Skeleton dev3 (#267)

* Compressed SWC format option. * lint/ruff corrections * Bulk skeleton generation. * New skeleton cache query endpoint * Ruff corrections * Trivial change * Removed debugging code. Labeled protected functions. * Ruff correctdions. * Ruff corrections. * Ruff changes that make the code definitively HARDER to read. Is this really what we want? * Converted print statements to logging calls. * Server version checking for new API endpoints. * Ruff conformity. * Added SkeletonClient.skeletons_exist(). * Ruff changes. * Skeleton Client documentation * update test docs to note confusing local server issue * generate_bulk_skeletons_async() now reports an upper bound estimate of the skeletonization time. * Documentation * Cleanup * Versions endpoint * Update changelog * SkeletonClient now only accepts 'dict' and 'swc' format requests. * ruff cleanup * ruff cleanup * ruff cleanup * Various handling of the new 'flatdict' skeleton format. * ruff compliance * Change default skeleton version to 3 (the current latest version) * ruff cleanup * Add packaging as a requirement * Bump version: 6.5.0 → 6.5.1 * version dependent networkx (#266) * Bump version: 6.5.1 → 6.5.2 * SkeletonClient changes, mainly simplification of output formats (#265) * Compressed SWC format option. * lint/ruff corrections * Bulk skeleton generation. * New skeleton cache query endpoint * Ruff corrections * Trivial change * Removed debugging code. Labeled protected functions. * Ruff correctdions. * Ruff corrections. * Ruff changes that make the code definitively HARDER to read. Is this really what we want? * Converted print statements to logging calls. * Server version checking for new API endpoints. * Ruff conformity. * Added SkeletonClient.skeletons_exist(). * Ruff changes. * Skeleton Client documentation * update test docs to note confusing local server issue * generate_bulk_skeletons_async() now reports an upper bound estimate of the skeletonization time. * Documentation * Cleanup * Versions endpoint * Update changelog * SkeletonClient now only accepts 'dict' and 'swc' format requests. * ruff cleanup * ruff cleanup * ruff cleanup * Various handling of the new 'flatdict' skeleton format. * ruff compliance --------- Co-authored-by: Casey Schneider-Mizell <[email protected]> * Bump version: 6.5.2 → 7.0.0 * added documentation of 7.0 changes to skeleton service * SkeletonClient changes, mainly simplification of output formats (#265) * Compressed SWC format option. * lint/ruff corrections * Bulk skeleton generation. * New skeleton cache query endpoint * Ruff corrections * Trivial change * Removed debugging code. Labeled protected functions. * Ruff correctdions. * Ruff corrections. * Ruff changes that make the code definitively HARDER to read. Is this really what we want? * Converted print statements to logging calls. * Server version checking for new API endpoints. * Ruff conformity. * Added SkeletonClient.skeletons_exist(). * Ruff changes. * Skeleton Client documentation * update test docs to note confusing local server issue * generate_bulk_skeletons_async() now reports an upper bound estimate of the skeletonization time. * Documentation * Cleanup * Versions endpoint * Update changelog * SkeletonClient now only accepts 'dict' and 'swc' format requests. * ruff cleanup * ruff cleanup * ruff cleanup * Various handling of the new 'flatdict' skeleton format. * ruff compliance --------- Co-authored-by: Casey Schneider-Mizell <[email protected]> * SkeletonClient changes, mainly simplification of output formats (#265) * Compressed SWC format option. * lint/ruff corrections * Bulk skeleton generation. * New skeleton cache query endpoint * Ruff corrections * Trivial change * Removed debugging code. Labeled protected functions. * Ruff correctdions. * Ruff corrections. * Ruff changes that make the code definitively HARDER to read. Is this really what we want? * Converted print statements to logging calls. * Server version checking for new API endpoints. * Ruff conformity. * Added SkeletonClient.skeletons_exist(). * Ruff changes. * Skeleton Client documentation * update test docs to note confusing local server issue * generate_bulk_skeletons_async() now reports an upper bound estimate of the skeletonization time. * Documentation * Cleanup * Versions endpoint * Update changelog * SkeletonClient now only accepts 'dict' and 'swc' format requests. * ruff cleanup * ruff cleanup * ruff cleanup * Various handling of the new 'flatdict' skeleton format. * ruff compliance --------- Co-authored-by: Casey Schneider-Mizell <[email protected]> * Change default skeleton version to 3 (the current latest version) * ruff cleanup * Old code somehow got reinstated. --------- Co-authored-by: Casey Schneider-Mizell <[email protected]> Co-authored-by: Ben Pedigo <[email protected]> Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com> Co-authored-by: Forrest Collman <[email protected]>
CAVEconnectome · Nov 21, 2024 · 6acbddf · 6acbddf
1 parent babb459
commit 6acbddf
Showing 1 changed file with 31 additions and 20 deletions.
diff --git a/caveclient/skeletonservice.py b/caveclient/skeletonservice.py
@@ -427,7 +427,7 @@ def get_skeleton(
         self,
         root_id: int,
         datastack_name: Optional[str] = None,
-        skeleton_version: Optional[int] = None,
+        skeleton_version: Optional[int] = 3,
         output_format: Literal[
             "dict",
             "swc",
@@ -447,7 +447,6 @@ def get_skeleton(
             The skeleton version to generate and retrieve. Options are documented in SkeletonService. Use 0 for Neuroglancer-compatibility. Use -1 for latest.
         output_format : string
             The format to retrieve. Options are:
-
             - 'dict': A dictionary
             - 'swc': A pandas DataFrame
 
