fix extra suffixes when demultiplexing

This ensures that filenames for reports and unidentified reads do not
get extra suffixes, when explicitly named and demultiplexing is enabled

Author: MikkelSchubert

src/userconfig.cpp

@@ -1384,23 +1384,23 @@ userconfig::get_output_filenames() const
 {
   output_files files;
 
-  files.settings_json = new_output_file("--out-json", { ".json" }).name;
-  files.settings_html = new_output_file("--out-html", { ".html" }).name;
+  files.settings_json = new_output_file("--out-json", { ".json" }, false).name;
+  files.settings_html = new_output_file("--out-html", { ".html" }, false).name;
 
   const std::string ext{ output_files::file_extension(out_file_format) };
   const std::string out1 = (interleaved_output ? "" : ".r1") + ext;
   const std::string out2 = (interleaved_output ? "" : ".r2") + ext;
 
   if (is_demultiplexing_enabled()) {
     files.unidentified_1 =
-      new_output_file("--out-unidentified1", { ".unidentified", out1 });
+      new_output_file("--out-unidentified1", { ".unidentified", out1 }, false);
 
     if (paired_ended_mode) {
       if (interleaved_output) {
         files.unidentified_2 = files.unidentified_1;
       } else {
-        files.unidentified_2 =
-          new_output_file("--out-unidentified2", { ".unidentified", out2 });
+        files.unidentified_2 = new_output_file(
+          "--out-unidentified2", { ".unidentified", out2 }, false);
       }
     }
   }
@@ -1451,11 +1451,12 @@ userconfig::get_output_filenames() const
 
 output_file
 userconfig::new_output_file(const std::string& key,
-                            const string_vec& values) const
+                            const string_vec& values,
+                            const bool sample_file) const
 {
   std::string out;
   if (argparser.is_set(key)) {
-    if (!is_demultiplexing_enabled()) {
+    if (!sample_file || !is_demultiplexing_enabled()) {
       const auto filename = argparser.value(key);
 
       return { filename, infer_output_format(filename) };

src/userconfig.hpp

@@ -242,8 +242,14 @@ class userconfig
   [[nodiscard]] output_format infer_output_format(
     const std::string& filename) const;
 
+  /**
+   * Generates filename + format for a output (FQ/SAM/BAM/HTML/JSON) file. If
+   * sample_file is true, the file is expected to belong to a sample. Otherwise
+   * it is expected to be a HTML/JSON/unidentified fastq file.
+   */
   [[nodiscard]] output_file new_output_file(const std::string& key,
-                                            const string_vec& values) const;
+                                            const string_vec& values,
+                                            bool sample_file = true) const;
 
   //! Argument parser setup to parse the arguments expected by AR
   argparse::parser argparser{};

tests/regression/demultiplexing/common/basenames/test_explicit.json

@@ -0,0 +1,60 @@
+{
+    "description": "Demultiplexing, trimming, and merging of PE reads with custom basenames and no default basename",
+    "exhaustive": true,
+    "arguments": [
+        "--out-prefix",
+        "/dev/null",
+        "--out-json",
+        "basename_1.json",
+        "--out-unidentified1",
+        "basename_1.unidentified.r1.fastq.gz",
+        "--out-unidentified2",
+        "basename_1.unidentified.r2.fastq.gz",
+        "--out-discarded",
+        "basename_2",
+        "--out-merged",
+        "basename_3",
+        "--out-singleton",
+        "basename_4",
+        "--out-file1",
+        "basename_5",
+        "--out-file2",
+        "basename_6",
+        "--out-html",
+        "basename_7.html",
+        "--barcode-list",
+        "barcodes.txt",
+        "--merge"
+    ],
+    "files": {
+        "input_1": [
+            "input_1.fastq"
+        ],
+        "input_2": [
+            "input_2.fastq"
+        ],
+        "barcodes": [
+            "barcodes.txt"
+        ],
+        "output": [
+            "basename_1.unidentified.r1.fastq.gz",
+            "basename_1.unidentified.r2.fastq.gz",
+            "basename_2.sample_1.discarded.fastq.gz",
+            "basename_2.sample_2.discarded.fastq.gz",
+            "basename_3.sample_1.merged.fastq.gz",
+            "basename_3.sample_2.merged.fastq.gz",
+            "basename_4.sample_1.singleton.fastq.gz",
+            "basename_4.sample_2.singleton.fastq.gz",
+            "basename_5.sample_1.r1.fastq.gz",
+            "basename_5.sample_2.r1.fastq.gz",
+            "basename_6.sample_1.r2.fastq.gz",
+            "basename_6.sample_2.r2.fastq.gz"
+        ],
+        "json": [
+            "basename_1.json"
+        ],
+        "html": [
+            "basename_7.html"
+        ]
+    }
+}