aldepyde 0.0.0a33__tar.gz → 0.0.0a37__tar.gz

{aldepyde-0.0.0a33 → aldepyde-0.0.0a37}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: aldepyde
-Version: 0.0.0a33
+Version: 0.0.0a37
 Summary: A package of chemistry and biochemical tools
 Author-email: Nate McMurray <nate.mcmurray13@gmail.com>
 License: MIT License

aldepyde-0.0.0a37/aldepyde/databases/SCOPe_Astral.py

@@ -0,0 +1,73 @@
+import io
+
+from aldepyde.databases._database import local_database
+import operator
+from contextlib import nullcontext
+import re
+
+class scop_parser(local_database):
+    op = {
+        "and": lambda a,b,c: a and b and c,
+        "or": lambda a,b,c: a or b or c
+    }
+
+    def fetch(self, url):
+        pass
+
+    def fetch_code(self, codes):
+        pass
+
+    def parse(self, text):
+        pass
+
+    def extract_all_scop(self):
+        pass
+
+    def partition_scope(self):
+        pass
+
+    def extract_all_astral(self):
+        lines = self.fp.readlines()
+        entry = b""
+        for line in lines:
+            if line.startswith(b">") and len(entry) > 0:
+                yield entry
+                entry = b""
+            entry += line
+        yield entry
+
+    # TODO allow a list of search parameters. Big challenge to make efficient, but could be cute
+    def partition_astral(self, destination:None|str=None, append=False, class_name:str=b'',contains_id:str=b'' , contains_desc:str=b'', mode="and") -> dict:
+        mode = mode.lower()
+        # Everything is a byte string in order to play nicely with future parent methods
+        if isinstance(class_name, str):
+            class_name = class_name.encode('utf-8')
+        if isinstance(contains_desc, str):
+            contains_desc = contains_desc.encode('utf-8')
+        if isinstance(contains_id, str):
+            contains_id = contains_id.encode('utf-8')
+        if mode != "and" and mode != "or":
+            raise ValueError("mode must be \"and\" or \"or\".")
+        logic = scop_parser.op[mode]
+        regex = re.compile(b">[a-zA-Z0-9_.]* *[a-l](.[0-9]+)?(.[0-9]+)?(.[0-9]+)?")
+        if append:
+            file_context = open(destination, 'ab') if destination is not None else nullcontext(io.BytesIO())
+        else:
+            file_context = open(destination, 'wb') if destination is not None else nullcontext(io.BytesIO())
+        with file_context as fp:
+            ret_dict = dict()
+            for line in self.extract_all_astral():
+                identifiers = regex.search(line).group().split()
+                id = identifiers[0]
+                cls = identifiers[1]
+                unmatched_spl = regex.sub(b'', line).split(b'\n')
+                desc = unmatched_spl[0]
+                sequence = unmatched_spl[1:]
+                if logic(class_name.lower() in cls.lower(), contains_id.lower() in id.lower(), contains_desc.lower() in desc.lower()):
+                    ret_dict[id] = { # Yes, I know '>' isn't part of the FASTA identifier. This keeps things more consistant
+                        "class" : cls,
+                        "description" : desc,
+                        "sequence" : b"".join(sequence)
+                    }
+                    fp.write(line)
+        return ret_dict

{aldepyde-0.0.0a33 → aldepyde-0.0.0a37}/aldepyde/databases/UniRef.py

@@ -1,42 +1,45 @@
 import zlib
-
-from aldepyde.databases.RemoteFileHandler import RemoteFileHandler
-from aldepyde.databases._database import _database
+from aldepyde.databases._database import streamable_database
 from aldepyde.utils import ProgressBar
-import os
-import gzip
 
-class uniref_parser(_database):
+class uniref_parser(streamable_database):
     def __init__(self):
         super().__init__()
 
