General CLI fixes when building clikt executable

build.gradle.kts

@@ -4,7 +4,7 @@ plugins {
 }
 
 group = "net.maizegenetics"
-version = "0.2.5"
+version = "0.2.6"
 
 repositories {
     mavenCentral()

pipeline_config.example.yaml

@@ -79,15 +79,19 @@ align_assemblies:
 # Converts MAF files from align_assemblies to compressed GVCF format
 # Automatically uses ref_fasta from align_assemblies and MAF output paths
 maf_to_gvcf:
-  sample_name: "optional_sample_name"    # Optional: Override sample name in GVCF
-  # input: "/custom/maf/files.txt"       # Optional: Custom MAF input (file, directory, or text list)
-  # output: "/custom/gvcf/output/"       # Optional: Custom output directory
+  # reference_file: "path/to/reference.fa"  # Optional: Reference FASTA file
+                                            #           Uses align_assemblies.ref_fasta if not specified
+  # maf_file: "/custom/maf/files.txt"       # Optional: MAF file, directory, or text list
+                                            #           Uses step 1 MAF outputs if not specified
+  # output_file: "sample.g.vcf.gz"          # Optional: Output GVCF filename (auto-generated if not specified)
+  # sample_name: "optional_sample_name"     # Optional: Override sample name in GVCF
+  # output_dir: "/custom/gvcf/output/"      # Optional: Custom output directory
 
 # Step 3: Downsample GVCF
 # Downsamples GVCF files at specified rates per chromosome
 # Automatically uses GVCF output directory from maf_to_gvcf
 downsample_gvcf:
-  ignore_contig: ""                      # Optional: Comma-separated patterns to ignore
+  ignore_contig: "__NO_MATCH__"          # Optional: Comma-separated patterns to ignore (currently needed)
   rates: "0.01,0.05,0.1,0.15,0.2"        # Optional: Comma-separated downsampling rates
   seed: 42                               # Optional: Random seed for reproducibility
   keep_ref: true                         # Optional: Keep reference blocks (true/false, default: true)
@@ -101,6 +105,7 @@ downsample_gvcf:
 convert_to_fasta:
   missing_records_as: "asRef"            # Optional: Missing records (asN, asRef, asNone)
   missing_genotype_as: "asN"             # Optional: Missing genotypes (asN, asRef, asNone)
+  ignore_contig: "__NO_MATCH__"          # Optional: Comma-separated patterns to skip (contigs matching will be ignored) (currently needed)
   # input: "/custom/gvcf/input/"         # Optional: Custom GVCF input (file, directory, or text list)
   # output: "/custom/fasta/output/"      # Optional: Custom output directory
 
@@ -142,9 +147,9 @@ pick_crossovers:
 # Automatically uses MAF files from align_assemblies (step 1)
 # Uses maf-convert tool (automatically downloaded if missing)
 create_chain_files:
-  jobs: 8                                # Optional: Number of parallel jobs (default: 8)
-  # input: "/custom/maf/input/"          # Optional: Custom MAF input (file, directory, or text list)
-  # output: "/custom/chain/output/"      # Optional: Custom output directory
+  jobs: 8                                     # Optional: Number of parallel jobs (default: 8)
+  # input: "/custom/maf/input/"               # Optional: Custom MAF input (file, directory, or text list)
+  # output: "/custom/chain/output/"           # Optional: Custom output directory
 
 # Step 8: Convert Coordinates
 # Converts crossover breakpoints from reference coordinates to assembly coordinates

src/main/kotlin/net/maizegenetics/commands/ConvertToFasta.kt

@@ -61,6 +61,11 @@ class ConvertToFasta : CliktCommand(name = "convert-to-fasta") {
     ).choice("asN", "asRef", "asNone", ignoreCase = true)
         .default("asN")
 
+    private val ignoreContig by option(
+        "--ignore-contig",
+        help = "Comma-separated list of string patterns to ignore (contigs containing these patterns will be skipped)"
+    ).default("")
+
     private val outputDirOption by option(
         "--output-dir", "-o",
         help = "Custom output directory (default: work_dir/output/04_fasta_results)"
@@ -153,6 +158,9 @@ class ConvertToFasta : CliktCommand(name = "convert-to-fasta") {
         logger.info("Reference FASTA: $refFasta")
         logger.info("Missing records as: $missingRecordsAs")
         logger.info("Missing genotype as: $missingGenotypeAs")
+        if (ignoreContig.isNotEmpty()) {
+            logger.info("Ignore contig patterns: $ignoreContig")
+        }
 