@@ -460,15 +459,10 @@ def get_skeleton(
         if not self.fc.l2cache.has_cache():
             raise NoL2CacheException("SkeletonClient requires an L2Cache.")
 
-        if output_format not in ["dict", "swc"]:
-            raise ValueError(f"Unknown output format: {output_format}")
-
-        if verbose_level >= 1:
-            logging.info(f"SkeletonService version: {self._server_version}")
-
-        if self._server_version < Version("0.6.0"):
-            logging.warning(
-                "SkeletonService version is less than 0.6.0. Please upgrade to the latest version."
+        valid_output_formats = ["dict", "swc"]
+        if output_format not in valid_output_formats:
+            raise ValueError(
+                f"Unknown output format: {output_format}. Valid options: {valid_output_formats}"
             )
 
         # The output formats were changed in server v0.6.0 and must be handled differently by the client
@@ -480,11 +474,18 @@ def get_skeleton(
         elif output_format == "swc":
             endpoint_format = "swccompressed"
 
-        if skeleton_version is None:
+        valid_skeleton_versions = [-1, 0, 1, 2, 3]
+        if skeleton_version not in valid_skeleton_versions:
+            raise ValueError(
+                f"Unknown skeleton version: {skeleton_version}. Valid options: {valid_skeleton_versions}"
+            )
+
+        if verbose_level >= 1:
+            logging.info(f"SkeletonService version: {self._server_version}")
+        if self._server_version < Version("0.6.0"):
             logging.warning(
                 "The optional nature of the 'skeleton_version' parameter will be deprecated in the future. Please specify a skeleton version."
             )
-            skeleton_version = -1
 
         # -1, to specify the latest version, was only added in server v0.6.1
         if self._server_version < Version("0.6.1") and skeleton_version == -1:
@@ -545,7 +546,7 @@ def get_bulk_skeletons(
         self,
         root_ids: List,
         datastack_name: Optional[str] = None,
-        skeleton_version: Optional[int] = None,
+        skeleton_version: Optional[int] = 3,
         output_format: Literal[
             "dict",
             "swc",
@@ -568,16 +569,26 @@ def get_bulk_skeletons(
         if not self.fc.l2cache.has_cache():
             raise NoL2CacheException("SkeletonClient requires an L2Cache.")
 
+        valid_output_formats = ["dict", "swc"]
+        if output_format not in valid_output_formats:
+            raise ValueError(
+                f"Unknown output format: {output_format}. Valid options: {valid_output_formats}"
+            )
+
+        # The output formats were changed in server v0.6.0 and must be handled differently by the client
         if output_format == "dict":
-            endpoint_format = "flatdict"
+            if self._server_version < Version("0.6.0"):
+                endpoint_format = "jsoncompressed"
+            else:
+                endpoint_format = "flatdict"
         elif output_format == "swc":
-            endpoint_format = "swc"
+            endpoint_format = "swccompressed"
 
-        if skeleton_version is None:
-            logging.warning(
-                "The optional nature of the 'skeleton_version' parameter will be deprecated in the future. Please specify a skeleton version."
+        valid_skeleton_versions = [-1, 0, 1, 2, 3]
+        if skeleton_version not in valid_skeleton_versions:
+            raise ValueError(
+                f"Unknown skeleton version: {skeleton_version}. Valid options: {valid_skeleton_versions}"
             )
-            skeleton_version = -1
 
         url = self._build_bulk_endpoint(
             root_ids,