fix: Views operations (#109)

* fix: Case sensitive info fields

* fix: Views operations

Author: mwiewior

Cargo.lock

@@ -1,6 +1,6 @@
 [package]
 name = "polars_bio"
-version = "0.7.3"
+version = "0.7.4"
 edition = "2021"
 
 [lib]
@@ -36,7 +36,7 @@ polars-python = { git = "https://github.com/mwiewior/polars.git" , rev = "9d4fca
 
 #exon ="0.32.4"
 exon = { git = "https://github.com/mwiewior/exon.git", rev="d134d923e6c592a9972d93215a12c759c70a7ed5"}
-datafusion-vcf = { git = "https://github.com/biodatageeks/datafusion-bio-formats.git/", rev = "4d7e72b4588025d3a020b1a55a16f6484ad846d9"}
+datafusion-vcf = { git = "https://github.com/biodatageeks/datafusion-bio-formats.git/", rev = "cc38b4ac3fe2d252245200422930fa879e92b92d"}
 async-trait = "0.1.86"
 futures = "0.3.31"
 coitrees = "0.4.0"

Cargo.toml

@@ -378,4 +378,4 @@ def from_polars(name: str, df: Union[pl.DataFrame, pl.LazyFrame]) -> None:
 def _cleanse_infos(t: Union[list[str], None]) -> Union[list[str], None]:
     if t is None:
         return None
-    return [x.upper().strip() for x in t]
+    return [x.strip() for x in t]

polars_bio/io.py

@@ -4,7 +4,7 @@ build-backend = "maturin"
 
 [project]
 name = "polars-bio"
-version = "0.7.3"
+version = "0.7.4"
 description = "Blazing fast genomic operations on large Python dataframes"
 authors = []
 requires-python = ">=3.9"

pyproject.toml

@@ -9,9 +9,9 @@ pub(crate) fn nearest_query(query_params: QueryParams) -> String {
             a.{} AS {}{}, -- pos_end
             b.{} AS {}{}, -- contig
             b.{} AS {}{}, -- pos_start
-            b.{} AS {}{},  -- pos_end
-            {}
+            b.{} AS {}{}  -- pos_end
             {}
+            {},
        CAST(
        CASE WHEN b.{} >= a.{}
             THEN
@@ -52,16 +52,16 @@ pub(crate) fn nearest_query(query_params: QueryParams) -> String {
                     "a".to_string(),
                     query_params.suffixes.0.clone(),
                 )
-                + ","
         } else {
             "".to_string()
         },
         if !query_params.other_columns_2.is_empty() {
-            format_non_join_tables(
-                query_params.other_columns_2.clone(),
-                "b".to_string(),
-                query_params.suffixes.1.clone(),
-            )
+            ",".to_string()
+                + &format_non_join_tables(
+                    query_params.other_columns_2.clone(),
+                    "b".to_string(),
+                    query_params.suffixes.1.clone(),
+                )
         } else {
             "".to_string()
         },
@@ -100,7 +100,7 @@ pub(crate) fn overlap_query(query_params: QueryParams) -> String {
                 {}
                 {}
             FROM
-                {} a, {} b
+                {} AS a, {} AS b
             WHERE
                 a.{}=b.{}
             AND
@@ -133,16 +133,16 @@ pub(crate) fn overlap_query(query_params: QueryParams) -> String {
                     "a".to_string(),
                     query_params.suffixes.0.clone(),
                 )
-                + ","
         } else {
             "".to_string()
         },
         if !query_params.other_columns_2.is_empty() {
-            format_non_join_tables(
-                query_params.other_columns_2.clone(),
-                "b".to_string(),
-                query_params.suffixes.1.clone(),
-            )
+            ",".to_string()
+                + &format_non_join_tables(
+                    query_params.other_columns_2.clone(),
+                    "b".to_string(),
+                    query_params.suffixes.1.clone(),
+                )
         } else {
             "".to_string()
         },

src/query.rs

@@ -17,6 +17,36 @@ def test_fields(self):
         assert self.df["cigar"][4] == "101M"
 
 
+class TestIOVCFInfo:
+    vcf_big = "gs://gcp-public-data--gnomad/release/2.1.1/liftover_grch38/vcf/genomes/gnomad.genomes.r2.1.1.sites.liftover_grch38.vcf.bgz"
+    vcf_infos_mixed_cases = (
+        pb.read_vcf(vcf_big, info_fields=["AF", "vep"], thread_num=1).limit(1).collect()
+    )
+
+    def test_count(self):
+        assert len(self.vcf_infos_mixed_cases) == 1
+
+
+class TestVCFViewsOperations:
+    def test_view(self):
+        vcf_big = "gs://gcp-public-data--gnomad/release/2.1.1/liftover_grch38/vcf/genomes/gnomad.genomes.r2.1.1.sites.liftover_grch38.vcf.bgz"
+        pb.register_vcf(vcf_big, "gnomad_big", info_fields=["AF", "vep"], thread_num=1)
+        pb.register_view(
+            "v_gnomad_big",
+            "SELECT chrom, start, end, split_part(vep, '|', 3) AS impact from gnomad_big where array_element(af,1)=0 and split_part(vep, '|', 3) in ('HIGH', 'MODERATE') limit 10",
+        )
+        vcf_sv = "gs://gcp-public-data--gnomad/release/4.1/genome_sv/gnomad.v4.1.sv.sites.vcf.gz"
+        pb.register_vcf(
+            vcf_sv, "gnomad_sv", thread_num=1, info_fields=["SVTYPE", "SVLEN"]
+        )
+        pb.register_view(
+            "v_gnomad_sv", "SELECT chrom, start, end FROM gnomad_sv limit 100"
+        )
+        assert len(pb.sql("SELECT * FROM v_gnomad_big").collect()) == 10
+        assert len(pb.nearest("v_gnomad_sv", "v_gnomad_big").collect()) == 100
+        assert len(pb.overlap("v_gnomad_sv", "v_gnomad_big").collect()) == 43
+
+
 class TestIOVCF:
     df_bgz = pb.read_vcf(f"{DATA_DIR}/io/vcf/vep.vcf.bgz").collect()
     df_none = pb.read_vcf(f"{DATA_DIR}/io/vcf/vep.vcf").collect()