_code/tests/tools/test_publish.py                | 3 ++-
 deep_code/tests/utils/test_dataset_stac_generator.py | 4 ++--
 deep_code/tests/utils/test_github_automation.py      | 5 +++--
 deep_code/tests/utils/test_osc_extension.py          | 2 ++
 deep_code/utils/dataset_stac_generator.py            | 2 +-
 deep_code/utils/github_automation.py                 | 3 ++-
 deep_code/utils/osc_extension.py                     | 6 +++---
 8 files changed, 16 insertions(+), 14 deletions(-)

diff --git a/deep_code/cli/publish.py b/deep_code/cli/publish.py
index 48b1e63..a9da0f3 100644
--- a/deep_code/cli/publish.py
+++ b/deep_code/cli/publish.py
@@ -10,10 +10,7 @@
 
 
 @click.command(name="publish-dataset")
-@click.argument(
-    "dataset_config",
-    type=click.Path(exists=True)
-)
+@click.argument("dataset_config", type=click.Path(exists=True))
 def publish_dataset(dataset_config):
     """Request publishing a dataset to the open science catalogue.
     """
diff --git a/deep_code/tests/tools/test_publish.py b/deep_code/tests/tools/test_publish.py
index 47c9961..ba306b8 100644
--- a/deep_code/tests/tools/test_publish.py
+++ b/deep_code/tests/tools/test_publish.py
@@ -1,5 +1,6 @@
+from unittest.mock import MagicMock, mock_open, patch
+
 import pytest
-from unittest.mock import patch, MagicMock, mock_open
 
 from deep_code.tools.publish import DatasetPublisher
 
diff --git a/deep_code/tests/utils/test_dataset_stac_generator.py b/deep_code/tests/utils/test_dataset_stac_generator.py
index 12321b2..f2ef71d 100644
--- a/deep_code/tests/utils/test_dataset_stac_generator.py
+++ b/deep_code/tests/utils/test_dataset_stac_generator.py
@@ -1,10 +1,10 @@
 import os
+import unittest
 from datetime import datetime
+from unittest.mock import MagicMock, patch
 
 import numpy as np
 from pystac import Collection
-import unittest
-from unittest.mock import patch, MagicMock
 from xarray import Dataset
 
 from deep_code.utils.dataset_stac_generator import OSCProductSTACGenerator
diff --git a/deep_code/tests/utils/test_github_automation.py b/deep_code/tests/utils/test_github_automation.py
index 58acc09..6a66868 100644
--- a/deep_code/tests/utils/test_github_automation.py
+++ b/deep_code/tests/utils/test_github_automation.py
@@ -1,7 +1,8 @@
+import json
 import unittest
-from unittest.mock import patch, MagicMock
 from pathlib import Path
-import json
+from unittest.mock import MagicMock, patch
+
 from deep_code.utils.github_automation import GitHubAutomation
 
 
diff --git a/deep_code/tests/utils/test_osc_extension.py b/deep_code/tests/utils/test_osc_extension.py
index 66300cc..c11e3f8 100644
--- a/deep_code/tests/utils/test_osc_extension.py
+++ b/deep_code/tests/utils/test_osc_extension.py
@@ -1,5 +1,7 @@
 import unittest
+
 from pystac import Collection, Extent, SpatialExtent, TemporalExtent
+
 from deep_code.utils.osc_extension import OscExtension
 
 
diff --git a/deep_code/utils/dataset_stac_generator.py b/deep_code/utils/dataset_stac_generator.py
index 21f4cf8..1bd0064 100644
--- a/deep_code/utils/dataset_stac_generator.py
+++ b/deep_code/utils/dataset_stac_generator.py
@@ -4,8 +4,8 @@
 # Permissions are hereby granted under the terms of the MIT License:
 # https://opensource.org/licenses/MIT.
 
-import os
 import logging
+import os
 from datetime import datetime, timezone
 
 import pandas as pd
diff --git a/deep_code/utils/github_automation.py b/deep_code/utils/github_automation.py
index d934d2a..4ad6e38 100644
--- a/deep_code/utils/github_automation.py
+++ b/deep_code/utils/github_automation.py
@@ -7,10 +7,11 @@
 import json
 import logging
 import os
-import requests
 import subprocess
 from pathlib import Path
 
+import requests
+
 
 class GitHubAutomation:
     """Automates GitHub operations needed to create a Pull Request.
diff --git a/deep_code/utils/osc_extension.py b/deep_code/utils/osc_extension.py
index 6aa7519..8a777de 100644
--- a/deep_code/utils/osc_extension.py
+++ b/deep_code/utils/osc_extension.py
@@ -7,10 +7,10 @@
 from typing import Literal
 
 import pystac
-from pystac import SpatialExtent, TemporalExtent, Extent
-from pystac.extensions.base import PropertiesExtension, ExtensionManagementMixin
+from pystac import Extent, SpatialExtent, TemporalExtent
+from pystac.extensions.base import ExtensionManagementMixin, PropertiesExtension
 
-from deep_code.constants import OSC_SCHEMA_URI, CF_SCHEMA_URI
+from deep_code.constants import CF_SCHEMA_URI, OSC_SCHEMA_URI
 
 
 class OscExtension(

From e1c5f472db328d701c5967dcd33cbf145ab78969 Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Mon, 20 Jan 2025 14:12:39 +0100
Subject: [PATCH 03/21] new constant for wf branch

---
 deep_code/constants.py     |  1 +
 deep_code/tools/publish.py | 17 ++++++++++++-----
 2 files changed, 13 insertions(+), 5 deletions(-)

diff --git a/deep_code/constants.py b/deep_code/constants.py
index d40d728..8a17bbc 100644
--- a/deep_code/constants.py
+++ b/deep_code/constants.py
@@ -13,3 +13,4 @@
     "https://gcmd.earthdata.nasa.gov/kms/concepts/concept_scheme/sciencekeywords"
 )
 OGC_API_RECORD_SPEC = "http://www.opengis.net/spec/ogcapi-records-1/1.0/req/record-core"
+WF_BRANCH_NAME = "add-new-workflow"
diff --git a/deep_code/tools/publish.py b/deep_code/tools/publish.py
index 27eff98..84ea29d 100644
--- a/deep_code/tools/publish.py
+++ b/deep_code/tools/publish.py
@@ -9,7 +9,8 @@
 import fsspec
 import yaml
 
-from deep_code.constants import OSC_BRANCH_NAME, OSC_REPO_NAME, OSC_REPO_OWNER
+from deep_code.constants import OSC_BRANCH_NAME, OSC_REPO_NAME, OSC_REPO_OWNER, \
+    WF_BRANCH_NAME
 from deep_code.utils.dataset_stac_generator import OSCProductSTACGenerator
 from deep_code.utils.github_automation import GitHubAutomation
 from deep_code.utils.ogc_record_generator import OSCWorkflowOGCApiRecordGenerator
@@ -156,14 +157,15 @@ def publish_workflow(self, workflow_config_path: str):
             logger.info("Automating GitHub tasks...")
             self.github_automation.fork_repository()
             self.github_automation.clone_repository()
-            OSC_NEW_BRANCH_NAME = OSC_BRANCH_NAME + "-" + workflow_id
-            self.github_automation.create_branch(OSC_NEW_BRANCH_NAME)
+            # WF_NEW_BRANCH_NAME = WF_BRANCH_NAME + "-" + workflow_id
+            WF_NEW_BRANCH_NAME = WF_BRANCH_NAME
+            self.github_automation.create_branch(WF_NEW_BRANCH_NAME)
             self.github_automation.add_file(file_path, ogc_record.to_dict())
             self.github_automation.commit_and_push(
-                OSC_NEW_BRANCH_NAME, f"Add new collection:{workflow_id}"
+                WF_NEW_BRANCH_NAME, f"Add new workflow:{workflow_id}"
             )
             pr_url = self.github_automation.create_pull_request(
-                OSC_NEW_BRANCH_NAME,
+                WF_NEW_BRANCH_NAME,
                 f"Add new collection",
                 "This PR adds a new workflow to the OSC repository.",
             )
@@ -171,3 +173,8 @@ def publish_workflow(self, workflow_config_path: str):
 
         finally:
             self.github_automation.clean_up()
+
+if __name__ == '__main__':
+    wp = WorkflowPublisher()
+    wp.publish_workflow("/home/tejas/bc/projects/deepesdl/deep-code/workflow-config"
+                        ".yaml")

From cb7c20616d5df1d10a9620136debe8354e933071 Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Mon, 20 Jan 2025 15:32:34 +0100
Subject: [PATCH 04/21] ordered imports

---
 deep_code/tools/publish.py              | 13 +++++--------
 deep_code/utils/ogc_record_generator.py |  9 +++------
 deep_code/utils/osc_extension.py        |  3 ++-
 3 files changed, 10 insertions(+), 15 deletions(-)

diff --git a/deep_code/tools/publish.py b/deep_code/tools/publish.py
index 84ea29d..a5f9c87 100644
--- a/deep_code/tools/publish.py
+++ b/deep_code/tools/publish.py
@@ -8,13 +8,14 @@
 
 import fsspec
 import yaml
+from utils.ogc_api_record import OgcRecord
 
-from deep_code.constants import OSC_BRANCH_NAME, OSC_REPO_NAME, OSC_REPO_OWNER, \
-    WF_BRANCH_NAME
+from deep_code.constants import (OSC_BRANCH_NAME, OSC_REPO_NAME,
+                                 OSC_REPO_OWNER, WF_BRANCH_NAME)
 from deep_code.utils.dataset_stac_generator import OSCProductSTACGenerator
 from deep_code.utils.github_automation import GitHubAutomation
-from deep_code.utils.ogc_record_generator import OSCWorkflowOGCApiRecordGenerator
-from utils.ogc_api_record import OgcRecord
+from deep_code.utils.ogc_record_generator import \
+    OSCWorkflowOGCApiRecordGenerator
 
 logger = logging.getLogger(__name__)
 logging.basicConfig(level=logging.INFO)
@@ -174,7 +175,3 @@ def publish_workflow(self, workflow_config_path: str):
         finally:
             self.github_automation.clean_up()
 
-if __name__ == '__main__':
-    wp = WorkflowPublisher()
-    wp.publish_workflow("/home/tejas/bc/projects/deepesdl/deep-code/workflow-config"
-                        ".yaml")
diff --git a/deep_code/utils/ogc_record_generator.py b/deep_code/utils/ogc_record_generator.py
index d323a93..25e8b7f 100644
--- a/deep_code/utils/ogc_record_generator.py
+++ b/deep_code/utils/ogc_record_generator.py
@@ -7,12 +7,9 @@
 from datetime import datetime, timezone
 
 from constants import DEFAULT_THEME_SCHEME
-from deep_code.utils.ogc_api_record import (
-    Contact,
-    RecordProperties,
-    Theme,
-    ThemeConcept,
-)
+
+from deep_code.utils.ogc_api_record import (Contact, RecordProperties, Theme,
+                                            ThemeConcept)
 
 
 class OSCWorkflowOGCApiRecordGenerator:
diff --git a/deep_code/utils/osc_extension.py b/deep_code/utils/osc_extension.py
index 8a777de..04bbb7f 100644
--- a/deep_code/utils/osc_extension.py
+++ b/deep_code/utils/osc_extension.py
@@ -8,7 +8,8 @@
 
 import pystac
 from pystac import Extent, SpatialExtent, TemporalExtent
-from pystac.extensions.base import ExtensionManagementMixin, PropertiesExtension
+from pystac.extensions.base import (ExtensionManagementMixin,
+                                    PropertiesExtension)
 
 from deep_code.constants import CF_SCHEMA_URI, OSC_SCHEMA_URI
 

From 2b7288d58511a6f06843a784cfae55eaa18d1335 Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Mon, 20 Jan 2025 15:32:52 +0100
Subject: [PATCH 05/21] unit tests

