patch version: fix missing imports

pgscatalog.match/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "pgscatalog.match"
-version = "0.3.2"
+version = "0.3.3"
 description = "Tools for matching variants in PGS scoring files and target variant information files"
 authors = ["Benjamin Wingfield <[email protected]>", "Samuel Lambert <[email protected]>", "Laurent Gil <[email protected]>"]
 readme = "README.md"

pgscatalog.match/src/pgscatalog/match/cli/_write.py

@@ -6,11 +6,11 @@
 import gzip
 import itertools
 
-from pgscatalog.core import MatchRateError
+from pgscatalog.core.lib.pgsexceptions import MatchRateError
 
-from ..lib.plinkscorefiles import PlinkScoreFiles
+from pgscatalog.match.lib.plinkscorefiles import PlinkScoreFiles
 
-from ._config import Config
+from pgscatalog.match.cli._config import Config
 
 
 def write_matches(matchresults, score_df):

pgscatalog.match/src/pgscatalog/match/cli/match_cli.py

@@ -9,16 +9,16 @@
 
 import polars as pl
 
-from .. import (
+from pgscatalog.match import (
     VariantFrame,
     ScoringFileFrame,
     match_variants,
     MatchResult,
     MatchResults,
 )
 
-from ._config import Config
-from ._write import write_matches
+from pgscatalog.match.cli._config import Config
+from pgscatalog.match.cli._write import write_matches
 
 logger = logging.getLogger(__name__)
 

pgscatalog.match/src/pgscatalog/match/cli/merge_cli.py

@@ -4,10 +4,10 @@
 import pathlib
 import shutil
 
-from ._config import Config
-from ._write import write_matches
+from pgscatalog.match.cli._config import Config
+from pgscatalog.match.cli._write import write_matches
 
-from ..lib import ScoringFileFrame, MatchResult, MatchResults
+from pgscatalog.match.lib import ScoringFileFrame, MatchResult, MatchResults
 
 import polars as pl
 

pgscatalog.match/src/pgscatalog/match/lib/__init__.py

@@ -1,9 +1,9 @@
 import logging
 
-from .variantframe import VariantFrame
-from .scoringfileframe import ScoringFileFrame, match_variants
-from .matchresult import MatchResult, MatchResults
-from .plinkscorefiles import PlinkScoreFiles
+from pgscatalog.match.lib.variantframe import VariantFrame
+from pgscatalog.match.lib.scoringfileframe import ScoringFileFrame, match_variants
+from pgscatalog.match.lib.matchresult import MatchResult, MatchResults
+from pgscatalog.match.lib.plinkscorefiles import PlinkScoreFiles
 
 log_fmt = "%(name)s: %(asctime)s %(levelname)-8s %(message)s"
 logging.basicConfig(format=log_fmt, datefmt="%Y-%m-%d %H:%M:%S")

pgscatalog.match/src/pgscatalog/match/lib/_arrow.py

@@ -6,12 +6,15 @@
 """
 import pathlib
 import tempfile
+from typing import Optional
 
 import polars as pl
 from functools import singledispatch
 
-import polars.exceptions
-from pgscatalog.core import NormalisedScoringFile, TargetVariants, TargetType
+from polars.io.csv import BatchedCsvReader
+
+from pgscatalog.match.lib.normalisedscoringfile import NormalisedScoringFile
+from pgscatalog.match.lib.targetvariants import TargetVariants, TargetType
 
 
 @singledispatch
@@ -25,7 +28,7 @@ def loose(path, tmpdir=None):
 
 
 @loose.register
-def _(path: NormalisedScoringFile, tmpdir=None):
+def _(path: NormalisedScoringFile, tmpdir=None):  # type: ignore
     """Write NormalisedScoringFiles to a list of arrow IPC files"""
     if tmpdir is None:
         tmpdir = tempfile.mkdtemp()
@@ -47,7 +50,7 @@ def _(path: NormalisedScoringFile, tmpdir=None):
 
 
 @loose.register
-def _(path: TargetVariants, tmpdir=None):
+def _(path: TargetVariants, tmpdir=None):  # type: ignore
     """Writes TargetVariants to a list of arrow IPC files"""
     if tmpdir is None:
         tmpdir = tempfile.mkdtemp()
@@ -63,7 +66,14 @@ def _(path: TargetVariants, tmpdir=None):
                 "column_5": pl.String,
                 "column_6": pl.String,
             }
-            new_colnames = ["#CHROM", "ID", "CM", "POS", "REF", "ALT"]
+            new_colnames: Optional[list[str]] = [
+                "#CHROM",
+                "ID",
+                "CM",
+                "POS",
+                "REF",
+                "ALT",
+            ]
             header = False
             comment = None
         case TargetType.PVAR:
@@ -94,7 +104,9 @@ def _(path: TargetVariants, tmpdir=None):
     return batch_read(reader, tmpdir=tmpdir, cols_keep=cols_keep)
 
