rcsb-embedding-model 0.0.32__py3-none-any.whl → 0.0.33__py3-none-any.whl

rcsb_embedding_model/utils/data.py

@@ -4,6 +4,7 @@ import gzip
 from io import StringIO, BytesIO
 
 import torch
+from requests import RequestException
 
 
 def collate_seq_embeddings(batch_list):
@@ -52,13 +53,10 @@ def stringio_from_url(url):
                 return StringIO(f.read())
         else:
             return StringIO(response.text)
-    except requests.exceptions.RequestException as e:
-        print(f"Error fetching URL: {e}")
-        return None
+    except RequestException as e:
+        raise RuntimeError(f"Error fetching URL: {e}")
     except (OSError, gzip.BadGzipFile) as e:
-        print(f"Error decompressing gzip file: {e}")
-        return None
-
+        raise RuntimeError(f"Error decompressing gzip file: {e}")
 
 
 def concatenate_tensors(file_list, max_residues, dim=0):

rcsb_embedding_model/utils/structure_parser.py

@@ -9,17 +9,20 @@ def get_structure_from_src(
         chain_id=None,
         assembly_id=None
 ):
-    if structure_format == "pdb":
-        pdb_file = PDBFile.read(src_structure)
-        structure = __get_pdb_structure(pdb_file, assembly_id)
-    elif structure_format == "mmcif":
-        cif_file = CIFFile.read(src_structure)
-        structure = __get_structure(cif_file, assembly_id)
-    elif structure_format == "binarycif":
-        cif_file = BinaryCIFFile.read(src_structure)
-        structure = __get_structure(cif_file, assembly_id)
-    else:
-        raise RuntimeError(f"Unknown file format {structure_format}")
+    try:
+        if structure_format == "pdb":
+            pdb_file = PDBFile.read(src_structure)
+            structure = __get_pdb_structure(pdb_file, assembly_id)
+        elif structure_format == "mmcif":
+            cif_file = CIFFile.read(src_structure)
+            structure = __get_structure(cif_file, assembly_id)
+        elif structure_format == "binarycif":
+            cif_file = BinaryCIFFile.read(src_structure)
+            structure = __get_structure(cif_file, assembly_id)
+        else:
+            raise RuntimeError(f"Unknown file format {structure_format}")
+    except Exception as e:
+        raise RuntimeError(f"Error reading structure from {src_structure}: {e}")
 
     if chain_id is not None:
         return structure[structure.chain_id == chain_id]
@@ -38,14 +41,17 @@ def get_protein_chains(structure, min_res_n=0):
 
 
 def get_assemblies(structure, structure_format="mmcif"):
-    if structure_format == "pdb":
-        return tuple(list_pdb_assemblies(PDBFile.read(structure)))
-    elif structure_format == "mmcif":
-        return tuple(list_assemblies(CIFFile.read(structure)).keys())
-    elif structure_format == "binarycif":
-        return tuple(list_assemblies(BinaryCIFFile.read(structure)))
-    else:
-        raise RuntimeError(f"Unknown file format {structure_format}")
+    try:
+        if structure_format == "pdb":
+            return tuple(list_pdb_assemblies(PDBFile.read(structure)))
+        elif structure_format == "mmcif":
+            return tuple(list_assemblies(CIFFile.read(structure)).keys())
+        elif structure_format == "binarycif":
+            return tuple(list_assemblies(BinaryCIFFile.read(structure)))
+        else:
+            raise RuntimeError(f"Unknown file format {structure_format}")
+    except Exception as e:
+        raise RuntimeError(f"Error reading assemblies from {structure}: {e}")
 
 
 def rename_atom_ch(atom_ch, ch="A"):

{rcsb_embedding_model-0.0.32.dist-info → rcsb_embedding_model-0.0.33.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rcsb-embedding-model
-Version: 0.0.32
+Version: 0.0.33
 Summary: Protein Embedding Model for Structure Search
 Project-URL: Homepage, https://github.com/rcsb/rcsb-embedding-model
 Project-URL: Issues, https://github.com/rcsb/rcsb-embedding-model/issues

{rcsb_embedding_model-0.0.32.dist-info → rcsb_embedding_model-0.0.33.dist-info}/RECORD

@@ -18,13 +18,13 @@ rcsb_embedding_model/modules/chain_module.py,sha256=KsZw2uagO4rpAKWv6ivqEMxIEzgt
 rcsb_embedding_model/modules/esm_module.py,sha256=otJRbCb319nCCob_4E1W_UClhkex9eDqcCyzWQO-vIs,740
 rcsb_embedding_model/modules/structure_module.py,sha256=4js02XzKvhc_G26ELsGhJ9SCi_wlvtVolObxfWt3BhE,1077
 rcsb_embedding_model/types/api_types.py,sha256=SCwALwvEb0KRKaoWKbuN7JyfOH-1whsI0Z4ki41dht8,1235
-rcsb_embedding_model/utils/data.py,sha256=BOjYdIRHrFqk8qFuKGrgCtVyfDupzgOVmH_0C-ecMvg,3813
+rcsb_embedding_model/utils/data.py,sha256=ThrcYycIizsV_Ycn6PPxF12JRr1m2K-v8TsIaVqx10A,3816
 rcsb_embedding_model/utils/model.py,sha256=xr3p02ohOgJ5UInwdIupN68Oq4yvNFhxobZRacS1adg,953
-rcsb_embedding_model/utils/structure_parser.py,sha256=z_Aid5QgGUAJkLL-fMzG2Uh2wPrHXmBdjhLn_YkPl-k,3035
+rcsb_embedding_model/utils/structure_parser.py,sha256=mS70NcliGETWPdr0rOiypiR-u_N_H6HeRoZKwVfQrAU,3342
 rcsb_embedding_model/utils/structure_provider.py,sha256=eWtxjkPpmRfmil_DKR1J6miaXR3lQ28DF5O0qrqSgGA,786
 rcsb_embedding_model/writer/batch_writer.py,sha256=rTFNasB0Xp4-XCNTXKeEWZxSrb7lvZytoRldJUWn9Jg,3312
-rcsb_embedding_model-0.0.32.dist-info/METADATA,sha256=JWSb2a_wZ2Y8YWzyM3vk_QG--ukrIGsI5Q1-R3bApjs,5351
-rcsb_embedding_model-0.0.32.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-rcsb_embedding_model-0.0.32.dist-info/entry_points.txt,sha256=MK11jTIEmaV-x4CkPX5IymDaVs7Ky_f2xxU8BJVZ_9Q,69
-rcsb_embedding_model-0.0.32.dist-info/licenses/LICENSE.md,sha256=oUaHiKgfBkChth_Sm67WemEvatO1U0Go8LHjaskXY0w,1522
-rcsb_embedding_model-0.0.32.dist-info/RECORD,,
+rcsb_embedding_model-0.0.33.dist-info/METADATA,sha256=eOV5-e3IlkBYzZlIduZSLGB1k7VwvliKsF6Jf0QZaOQ,5351
+rcsb_embedding_model-0.0.33.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+rcsb_embedding_model-0.0.33.dist-info/entry_points.txt,sha256=MK11jTIEmaV-x4CkPX5IymDaVs7Ky_f2xxU8BJVZ_9Q,69
+rcsb_embedding_model-0.0.33.dist-info/licenses/LICENSE.md,sha256=oUaHiKgfBkChth_Sm67WemEvatO1U0Go8LHjaskXY0w,1522
+rcsb_embedding_model-0.0.33.dist-info/RECORD,,