---
 deep_code/tests/utils/test_ogc_api_record.py  | 102 ++++++++++++++++++
 .../tests/utils/test_ogc_record_generator.py  |  65 +++++++++++
 2 files changed, 167 insertions(+)
 create mode 100644 deep_code/tests/utils/test_ogc_api_record.py
 create mode 100644 deep_code/tests/utils/test_ogc_record_generator.py

diff --git a/deep_code/tests/utils/test_ogc_api_record.py b/deep_code/tests/utils/test_ogc_api_record.py
new file mode 100644
index 0000000..a48aea4
--- /dev/null
+++ b/deep_code/tests/utils/test_ogc_api_record.py
@@ -0,0 +1,102 @@
+import unittest
+
+from deep_code.constants import OGC_API_RECORD_SPEC
+from deep_code.utils.ogc_api_record import (Contact, JupyterKernelInfo,
+                                            OgcRecord, RecordProperties, Theme,
+                                            ThemeConcept)
+
+
+class TestClasses(unittest.TestCase):
+
+    def test_contact_initialization(self):
+        contact = Contact(
+            name="Person-X",
+            organization="Organization X",
+            position="Researcher",
+            links=[{"url": "http://example.com", "type": "website"}],
+            contactInstructions="Contact via email",
+            roles=["developer", "reviewer"]
+        )
+
+        self.assertEqual(contact.name, "Person-X")
+        self.assertEqual(contact.organization, "Organization X")
+        self.assertEqual(contact.position, "Researcher")
+        self.assertEqual(len(contact.links), 1)
+        self.assertEqual(contact.contactInstructions, "Contact via email")
+        self.assertIn("developer", contact.roles)
+
+    def test_theme_concept_initialization(self):
+        theme_concept = ThemeConcept(id="concept1")
+        self.assertEqual(theme_concept.id, "concept1")
+
+    def test_theme_initialization(self):
+        theme_concepts = [ThemeConcept(id="concept1"), ThemeConcept(id="concept2")]
+        theme = Theme(concepts=theme_concepts, scheme="http://example.com/scheme")
+
+        self.assertEqual(len(theme.concepts), 2)
+        self.assertEqual(theme.scheme, "http://example.com/scheme")
+
+    def test_jupyter_kernel_info_initialization(self):
+        kernel_info = JupyterKernelInfo(name="Python", python_version=3.9, env_file="env.yml")
+
+        self.assertEqual(kernel_info.name, "Python")
+        self.assertEqual(kernel_info.python_version, 3.9)
+        self.assertEqual(kernel_info.env_file, "env.yml")
+
+    def test_record_properties_initialization(self):
+        kernel_info = JupyterKernelInfo(name="Python", python_version=3.9, env_file="env.yml")
+        contacts = [Contact(name="Jane Doe", organization="Org Y")]
+        themes = [Theme(concepts=[ThemeConcept(id="concept1")], scheme="scheme1")]
+
+        record_properties = RecordProperties(
+            created="2025-01-01",
+            type="dataset",
+            title="Sample Dataset",
+            description="A sample dataset",
+            jupyter_kernel_info=kernel_info,
+            updated="2025-01-02",
+            contacts=contacts,
+            themes=themes,
+            keywords=["sample", "test"],
+            formats=[{"format": "JSON"}],
+            license="CC-BY"
+        )
+
+        self.assertEqual(record_properties.created, "2025-01-01")
+        self.assertEqual(record_properties.updated, "2025-01-02")
+        self.assertEqual(record_properties.type, "dataset")
+        self.assertEqual(record_properties.title, "Sample Dataset")
+        self.assertEqual(record_properties.description, "A sample dataset")
+        self.assertEqual(record_properties.jupyter_kernel_info.name, "Python")
+        self.assertEqual(len(record_properties.contacts), 1)
+        self.assertEqual(len(record_properties.themes), 1)
+        self.assertIn("sample", record_properties.keywords)
+        self.assertEqual(record_properties.license, "CC-BY")
+
+    def test_ogc_record_initialization(self):
+        kernel_info = JupyterKernelInfo(name="Python", python_version=3.9, env_file="env.yml")
+        properties = RecordProperties(
+            created="2025-01-01",
+            type="dataset",
+            title="Sample Dataset",
+            description="A sample dataset",
+            jupyter_kernel_info=kernel_info
+        )
+
+        ogc_record = OgcRecord(
+            id="record1",
+            type="Feature",
+            time={"start": "2025-01-01T00:00:00Z", "end": "2025-01-02T00:00:00Z"},
+            properties=properties,
+            links=[{"href": "http://example.com", "rel": "self"}],
+            linkTemplates=[{"template": "http://example.com/{id}"}]
+        )
+
+        self.assertEqual(ogc_record.id, "record1")
+        self.assertEqual(ogc_record.type, "Feature")
+        self.assertEqual(ogc_record.time["start"], "2025-01-01T00:00:00Z")
+        self.assertEqual(ogc_record.properties.title, "Sample Dataset")
+        self.assertEqual(len(ogc_record.links), 1)
+        self.assertEqual(ogc_record.linkTemplates[0]["template"], "http://example.com/{id}")
+        self.assertEqual(ogc_record.conformsTo[0], OGC_API_RECORD_SPEC)
+
diff --git a/deep_code/tests/utils/test_ogc_record_generator.py b/deep_code/tests/utils/test_ogc_record_generator.py
new file mode 100644
index 0000000..dc8c816
--- /dev/null
+++ b/deep_code/tests/utils/test_ogc_record_generator.py
@@ -0,0 +1,65 @@
+import unittest
+from datetime import datetime, timezone
+
+from deep_code.constants import DEFAULT_THEME_SCHEME
+from deep_code.utils.ogc_record_generator import OSCWorkflowOGCApiRecordGenerator
+
+
+class TestOSCWorkflowOGCApiRecordGenerator(unittest.TestCase):
+
+    def test_build_contact_objects(self):
+        contacts_list = [
+            {"name": "Alice", "organization": "Org A", "position": "Researcher"},
+            {"name": "Bob", "organization": "Org B", "position": "Developer"}
+        ]
+
+        result = OSCWorkflowOGCApiRecordGenerator.build_contact_objects(contacts_list)
+
+        self.assertEqual(len(result), 2)
+        self.assertEqual(result[0].name, "Alice")
+        self.assertEqual(result[0].organization, "Org A")
+        self.assertEqual(result[0].position, "Researcher")
+        self.assertEqual(result[1].name, "Bob")
+        self.assertEqual(result[1].organization, "Org B")
+        self.assertEqual(result[1].position, "Developer")
+
+    def test_build_theme(self):
+        osc_themes = ["theme1", "theme2"]
+
+        theme = OSCWorkflowOGCApiRecordGenerator.build_theme(osc_themes)
+
+        self.assertEqual(len(theme.concepts), 2)
+        self.assertEqual(theme.concepts[0].id, "theme1")
+        self.assertEqual(theme.concepts[1].id, "theme2")
+        self.assertEqual(theme.scheme, DEFAULT_THEME_SCHEME)
+
+    def test_build_record_properties(self):
+        generator = OSCWorkflowOGCApiRecordGenerator()
+        properties = {
+            "title": "Test Workflow",
+            "description": "A test description",
+            "themes": ["theme1"],
+            "jupyter_kernel_info": {
+                "name": "deepesdl-xcube-1.7.1",
+                "python_version": 3.11,
+                "env_file": "https://git/env.yml"
+            }
+        }
+        contacts = [
+            {"name": "Alice", "organization": "Org A", "position": "Researcher"}
+        ]
+
+        record_properties = generator.build_record_properties(properties, contacts)
+
+        now_iso = datetime.now(timezone.utc).isoformat()
+
+        self.assertEqual(record_properties.title, "Test Workflow")
+        self.assertEqual(record_properties.description, "A test description")
+        self.assertEqual(len(record_properties.contacts), 1)
+        self.assertEqual(record_properties.contacts[0].name, "Alice")
+        self.assertEqual(len(record_properties.themes), 1)
+        self.assertEqual(record_properties.themes[0].concepts[0].id, "theme1")
+        self.assertEqual(record_properties.type, "workflow")
+        self.assertTrue("created" in record_properties.__dict__)
+        self.assertTrue("updated" in record_properties.__dict__)
+

From 2f3cd3fb719a63d1567a6d90b612eb72672e31d8 Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Mon, 20 Jan 2025 15:42:47 +0100
Subject: [PATCH 06/21] refactor

---
 deep_code/tests/utils/test_ogc_api_record.py  | 37 ++++++++++++-------
 .../tests/utils/test_ogc_record_generator.py  |  8 ++--
 deep_code/tools/publish.py                    | 15 ++++----
 deep_code/utils/ogc_record_generator.py       | 12 ++++--
 deep_code/utils/osc_extension.py              |  3 +-
 5 files changed, 44 insertions(+), 31 deletions(-)

diff --git a/deep_code/tests/utils/test_ogc_api_record.py b/deep_code/tests/utils/test_ogc_api_record.py
index a48aea4..52640fe 100644
--- a/deep_code/tests/utils/test_ogc_api_record.py
+++ b/deep_code/tests/utils/test_ogc_api_record.py
@@ -1,13 +1,17 @@
 import unittest
 
 from deep_code.constants import OGC_API_RECORD_SPEC
-from deep_code.utils.ogc_api_record import (Contact, JupyterKernelInfo,
-                                            OgcRecord, RecordProperties, Theme,
-                                            ThemeConcept)
+from deep_code.utils.ogc_api_record import (
+    Contact,
+    JupyterKernelInfo,
+    OgcRecord,
+    RecordProperties,
+    Theme,
+    ThemeConcept,
+)
 
 
 class TestClasses(unittest.TestCase):
-
     def test_contact_initialization(self):
         contact = Contact(
             name="Person-X",
@@ -15,7 +19,7 @@ def test_contact_initialization(self):
             position="Researcher",
             links=[{"url": "http://example.com", "type": "website"}],
             contactInstructions="Contact via email",
-            roles=["developer", "reviewer"]
+            roles=["developer", "reviewer"],
         )
 
         self.assertEqual(contact.name, "Person-X")
@@ -37,14 +41,18 @@ def test_theme_initialization(self):
         self.assertEqual(theme.scheme, "http://example.com/scheme")
 
     def test_jupyter_kernel_info_initialization(self):
-        kernel_info = JupyterKernelInfo(name="Python", python_version=3.9, env_file="env.yml")
+        kernel_info = JupyterKernelInfo(
+            name="Python", python_version=3.9, env_file="env.yml"
+        )
 
         self.assertEqual(kernel_info.name, "Python")
         self.assertEqual(kernel_info.python_version, 3.9)
         self.assertEqual(kernel_info.env_file, "env.yml")
 
     def test_record_properties_initialization(self):
-        kernel_info = JupyterKernelInfo(name="Python", python_version=3.9, env_file="env.yml")
+        kernel_info = JupyterKernelInfo(
+            name="Python", python_version=3.9, env_file="env.yml"
+        )
         contacts = [Contact(name="Jane Doe", organization="Org Y")]
         themes = [Theme(concepts=[ThemeConcept(id="concept1")], scheme="scheme1")]
 
