Allow cat_cat to retain double file extensions for .gz files (nf-core…

…#4230)

* Retain double file extensions for .gz files

- Adds logic to capture file extensions that
preceed `.gz`, e.g. `.fasta.gz` and use this
double extension as the suffix for the output
file.
- Updates tests accordingly.

* Re-add cat/cat entry to pytest_modules.yml

* Simplify code to extract double .gz file extensions

Instead of a conditional check using a regex capture group
and a nested lastIndexOf() extraction for double extensions `.xxxxx.gz`,
this update searches for a regex match and directly adds it as a suffix
if there is a match, and uses the existing lastIndexOf() method
to capture a single file extension otherwise.

Co-authored-by: Matthias Hörtenhuber <[email protected]>

* [automated] Fix linting with Prettier

* Moved the function outside of the process block

* bugfix: missing closing bracket

* Reordered the lines to make the diff shorter

* Changed the tests to regular snapshots so that we can see and check the output file name

---------

Co-authored-by: Matthias Hörtenhuber <[email protected]>
Co-authored-by: nf-core-bot <[email protected]>
Co-authored-by: Matthieu Muffato <[email protected]>

modules/nf-core/cat/cat/main.nf

@@ -22,6 +22,8 @@ process CAT_CAT {
     def args2 = task.ext.args2 ?: ''
     def file_list = files_in.collect { it.toString() }
 
+    // choose appropriate concatenation tool depending on input and output format
+
     // | input     | output     | command1 | command2 |
     // |-----------|------------|----------|----------|
     // | gzipped   | gzipped    | cat      |          |
@@ -30,7 +32,7 @@ process CAT_CAT {
     // | ungzipped | gzipped    | cat      | pigz     |
 
     // Use input file ending as default
-    prefix   = task.ext.prefix ?: "${meta.id}${file_list[0].substring(file_list[0].lastIndexOf('.'))}"
+    prefix   = task.ext.prefix ?: "${meta.id}${getFileSuffix(file_list[0])}"
     out_zip  = prefix.endsWith('.gz')
     in_zip   = file_list[0].endsWith('.gz')
     command1 = (in_zip && !out_zip) ? 'zcat' : 'cat'
@@ -68,3 +70,10 @@ process CAT_CAT {
     END_VERSIONS
     """
 }
+
+// for .gz files also include the second to last extension if it is present. E.g., .fasta.gz
+def getFileSuffix(filename) {
+    def match = filename =~ /^.*?((\.\w{1,5})?(\.\w{1,5}\.gz$))/
+    return match ? match[0][1] : filename.substring(filename.lastIndexOf('.'))
+}
+

modules/nf-core/cat/cat/tests/main.nf.test

@@ -83,8 +83,7 @@ nextflow_process {
             def lines = path(process.out.file_out.get(0).get(1)).linesGzip
             assertAll(
                 { assert process.success },
-                { assert snapshot(lines[0..5]).match("test_cat_zipped_zipped_lines") },
-                { assert snapshot(lines.size()).match("test_cat_zipped_zipped_size")}
+                { assert snapshot(process.out).match() }
             )
         }
     }
@@ -142,8 +141,7 @@ nextflow_process {
             def lines = path(process.out.file_out.get(0).get(1)).linesGzip
             assertAll(
                 { assert process.success },
-                { assert snapshot(lines[0..5]).match("test_cat_unzipped_zipped_lines") },
-                { assert snapshot(lines.size()).match("test_cat_unzipped_zipped_size")}
+                { assert snapshot(process.out).match() }
             )
         }
     }