Issue #61/migrate sdfield

rdock-utils/pyproject.toml

@@ -1,3 +1,15 @@
+[project]
+name = "rdock-utils"
+version = "0.1.0"
+description = "Utilities for working with RDock and operating on SD files"
+requires-python = ">=3.10.0"
+
+[project.scripts]
+sdfield = "rdock_utils.sdfield:main"
+
+[project.urls]
+Repository = "https://github.com/CBDD/rDock.git"
+
 [tool.black]
 line-length = 119
 target-version = ['py312']

rdock-utils/rdock_utils/parser.py

@@ -0,0 +1,76 @@
+# Standard Library
+import logging
+from io import StringIO
+from typing import Any, TextIO
+
+logger = logging.getLogger("SDParser")
+
+
+class FastSDMol:
+    def __init__(self, lines: list[str], data: dict[str, str]) -> None:
+        self.lines = lines
+        self.data = data
+
+    @classmethod
+    def read(cls, source: TextIO) -> "FastSDMol | None":
+        lines: list[str] = []
+        data: dict[str, str] = {}
+        terminator_found = False
+        for line in source:
+            if line.startswith("$$$$"):
+                terminator_found = True
+                break
+            if not line.startswith(">"):
+                lines.append(line)
+                continue
+
+            # dealing with fields
+            field_name = cls.parse_field_name(line)
+            field_value = source.readline()
+            if field_value.startswith("$$$$"):
+                terminator_found = True
+                logger.warning(
+                    f"found end of molecule {lines[0]} while looking for field {field_name} value."
+                    " defaulting to empty string."
+                )
+                data[field_name] = ""
+                break
+            data[field_name] = field_value.strip("\n")
+            discard_line = source.readline()
+            if discard_line.startswith("$$$$"):
+                terminator_found = True
+                logger.warning(f"found end of molecule {lines[0]} while expecting empty line after field {field_name}")
+                break
+
+        if not terminator_found and all(line.strip() == "" for line in lines):
+            return None
+
+        if len(lines) >= 4:
+            return cls(lines, data)
+
+        # if we've reached this point, we have an invalid molecule
+        raise ValueError(f"invalid molecule: {lines}")
+
+    @staticmethod
+    def parse_field_name(field_line: str) -> str:
+        field_start = field_line.find("<") + 1
+        field_end = field_line.find(">", 1)
+        return field_line[field_start:field_end]
+
+    @staticmethod
+    def str_field(field_name: str, field_value: Any) -> str:
+        return f">  <{field_name}>\n{field_value}\n\n"
+
+    def __repr__(self) -> str:
+        str_io = StringIO()
+        self.write(str_io)
+        return str_io.getvalue()
+
+    def __str__(self) -> str:
+        return f"<Molecule {self.lines[0]}>"
+
+    def write(self, dest: TextIO) -> None:
+        dest.writelines(self.lines)
+        for field_name, field_value in self.data.items():
+            dest.write(self.str_field(field_name, field_value))
+        dest.write("$$$$")

rdock-utils/rdock_utils/sdfield.py

@@ -0,0 +1,55 @@
+# Standard Library
+import argparse
+import sys
+from logging import getLogger
+from typing import Iterable, TextIO
+
+# Local imports
+from .parser import FastSDMol
+
+logger = getLogger("sdfield")
+
+
+def get_parser() -> argparse.ArgumentParser:
+    parser = argparse.ArgumentParser(description="Adding fields to SD files")
+    parser.add_argument("fieldname", type=str, help="name of the field to be added")
+    parser.add_argument("value", type=str, help="value of the field to be added")
+    infile_help = "input file[s] to be processed. if not provided, stdin is used."
+    parser.add_argument("infile", type=str, nargs="*", help=infile_help)
+    outfile_help = "output file. if not provided, stdout is used."
+    parser.add_argument("-o", "--outfile", default=None, type=str, help=outfile_help)
+
+    return parser
+
+
+def inputs_generator(inputs: list[str]) -> Iterable[TextIO]:
+    if not inputs:
+        yield sys.stdin
+    else:
+        for infile in inputs:
+            yield open(infile, "r")
+
+
+def read_molecules(file: TextIO) -> Iterable[FastSDMol]:
+    while True:
+        try:
+            mol = FastSDMol.read(file)
+            if mol is None:
+                break
+            yield mol
+        except ValueError as e:
+            logger.warning(f"error reading molecule: {e}")
+
+
+def main(argv: list[str] | None = None) -> None:
+    parser = get_parser()
+    args = parser.parse_args(argv)
+    inputs = inputs_generator(args.infile)
+    for source in inputs:
+        for molecule in read_molecules(source):
+            molecule.data[args.fieldname] = args.value
+            print(repr(molecule))
+
+
+if __name__ == "__main__":
+    main()

rdock-utils/setup.py

@@ -1,20 +1,12 @@
 #!/usr/bin/env python
 
 # Dependencies
-from setuptools import setup
-
-raise Exception("Please make sure you have modified all necessary attributes before pip installing the package")
+from setuptools import find_packages, setup
 
 setup(
     name="rdock-utils",
     version="0.01",
-    description="",
-    author="",
-    author_email="",
-    url="",
-    packages=[],
-    # inlcude_package_data=True,
-    # package_data={'package.module':[folder/with/data/*]}
-    # scripts=[],
+    url="https://github.com/CBDD/rDock.git",
+    packages=find_packages(include=["rdock_utils"]),
     install_requires=[],
 )