@@ -59,7 +67,7 @@ def test_record_properties_initialization(self):
             themes=themes,
             keywords=["sample", "test"],
             formats=[{"format": "JSON"}],
-            license="CC-BY"
+            license="CC-BY",
         )
 
         self.assertEqual(record_properties.created, "2025-01-01")
@@ -74,13 +82,15 @@ def test_record_properties_initialization(self):
         self.assertEqual(record_properties.license, "CC-BY")
 
     def test_ogc_record_initialization(self):
-        kernel_info = JupyterKernelInfo(name="Python", python_version=3.9, env_file="env.yml")
+        kernel_info = JupyterKernelInfo(
+            name="Python", python_version=3.9, env_file="env.yml"
+        )
         properties = RecordProperties(
             created="2025-01-01",
             type="dataset",
             title="Sample Dataset",
             description="A sample dataset",
-            jupyter_kernel_info=kernel_info
+            jupyter_kernel_info=kernel_info,
         )
 
         ogc_record = OgcRecord(
@@ -89,7 +99,7 @@ def test_ogc_record_initialization(self):
             time={"start": "2025-01-01T00:00:00Z", "end": "2025-01-02T00:00:00Z"},
             properties=properties,
             links=[{"href": "http://example.com", "rel": "self"}],
-            linkTemplates=[{"template": "http://example.com/{id}"}]
+            linkTemplates=[{"template": "http://example.com/{id}"}],
         )
 
         self.assertEqual(ogc_record.id, "record1")
@@ -97,6 +107,7 @@ def test_ogc_record_initialization(self):
         self.assertEqual(ogc_record.time["start"], "2025-01-01T00:00:00Z")
         self.assertEqual(ogc_record.properties.title, "Sample Dataset")
         self.assertEqual(len(ogc_record.links), 1)
-        self.assertEqual(ogc_record.linkTemplates[0]["template"], "http://example.com/{id}")
+        self.assertEqual(
+            ogc_record.linkTemplates[0]["template"], "http://example.com/{id}"
+        )
         self.assertEqual(ogc_record.conformsTo[0], OGC_API_RECORD_SPEC)
-
diff --git a/deep_code/tests/utils/test_ogc_record_generator.py b/deep_code/tests/utils/test_ogc_record_generator.py
index dc8c816..f4fe372 100644
--- a/deep_code/tests/utils/test_ogc_record_generator.py
+++ b/deep_code/tests/utils/test_ogc_record_generator.py
@@ -6,11 +6,10 @@
 
 
 class TestOSCWorkflowOGCApiRecordGenerator(unittest.TestCase):
-
     def test_build_contact_objects(self):
         contacts_list = [
             {"name": "Alice", "organization": "Org A", "position": "Researcher"},
-            {"name": "Bob", "organization": "Org B", "position": "Developer"}
+            {"name": "Bob", "organization": "Org B", "position": "Developer"},
         ]
 
         result = OSCWorkflowOGCApiRecordGenerator.build_contact_objects(contacts_list)
@@ -42,8 +41,8 @@ def test_build_record_properties(self):
             "jupyter_kernel_info": {
                 "name": "deepesdl-xcube-1.7.1",
                 "python_version": 3.11,
-                "env_file": "https://git/env.yml"
-            }
+                "env_file": "https://git/env.yml",
+            },
         }
         contacts = [
             {"name": "Alice", "organization": "Org A", "position": "Researcher"}
@@ -62,4 +61,3 @@ def test_build_record_properties(self):
         self.assertEqual(record_properties.type, "workflow")
         self.assertTrue("created" in record_properties.__dict__)
         self.assertTrue("updated" in record_properties.__dict__)
-
diff --git a/deep_code/tools/publish.py b/deep_code/tools/publish.py
index a5f9c87..4c83bb9 100644
--- a/deep_code/tools/publish.py
+++ b/deep_code/tools/publish.py
@@ -8,14 +8,17 @@
 
 import fsspec
 import yaml
-from utils.ogc_api_record import OgcRecord
 
-from deep_code.constants import (OSC_BRANCH_NAME, OSC_REPO_NAME,
-                                 OSC_REPO_OWNER, WF_BRANCH_NAME)
+from deep_code.constants import (
+    OSC_BRANCH_NAME,
+    OSC_REPO_NAME,
+    OSC_REPO_OWNER,
+    WF_BRANCH_NAME,
+)
 from deep_code.utils.dataset_stac_generator import OSCProductSTACGenerator
 from deep_code.utils.github_automation import GitHubAutomation
-from deep_code.utils.ogc_record_generator import \
-    OSCWorkflowOGCApiRecordGenerator
+from deep_code.utils.ogc_api_record import OgcRecord
+from deep_code.utils.ogc_record_generator import OSCWorkflowOGCApiRecordGenerator
 
 logger = logging.getLogger(__name__)
 logging.basicConfig(level=logging.INFO)
@@ -140,7 +143,6 @@ def publish_workflow(self, workflow_config_path: str):
             workflow_id = workflow_config.get("workflow_id")
             properties_list = workflow_config.get("properties", [])
 
-
             contacts = workflow_config.get("contact", [])
             rg = OSCWorkflowOGCApiRecordGenerator()
             wf_record_properties = rg.build_record_properties(properties_list, contacts)
@@ -174,4 +176,3 @@ def publish_workflow(self, workflow_config_path: str):
 
         finally:
             self.github_automation.clean_up()
-
diff --git a/deep_code/utils/ogc_record_generator.py b/deep_code/utils/ogc_record_generator.py
index 25e8b7f..481663f 100644
--- a/deep_code/utils/ogc_record_generator.py
+++ b/deep_code/utils/ogc_record_generator.py
@@ -6,15 +6,19 @@
 
 from datetime import datetime, timezone
 
-from constants import DEFAULT_THEME_SCHEME
-
-from deep_code.utils.ogc_api_record import (Contact, RecordProperties, Theme,
-                                            ThemeConcept)
+from deep_code.constants import DEFAULT_THEME_SCHEME
+from deep_code.utils.ogc_api_record import (
+    Contact,
+    RecordProperties,
+    Theme,
+    ThemeConcept,
+)
 
 
 class OSCWorkflowOGCApiRecordGenerator:
     """Generates OGC API record for a workflow
     """
+
     @staticmethod
     def build_contact_objects(contacts_list: list[dict]) -> list[Contact]:
         """Build a list of Contact objects from a list of contact dictionaries.
diff --git a/deep_code/utils/osc_extension.py b/deep_code/utils/osc_extension.py
index 04bbb7f..8a777de 100644
--- a/deep_code/utils/osc_extension.py
+++ b/deep_code/utils/osc_extension.py
@@ -8,8 +8,7 @@
 
 import pystac
 from pystac import Extent, SpatialExtent, TemporalExtent
-from pystac.extensions.base import (ExtensionManagementMixin,
-                                    PropertiesExtension)
+from pystac.extensions.base import ExtensionManagementMixin, PropertiesExtension
 
 from deep_code.constants import CF_SCHEMA_URI, OSC_SCHEMA_URI
 

From 81256e09791723d3b4f47f0cb6f0d8b203232003 Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Mon, 20 Jan 2025 16:21:21 +0100
Subject: [PATCH 07/21] adjusted xrlint imports

---
 deep_code/utils/ogc_api_record.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/deep_code/utils/ogc_api_record.py b/deep_code/utils/ogc_api_record.py
index da783e0..437c2c8 100644
--- a/deep_code/utils/ogc_api_record.py
+++ b/deep_code/utils/ogc_api_record.py
@@ -1,6 +1,7 @@
 from typing import Any, Optional
 
-from xrlint.util.codec import JsonSerializable, MappingConstructible
+from xrlint.util.constructible import MappingConstructible
+from xrlint.util.serializable import JsonSerializable
 
 from deep_code.constants import OGC_API_RECORD_SPEC
 

From c32a09fdf324c80269efa7f3d1ef559dec22990a Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Mon, 20 Jan 2025 16:57:10 +0100
Subject: [PATCH 08/21] added cli command

---
 deep_code/cli/main.py      |  4 +++-
 deep_code/cli/publish.py   | 10 +++++++++-
 deep_code/tools/publish.py |  9 ++++++---
 3 files changed, 18 insertions(+), 5 deletions(-)

diff --git a/deep_code/cli/main.py b/deep_code/cli/main.py
index be88985..af140a4 100644
--- a/deep_code/cli/main.py
+++ b/deep_code/cli/main.py
@@ -6,7 +6,7 @@
 
 import click
 
-from deep_code.cli.publish import publish_dataset
+from deep_code.cli.publish import publish_dataset, publish_workflow
 
 
 @click.group()
@@ -16,5 +16,7 @@ def main():
 
 
 main.add_command(publish_dataset)
+main.add_command(publish_workflow)
+
 if __name__ == "__main__":
     main()
diff --git a/deep_code/cli/publish.py b/deep_code/cli/publish.py
index a9da0f3..a3d81d0 100644
--- a/deep_code/cli/publish.py
+++ b/deep_code/cli/publish.py
@@ -6,7 +6,7 @@
 
 import click
 
-from deep_code.tools.publish import DatasetPublisher
+from deep_code.tools.publish import DatasetPublisher, WorkflowPublisher
 
 
 @click.command(name="publish-dataset")
@@ -16,3 +16,11 @@ def publish_dataset(dataset_config):
     """
     publisher = DatasetPublisher()
     publisher.publish_dataset(dataset_config_path=dataset_config)
+
+
+@click.command(name="publish-workflow")
+@click.argument("workflow_metadata", type=click.Path(exists=True))
+def publish_workflow(workflow_metadata):
+
+    workflow_publisher = WorkflowPublisher()
+    workflow_publisher.publish_workflow(workflow_config_path=workflow_metadata)
diff --git a/deep_code/tools/publish.py b/deep_code/tools/publish.py
index 4c83bb9..89a041e 100644
--- a/deep_code/tools/publish.py
+++ b/deep_code/tools/publish.py
@@ -133,6 +133,10 @@ def __init__(self):
             self.github_username, self.github_token, OSC_REPO_OWNER, OSC_REPO_NAME
         )
 
+    @staticmethod
+    def _normalize_name(name: str | None) -> str | None:
+        return name.replace(" ", "-").lower() if name else None
+
     def publish_workflow(self, workflow_config_path: str):
 
         with fsspec.open(workflow_config_path, "r") as file:
@@ -140,7 +144,7 @@ def publish_workflow(self, workflow_config_path: str):
 
         try:
             logger.info("Generating OGC API Record for the workflow...")
-            workflow_id = workflow_config.get("workflow_id")
+            workflow_id = self._normalize_name(workflow_config.get("workflow_id"))
             properties_list = workflow_config.get("properties", [])
 
             contacts = workflow_config.get("contact", [])
@@ -160,8 +164,7 @@ def publish_workflow(self, workflow_config_path: str):
             logger.info("Automating GitHub tasks...")
             self.github_automation.fork_repository()
             self.github_automation.clone_repository()
-            # WF_NEW_BRANCH_NAME = WF_BRANCH_NAME + "-" + workflow_id
-            WF_NEW_BRANCH_NAME = WF_BRANCH_NAME
+            WF_NEW_BRANCH_NAME = WF_BRANCH_NAME + "-" + workflow_id
             self.github_automation.create_branch(WF_NEW_BRANCH_NAME)
             self.github_automation.add_file(file_path, ogc_record.to_dict())
             self.github_automation.commit_and_push(

From bd72aa83e5222fc07ac6690fa982573052c94fcf Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Tue, 21 Jan 2025 10:12:12 +0100
Subject: [PATCH 09/21] git merge tejas-xxx-build-varibale-reference-osc

---
 deep_code/tests/tools/test_publish.py         |  31 ++-
 .../utils/test_dataset_stac_generator.py      |  40 ++--
 deep_code/tools/publish.py                    | 130 +++++++++--
 deep_code/utils/dataset_stac_generator.py     | 208 ++++++++++++++----
 4 files changed, 325 insertions(+), 84 deletions(-)

diff --git a/deep_code/tests/tools/test_publish.py b/deep_code/tests/tools/test_publish.py
index ba306b8..4a13581 100644
--- a/deep_code/tests/tools/test_publish.py
+++ b/deep_code/tests/tools/test_publish.py
@@ -55,22 +55,21 @@ def test_publish_dataset_success(
         mock_subprocess_run,
         mock_chdir,
     ):
-
         #  Mock the YAML reads
         git_yaml_content = """
-          github-username: test-user
-          github-token: test-token
-          """
+             github-username: test-user
+             github-token: test-token
+             """
         dataset_yaml_content = """
-          dataset-id: test-dataset
-          collection-id: test-collection
-          documentation-link: http://example.com/doc
-          access-link: http://example.com/access
-          dataset-status: ongoing
-          dataset-region: Global
-          dataset-theme: ["climate"]
-          cf-parameter: []
-          """
+             dataset_id: test-dataset
+             collection_id: test-collection
+             documentation_link: http://example.com/doc
+             access_link: http://example.com/access
+             dataset_status: ongoing
+             dataset_region: Global
+             osc_theme: ["climate"]
+             cf_parameter: []
+             """
         mock_fsspec_open.side_effect = [
             mock_open(read_data=git_yaml_content)(),
             mock_open(read_data=dataset_yaml_content)(),
@@ -103,8 +102,8 @@ def test_publish_dataset_success(
             "links": [],
             "stac_version": "1.0.0",
         }
-        with patch("deep_code.tools.publish.OSCProductSTACGenerator") as mock_generator:
-            mock_generator.return_value.build_stac_collection.return_value = (
+        with patch("deep_code.tools.publish.OSCDatasetSTACGenerator") as mock_generator:
+            mock_generator.return_value.build_dataset_stac_collection.return_value = (
                 mock_collection
             )
 
