Detect conda explicit files by content instead of extension

modules/nextflow/src/main/groovy/nextflow/conda/CondaCache.groovy

@@ -23,6 +23,7 @@ import java.nio.file.Paths
 import java.util.concurrent.ConcurrentHashMap
 
 import groovy.transform.CompileStatic
+import groovy.transform.Memoized
 import groovy.transform.PackageScope
 import groovy.util.logging.Slf4j
 import groovyx.gpars.dataflow.DataflowVariable
@@ -162,8 +163,53 @@ class CondaCache {
         (str.endsWith('.yml') || str.endsWith('.yaml')) && !str.contains('\n')
     }
 
-    boolean isTextFilePath(String str) {
-        str.endsWith('.txt') && !str.contains('\n')
+    /**
+     * Check if the given string is a path to a conda explicit file
+     * by verifying it contains the @EXPLICIT marker in the first 20 lines
+     *
+     * @param str The conda environment string
+     * @return {@code true} if it's a path to an explicit file, {@code false} otherwise
+     */
+    @PackageScope
+    @Memoized // <-- annotate as "Memoized" to avoid parsing multiple time the same file
+    boolean isExplicitFile(String str) {
+        if( str.contains('\n') )
+            return false
+        try {
+            final path = str as Path
+            if( !path.exists() )
+                return false
+            return containsExplicitMarker(path)
+        }
+        catch( Exception e ) {
+            log.debug "Failed to check explicit file: $str - ${e.message}"
+            return false
+        }
+    }
+
+    /**
+     * Check if a file contains the @EXPLICIT marker in the first 20 lines
+     *
+     * @param path The file path to check
+     * @return {@code true} if the marker is found, {@code false} otherwise
+     */
+    private boolean containsExplicitMarker(Path path) {
+        try {
+            try(final reader = path.newReader()) {
+                for( int i = 0; i < 20; i++ ) {
+                    final line = reader.readLine()
+                    if( line == null )
+                        break
+                    if( line.trim() == '@EXPLICIT' )
+                        return true
+                }
+                return false
+            }
+        }
+        catch( Exception e ) {
+            log.debug "Failed to check for @EXPLICIT marker in file: $path - ${e.message}"
+            return false
+        }
     }
 
     /**
@@ -196,16 +242,14 @@ class CondaCache {
                 throw new IllegalArgumentException("Error parsing Conda environment YAML file: $condaEnv -- Check the log file for details", e)
             }
         }
-        else if( isTextFilePath(condaEnv) )  {
+        // check if it's a conda explicit file (contains @EXPLICIT marker)
+        else if( isExplicitFile(condaEnv) )  {
             try {
                 final path = condaEnv as Path
                 content = path.text
             }
-            catch( NoSuchFileException e ) {
-                throw new IllegalArgumentException("Conda environment file does not exist: $condaEnv")
-            }
             catch( Exception e ) {
-                throw new IllegalArgumentException("Error parsing Conda environment text file: $condaEnv -- Check the log file for details", e)
+                throw new IllegalArgumentException("Error reading Conda explicit file: $condaEnv -- Check the log file for details", e)
             }
         }
         // it's interpreted as user provided prefix directory
@@ -284,7 +328,7 @@ class CondaCache {
             final yesOpt = binaryName=="mamba" || binaryName == "micromamba"  ? '--yes ' : ''
             cmd = "${binaryName} env create ${yesOpt}--prefix ${Escape.path(prefixPath)} --file ${target}"
         }
-        else if( isTextFilePath(condaEnv) ) {
+        else if( isExplicitFile(condaEnv) ) {
             cmd = "${binaryName} create ${opts}--yes --quiet --prefix ${Escape.path(prefixPath)} --file ${Escape.path(makeAbsolute(condaEnv))}"
         }
 

modules/nextflow/src/test/groovy/nextflow/conda/CondaCacheTest.groovy

@@ -46,17 +46,49 @@ class CondaCacheTest extends Specification {
         cache.isYamlFilePath('env.yaml')
     }
 
-    def 'should text file' () {
+    def 'should detect explicit file by content' () {
 
         given:
+        def folder = Files.createTempDirectory('test')
         def cache = new CondaCache()
 
+        // File with @EXPLICIT marker
+        def explicitFile = folder.resolve('packages.txt')
+        explicitFile.text = '''\
+            # This file was created by conda
+            @EXPLICIT
+            https://repo.anaconda.com/pkgs/main/linux-64/bwa-0.7.17.tar.bz2
+            '''.stripIndent()
+
+        // File without @EXPLICIT marker
+        def regularFile = folder.resolve('regular.txt')
+        regularFile.text = '''\
+            bwa=0.7.17
+            samtools=1.9
+            '''.stripIndent()
+
+        // YAML file (should not match)
+        def yamlFile = folder.resolve('env.yaml')
+        yamlFile.text = '''\
+            dependencies:
+              - bwa=0.7.17
+            '''.stripIndent()
+
         expect:
-        !cache.isTextFilePath('foo=1.0')
-        !cache.isTextFilePath('env.yaml')
-        !cache.isTextFilePath('foo.txt\nbar.txt')
-        cache.isTextFilePath('env.txt')
-        cache.isTextFilePath('foo/bar/env.txt')
+        // Non-existent file
+        !cache.isExplicitFile('foo=1.0')
+        !cache.isExplicitFile('nonexistent.txt')
+        // String with newline
+        !cache.isExplicitFile('foo.txt\nbar.txt')
+        // File with @EXPLICIT marker
+        cache.isExplicitFile(explicitFile.toString())
+        // File without @EXPLICIT marker
+        !cache.isExplicitFile(regularFile.toString())
+        // YAML file
+        !cache.isExplicitFile(yamlFile.toString())
+
+        cleanup:
+        folder?.deleteDir()
     }
 