         // Collect GVCF files
         val gvcfFiles = collectGvcfFiles()
@@ -211,8 +219,8 @@ class ConvertToFasta : CliktCommand(name = "convert-to-fasta") {
             "FASTA file"
         )
 
-        // Clean up temp directory
-        if (tempDir.exists() && preparedFiles.any { it.parent == tempDir }) {
+        // Clean up temp directory - always clean up if it exists
+        if (tempDir.exists()) {
             logger.info("Cleaning up temporary uncompressed files")
             try {
                 tempDir.deleteRecursively()
@@ -246,6 +254,9 @@ class ConvertToFasta : CliktCommand(name = "convert-to-fasta") {
             add("--fasta-file=${refFasta.toAbsolutePath()}")
             add("--missing-records-as=$missingRecordsAs")
             add("--missing-genotype-as=$missingGenotypeAs")
+            if (ignoreContig.isNotEmpty()) {
+                add("--ignore-contig=$ignoreContig")
+            }
         }
 
         // Run MLImpute ConvertToFasta command

src/main/kotlin/net/maizegenetics/commands/DownsampleGvcf.kt

@@ -233,7 +233,8 @@ class DownsampleGvcf : CliktCommand(name = "downsample-gvcf") {
         logger.info("Output directory: $outputDir")
 
         // Clean up temp directory if not keeping uncompressed files
-        if (!keepUncompressed && tempDir.exists() && mlimputeInputDir == tempDir) {
+        // Always clean up if temp directory exists and was used or created
+        if (!keepUncompressed && tempDir.exists()) {
             logger.info("Cleaning up temporary uncompressed files")
             try {
                 tempDir.deleteRecursively()

src/main/kotlin/net/maizegenetics/commands/MafToGvcf.kt

@@ -143,15 +143,29 @@ class MafToGvcf : CliktCommand(name = "maf-to-gvcf") {
         logger.info("Sample name: $finalSampleName")
 
         // Determine output file name
-        val outputFileName = if (outputFile != null && isSingleMaf) {
-            outputFile!!.fileName
+        // Note: biokotlin-tools automatically compresses and adds .gz extension
+        val outputFileName: Path
+        val expectedOutputPath: Path
+
+        if (outputFile != null && isSingleMaf) {
+            val userFileName = outputFile!!.fileName.toString()
+            // If user specified .gz extension, strip it since biokotlin-tools adds it
+            if (userFileName.endsWith(".gz")) {
+                outputFileName = Path.of(userFileName.removeSuffix(".gz"))
+                expectedOutputPath = outputDir.resolve(userFileName)
+            } else {
+                outputFileName = outputFile!!.fileName
+                expectedOutputPath = outputDir.resolve("${userFileName}.gz")
+            }
         } else {
-            // Auto-generate output filename based on MAF filename (compressed)
-            Path.of("${mafBaseName}.g.vcf.gz")
+            // Auto-generate output filename based on MAF filename
+            // biokotlin-tools will add .gz when compressing
+            outputFileName = Path.of("${mafBaseName}.g.vcf")
+            expectedOutputPath = outputDir.resolve("${mafBaseName}.g.vcf.gz")
         }
 
         val fullOutputPath = outputDir.resolve(outputFileName)
-        logger.info("Output file: $fullOutputPath")
+        logger.info("Output file: $expectedOutputPath")
 
         // Run biokotlin-tools maf-to-gvcf-converter through pixi to use Java 21
         logger.info("Running biokotlin-tools maf-to-gvcf-converter")
@@ -173,7 +187,7 @@ class MafToGvcf : CliktCommand(name = "maf-to-gvcf") {
 
         logger.info("Conversion completed for: ${mafFile.name}")
 
-        // Return the GVCF file path
-        return fullOutputPath
+        // Return the GVCF file path (with .gz extension added by biokotlin-tools)
+        return expectedOutputPath
     }
 }