@@ -112,7 +111,7 @@ def test_publish_dataset_success(
             publisher = DatasetPublisher()
             publisher.publish_dataset("/fake/path/to/dataset-config.yaml")
 
-        # 6Assert that we called git clone with /tmp/temp_repo
+        # Assert that we called git clone with /tmp/temp_repo
         # Because expanduser("~") is now patched to /tmp, the actual path is /tmp/temp_repo
         auth_url = "https://test-user:test-token@github.com/test-user/open-science-catalog-metadata-testing.git"
         mock_subprocess_run.assert_any_call(
diff --git a/deep_code/tests/utils/test_dataset_stac_generator.py b/deep_code/tests/utils/test_dataset_stac_generator.py
index f2ef71d..d4444c1 100644
--- a/deep_code/tests/utils/test_dataset_stac_generator.py
+++ b/deep_code/tests/utils/test_dataset_stac_generator.py
@@ -7,7 +7,7 @@
 from pystac import Collection
 from xarray import Dataset
 
-from deep_code.utils.dataset_stac_generator import OSCProductSTACGenerator
+from deep_code.utils.dataset_stac_generator import OSCDatasetSTACGenerator
 
 
 class TestOSCProductSTACGenerator(unittest.TestCase):
@@ -28,15 +28,31 @@ def setUp(self, mock_data_store):
             },
             attrs={"description": "Mock dataset for testing.", "title": "Mock Dataset"},
             data_vars={
-                "var1": (("time", "lat", "lon"), np.random.rand(2, 5, 10)),
-                "var2": (("time", "lat", "lon"), np.random.rand(2, 5, 10)),
+                "var1": (
+                    ("time", "lat", "lon"),
+                    np.random.rand(2, 5, 10),
+                    {
+                        "description": "dummy",
+                        "standard_name": "var1",
+                        "gcmd_keyword_url": "https://dummy",
+                    },
+                ),
+                "var2": (
+                    ("time", "lat", "lon"),
+                    np.random.rand(2, 5, 10),
+                    {
+                        "description": "dummy",
+                        "standard_name": "var2",
+                        "gcmd_keyword_url": "https://dummy",
+                    },
+                ),
             },
         )
         mock_store = MagicMock()
         mock_store.open_data.return_value = self.mock_dataset
         mock_data_store.return_value = mock_store
 
-        self.generator = OSCProductSTACGenerator(
+        self.generator = OSCDatasetSTACGenerator(
             dataset_id="mock-dataset-id",
             collection_id="mock-collection-id",
             access_link="s3://mock-bucket/mock-dataset",
@@ -66,7 +82,7 @@ def test_get_temporal_extent(self):
 
     def test_get_variables(self):
         """Test variable extraction."""
-        variables = self.generator._get_variables()
+        variables = self.generator.get_variable_ids()
         self.assertEqual(variables, ["var1", "var2"])
 
     def test_get_general_metadata(self):
@@ -78,7 +94,7 @@ def test_get_general_metadata(self):
     @patch("pystac.Collection.set_self_href")
     def test_build_stac_collection(self, mock_set_self_href, mock_add_link):
         """Test STAC collection creation."""
-        collection = self.generator.build_stac_collection()
+        collection = self.generator.build_dataset_stac_collection()
         self.assertIsInstance(collection, Collection)
         self.assertEqual(collection.id, "mock-collection-id")
         self.assertEqual(collection.description, "Mock dataset for testing.")
@@ -104,8 +120,6 @@ def test_invalid_temporal_extent(self):
         with self.assertRaises(ValueError):
             self.generator._get_temporal_extent()
 
-
-class TestOpenDataset(unittest.TestCase):
     @patch("deep_code.utils.dataset_stac_generator.new_data_store")
     @patch("deep_code.utils.dataset_stac_generator.logging.getLogger")
     def test_open_dataset_success_public_store(self, mock_logger, mock_new_data_store):
@@ -113,10 +127,10 @@ def test_open_dataset_success_public_store(self, mock_logger, mock_new_data_stor
         # Create a mock store and mock its `open_data` method
         mock_store = MagicMock()
         mock_new_data_store.return_value = mock_store
-        mock_store.open_data.return_value = "mock_dataset"
+        mock_store.open_data.return_value = self.mock_dataset
 
         # Instantiate the generator (this will implicitly call _open_dataset)
-        generator = OSCProductSTACGenerator("mock-dataset-id", "mock-collection-id")
+        generator = OSCDatasetSTACGenerator("mock-dataset-id", "mock-collection-id")
 
         # Validate that the dataset is assigned correctly
         self.assertEqual(generator.dataset, "mock_dataset")
@@ -151,13 +165,13 @@ def test_open_dataset_success_authenticated_store(
             mock_store,
             # Second call (authenticated store) returns a mock store
         ]
-        mock_store.open_data.return_value = "mock_dataset"
+        mock_store.open_data.return_value = self.mock_dataset
 
         os.environ["S3_USER_STORAGE_BUCKET"] = "mock-bucket"
         os.environ["S3_USER_STORAGE_KEY"] = "mock-key"
         os.environ["S3_USER_STORAGE_SECRET"] = "mock-secret"
 
-        generator = OSCProductSTACGenerator("mock-dataset-id", "mock-collection-id")
+        generator = OSCDatasetSTACGenerator("mock-dataset-id", "mock-collection-id")
 
         # Validate that the dataset was successfully opened with the authenticated store
         self.assertEqual(generator.dataset, "mock_dataset")
@@ -195,7 +209,7 @@ def test_open_dataset_failure(self, mock_logger, mock_new_data_store):
         os.environ["S3_USER_STORAGE_SECRET"] = "mock-secret"
 
         with self.assertRaises(ValueError) as context:
-            OSCProductSTACGenerator("mock-dataset-id", "mock-collection-id")
+            OSCDatasetSTACGenerator("mock-dataset-id", "mock-collection-id")
 
         self.assertIn(
             "Failed to open Zarr dataset with ID mock-dataset-id",
diff --git a/deep_code/tools/publish.py b/deep_code/tools/publish.py
index 89a041e..cc13353 100644
--- a/deep_code/tools/publish.py
+++ b/deep_code/tools/publish.py
@@ -5,6 +5,7 @@
 # https://opensource.org/licenses/MIT.
 
 import logging
+from pathlib import Path
 
 import fsspec
 import yaml
@@ -15,7 +16,7 @@
     OSC_REPO_OWNER,
     WF_BRANCH_NAME,
 )
-from deep_code.utils.dataset_stac_generator import OSCProductSTACGenerator
+from deep_code.utils.dataset_stac_generator import OSCDatasetSTACGenerator
 from deep_code.utils.github_automation import GitHubAutomation
 from deep_code.utils.ogc_api_record import OgcRecord
 from deep_code.utils.ogc_record_generator import OSCWorkflowOGCApiRecordGenerator
@@ -24,6 +25,76 @@
 logging.basicConfig(level=logging.INFO)
 
 
+class BasePublisher:
+    """
+    Base class providing common GitHub automation steps:
+      - Reading credentials from `.gitaccess`
+      - Setting up GitHubAutomation
+      - Forking, cloning, creating a branch
+      - Adding files, committing, pushing
+      - Creating a pull request
+    """
+
+    def __init__(self):
+        with fsspec.open(".gitaccess", "r") as file:
+            git_config = yaml.safe_load(file) or {}
+
+        self.github_username = git_config.get("github-username")
+        self.github_token = git_config.get("github-token")
+        if not self.github_username or not self.github_token:
+            raise ValueError("GitHub credentials are missing in `.gitaccess` file.")
+
+        self.github_automation = GitHubAutomation(
+            self.github_username, self.github_token, OSC_REPO_OWNER, OSC_REPO_NAME
+        )
+
+    def publish_file(
+        self,
+        branch_name: str,
+        file_path: str,
+        file_content: dict,
+        commit_message: str,
+        pr_title: str,
+        pr_body: str,
+    ) -> str:
+        """
+        Publish a single file to GitHub in a new branch and open a PR.
+
+        Args:
+            branch_name: Name of the branch to create (e.g. 'osc-branch-collectionid').
+            file_path: File path in the repo (e.g. 'products/.../collection.json').
+            file_content: The JSON/dict content to commit.
+            commit_message: Commit message.
+            pr_title: Title for the pull request.
+            pr_body: Body of the pull request.
+
+        Returns:
+            str: The URL of the created pull request.
+        """
+        try:
+            logger.info("Starting GitHub automation...")
+            self.github_automation.fork_repository()
+            self.github_automation.clone_repository()
+            self.github_automation.create_branch(branch_name)
+
+            # Add the file
+            self.github_automation.add_file(file_path, file_content)
+
+            # Commit and push
+            self.github_automation.commit_and_push(branch_name, commit_message)
+
+            # Create pull request
+            pr_url = self.github_automation.create_pull_request(
+                branch_name, pr_title, pr_body
+            )
+            logger.info(f"Pull request created at: {pr_url}")
+            return pr_url
+
+        finally:
+            # Always clean up local clone
+            self.github_automation.clean_up()
+
+
 class DatasetPublisher:
     """
     Publishes products to the OSC GitHub repository.
@@ -58,14 +129,14 @@ def publish_dataset(self, dataset_config_path: str):
         with fsspec.open(dataset_config_path, "r") as file:
             dataset_config = yaml.safe_load(file)
 