 
-def batch_read(reader, tmpdir, cols_keep) -> list[pathlib.Path]:
+def batch_read(
+    reader: BatchedCsvReader, tmpdir: pathlib.Path, cols_keep: list[str]
+) -> list[pathlib.Path]:
     """Read a CSV in batches and write them to temporary files"""
     arrowpaths = []
     # batch_size should be >= thread pool size, so tasks will be distributed amongst workers

pgscatalog.match/src/pgscatalog/match/lib/_match/label.py

@@ -12,18 +12,18 @@
 import polars as pl
 from xopen import xopen
 
-from .preprocess import complement_valid_alleles
+from pgscatalog.match.lib._match.preprocess import complement_valid_alleles
 
 logger = logging.getLogger(__name__)
 
 
 def label_matches(
     df: pl.LazyFrame,
-    keep_first_match,
-    remove_ambiguous,
-    remove_multiallelic,
-    skip_flip,
-    filter_IDs,
+    keep_first_match: bool,
+    remove_ambiguous: bool,
+    remove_multiallelic: bool,
+    skip_flip: bool,
+    filter_IDs: pathlib.Path,
 ) -> pl.LazyFrame:
     """Label match candidates with additional metadata. Column definitions:
 
@@ -220,7 +220,9 @@ def _label_duplicate_id(df: pl.LazyFrame, keep_first_match: bool) -> pl.LazyFram
     )
 
 
-def _label_biallelic_ambiguous(df: pl.LazyFrame, remove_ambiguous) -> pl.LazyFrame:
+def _label_biallelic_ambiguous(
+    df: pl.LazyFrame, remove_ambiguous: bool
+) -> pl.LazyFrame:
     """
     Identify ambiguous variants (A/T & C/G SNPs)
 

pgscatalog.match/src/pgscatalog/match/lib/_plinkframe.py

@@ -6,7 +6,7 @@
 import gzip
 import pathlib
 
-from ._match.plink import plinkify, pivot_score
+from pgscatalog.match.lib._match.plink import plinkify, pivot_score
 
 
 class PlinkFrame:

pgscatalog.match/src/pgscatalog/match/lib/matchresult.py

@@ -3,12 +3,12 @@
 
 import polars as pl
 
-from pgscatalog.core import ZeroMatchesError, MatchRateError
+from pgscatalog.core.lib.pgsexceptions import ZeroMatchesError, MatchRateError
 
-from ._plinkframe import PlinkFrames
-from ._match.label import label_matches
-from ._match.filter import filter_scores
-from ._match.log import make_logs, check_log_count, make_summary_log
+from pgscatalog.match.lib._plinkframe import PlinkFrames
+from pgscatalog.match.lib._match.label import label_matches
+from pgscatalog.match.lib._match.filter import filter_scores
+from pgscatalog.match.lib._match.log import make_logs, check_log_count, make_summary_log
 
 logger = logging.getLogger(__name__)
 

pgscatalog.match/src/pgscatalog/match/lib/normalisedscoringfile.py

@@ -0,0 +1,57 @@
+import csv
+
+from xopen import xopen
+
+from pgscatalog.core.lib.models import ScoreVariant
+
+
+def _read_normalised_rows(path):
+    with xopen(path) as f:
+        reader = csv.DictReader(f, delimiter="\t")
+        for row in reader:
+            yield ScoreVariant(**row)
+
+
+class NormalisedScoringFile:
+    """This class represents a ScoringFile that's been normalised to have a consistent format
+
+    Its main purpose is to provide a convenient way to iterate over variants
+
+    # TODO: replace with a pydantic model in pgscatalog.core
+    """
+
+    def __init__(self, path):
+        try:
+            with xopen(path):
+                pass
+        except TypeError:
+            self.is_path = False
+            self.path = str(path)
+        else:
+            self.is_path = True
+            self.path = path
+        finally:
+            # either a ScoringFile or a path to a combined file
+            self._scoringfile = path
+
+    def __iter__(self):
+        yield from self.variants
+
+    @property
+    def variants(self):
+        if self.is_path:
+            # get a fresh generator from the file
+            self._variants = _read_normalised_rows(self._scoringfile)
+        else:
+            # get a fresh generator from the normalise() method
+            self._variants = self._scoringfile.normalise()
+
+        return self._variants
+
+    def __repr__(self):
+        if self.is_path:
+            x = f"{repr(str(self._scoringfile))}"
+        else:
+            x = f"{repr(self._scoringfile)}"
+
+        return f"{type(self).__name__}({x})"

pgscatalog.match/src/pgscatalog/match/lib/scoringfileframe.py

@@ -4,12 +4,12 @@
 
 import polars as pl
 
-from pgscatalog.core import NormalisedScoringFile
+from pgscatalog.match.lib.normalisedscoringfile import NormalisedScoringFile
 
-from ._arrow import loose
-from ._match.preprocess import complement_valid_alleles
-from ._match.match import get_all_matches
-from .matchresult import MatchResult
+from pgscatalog.match.lib._arrow import loose
+from pgscatalog.match.lib._match.preprocess import complement_valid_alleles
+from pgscatalog.match.lib._match.match import get_all_matches
+from pgscatalog.match.lib.matchresult import MatchResult
 
 logger = logging.getLogger(__name__)