 
@@ -147,27 +179,28 @@ class CondaCacheTest extends Specification {
 
     }
 
-    def 'should create conda env prefix path for a text env file' () {
+    def 'should create conda env prefix path for a conda explicit file' () {
 
         given:
         def folder = Files.createTempDirectory('test')
         def cache = Spy(CondaCache)
         def BASE = Paths.get('/conda/envs')
         def ENV = folder.resolve('bar.txt')
-        ENV.text = '''
-                star=2.5.4a
-                bwa=0.7.15
-                multiqc=1.2.3
-                '''
-                .stripIndent(true)  // https://issues.apache.org/jira/browse/GROOVY-9423
+        ENV.text = '''\
+                # This file was created by conda
+                @EXPLICIT
+                https://repo.anaconda.com/pkgs/main/linux-64/star-2.5.4a.tar.bz2
+                https://repo.anaconda.com/pkgs/main/linux-64/bwa-0.7.15.tar.bz2
+                https://repo.anaconda.com/pkgs/main/linux-64/multiqc-1.2.3.tar.bz2
+                '''.stripIndent()
 
         when:
         def prefix = cache.condaPrefixPath(ENV.toString())
         then:
         1 * cache.isYamlFilePath(ENV.toString())
-        1 * cache.isTextFilePath(ENV.toString())
+        1 * cache.isExplicitFile(ENV.toString())
         1 * cache.getCacheDir() >> BASE
-        prefix.toString() == "/conda/envs/env-85371202d8820331ff19ae89c0595497"
+        prefix.toString() == "/conda/envs/env-24d602a5eecba868858ab48a41e2c9bd"
 
         cleanup:
         folder?.deleteDir()
@@ -323,7 +356,7 @@ class CondaCacheTest extends Specification {
         def result = cache.createLocalCondaEnv0(ENV,PREFIX)
         then:
         1 * cache.isYamlFilePath(ENV)
-        1 * cache.isTextFilePath(ENV)
+        1 * cache.isExplicitFile(ENV)
         0 * cache.makeAbsolute(_)
         1 * cache.runCommand( "conda create --this --that --yes --quiet --prefix $PREFIX $ENV" ) >> null
         result == PREFIX
@@ -340,7 +373,7 @@ class CondaCacheTest extends Specification {
         def result = cache.createLocalCondaEnv0(ENV, PREFIX)
         then:
         1 * cache.isYamlFilePath(ENV)
-        1 * cache.isTextFilePath(ENV)
+        1 * cache.isExplicitFile(ENV)
         0 * cache.makeAbsolute(_)
         1 * cache.runCommand("mamba create --this --that --yes --quiet --prefix $PREFIX $ENV") >> null
         result == PREFIX
@@ -357,7 +390,7 @@ class CondaCacheTest extends Specification {
         def result = cache.createLocalCondaEnv0(ENV, PREFIX)
         then:
         1 * cache.isYamlFilePath(ENV)
-        1 * cache.isTextFilePath(ENV)
+        1 * cache.isExplicitFile(ENV)
         0 * cache.makeAbsolute(_)
         1 * cache.runCommand("micromamba create --this --that --yes --quiet --prefix $PREFIX $ENV") >> null
         result == PREFIX
@@ -374,7 +407,7 @@ class CondaCacheTest extends Specification {
         def result = cache.createLocalCondaEnv0(ENV, PREFIX)
         then:
         1 * cache.isYamlFilePath(ENV)
-        1 * cache.isTextFilePath(ENV)
+        1 * cache.isExplicitFile(ENV)
         0 * cache.makeAbsolute(_)
         1 * cache.runCommand("conda create --yes --quiet --prefix /foo/bar -c bioconda -c defaults bwa=1.1.1") >> null
         result == PREFIX
@@ -391,7 +424,7 @@ class CondaCacheTest extends Specification {
         def result = cache.createLocalCondaEnv0(ENV, PREFIX)
         then:
         1 * cache.isYamlFilePath(ENV)
-        0 * cache.isTextFilePath(ENV)
+        0 * cache.isExplicitFile(ENV)
         1 * cache.makeAbsolute(ENV) >> Paths.get('/usr/base').resolve(ENV)
         1 * cache.runCommand( "conda env create --prefix $PREFIX --file /usr/base/foo.yml" ) >> null
         result == PREFIX
@@ -409,17 +442,29 @@ class CondaCacheTest extends Specification {
         def result = cache.createLocalCondaEnv0(ENV, PREFIX)
         then:
         1 * cache.isYamlFilePath(ENV)
-        0 * cache.isTextFilePath(ENV)
+        0 * cache.isExplicitFile(ENV)
         1 * cache.makeAbsolute(ENV) >> Paths.get('/usr/base').resolve(ENV)
         1 * cache.runCommand( "micromamba env create --yes --prefix $PREFIX --file /usr/base/foo.yml" ) >> null
         result == PREFIX
 