-        dataset_id = dataset_config.get("dataset-id")
-        collection_id = dataset_config.get("collection-id")
-        documentation_link = dataset_config.get("documentation-link")
-        access_link = dataset_config.get("access-link")
-        dataset_status = dataset_config.get("dataset-status")
-        osc_region = dataset_config.get("dataset-region")
-        dataset_theme = dataset_config.get("dataset-theme")
-        cf_params = dataset_config.get("cf-parameter")
+        dataset_id = dataset_config.get("dataset_id")
+        collection_id = dataset_config.get("collection_id")
+        documentation_link = dataset_config.get("documentation_link")
+        access_link = dataset_config.get("access_link")
+        dataset_status = dataset_config.get("dataset_status")
+        osc_region = dataset_config.get("osc_region")
+        osc_themes = dataset_config.get("osc_themes")
+        cf_params = dataset_config.get("cf_parameter")
 
         if not dataset_id or not collection_id:
             raise ValueError(
@@ -75,17 +146,20 @@ def publish_dataset(self, dataset_config_path: str):
 
         try:
             logger.info("Generating STAC collection...")
-            generator = OSCProductSTACGenerator(
+            generator = OSCDatasetSTACGenerator(
                 dataset_id=dataset_id,
                 collection_id=collection_id,
                 documentation_link=documentation_link,
                 access_link=access_link,
                 osc_status=dataset_status,
                 osc_region=osc_region,
-                osc_themes=dataset_theme,
+                osc_themes=osc_themes,
                 cf_params=cf_params,
             )
-            collection = generator.build_stac_collection()
+            # get variables from the datasets
+            variable_ids = generator.get_variable_ids()
+            # build STAC collection for the dataset
+            ds_collection = generator.build_dataset_stac_collection()
 
             file_path = f"products/{collection_id}/collection.json"
             logger.info("Automating GitHub tasks...")
@@ -93,13 +167,41 @@ def publish_dataset(self, dataset_config_path: str):
             self.github_automation.clone_repository()
             OSC_NEW_BRANCH_NAME = OSC_BRANCH_NAME + "-" + collection_id
             self.github_automation.create_branch(OSC_NEW_BRANCH_NAME)
-            self.github_automation.add_file(file_path, collection.to_dict())
+
+            for var_id in variable_ids:
+                var_file_path = f"variables/{var_id}/catalog.json"
+                if not self.github_automation.file_exists(var_file_path):
+                    logger.info(
+                        f"Variable catalog for {var_id} does not exist. Creating..."
+                    )
+                    var_metadata = generator.variables_metadata.get(var_id)
+                    var_catalog = generator.build_variable_catalog(var_metadata)
+                    self.github_automation.add_file(
+                        var_file_path, var_catalog.to_dict()
+                    )
+                else:
+                    logger.info(
+                        f"Variable catalog already exists for {var_id}. so add the "
+                        f"product as child link..."
+                    )
+                    full_path = (
+                        Path(self.github_automation.local_clone_dir) / var_file_path
+                    )
+                    self.github_automation.add_file(
+                        var_file_path,
+                        generator.update_existing_variable_catalog(
+                            full_path, var_id
+                        ).to_dict(),
+                    )
+
+            self.github_automation.add_file(file_path, ds_collection.to_dict())
+
             self.github_automation.commit_and_push(
                 OSC_NEW_BRANCH_NAME, f"Add new collection:{collection_id}"
             )
             pr_url = self.github_automation.create_pull_request(
                 OSC_NEW_BRANCH_NAME,
-                f"Add new collection",
+                f"Add new dataset collection",
                 "This PR adds a new collection to the repository.",
             )
 
diff --git a/deep_code/utils/dataset_stac_generator.py b/deep_code/utils/dataset_stac_generator.py
index 1bd0064..5b04c86 100644
--- a/deep_code/utils/dataset_stac_generator.py
+++ b/deep_code/utils/dataset_stac_generator.py
@@ -9,13 +9,13 @@
 from datetime import datetime, timezone
 
 import pandas as pd
-from pystac import Collection, Extent, Link, SpatialExtent, TemporalExtent
+from pystac import Catalog, Collection, Extent, Link, SpatialExtent, TemporalExtent
 from xcube.core.store import new_data_store
 
 from deep_code.utils.osc_extension import OscExtension
 
 
-class OSCProductSTACGenerator:
+class OSCDatasetSTACGenerator:
     """Generates OSC STAC Collections for a product from Zarr datasets.
 
     Args:
@@ -53,6 +53,7 @@ def __init__(
         self.cf_params = cf_params or {}
         self.logger = logging.getLogger(__name__)
         self.dataset = self._open_dataset()
+        self.variables_metadata = self.get_variables_metadata()
 
     def _open_dataset(self):
         """Open the dataset using a S3 store as a xarray Dataset."""
@@ -170,29 +171,6 @@ def _get_temporal_extent(self) -> TemporalExtent:
     def _normalize_name(name: str | None) -> str | None:
         return name.replace(" ", "-").lower() if name else None
 
-    def _get_variables(self) -> list[str]:
-        """Extracts variable names or descriptions from the dataset.
-
-        Variables are prioritized based on their `long_name` or `standard_name`
-        attributes. If neither is available, the variable's key from
-        `dataset.data_vars.keys()` is used.
-
-        Returns:
-            A list of variable names or descriptions.
-        """
-        variables = []
-        for var_name, variable in self.dataset.data_vars.items():
-            long_name = self._normalize_name(variable.attrs.get("long_name"))
-            standard_name = self._normalize_name(variable.attrs.get("standard_name"))
-            if not long_name and not standard_name:
-                self.logger.error(
-                    f"Metadata missing for variable '{var_name}': 'long_name' and "
-                    f"'standard_name' attributes are not available."
-                )
-            # Prioritize 'long_name', fallback to 'standard_name', then use variable key
-            variables.append(long_name or standard_name or var_name)
-        return variables
-
     def _get_general_metadata(self) -> dict:
         return {
             "description": self.dataset.attrs.get(
@@ -200,36 +178,173 @@ def _get_general_metadata(self) -> dict:
             )
         }
 
-    def _get_variable_metadata(self, var_name, var_data) -> dict:
-        """Extract metadata from a single variable's attributes.
+    def extract_metadata_for_variable(self, variable_data) -> dict:
+        """Extract metadata for a single variable."""
+        long_name = variable_data.attrs.get("long_name")
+        standard_name = variable_data.attrs.get("standard_name")
+        title = long_name or standard_name or variable_data.name
+        description = variable_data.attrs.get("description", "No variable description")
+        gcmd_keyword_url = variable_data.attrs.get("gcmd_keyword_url")
+        return {
+            "variable_id": self._normalize_name(title),
+            "description": description,
+            "gcmd_keyword_url": gcmd_keyword_url,
+        }
+
+    def get_variable_ids(self) -> list[str]:
+        """Get variable IDs for all variables in the dataset."""
+        return list(self.variables_metadata.keys())
+
+    def get_variables_metadata(self) -> dict[str, dict]:
+        """Extract metadata for all variables in the dataset."""
+        variables_metadata = {}
+        for var_name, variable in self.dataset.data_vars.items():
+            var_metadata = self.extract_metadata_for_variable(variable)
+            variables_metadata[var_metadata.get("variable_id")] = var_metadata
+        return variables_metadata
+
+    def _add_gcmd_link_to_var_catalog(
+        self, var_catalog: Catalog, var_metadata: dict
+    ) -> None:
+        """
+        Checks for a GCMD keyword URL in var_metadata, adds a 'via' link to the catalog
+        pointing to the GCMD Keyword Viewer.
 
         Args:
-            var_name: The raw variable name in the dataset.
-            var_data: An xarray DataArray containing variable data and attrs.
+            var_catalog: The PySTAC Catalog to which we want to add the link.
+            var_metadata: Dictionary containing metadata about the variable,
+                          including 'gcmd_keyword_url'.
+        """
+        gcmd_keyword_url = var_metadata.get("gcmd_keyword_url")
+        if not gcmd_keyword_url:
+            self.logger.debug(
+                f"No gcmd_keyword_url in var_metadata. Skipping adding GCMD link in "
+                f'the {var_metadata.get("variable_id")} catalog'
+            )
+            return
+        var_catalog.add_link(
+            Link(
+                rel="via",
+                target=gcmd_keyword_url,
+                title="Description",
+                media_type="text/html",
+            )
+        )
+        self.logger.info(
+            f'Added GCMD link for {var_metadata.get("variable_id")} '
+            f"catalog {gcmd_keyword_url}."
+        )
+
+    def build_variable_catalog(self, var_metadata) -> Catalog:
+        """Build an OSC STAC Catalog for the variables in the dataset.
 
         Returns:
-            A dict with 'id', 'title', and 'description'.
+            A pystac.Catalog object.
         """
-        long_name = var_data.attrs.get("long_name")
-        standard_name = var_data.attrs.get("standard_name")
-        title = long_name or standard_name or var_name
+        var_id = var_metadata.get("variable_id")
+        concepts = [{"id": theme} for theme in self.osc_themes]
 
-        normalized_title = self._normalize_name(title)
+        themes = [
+            {
+                "scheme": "https://github.com/stac-extensions/osc#theme",
+                "concepts": concepts,
+            }
+        ]
 
-        description = var_data.attrs.get("description", "No variable description")
+        now_iso = datetime.now(timezone.utc).isoformat()
 
-        return {"id": var_name, "title": normalized_title, "description": description}
+        # Create a PySTAC Catalog object
+        var_catalog = Catalog(
+            id=var_id,
+            description=var_metadata.get("description"),
+            title=var_id,
+            stac_extensions=[
+                "https://stac-extensions.github.io/themes/v1.0.0/schema.json"
+            ],
+        )
 
-    def build_stac_collection(self) -> Collection:
-        """
-        Build an OSC STAC Collection for the dataset.
+        var_catalog.stac_version = "1.0.0"
+        var_catalog.extra_fields["updated"] = now_iso
+        var_catalog.keywords = []
+
+        # Add the 'themes' block (from your example JSON)
+        var_catalog.extra_fields["themes"] = themes
 
-        :return: A pystac.Collection object.
+        var_catalog.remove_links("root")
+        # Add relevant links
+        var_catalog.add_link(
+            Link(
+                rel="root",
+                target="../../catalog.json",
+                media_type="application/json",
+                title="Open Science Catalog",
+            )
+        )
+
+        # 'child' link: points to the product (or one of its collections) using this variable
+        var_catalog.add_link(
+            Link(
+                rel="child",
+                target=f"../../products/{self.collection_id}/collection.json",
+                media_type="application/json",
+                title=self.collection_id,
+            )
+        )
+
+        # 'parent' link: back up to the variables overview
+        var_catalog.add_link(
+            Link(
+                rel="parent",
+                target="../catalog.json",
+                media_type="application/json",
+                title="Variables",
+            )
+        )
+        # Add gcmd link for the variable definition
+        self._add_gcmd_link_to_var_catalog(var_catalog, var_metadata)
+
+        self_href = (
+            f"https://esa-earthcode.github.io/open-science-catalog-metadata/variables"
+            f"/{var_id}/catalog.json"
+        )
+        # 'self' link: the direct URL where this JSON is hosted
+        var_catalog.set_self_href(self_href)
+
+        return var_catalog
+
+    def update_existing_variable_catalog(self, var_file_path, var_id) -> Catalog:
+        existing_catalog = Catalog.from_file(var_file_path)
+        now_iso = datetime.now(timezone.utc).isoformat()
+        existing_catalog.extra_fields["updated"] = now_iso
+
+        # add 'child' link as the product
+        existing_catalog.add_link(
+            Link(
+                rel="child",
+                target=f"../../products/{self.collection_id}/collection.json",
+                media_type="application/json",
+                title=self.collection_id,
+            )
+        )
+        self_href = (
+            f"https://esa-earthcode.github.io/open-science-catalog-metadata/variables"
+            f"/{var_id}/catalog.json"
+        )
+        # 'self' link: the direct URL where this JSON is hosted
+        existing_catalog.set_self_href(self_href)
+
+        return existing_catalog
+
+    def build_dataset_stac_collection(self) -> Collection:
+        """Build an OSC STAC Collection for the dataset.
+
+        Returns:
+            A pystac.Collection object.
         """
         try:
             spatial_extent = self._get_spatial_extent()
             temporal_extent = self._get_temporal_extent()
-            variables = self._get_variables()
+            variables = self.get_variable_ids()
             general_metadata = self._get_general_metadata()
         except ValueError as e:
             raise ValueError(f"Metadata extraction failed: {e}")
@@ -260,6 +375,7 @@ def build_stac_collection(self) -> Collection:
         now_iso = datetime.now(timezone.utc).isoformat()
         collection.extra_fields["created"] = now_iso
         collection.extra_fields["updated"] = now_iso
+        collection.title = self.collection_id
 
         # Remove any existing root link and re-add it properly
         collection.remove_links("root")
@@ -284,6 +400,16 @@ def build_stac_collection(self) -> Collection:
                 title="Products",
             )
         )
+        # Add variables ref
+        for var in variables:
+            collection.add_link(
+                Link(
+                    rel="related",
+                    target=f"../../varibales/{var}/catalog.json",
+                    media_type="application/json",
+                    title="Variable: " + var,
+                )
+            )
 
         self_href = (
             "https://esa-earthcode.github.io/"

From c96cd1d58a6f50b650001313c25204098918d9cd Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Tue, 21 Jan 2025 10:14:52 +0100
Subject: [PATCH 10/21] update README.md

---
 README.md | 27 ++++++++++++++++-----------
 1 file changed, 16 insertions(+), 11 deletions(-)

diff --git a/README.md b/README.md
index 1d4b381..74ffd87 100644
--- a/README.md
+++ b/README.md
@@ -9,13 +9,16 @@
 and Python API providing utilities that aid integration of DeepESDL datasets, 
 experiments with EarthCODE.
 
