Merge pull request #72 from nvnieuwk/duplicate-checking

Duplicate checking
nextflow-io · Jul 10, 2023 · a288a90 · a288a90
2 parents b4c5810 + cf258c8
commit a288a90
Show file tree

Hide file tree

Showing 5 changed files with 63 additions and 7 deletions.
diff --git a/docs/samplesheets/fromSamplesheet.md b/docs/samplesheets/fromSamplesheet.md
@@ -23,9 +23,10 @@ This parameter be described in the Nextflow parameter schema using as a file, wi
 
 The given sample sheet schema specified in the `schema` key is then loaded and used for validation and sample sheet generation.
 
-An additional function optional argument can be used:
+Some additional function optional arguments can be used:
 
-- File name for the pipeline parameters schema. (Default: `nextflow_schema.json`)
+- `schema_filename`: File name for the pipeline parameters schema. (Default: `nextflow_schema.json`)
+- `skip_duplicate_check`: Skip the checking for duplicates. Can also be skipped with the `--validationSkipDuplicateCheck` parameter. (Default: `false`)
 
 ```groovy
 Channel.fromSamplesheet('input')
@@ -35,6 +36,7 @@ Channel.fromSamplesheet('input')
 Channel.fromSamplesheet(
   'input',
   schema_filename: 'custom_nextflow_schema.json',
+  skip_duplicate_check: false
 )
 ```
 

diff --git a/plugins/nf-validation/src/main/nextflow/validation/SamplesheetConverter.groovy b/plugins/nf-validation/src/main/nextflow/validation/SamplesheetConverter.groovy
@@ -36,6 +36,8 @@ class SamplesheetConverter {
     private static List<String> schemaErrors = []
     private static List<String> warnings = []
 
+    private static List<Map> rows = []
+
     static boolean hasErrors() { errors.size()>0 }
     static Set<String> getErrors() { errors.sort().collect { "\t${it}".toString() } as Set }
 
@@ -54,7 +56,8 @@ class SamplesheetConverter {
 
     static List convertToList(
         Path samplesheetFile, 
-        Path schemaFile
+        Path schemaFile,
+        Boolean skipDuplicateCheck
         ) {
 
         def Map schemaMap = (Map) new JsonSlurper().parseText(schemaFile.text)
@@ -78,14 +81,15 @@ class SamplesheetConverter {
         def Map<String,List<String>> booleanUniques = [:]
         def Map<String,List<Map<String,String>>> listUniques = [:]
         def Boolean headerCheck = true
+        this.rows = []
         resetCount()
 
         def List outputs = samplesheetList.collect { Map<String,String> fullRow ->
             increaseCount()
 
             Map<String,String> row = fullRow.findAll { it.value != "" }
             def Set rowKeys = row.keySet()
-            def String yamlInfo = fileType == "yaml" ? " for sample ${this.getCount()}." : ""
+            def String yamlInfo = fileType == "yaml" ? " for entry ${this.getCount()}." : ""
 
             // Check the header (CSV/TSV) or present fields (YAML)
             if(headerCheck) {
@@ -99,6 +103,13 @@ class SamplesheetConverter {
                 }
             }
 
+            // Check for row uniqueness
+            if(!skipDuplicateCheck && this.rows.contains(row)) {
+                def Integer firstDuplicate = this.rows.findIndexOf { it == row }
+                this.errors << "The samplesheet contains duplicate rows for entry ${firstDuplicate + 1} and entry ${getCount()} (${row})".toString()
+            }
+            this.rows.add(row)
+
             def Map meta = [:]
             def ArrayList output = []
 

diff --git a/plugins/nf-validation/src/main/nextflow/validation/SchemaValidator.groovy b/plugins/nf-validation/src/main/nextflow/validation/SchemaValidator.groovy
@@ -156,6 +156,7 @@ class SchemaValidator extends PluginExtensionPoint {
 