     }
 
-    def 'should create a conda env with a text file' () {
+    def 'should create a conda env with an explicit file' () {
 
         given:
-        def ENV = 'foo.txt'
+        def folder = Files.createTempDirectory('test')
+        def envFile = folder.resolve('explicit.txt')
+        envFile.text = '''\
+            # This file may be used to create an environment using:
+            # $ conda create --name <env> --file <this file>
+            # platform: linux-64
+            @EXPLICIT
+            https://conda.anaconda.org/conda-forge/linux-64/_libgcc_mutex-0.1-conda_forge.tar.bz2#d7c89558ba9fa0495403155b64376d81
+            https://conda.anaconda.org/conda-forge/linux-64/libgomp-15.2.0-h767d61c_7.conda#f7b4d76975aac7e5d9e6ad13845f92fe
+            https://conda.anaconda.org/conda-forge/linux-64/_openmp_mutex-4.5-2_gnu.tar.bz2#73aaf86a425cc6e73fcf236a5a46396d
+            https://conda.anaconda.org/conda-forge/linux-64/libgcc-15.2.0-h767d61c_7.conda#c0374badb3a5d4b1372db28d19462c53
+            '''.stripIndent()
+        def ENV = envFile.toString()
         def PREFIX = Paths.get('/conda/envs/my-env')
         and:
         def cache = Spy(new CondaCache(createOptions: '--this --that'))
@@ -428,17 +473,30 @@ class CondaCacheTest extends Specification {
         def result = cache.createLocalCondaEnv0(ENV, PREFIX)
         then:
         1 * cache.isYamlFilePath(ENV)
-        1 * cache.isTextFilePath(ENV)
-        1 * cache.makeAbsolute(ENV) >> Paths.get('/usr/base').resolve(ENV)
-        1 * cache.runCommand( "conda create --this --that --yes --quiet --prefix $PREFIX --file /usr/base/foo.txt" ) >> null
+        1 * cache.isExplicitFile(ENV)
+        1 * cache.makeAbsolute(ENV) >> envFile.toAbsolutePath()
+        1 * cache.runCommand( "conda create --this --that --yes --quiet --prefix $PREFIX --file ${envFile.toAbsolutePath()}" ) >> null
         result == PREFIX
 
+        cleanup:
+        folder?.deleteDir()
     }
 
-    def 'should create a conda env with a text file - using micromamba' () {
+    def 'should create a conda env with an explicit file - using micromamba' () {
 
         given:
-        def ENV = 'foo.txt'
+        def folder = Files.createTempDirectory('test')
+        def envFile = folder.resolve('explicit.txt')
+        envFile.text = '''\
+            # This file may be used to create an environment using:
+            # $ conda create --name <env> --file <this file>
+            # platform: linux-64
+            @EXPLICIT
+            https://conda.anaconda.org/conda-forge/linux-64/_libgcc_mutex-0.1-conda_forge.tar.bz2#d7c89558ba9fa0495403155b64376d81
+            https://conda.anaconda.org/conda-forge/linux-64/libgomp-15.2.0-h767d61c_7.conda#f7b4d76975aac7e5d9e6ad13845f92fe
+            https://conda.anaconda.org/conda-forge/linux-64/_openmp_mutex-4.5-2_gnu.tar.bz2#73aaf86a425cc6e73fcf236a5a46396d
+            '''.stripIndent()
+        def ENV = envFile.toString()
         def PREFIX = Paths.get('/conda/envs/my-env')
         and:
         def cache = Spy(new CondaCache(useMicromamba: true, createOptions: '--this --that'))
@@ -447,11 +505,13 @@ class CondaCacheTest extends Specification {
         def result = cache.createLocalCondaEnv0(ENV, PREFIX)
         then:
         1 * cache.isYamlFilePath(ENV)
-        1 * cache.isTextFilePath(ENV)
-        1 * cache.makeAbsolute(ENV) >> Paths.get('/usr/base').resolve(ENV)
-        1 * cache.runCommand( "micromamba create --this --that --yes --quiet --prefix $PREFIX --file /usr/base/foo.txt" ) >> null
+        1 * cache.isExplicitFile(ENV)
+        1 * cache.makeAbsolute(ENV) >> envFile.toAbsolutePath()
+        1 * cache.runCommand( "micromamba create --this --that --yes --quiet --prefix $PREFIX --file ${envFile.toAbsolutePath()}" ) >> null
         result == PREFIX
 
+        cleanup:
+        folder?.deleteDir()
     }
 
     def 'should get options from the config' () {