Merge pull request #118 from nvnieuwk/feat/format-filepathpattern

format file-path-pattern

Author: nvnieuwk

CHANGELOG.md

@@ -11,6 +11,10 @@
 - Floats and doubles should now be created when using the `number` type in the schema ([#113](https://github.com/nextflow-io/nf-validation/pull/113/))
 - When `0` is used as a default value in the schema, a `0` will now be used as the value in the `.fromSamplesheet()` channel instead of `null` ([#114](https://github.com/nextflow-io/nf-validation/pull/114))
 
+## New features
+
+- Added `file-path-pattern` format to check every file fetched using a glob pattern. Using a glob is now also possible in the samplesheet and will create a list of all files found using that glob pattern. ([#118](https://github.com/nextflow-io/nf-validation/pull/118))
+
 # Version 1.0.0 - Tonkotsu
 
 The nf-validation plugin is now in production use across many pipelines and has (we hope) now reached a point of relative stability. The bump to major version v1.0.0 signifies that it is suitable for use in production pipelines.

docs/nextflow_schema/nextflow_schema_specification.md

@@ -342,13 +342,16 @@ Example usage is as follows:
 The available `format` types are below:
 
 `file-path`
-: States if the provided value is a file. Does not check its existence, but it does check that the path is not a directory.
+: States that the provided value is a file. Does not check its existence, but it does check that the path is not a directory.
 
 `directory-path`
-: States if the provided value is a directory. Does not check its existence, but if it exists, it does check that the path is not a file.
+: States that the provided value is a directory. Does not check its existence, but if it exists, it does check that the path is not a file.
 
 `path`
-: States if the provided value is a path (file or directory). Does not check its existence.
+: States that the provided value is a path (file or directory). Does not check its existence.
+
+`file-path-pattern`
+: States that the provided value is a globbing pattern that will be used to fetch files. Checks that the pattern is valid and that at least one file is found.
 
 ### `exists`
 

docs/nextflow_schema/sample_sheet_schema_specification.md

@@ -78,6 +78,8 @@ Please refer to the [Nextflow schema specification](../nextflow_schema/nextflow_
     Be sure to set `"type": "string"` and `"format": "file-path"` for these properties,
     so that nf-validation correctly returns this sample sheet field as a `Nextflow.file` object.
 
+When using the `file-path-pattern` format for a globbing pattern, a list will be created with all files found by the globbing pattern. See [here](../nextflow_schema/nextflow_schema_specification.md#file-path-pattern) for more information.
+
 ## Sample sheet keys
 
 Below are the properties that are specific to sample sheet schema.

plugins/nf-validation/src/main/nextflow/validation/FormatValidators/FilePathPatternValidator.groovy

@@ -0,0 +1,34 @@
+package nextflow.validation
+
+import java.nio.file.Path
+import groovy.util.logging.Slf4j
+
+import org.everit.json.schema.FormatValidator
+import nextflow.Nextflow
+
+@Slf4j
+public class FilePathPatternValidator implements FormatValidator {
+
+    @Override
+    public Optional<String> validate(final String subject) {
+        if (subject.startsWith('s3://')) {
+            log.debug("S3 paths are not supported by 'FilePathPatternValidator': '${subject}'")
+            return Optional.empty()
+        }
+        ArrayList files = Nextflow.file(subject) as ArrayList
+        ArrayList errors = []
+
+        if(files.size() == 0) {
+            return Optional.of("No files were found using the globbing pattern '${subject}'" as String)
+        }
+        for( file : files ) {
+            if (file.isDirectory()) {
+                errors.add("'${file.toString()}' is not a file, but a directory" as String)
+            }
+        }
+        if(errors.size > 0) {
+            return Optional.of(errors.join('\n'))
+        }
+        return Optional.empty()
+    }
+}

plugins/nf-validation/src/main/nextflow/validation/SamplesheetConverter.groovy

@@ -258,10 +258,17 @@ class SamplesheetConverter {
 
             // Check and convert to the desired format
             def String format = field['value']['format']
-            if(format && format.contains("path")) {
-                def Path inputFile = Nextflow.file(input) as Path
-                return inputFile
+            if(format) {
+                if(format == "file-path-pattern") {
+                    def ArrayList inputFiles = Nextflow.file(input) as ArrayList
+                    return inputFiles
+                }
+                if(format.contains("path")) {
+                    def Path inputFile = Nextflow.file(input) as Path
+                    return inputFile
+                }
             }
+            
 
             // Return the plain string value
             return result

plugins/nf-validation/src/main/nextflow/validation/SchemaValidator.groovy

@@ -351,6 +351,7 @@ class SchemaValidator extends PluginExtensionPoint {
                 .addFormatValidator("file-path", new FilePathValidator())
                 .addFormatValidator("directory-path", new DirectoryPathValidator())
                 .addFormatValidator("path", new PathValidator())
+                .addFormatValidator("file-path-pattern", new FilePathPatternValidator())
                 .build()
         final schema = schemaLoader.load().build()
 
@@ -526,6 +527,7 @@ class SchemaValidator extends PluginExtensionPoint {
             .addFormatValidator("file-path", new FilePathValidator())
             .addFormatValidator("directory-path", new DirectoryPathValidator())
             .addFormatValidator("path", new PathValidator())
+            .addFormatValidator("file-path-pattern", new FilePathPatternValidator())
             .build()
         final schema = schemaLoader.load().build()