Merge pull request #5561 from martenson/backport-fastgz

[18.01] add fasta.gz support

config/datatypes_conf.xml.sample

@@ -79,6 +79,9 @@
       <converter file="fasta_to_fai.xml" target_datatype="fai"/>
       <display file="igv/genome_fasta.xml" inherit="true"/>
     </datatype>
+    <datatype extension="fasta.gz" type="galaxy.datatypes.sequence:FastaGz" display_in_upload="true">
+        <converter file="fastagz_to_fasta.xml" target_datatype="fasta"/>
+    </datatype>
     <datatype extension="fastq" type="galaxy.datatypes.sequence:Fastq" display_in_upload="true" description="FASTQ format is a text-based format for storing both a biological sequence (usually nucleotide sequence) and its corresponding quality scores." description_url="https://wiki.galaxyproject.org/Learn/Datatypes#Fastq">
         <converter file="fastq_to_fqtoc.xml" target_datatype="fqtoc"/>
     </datatype>
@@ -766,6 +769,7 @@
     <!-- TODO: see molecules.py <sniffer type="galaxy.datatypes.molecules:SMILES"/>-->
     <sniffer type="galaxy.datatypes.phylip:Phylip"/>
     <sniffer type="galaxy.datatypes.sequence:Fasta"/>
+    <sniffer type="galaxy.datatypes.sequence:FastaGz"/>
     <sniffer type="galaxy.datatypes.sequence:FastqSanger"/>
     <sniffer type="galaxy.datatypes.sequence:FastqSangerGz"/>
     <sniffer type="galaxy.datatypes.sequence:FastqSangerBz2"/>

lib/galaxy/datatypes/converters/fastagz_to_fasta.xml

@@ -0,0 +1,11 @@
+<tool id="CONVERTER_fastagz_to_fasta" name="Convert fasta.gz files to fasta" version="1.0.0" hidden="true">
+  <command>gzip -dcf '$input1' > '$output1'</command>
+  <inputs>
+    <param format="fasta.gz" name="input1" type="data" label="Choose FASTA file"/>
+  </inputs>
+  <outputs>
+    <data format="fasta" name="output1"/>
+  </outputs>
+  <help>
+  </help>
+</tool>

lib/galaxy/datatypes/sequence.py

@@ -39,6 +39,7 @@ if sys.version_info > (3,):
 log = logging.getLogger(__name__)
 
 SNIFF_COMPRESSED_FASTQS = os.environ.get("GALAXY_ENABLE_BETA_COMPRESSED_FASTQ_SNIFFING", "0") == "1"
+SNIFF_COMPRESSED_FASTAS = os.environ.get("GALAXY_ENABLE_BETA_COMPRESSED_FASTA_SNIFFING", "0") == "1"
 
 
 class SequenceSplitLocations(data.Text):
@@ -311,6 +312,21 @@ class Alignment(data.Text):
         raise NotImplementedError("Can't split generic alignment files")
 
 
+class FastaGz(Sequence, Binary):
+    """Class representing a generic compressed FASTA sequence"""
+    edam_format = "format_1929"
+    file_ext = "fasta.gz"
+    compressed = True
+
+    def sniff(self, filename):
+        """Determines whether the file is in gzip-compressed FASTA format"""
+        if not SNIFF_COMPRESSED_FASTAS:
+            return False
+        if not is_gzip(filename):
+            return False
+        return Sequence.sniff(self, filename)
+
+
 class Fasta(Sequence):
     """Class representing a FASTA sequence"""
     edam_format = "format_1929"

lib/galaxy/datatypes/sniff.py

@@ -487,7 +487,7 @@ def handle_uploaded_dataset_file(filename, datatypes_registry, ext='auto'):
 AUTO_DETECT_EXTENSIONS = ['auto']  # should 'data' also cause auto detect?
 DECOMPRESSION_FUNCTIONS = dict(gzip=gzip.GzipFile, bz2=bz2.BZ2File)
 COMPRESSION_CHECK_FUNCTIONS = [('gzip', is_gzip), ('bz2', is_bz2)]
-COMPRESSION_DATATYPES = dict(gzip=['bam', 'fastq.gz', 'fastqsanger.gz', 'fastqillumina.gz', 'fastqsolexa.gz', 'fastqcssanger.gz'], bz2=['fastq.bz2', 'fastqsanger.bz2', 'fastqillumina.bz2', 'fastqsolexa.bz2', 'fastqcssanger.bz2'])
+COMPRESSION_DATATYPES = dict(gzip=['bam', 'fasta.gz', 'fastq.gz', 'fastqsanger.gz', 'fastqillumina.gz', 'fastqsolexa.gz', 'fastqcssanger.gz'], bz2=['fastq.bz2', 'fastqsanger.bz2', 'fastqillumina.bz2', 'fastqsolexa.bz2', 'fastqcssanger.bz2'])
 COMPRESSED_EXTENSIONS = []
 for exts in COMPRESSION_DATATYPES.values():
     COMPRESSED_EXTENSIONS.extend(exts)