+The first release will focus on implementing the publish feature of DeepESDL 
+experiments/workflow as OGC API record and Datasets as an OSC stac collection.
+
 ## Setup
 
 ## Install
 `deep-code` will be available in PyPI and conda-forge. Till the stable release,
 developers/contributors can follow the below steps to install deep-code.
 
-## Installing from the repository for Developer
+## Installing from the repository for Developers/Contributors
 
 To install deep-code directly from the git repository, clone the repository, and execute the steps below:
 
@@ -72,16 +75,18 @@ github-token: personal access token
 #### dataset-config.yaml example
 
 ```
-dataset-id: hydrology-1D-0.009deg-100x60x60-3.0.2.zarr 
-collection-id: hydrology
-
-#non-mandatory
-documentation-link: https://deepesdl.readthedocs.io/en/latest/datasets/hydrology-1D-0-009deg-100x60x60-3-0-2-zarr/
-access-link: s3://test
-dataset-status: completed
-dataset-region: global
-dataset-theme: ["ocean", "environment"]
-cf-parameter: [{"Name" : "hydrology"}]
+dataset_id: hydrology-1D-0.009deg-100x60x60-3.0.2.zarr
+collection_id: hydrology
+osc_themes:
+  - Land
+  - Oceans
+# non-mandatory
+documentation_link: https://deepesdl.readthedocs.io/en/latest/datasets/hydrology-1D-0.009deg-100x60x60-3.0.2.zarr/
+access_link: s3://test
+dataset_status: completed
+osc_region: global
+cf_parameter:
+  - name: hydrology
 ```
 
 dataset-id has to be a valid dataset-id from `deep-esdl-public` s3 or your team bucket.
\ No newline at end of file

From 2491433ed1e0cfd2b114270fe1fe4d746e8cc47f Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Tue, 21 Jan 2025 10:17:28 +0100
Subject: [PATCH 11/21] updated github utility class

---
 deep_code/utils/github_automation.py | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/deep_code/utils/github_automation.py b/deep_code/utils/github_automation.py
index 4ad6e38..4f3425e 100644
--- a/deep_code/utils/github_automation.py
+++ b/deep_code/utils/github_automation.py
@@ -114,3 +114,9 @@ def clean_up(self):
             subprocess.run(["rm", "-rf", self.local_clone_dir])
         except subprocess.CalledProcessError as e:
             raise RuntimeError(f"Failed to clean-up local repository: {e}")
+
+    def file_exists(self, file_path) -> bool:
+        full_path = Path(self.local_clone_dir) / file_path
+        exists = os.path.isfile(full_path)
+        logging.debug(f"Checking existence of {full_path}: {exists}")
+        return exists
\ No newline at end of file

From ba7d25b047badda2341cdb5cc829223556dbd0fd Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Tue, 21 Jan 2025 10:45:04 +0100
Subject: [PATCH 12/21] introduced BasePublisher class to keep code DRY

---
 deep_code/tests/tools/test_publish.py |   6 +-
 deep_code/tools/publish.py            | 302 +++++++++++---------------
 2 files changed, 123 insertions(+), 185 deletions(-)

diff --git a/deep_code/tests/tools/test_publish.py b/deep_code/tests/tools/test_publish.py
index 4a13581..02c6bde 100644
--- a/deep_code/tests/tools/test_publish.py
+++ b/deep_code/tests/tools/test_publish.py
@@ -13,7 +13,7 @@ def test_init_missing_credentials(self, mock_fsspec_open):
         )()
 
         with pytest.raises(
-            ValueError, match="GitHub credentials are missing in the `.gitaccess` file."
+            ValueError, match="GitHub credentials are missing in `.gitaccess` file."
         ):
             DatasetPublisher()
 
@@ -34,9 +34,7 @@ def test_publish_dataset_missing_ids(self, mock_fsspec_open):
         publisher = DatasetPublisher()
 
         with pytest.raises(
-            ValueError,
-            match="Dataset ID or Collection ID is missing in the "
-            "dataset-config.yaml file.",
+            ValueError, match="Dataset ID or Collection ID missing in config."
         ):
             publisher.publish_dataset("/path/to/dataset-config.yaml")
 
diff --git a/deep_code/tools/publish.py b/deep_code/tools/publish.py
index cc13353..e8bec94 100644
--- a/deep_code/tools/publish.py
+++ b/deep_code/tools/publish.py
@@ -27,18 +27,14 @@
 
 class BasePublisher:
     """
-    Base class providing common GitHub automation steps:
-      - Reading credentials from `.gitaccess`
-      - Setting up GitHubAutomation
-      - Forking, cloning, creating a branch
-      - Adding files, committing, pushing
-      - Creating a pull request
+    Base class providing:
+      - Reading .gitaccess for credentials
+      - Common GitHub automation steps (fork, clone, branch, file commit, pull request)
     """
 
     def __init__(self):
         with fsspec.open(".gitaccess", "r") as file:
             git_config = yaml.safe_load(file) or {}
-
         self.github_username = git_config.get("github-username")
         self.github_token = git_config.get("github-token")
         if not self.github_username or not self.github_token:
@@ -48,37 +44,36 @@ def __init__(self):
             self.github_username, self.github_token, OSC_REPO_OWNER, OSC_REPO_NAME
         )
 
-    def publish_file(
+    def publish_files(
         self,
         branch_name: str,
-        file_path: str,
-        file_content: dict,
+        file_dict: dict[str, dict],
         commit_message: str,
         pr_title: str,
         pr_body: str,
     ) -> str:
-        """
-        Publish a single file to GitHub in a new branch and open a PR.
+        """Publish multiple files to a new branch and open a PR.
 
         Args:
-            branch_name: Name of the branch to create (e.g. 'osc-branch-collectionid').
-            file_path: File path in the repo (e.g. 'products/.../collection.json').
-            file_content: The JSON/dict content to commit.
-            commit_message: Commit message.
-            pr_title: Title for the pull request.
-            pr_body: Body of the pull request.
+            branch_name: Branch name to create (e.g. "osc-branch-collectionid").
+            file_dict: { file_path: file_content_dict } for each file to commit.
+            commit_message: Commit message for all changes.
+            pr_title: Title of the pull request.
+            pr_body: Description/body of the pull request.
 
         Returns:
-            str: The URL of the created pull request.
+            URL of the created pull request.
         """
         try:
-            logger.info("Starting GitHub automation...")
+            logger.info("Forking and cloning repository...")
             self.github_automation.fork_repository()
             self.github_automation.clone_repository()
             self.github_automation.create_branch(branch_name)
 
-            # Add the file
-            self.github_automation.add_file(file_path, file_content)
+            # Add each file to the branch
+            for file_path, content in file_dict.items():
+                logger.info(f"Adding {file_path} to {branch_name}")
+                self.github_automation.add_file(file_path, content)
 
             # Commit and push
             self.github_automation.commit_and_push(branch_name, commit_message)
@@ -87,47 +82,23 @@ def publish_file(
             pr_url = self.github_automation.create_pull_request(
                 branch_name, pr_title, pr_body
             )
-            logger.info(f"Pull request created at: {pr_url}")
+            logger.info(f"Pull request created: {pr_url}")
             return pr_url
 
         finally:
-            # Always clean up local clone
+            # Cleanup local clone
             self.github_automation.clean_up()
 
 
-class DatasetPublisher:
+class DatasetPublisher(BasePublisher):
+    """Publishes products (datasets) to the OSC GitHub repository.
+    Inherits from BasePublisher for GitHub publishing logic.
     """
-    Publishes products to the OSC GitHub repository.
-
-    Credentials must be provided via a hidden file named `.gitaccess`, located in
-    the root of the repository. This file is expected to contain YAML of the form:
-
-        github-username: "YOUR_GITHUB_USERNAME"
-        github-token: "YOUR_GITHUB_PERSONAL_ACCESS_TOKEN"
-    """
-
-    def __init__(self):
-        with fsspec.open(".gitaccess", "r") as file:
-            git_config = yaml.safe_load(file) or {}
-
-        self.github_username = git_config.get("github-username")
-        self.github_token = git_config.get("github-token")
-
-        if not self.github_username or not self.github_token:
-            raise ValueError("GitHub credentials are missing in the `.gitaccess` file.")
-
-        self.github_automation = GitHubAutomation(
-            self.github_username, self.github_token, OSC_REPO_OWNER, OSC_REPO_NAME
-        )
 
     def publish_dataset(self, dataset_config_path: str):
-        """Publish a product collection to the specified GitHub repository.
-
-        Args:
-             dataset_config_path: Path to the YAML file containing dataset config
-        """
+        """Publish a product collection to the specified GitHub repository."""
         with fsspec.open(dataset_config_path, "r") as file:
-            dataset_config = yaml.safe_load(file)
+            dataset_config = yaml.safe_load(file) or {}
 
         dataset_id = dataset_config.get("dataset_id")
         collection_id = dataset_config.get("collection_id")
@@ -139,145 +110,114 @@ def publish_dataset(self, dataset_config_path: str):
         cf_params = dataset_config.get("cf_parameter")
 
         if not dataset_id or not collection_id:
-            raise ValueError(
-                "Dataset ID or Collection ID is missing in the dataset-config.yaml "
-                "file."
-            )
-
-        try:
-            logger.info("Generating STAC collection...")
-            generator = OSCDatasetSTACGenerator(
-                dataset_id=dataset_id,
-                collection_id=collection_id,
-                documentation_link=documentation_link,
-                access_link=access_link,
-                osc_status=dataset_status,
-                osc_region=osc_region,
-                osc_themes=osc_themes,
-                cf_params=cf_params,
-            )
-            # get variables from the datasets
-            variable_ids = generator.get_variable_ids()
-            # build STAC collection for the dataset
-            ds_collection = generator.build_dataset_stac_collection()
-
-            file_path = f"products/{collection_id}/collection.json"
-            logger.info("Automating GitHub tasks...")
-            self.github_automation.fork_repository()
-            self.github_automation.clone_repository()
-            OSC_NEW_BRANCH_NAME = OSC_BRANCH_NAME + "-" + collection_id
-            self.github_automation.create_branch(OSC_NEW_BRANCH_NAME)
-
-            for var_id in variable_ids:
-                var_file_path = f"variables/{var_id}/catalog.json"
-                if not self.github_automation.file_exists(var_file_path):
-                    logger.info(
-                        f"Variable catalog for {var_id} does not exist. Creating..."
-                    )
-                    var_metadata = generator.variables_metadata.get(var_id)
-                    var_catalog = generator.build_variable_catalog(var_metadata)
-                    self.github_automation.add_file(
-                        var_file_path, var_catalog.to_dict()
-                    )
-                else:
-                    logger.info(
-                        f"Variable catalog already exists for {var_id}. so add the "
-                        f"product as child link..."
-                    )
-                    full_path = (
-                        Path(self.github_automation.local_clone_dir) / var_file_path
-                    )
-                    self.github_automation.add_file(
-                        var_file_path,
-                        generator.update_existing_variable_catalog(
-                            full_path, var_id
-                        ).to_dict(),
-                    )
-
-            self.github_automation.add_file(file_path, ds_collection.to_dict())
-
-            self.github_automation.commit_and_push(
-                OSC_NEW_BRANCH_NAME, f"Add new collection:{collection_id}"
-            )
-            pr_url = self.github_automation.create_pull_request(
-                OSC_NEW_BRANCH_NAME,
-                f"Add new dataset collection",
-                "This PR adds a new collection to the repository.",
-            )
-
-            logger.info(f"Pull request created: {pr_url}")
-
-        finally:
-            self.github_automation.clean_up()
-
-
-class WorkflowPublisher:
-    """Publishes workflow to the OSC GitHub repository.
-
-        Credentials must be provided via a hidden file named `.gitaccess`, located in
-        the root of the repository. This file is expected to contain YAML of the form:
-
-            github-username: "YOUR_GITHUB_USERNAME"
-            github-token: "YOUR_GITHUB_PERSONAL_ACCESS_TOKEN"
-        """
+            raise ValueError("Dataset ID or Collection ID missing in config.")
+
+        logger.info("Generating STAC collection...")
+
+        generator = OSCDatasetSTACGenerator(
+            dataset_id=dataset_id,
+            collection_id=collection_id,
+            documentation_link=documentation_link,
+            access_link=access_link,
+            osc_status=dataset_status,
+            osc_region=osc_region,
+            osc_themes=osc_themes,
+            cf_params=cf_params,
+        )
 
