issue #85 - data release helper code

CHANGELOG.md

@@ -1,8 +1,14 @@
-## [unreleased]
+## [0.2.26] 2024-08-15
+
+Bumped version to 0.2.26 to catch up with data release. Only new client functionality is #81 'data_release' helper functions
+
+All other changes in this release were for data (and contained in data_v0.2.26)
 
 ### Added
 
-- New GFFs: RefSeq RS_2023_10, Ensembl 110, 111
+- #81 New 'data_release' code eg 'get_latest_combo_file_urls' that looks on GitHub to find latest data
+- New GFFs: RefSeq RS_2023_10, Ensembl 111, 112
+- #79 - RefSeq MT transcripts
 - #66 - We now store 'Note' field (thanks holtgrewe for suggestion)
 - Added requirements.txt for 'generate_transcript_data' sections
 - client / JSON data schema version compatability check
@@ -15,6 +21,7 @@
 - #64 - Split code/data versions. json.gz are now labelled according to data schema version (thanks holtgrewe)
 - Renamed 'CHM13v2.0' to 'T2T-CHM13v2.0' so it could work with biocommons bioutils
 - #72 - Correctly handle ncRNA_gene genes (thanks holtgrewe for reporting)
+- #73 - HGNC ID was missing for some chrMT genes in Ensembl
 
 ## [0.2.21] - 2023-08-14
 
@@ -209,7 +216,8 @@
 
 - Initial commit
 
-[unreleased]: https://github.com/SACGF/cdot/compare/v0.2.21...HEAD
+[unreleased]: https://github.com/SACGF/cdot/compare/v0.2.26...HEAD
+[0.2.26]: https://github.com/SACGF/cdot/compare/v0.2.21...v0.2.26
 [0.2.21]: https://github.com/SACGF/cdot/compare/v0.2.20...v0.2.21
 [0.2.20]: https://github.com/SACGF/cdot/compare/v0.2.19...v0.2.20
 [0.2.19]: https://github.com/SACGF/cdot/compare/v0.2.18...v0.2.19

cdot/__init__.py

@@ -1,4 +1,4 @@
-__version__ = "0.2.21"
+__version__ = "0.2.26"
 
 
 def get_data_schema_int(version: str) -> int:

cdot/data_release.py

@@ -0,0 +1,56 @@
+import re
+import requests
+import cdot
+
+from cdot import get_data_schema_int
+
+
+def get_latest_data_release_tag_name():
+    latest_data_release = get_latest_data_release()
+    return latest_data_release.get('tag_name')
+
+def _get_version_from_tag_name(tag_name, data_version=False):
+    """ Returns None if doesn't match required prefix """
+    release_prefix = "v"
+    if data_version:
+        release_prefix = "data_" + release_prefix
+
+    if not tag_name.startswith(release_prefix):
+        return None
+    return tag_name.lstrip(release_prefix)
+
+
+def get_latest_data_release():
+    client_data_schema = get_data_schema_int(cdot.__version__)
+
+    url = "https://api.github.com/repos/SACGF/cdot/releases"
+    response = requests.get(url)
+    json_data = response.json()
+    for release in json_data:
+        tag_name = release['tag_name']  # Should look like 'v0.2.25' for code or 'data_v0.2.25' for data
+        # We require a data version
+        data_version = _get_version_from_tag_name(tag_name, data_version=True)
+        if data_version is None:
+            continue
+
+        data_schema = get_data_schema_int(data_version)
+        if data_schema != client_data_schema:
+            continue
+        return release
+    return {}
+
+def get_latest_combo_file_urls(annotation_consortia, genome_builds):
+    # lower case everything to be case insensitive
+    annotation_consortia = {x.lower() for x in annotation_consortia}
+    genome_builds = {x.lower() for x in genome_builds}
+
+    file_urls = []
+    if latest_data_release := get_latest_data_release():
+        for asset in latest_data_release["assets"]:
+            browser_download_url = asset["browser_download_url"]
+            filename = browser_download_url.rsplit("/")[-1]
+            if m := re.match(r"cdot-(\d+\.\d+\.\d+)\.(refseq|ensembl)\.(.+)\.json\.gz", filename):
+                _version, annotation_consortium, genome_build = m.groups()
+                if annotation_consortium.lower() in annotation_consortia and genome_build.lower() in genome_builds:
+                    file_urls.append(browser_download_url)
+    return file_urls