         // Set defaults for optional inputs
         def String schemaFilename = options?.containsKey('schema_filename') ? options.schema_filename as String : 'nextflow_schema.json'
+        def Boolean skipDuplicateCheck = options?.containsKey('skip_duplicate_check') ? options.skip_duplicate_check as Boolean : params.validationSkipDuplicateCheck ? params.validationSkipDuplicateCheck as Boolean : false
 
         def slurper = new JsonSlurper()
         def Map parsed = (Map) slurper.parse( Path.of(getSchemaPath(baseDir, schemaFilename)) )
@@ -194,7 +195,7 @@ class SchemaValidator extends PluginExtensionPoint {
 
         // Convert to channel
         final channel = CH.create()
-        List arrayChannel = SamplesheetConverter.convertToList(samplesheetFile, schemaFile)
+        List arrayChannel = SamplesheetConverter.convertToList(samplesheetFile, schemaFile, skipDuplicateCheck)
         session.addIgniter {
             arrayChannel.each { 
                 channel.bind(it) 
@@ -227,6 +228,9 @@ class SchemaValidator extends PluginExtensionPoint {
         if( !params.containsKey("validationSchemaIgnoreParams") ) {
             params.validationSchemaIgnoreParams = false
         }
+        if( !params.containsKey("validationSkipDuplicateCheck") ) {
+            params.validationSkipDuplicateCheck = false
+        }
 
         return params
     }
@@ -236,7 +240,15 @@ class SchemaValidator extends PluginExtensionPoint {
     // Add expected params
     //
     List addExpectedParams() {
-        def List expectedParams = ["validationFailUnrecognisedParams", "validationLenientMode", "monochrome_logs", "help", "validationShowHiddenParams", "validationSchemaIgnoreParams"]
+        def List expectedParams = [
+            "validationFailUnrecognisedParams",
+            "validationLenientMode",
+            "monochrome_logs",
+            "help",
+            "validationShowHiddenParams",
+            "validationSchemaIgnoreParams",
+            "validationSkipDuplicateCheck"
+        ]
 
         return expectedParams
     }

diff --git a/plugins/nf-validation/src/test/nextflow/validation/SamplesheetConverterTest.groovy b/plugins/nf-validation/src/test/nextflow/validation/SamplesheetConverterTest.groovy
@@ -214,7 +214,7 @@ class SamplesheetConverterTest extends Dsl2Spec{
         !stdout
     }
 
-        def 'errors before channel conversion' () {
+    def 'errors before channel conversion' () {
         given:
         def SCRIPT_TEXT = '''
             include { fromSamplesheet } from 'plugin/nf-validation'
@@ -250,4 +250,31 @@ class SamplesheetConverterTest extends Dsl2Spec{
         errorMessages[12] == "* -- Entry 3 - field_2: expected type: Integer, found: String (false)"
         !stdout
     }
+
+    def 'duplicates' () {
+        given:
+        def SCRIPT_TEXT = '''
+            include { fromSamplesheet } from 'plugin/nf-validation'
+
+            params.input = 'src/testResources/duplicate.csv'
+
+            workflow {
+                Channel.fromSamplesheet("input", schema_filename:"src/testResources/nextflow_schema_with_samplesheet_converter.json").view()
+            }
+        '''
+
+        when:
+        dsl_eval(SCRIPT_TEXT)
+        def stdout = capture
+                .toString()
+                .readLines()
+                .findResults {it.startsWith('[[') ? it : null }
+
+        then:
+        def error = thrown(SchemaValidationException)
+        def errorMessages = error.message.readLines() 
+        errorMessages[0] == "Samplesheet errors:"
+        errorMessages[4] == "\tThe samplesheet contains duplicate rows for entry 2 and entry 3 ([field_4:string1, field_5:25, field_6:false])"
+        !stdout
+    }
 }
diff --git a/plugins/nf-validation/src/testResources/duplicate.csv b/plugins/nf-validation/src/testResources/duplicate.csv
@@ -0,0 +1,4 @@
+field_1,field_2,field_3,field_4,field_5,field_6,field_7,field_8,field_9,field_10,field_11
+fullField,10,true,string1,25,false,src/testResources/test.txt,src/testResources/testDir,src/testResources/test.txt,unique1,1
+,,,string1,25,false,,,,,,
+,,,string1,25,false,,,,,,