     # TODO single entry parsing
     # TODO store metadata upon request
+    # TODO implement abstract methods
 
     @staticmethod
-    def stream_uniref_gz(filepath, chunk_size=8192, use_progress_bar=False):
-        raw_stream, size = _database.open_stream(filepath)
+    def stream_uniref_gz(filepath, chunk_size=8192, use_progress_bar=False, stitch=False):
+        raw_stream, size = streamable_database.open_stream(filepath)
         pbar = ProgressBar(size//chunk_size) if use_progress_bar else None
         decompressor = zlib.decompressobj(16 + zlib.MAX_WBITS)
         try:
-            while True:
-                comp_chunk = raw_stream.read(chunk_size)
-                if not comp_chunk:
-                    break
-                if pbar is not None:
-                    pbar.update()
-                decomp_chunk = decompressor.decompress(comp_chunk)
-                if decomp_chunk:
-                    yield decomp_chunk
-            final = decompressor.flush()
-            if final:
-                yield final
+            if not stitch:
+                while True:
+                    comp_chunk = raw_stream.read(chunk_size)
+                    if not comp_chunk:
+                        break
+                    if pbar is not None:
+                        pbar.update()
+                    decomp_chunk = decompressor.decompress(comp_chunk)
+                    if decomp_chunk:
+                        yield decomp_chunk
+                final = decompressor.flush()
+                if final:
+                    yield final
+            else:
+                # Really hacky solution for now
+                # TODO Clean this up
+                yield from uniref_parser.stitch_streamed_sequences(
+                    uniref_parser.stream_uniref_gz(filepath=filepath, chunk_size=chunk_size, use_progress_bar=use_progress_bar, stitch=False))
         finally:
             raw_stream.close()
 
     @staticmethod
     def download_file(url, destination, chunk_size=8192, use_progress_bar=False):
-        raw_stream, size = _database.open_stream(url)
+        raw_stream, size = streamable_database.open_stream(url)
         pbar = ProgressBar(size // chunk_size) if use_progress_bar else None
         with open(destination, 'wb') as fp:
             while True:
@@ -48,8 +51,6 @@ class uniref_parser(_database):
                 fp.write(chunk)
 
 
-
-
     @staticmethod
     def stitch_streamed_sequences(stream, as_str=True):
         buffer = b''

{aldepyde-0.0.0a33 → aldepyde-0.0.0a37}/aldepyde/databases/_database.py

@@ -5,7 +5,7 @@ import os
 from typing import Tuple, BinaryIO
 from io import TextIOWrapper
 
-class _database(ABC):
+class streamable_database(ABC):
 
     def __init__(self):
         pass
@@ -38,4 +38,35 @@ class _database(ABC):
     # Yes, I know the first conditionals do the same thing
 
     def __call__(self):
-        pass
+        pass
+
+class local_database(ABC):
+
+    def __init__(self, filepath=None, as_fp=False):
+        self.fp = None
+        self.as_fp = as_fp
+        self.size = None
+        self.load_path(filepath)
+
+    def load_path(self, filepath):
+        self.filepath = filepath
+
+    def get_pointer(self):
+        return self.fp
+
+    def __enter__(self):
+        self.fp, self.size = local_database.open_stream(self.filepath)
+        if self.as_fp:
+            return self.fp
+        else:
+            return self
+
+    def __exit__(self, exc_type, exc_val, exc_tb):
+        if self.fp is not None:
+            self.fp.close()
+        self.fp = None
+
+    @staticmethod
+    def open_stream(source:str) -> Tuple[BinaryIO, int] | None:
+        size = os.path.getsize(source)
+        return open(source, 'rb'), size

{aldepyde-0.0.0a33 → aldepyde-0.0.0a37}/aldepyde.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: aldepyde
-Version: 0.0.0a33
+Version: 0.0.0a37
 Summary: A package of chemistry and biochemical tools
 Author-email: Nate McMurray <nate.mcmurray13@gmail.com>
 License: MIT License

{aldepyde-0.0.0a33 → aldepyde-0.0.0a37}/aldepyde.egg-info/SOURCES.txt

@@ -31,6 +31,7 @@ aldepyde/data/RemoteFileHandler.py
 aldepyde/data/__init__.py
 aldepyde/databases/PDB.py
 aldepyde/databases/RemoteFileHandler.py
+aldepyde/databases/SCOPe_Astral.py
 aldepyde/databases/UniRef.py
 aldepyde/databases/__init__.py
 aldepyde/databases/_database.py

{aldepyde-0.0.0a33 → aldepyde-0.0.0a37}/pyproject.toml

@@ -5,7 +5,7 @@ build-backend = "setuptools.build_meta"
 "aldepyde" = ["json/*.json"]
 [project]
 name = "aldepyde"
-version = "0.0.0a33"
+version = "0.0.0a37"
 authors = [
   { name="Nate McMurray", email="nate.mcmurray13@gmail.com" },
 ]