-    def __init__(self):
-        with fsspec.open(".gitaccess", "r") as file:
-            git_config = yaml.safe_load(file) or {}
+        variable_ids = generator.get_variable_ids()
+        ds_collection = generator.build_dataset_stac_collection()
+
+        # Prepare a dictionary of file paths and content
+        file_dict = {}
+        product_path = f"products/{collection_id}/collection.json"
+        file_dict[product_path] = ds_collection.to_dict()
+
+        # Add or update variable files
+        for var_id in variable_ids:
+            var_file_path = f"variables/{var_id}/catalog.json"
+            if not self.github_automation.file_exists(var_file_path):
+                logger.info(
+                    f"Variable catalog for {var_id} does not exist. Creating..."
+                )
+                var_metadata = generator.variables_metadata.get(var_id)
+                var_catalog = generator.build_variable_catalog(var_metadata)
+                file_dict[var_file_path] = var_catalog.to_dict()
+            else:
+                logger.info(
+                    f"Variable catalog already exists for {var_id}, adding product link."
+                )
+                full_path = Path(self.github_automation.local_clone_dir) / var_file_path
+                updated_catalog = generator.update_existing_variable_catalog(
+                    full_path, var_id
+                )
+                file_dict[var_file_path] = updated_catalog.to_dict()
+
+        # Create branch name, commit message, PR info
+        branch_name = f"{OSC_BRANCH_NAME}-{collection_id}"
+        commit_message = f"Add new dataset collection: {collection_id}"
+        pr_title = "Add new dataset collection"
+        pr_body = "This PR adds a new dataset collection to the repository."
+
+        # Publish all files in one go
+        pr_url = self.publish_files(
+            branch_name=branch_name,
+            file_dict=file_dict,
+            commit_message=commit_message,
+            pr_title=pr_title,
+            pr_body=pr_body,
+        )
 
-        self.github_username = git_config.get("github-username")
-        self.github_token = git_config.get("github-token")
+        logger.info(f"Pull request created: {pr_url}")
 
-        if not self.github_username or not self.github_token:
-            raise ValueError("GitHub credentials are missing in the `.gitaccess` file.")
 
-        self.github_automation = GitHubAutomation(
-            self.github_username, self.github_token, OSC_REPO_OWNER, OSC_REPO_NAME
-        )
+class WorkflowPublisher(BasePublisher):
+    """Publishes workflows to the OSC GitHub repository."""
 
     @staticmethod
     def _normalize_name(name: str | None) -> str | None:
         return name.replace(" ", "-").lower() if name else None
 
     def publish_workflow(self, workflow_config_path: str):
-
         with fsspec.open(workflow_config_path, "r") as file:
-            workflow_config = yaml.safe_load(file)
+            workflow_config = yaml.safe_load(file) or {}
+
+        workflow_id = self._normalize_name(workflow_config.get("workflow_id"))
+        if not workflow_id:
+            raise ValueError("workflow_id is missing in workflow config.")
+
+        properties_list = workflow_config.get("properties", [])
+        contacts = workflow_config.get("contact", [])
+        links = workflow_config.get("links", [])
+
+        logger.info("Generating OGC API Record for the workflow...")
+        rg = OSCWorkflowOGCApiRecordGenerator()
+        wf_record_properties = rg.build_record_properties(properties_list, contacts)
+
+        ogc_record = OgcRecord(
+            id=workflow_id,
+            type="Feature",
+            time={},
+            properties=wf_record_properties,
+            links=links,
+        )
 
-        try:
-            logger.info("Generating OGC API Record for the workflow...")
-            workflow_id = self._normalize_name(workflow_config.get("workflow_id"))
-            properties_list = workflow_config.get("properties", [])
-
-            contacts = workflow_config.get("contact", [])
-            rg = OSCWorkflowOGCApiRecordGenerator()
-            wf_record_properties = rg.build_record_properties(properties_list, contacts)
-
-            links = workflow_config.get("links")
-            ogc_record = OgcRecord(
-                id=workflow_id,
-                type="Feature",
-                time={},
-                properties=wf_record_properties,
-                links=links,
-            )
+        file_path = f"workflow/{workflow_id}/collection.json"
 
-            file_path = f"workflow/{workflow_id}/collection.json"
-            logger.info("Automating GitHub tasks...")
-            self.github_automation.fork_repository()
-            self.github_automation.clone_repository()
-            WF_NEW_BRANCH_NAME = WF_BRANCH_NAME + "-" + workflow_id
-            self.github_automation.create_branch(WF_NEW_BRANCH_NAME)
-            self.github_automation.add_file(file_path, ogc_record.to_dict())
-            self.github_automation.commit_and_push(
-                WF_NEW_BRANCH_NAME, f"Add new workflow:{workflow_id}"
-            )
-            pr_url = self.github_automation.create_pull_request(
-                WF_NEW_BRANCH_NAME,
-                f"Add new collection",
-                "This PR adds a new workflow to the OSC repository.",
-            )
-            logger.info(f"Pull request created: {pr_url}")
+        # Prepare the single file dict
+        file_dict = {file_path: ogc_record.to_dict()}
 
-        finally:
-            self.github_automation.clean_up()
+        branch_name = f"{WF_BRANCH_NAME}-{workflow_id}"
+        commit_message = f"Add new workflow: {workflow_id}"
+        pr_title = "Add new workflow"
+        pr_body = "This PR adds a new workflow to the OSC repository."
+
+        pr_url = self.publish_files(
+            branch_name=branch_name,
+            file_dict=file_dict,
+            commit_message=commit_message,
+            pr_title=pr_title,
+            pr_body=pr_body,
+        )
+
+        logger.info(f"Pull request created: {pr_url}")

From b2e14c83cb8ef05567377035e68244c2949f1589 Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Tue, 21 Jan 2025 10:45:14 +0100
Subject: [PATCH 13/21] refactor

---
 deep_code/utils/github_automation.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/deep_code/utils/github_automation.py b/deep_code/utils/github_automation.py
index 4f3425e..dbecfe4 100644
--- a/deep_code/utils/github_automation.py
+++ b/deep_code/utils/github_automation.py
@@ -119,4 +119,4 @@ def file_exists(self, file_path) -> bool:
         full_path = Path(self.local_clone_dir) / file_path
         exists = os.path.isfile(full_path)
         logging.debug(f"Checking existence of {full_path}: {exists}")
-        return exists
\ No newline at end of file
+        return exists

From 30e320a62df2de4a1e07fa1ef0cb1ae58fece6df Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Tue, 21 Jan 2025 10:51:57 +0100
Subject: [PATCH 14/21] updated README.md

---
 README.md | 45 ++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 44 insertions(+), 1 deletion(-)

diff --git a/README.md b/README.md
index 74ffd87..364b703 100644
--- a/README.md
+++ b/README.md
@@ -89,4 +89,47 @@ cf_parameter:
   - name: hydrology
 ```
 
-dataset-id has to be a valid dataset-id from `deep-esdl-public` s3 or your team bucket.
\ No newline at end of file
+dataset-id has to be a valid dataset-id from `deep-esdl-public` s3 or your team bucket.
+
+### deep-code publish-workflow
+
+Publish a workflow/experiment to the EarthCODE open-science catalog.
+
+```commandline
+deep-code publish-workflow /path/to/workflow-config.yaml
+ ```
+#### workflow-config.yaml example
+
+```
+workflow_id: "4D Med hydrology cube generation"
+properties:
+  title: "Hydrology cube generation recipe"
+  description: "4D Med cube generation"
+  keywords:
+    - Earth Science
+  themes:
+      - Atmosphere
+      - Ocean
+      - Evaporation
+  license: proprietary
+  jupyter_kernel_info:
+    name: deepesdl-xcube-1.7.1
+    python_version: 3.11
+    env_file: https://git/env.yml
+links:
+  - rel: "documentation"
+    type: "application/json"
+    title: "4DMed Hydrology Cube Generation Recipe"
+    href: "https://github.com/deepesdl/cube-gen/tree/main/hydrology/README.md"
+  - rel: "jupyter-notebook"
+    type: "application/json"
+    title: "Workflow Jupyter Notebook"
+    href: "https://github.com/deepesdl/cube-gen/blob/main/hydrology/notebooks/reading_hydrology.ipynb"
+contact:
+  - name: Tejas Morbagal Harish
+    organization: Brockmann Consult GmbH
+    links:
+      - rel: "about"
+        type: "text/html"
+        href: "https://www.brockmann-consult.de/"
+```

From 1c148d38f806ee06b4b2929f9eee182f21f17f2d Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Tue, 21 Jan 2025 11:02:14 +0100
Subject: [PATCH 15/21] refactor

---
 deep_code/tests/tools/test_publish.py | 4 ++--
 deep_code/tools/publish.py            | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/deep_code/tests/tools/test_publish.py b/deep_code/tests/tools/test_publish.py
index 02c6bde..00ac637 100644
--- a/deep_code/tests/tools/test_publish.py
+++ b/deep_code/tests/tools/test_publish.py
@@ -13,7 +13,7 @@ def test_init_missing_credentials(self, mock_fsspec_open):
         )()
 
         with pytest.raises(
-            ValueError, match="GitHub credentials are missing in `.gitaccess` file."
+            ValueError, match="GitHub credentials are missing in the `.gitaccess` file."
         ):
             DatasetPublisher()
 
@@ -34,7 +34,7 @@ def test_publish_dataset_missing_ids(self, mock_fsspec_open):
         publisher = DatasetPublisher()
 
         with pytest.raises(
-            ValueError, match="Dataset ID or Collection ID missing in config."
+            ValueError, match="Dataset ID or Collection ID missing in the config."
         ):
             publisher.publish_dataset("/path/to/dataset-config.yaml")
 
diff --git a/deep_code/tools/publish.py b/deep_code/tools/publish.py
index e8bec94..e91dd90 100644
--- a/deep_code/tools/publish.py
+++ b/deep_code/tools/publish.py
@@ -38,7 +38,7 @@ def __init__(self):
         self.github_username = git_config.get("github-username")
         self.github_token = git_config.get("github-token")
         if not self.github_username or not self.github_token:
-            raise ValueError("GitHub credentials are missing in `.gitaccess` file.")
+            raise ValueError("GitHub credentials are missing in the `.gitaccess` file.")
 
         self.github_automation = GitHubAutomation(
             self.github_username, self.github_token, OSC_REPO_OWNER, OSC_REPO_NAME
@@ -110,7 +110,7 @@ def publish_dataset(self, dataset_config_path: str):
         cf_params = dataset_config.get("cf_parameter")
 
         if not dataset_id or not collection_id:
-            raise ValueError("Dataset ID or Collection ID missing in config.")
+            raise ValueError("Dataset ID or Collection ID missing in the config.")
 
         logger.info("Generating STAC collection...")
 

From e21c64f5e4e89afceaaa18313aaa2fea97acee43 Mon Sep 17 00:00:00 2001
From: Tejas Morbagal Harish <tejas.morbagalharish@brockmann-consult.de>
Date: Mon, 27 Jan 2025 11:51:06 +0100
Subject: [PATCH 16/21] Update deep_code/utils/dataset_stac_generator.py

Co-authored-by: Norman Fomferra <norman.fomferra@brockmann-consult.de>
---
 deep_code/utils/dataset_stac_generator.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/deep_code/utils/dataset_stac_generator.py b/deep_code/utils/dataset_stac_generator.py
index 5b04c86..e3c4eb4 100644
--- a/deep_code/utils/dataset_stac_generator.py
+++ b/deep_code/utils/dataset_stac_generator.py
@@ -15,7 +15,7 @@
 from deep_code.utils.osc_extension import OscExtension
 
 
-class OSCDatasetSTACGenerator:
+class OscDatasetStacGenerator:
     """Generates OSC STAC Collections for a product from Zarr datasets.
 
     Args:

From 560bcde22cda4ceb9c006d5e560b2881f8cdb592 Mon Sep 17 00:00:00 2001
From: Tejas Morbagal Harish <tejas.morbagalharish@brockmann-consult.de>
Date: Mon, 27 Jan 2025 11:51:19 +0100
Subject: [PATCH 17/21] Update
 deep_code/tests/utils/test_dataset_stac_generator.py

Co-authored-by: Norman Fomferra <norman.fomferra@brockmann-consult.de>
---
 deep_code/tests/utils/test_dataset_stac_generator.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/deep_code/tests/utils/test_dataset_stac_generator.py b/deep_code/tests/utils/test_dataset_stac_generator.py
index d4444c1..55b0d7e 100644
--- a/deep_code/tests/utils/test_dataset_stac_generator.py
+++ b/deep_code/tests/utils/test_dataset_stac_generator.py
@@ -7,7 +7,7 @@
 from pystac import Collection
 from xarray import Dataset
 
-from deep_code.utils.dataset_stac_generator import OSCDatasetSTACGenerator
+from deep_code.utils.dataset_stac_generator import OscDatasetStacGenerator
 
 
 class TestOSCProductSTACGenerator(unittest.TestCase):

From 06dcf3bc351ef6cf68e68c15b38b9f4c7c4fef03 Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Mon, 27 Jan 2025 11:57:40 +0100
Subject: [PATCH 18/21] refactor classname using camelcase

---
 deep_code/tests/tools/test_publish.py                | 2 +-
 deep_code/tests/utils/test_dataset_stac_generator.py | 8 ++++----
 deep_code/tools/publish.py                           | 4 ++--
 3 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/deep_code/tests/tools/test_publish.py b/deep_code/tests/tools/test_publish.py
index 00ac637..3e0f5e8 100644
--- a/deep_code/tests/tools/test_publish.py
+++ b/deep_code/tests/tools/test_publish.py
@@ -100,7 +100,7 @@ def test_publish_dataset_success(
             "links": [],
             "stac_version": "1.0.0",
         }
-        with patch("deep_code.tools.publish.OSCDatasetSTACGenerator") as mock_generator:
+        with patch("deep_code.tools.publish.OscDatasetStacGenerator") as mock_generator:
             mock_generator.return_value.build_dataset_stac_collection.return_value = (
                 mock_collection
             )
diff --git a/deep_code/tests/utils/test_dataset_stac_generator.py b/deep_code/tests/utils/test_dataset_stac_generator.py
index 55b0d7e..64285a7 100644
--- a/deep_code/tests/utils/test_dataset_stac_generator.py
+++ b/deep_code/tests/utils/test_dataset_stac_generator.py
@@ -52,7 +52,7 @@ def setUp(self, mock_data_store):
         mock_store.open_data.return_value = self.mock_dataset
         mock_data_store.return_value = mock_store
 
-        self.generator = OSCDatasetSTACGenerator(
+        self.generator = OscDatasetStacGenerator(
             dataset_id="mock-dataset-id",
             collection_id="mock-collection-id",
             access_link="s3://mock-bucket/mock-dataset",
@@ -130,7 +130,7 @@ def test_open_dataset_success_public_store(self, mock_logger, mock_new_data_stor
         mock_store.open_data.return_value = self.mock_dataset
 
         # Instantiate the generator (this will implicitly call _open_dataset)
-        generator = OSCDatasetSTACGenerator("mock-dataset-id", "mock-collection-id")
+        generator = OscDatasetStacGenerator("mock-dataset-id", "mock-collection-id")
 
         # Validate that the dataset is assigned correctly
         self.assertEqual(generator.dataset, "mock_dataset")
@@ -171,7 +171,7 @@ def test_open_dataset_success_authenticated_store(
         os.environ["S3_USER_STORAGE_KEY"] = "mock-key"
         os.environ["S3_USER_STORAGE_SECRET"] = "mock-secret"
 
-        generator = OSCDatasetSTACGenerator("mock-dataset-id", "mock-collection-id")
+        generator = OscDatasetStacGenerator("mock-dataset-id", "mock-collection-id")
 
         # Validate that the dataset was successfully opened with the authenticated store
         self.assertEqual(generator.dataset, "mock_dataset")
@@ -209,7 +209,7 @@ def test_open_dataset_failure(self, mock_logger, mock_new_data_store):
         os.environ["S3_USER_STORAGE_SECRET"] = "mock-secret"
 
         with self.assertRaises(ValueError) as context:
-            OSCDatasetSTACGenerator("mock-dataset-id", "mock-collection-id")
+            OscDatasetStacGenerator("mock-dataset-id", "mock-collection-id")
 
         self.assertIn(
             "Failed to open Zarr dataset with ID mock-dataset-id",
diff --git a/deep_code/tools/publish.py b/deep_code/tools/publish.py
index e91dd90..8e5f8c9 100644
--- a/deep_code/tools/publish.py
+++ b/deep_code/tools/publish.py
@@ -16,7 +16,7 @@
     OSC_REPO_OWNER,
     WF_BRANCH_NAME,
 )
-from deep_code.utils.dataset_stac_generator import OSCDatasetSTACGenerator
+from deep_code.utils.dataset_stac_generator import OscDatasetStacGenerator
 from deep_code.utils.github_automation import GitHubAutomation
 from deep_code.utils.ogc_api_record import OgcRecord
 from deep_code.utils.ogc_record_generator import OSCWorkflowOGCApiRecordGenerator
@@ -114,7 +114,7 @@ def publish_dataset(self, dataset_config_path: str):
 
         logger.info("Generating STAC collection...")
 
-        generator = OSCDatasetSTACGenerator(
+        generator = OscDatasetStacGenerator(
             dataset_id=dataset_id,
             collection_id=collection_id,
             documentation_link=documentation_link,

From 949a669844b5820537fb5de8e95d2afb0c7bfb7e Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Mon, 27 Jan 2025 12:03:07 +0100
Subject: [PATCH 19/21] refactor naming of variable attrs

---
 deep_code/utils/dataset_stac_generator.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/deep_code/utils/dataset_stac_generator.py b/deep_code/utils/dataset_stac_generator.py
index e3c4eb4..3d4da00 100644
--- a/deep_code/utils/dataset_stac_generator.py
+++ b/deep_code/utils/dataset_stac_generator.py
@@ -182,11 +182,11 @@ def extract_metadata_for_variable(self, variable_data) -> dict:
         """Extract metadata for a single variable."""
         long_name = variable_data.attrs.get("long_name")
         standard_name = variable_data.attrs.get("standard_name")
-        title = long_name or standard_name or variable_data.name
-        description = variable_data.attrs.get("description", "No variable description")
+        variable_id = standard_name or variable_data.name
+        description = variable_data.attrs.get("description", long_name)
         gcmd_keyword_url = variable_data.attrs.get("gcmd_keyword_url")
         return {
-            "variable_id": self._normalize_name(title),
+            "variable_id": self._normalize_name(variable_id),
             "description": description,
             "gcmd_keyword_url": gcmd_keyword_url,
         }

From 901d7f81af491a9413d28be4701e480faab57683 Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Mon, 27 Jan 2025 12:29:15 +0100
Subject: [PATCH 20/21] use composition over inheritance

---
 deep_code/tools/publish.py | 24 +++++++++++++++++-------
 1 file changed, 17 insertions(+), 7 deletions(-)

diff --git a/deep_code/tools/publish.py b/deep_code/tools/publish.py
index 8e5f8c9..ba762e1 100644
--- a/deep_code/tools/publish.py
+++ b/deep_code/tools/publish.py
@@ -25,7 +25,7 @@
 logging.basicConfig(level=logging.INFO)
 
 
-class BasePublisher:
+class GitHubPublisher:
     """
     Base class providing:
       - Reading .gitaccess for credentials
@@ -90,11 +90,15 @@ def publish_files(
             self.github_automation.clean_up()
 
 
-class DatasetPublisher(BasePublisher):
+class DatasetPublisher:
     """Publishes products (datasets) to the OSC GitHub repository.
     Inherits from BasePublisher for GitHub publishing logic.
     """
 
+    def __init__(self):
+        # Composition
+        self.gh_publisher = GitHubPublisher()
+
     def publish_dataset(self, dataset_config_path: str):
         """Publish a product collection to the specified GitHub repository."""
         with fsspec.open(dataset_config_path, "r") as file:
@@ -136,7 +140,7 @@ def publish_dataset(self, dataset_config_path: str):
         # Add or update variable files
         for var_id in variable_ids:
             var_file_path = f"variables/{var_id}/catalog.json"
-            if not self.github_automation.file_exists(var_file_path):
+            if not self.gh_publisher.github_automation.file_exists(var_file_path):
                 logger.info(
                     f"Variable catalog for {var_id} does not exist. Creating..."
                 )
@@ -147,7 +151,10 @@ def publish_dataset(self, dataset_config_path: str):
                 logger.info(
                     f"Variable catalog already exists for {var_id}, adding product link."
                 )
-                full_path = Path(self.github_automation.local_clone_dir) / var_file_path
+                full_path = (
+                    Path(self.gh_publisher.github_automation.local_clone_dir)
+                    / var_file_path
+                )
                 updated_catalog = generator.update_existing_variable_catalog(
                     full_path, var_id
                 )
@@ -160,7 +167,7 @@ def publish_dataset(self, dataset_config_path: str):
         pr_body = "This PR adds a new dataset collection to the repository."
 
         # Publish all files in one go
-        pr_url = self.publish_files(
+        pr_url = self.gh_publisher.publish_files(
             branch_name=branch_name,
             file_dict=file_dict,
             commit_message=commit_message,
@@ -171,9 +178,12 @@ def publish_dataset(self, dataset_config_path: str):
         logger.info(f"Pull request created: {pr_url}")
 
 
-class WorkflowPublisher(BasePublisher):
+class WorkflowPublisher:
     """Publishes workflows to the OSC GitHub repository."""
 
+    def __init__(self):
+        self.gh_publisher = GitHubPublisher()
+
     @staticmethod
     def _normalize_name(name: str | None) -> str | None:
         return name.replace(" ", "-").lower() if name else None
@@ -212,7 +222,7 @@ def publish_workflow(self, workflow_config_path: str):
         pr_title = "Add new workflow"
         pr_body = "This PR adds a new workflow to the OSC repository."
 
-        pr_url = self.publish_files(
+        pr_url = self.gh_publisher.publish_files(
             branch_name=branch_name,
             file_dict=file_dict,
             commit_message=commit_message,

From 5503cd85c7cf13cd3e6ff6833576b95a73bbe569 Mon Sep 17 00:00:00 2001
From: tejas <tejas.morbagalharish@brockmann-consult.de>
Date: Mon, 27 Jan 2025 12:37:59 +0100
Subject: [PATCH 21/21] regactor WF_BRANCH_NAME to
 add-new-workflow-from-deepesdl

---
 deep_code/constants.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/deep_code/constants.py b/deep_code/constants.py
index 8a17bbc..992ddf4 100644
--- a/deep_code/constants.py
+++ b/deep_code/constants.py
@@ -13,4 +13,4 @@
     "https://gcmd.earthdata.nasa.gov/kms/concepts/concept_scheme/sciencekeywords"
 )
 OGC_API_RECORD_SPEC = "http://www.opengis.net/spec/ogcapi-records-1/1.0/req/record-core"
-WF_BRANCH_NAME = "add-new-workflow"
+WF_BRANCH_NAME = "add-new-workflow-from-deepesdl"