Merge pull request #18204 from martenson/tight-axt

[24.1] Tighten axt sniffer

lib/galaxy/config/sample/datatypes_conf.xml.sample

@@ -32,7 +32,7 @@
     <datatype extension="gfa2" auto_compressed_types="gz" type="galaxy.datatypes.text:Gfa2" mimetype="text/plain" display_in_upload="true"/>
     <datatype extension="asn1" type="galaxy.datatypes.data:GenericAsn1" mimetype="text/plain" display_in_upload="true"/>
     <datatype extension="asn1-binary" type="galaxy.datatypes.binary:GenericAsn1Binary" mimetype="application/octet-stream" display_in_upload="true"/>
-    <datatype extension="axt" type="galaxy.datatypes.sequence:Axt" display_in_upload="true" description="blastz pairwise alignment format.  Each alignment block in an axt file contains three lines: a summary line and 2 sequence lines.  Blocks are separated from one another by blank lines.  The summary line contains chromosomal position and size information about the alignment. It consists of 9 required fields." description_url="https://wiki.galaxyproject.org/Learn/Datatypes#Axt"/>
+    <datatype extension="axt" type="galaxy.datatypes.sequence:Axt" display_in_upload="true" description="A pairwise alignment format." description_url="https://genome.ucsc.edu/goldenPath/help/axt.html"/>
     <datatype extension="fli" type="galaxy.datatypes.tabular:FeatureLocationIndex" display_in_upload="false"/>
     <datatype extension="bam" type="galaxy.datatypes.binary:Bam" mimetype="application/octet-stream" display_in_upload="true" description="A binary file compressed in the BGZF format with a '.bam' file extension." description_url="https://wiki.galaxyproject.org/Learn/Datatypes#BAM">
       <converter file="bam_to_bai.xml" target_datatype="bai"/>

lib/galaxy/datatypes/chain.py

@@ -9,6 +9,7 @@
 from galaxy.datatypes.sniff import (
     build_sniff_from_prefix,
     FilePrefix,
+    get_headers,
 )
 from galaxy.util import (
     commaify,
@@ -91,41 +92,31 @@ def sniff_prefix(self, file_prefix: FilePrefix) -> bool:
         >>> fname = get_test_fname( '1.chain' )
         >>> Chain().sniff( fname )
         True
+        >>> fname = get_test_fname( '2.chain' )
+        >>> Chain().sniff( fname )
+        True
         >>>
         """
-        fh = file_prefix.string_io()
-        for line in fh:
-            line = line.strip()
-            if line:  # first non-empty line
-                if line.startswith("chain"):
-                    tokens = line.split()
-                    if not (
-                        len(tokens) in [12, 13]
-                        and tokens[4] in self.strands
-                        and tokens[9] in self.strands
-                        and tokens[3].isdigit()
-                        and tokens[5].isdigit()
-                        and tokens[6].isdigit()
-                    ):
-                        return False
-                    prior_token_len = 0
-                    for line in fh:
-                        line = line.strip()
-                        if line == "":
-                            break
-                        tokens = line.split()
-                        if prior_token_len == 1:
-                            return False
-                        if len(tokens) not in [1, 3]:
-                            return False
-                        if not all(token.isdigit() for token in tokens):
-                            return False
-                        prior_token_len = len(tokens)
-                    if prior_token_len == 1:
-                        return True
-                else:
-                    return False
-        return False
+        headers = get_headers(file_prefix, None, count=2, comment_designator="#")
+        if not (
+            len(headers) == 2
+            and len(headers[0]) in [12, 13]
+            and headers[0][0] == "chain"
+            and headers[0][1].isdecimal()
+            and headers[0][3].isdecimal()
+            and headers[0][4] in self.strands
+            and headers[0][5].isdecimal()
+            and headers[0][6].isdecimal()
+            and headers[0][8].isdecimal()
+            and headers[0][9] in self.strands
+            and headers[0][10].isdecimal()
+            and headers[0][11].isdecimal()
+            and headers[1][0].isdecimal()
+            and len(headers[1]) in [1, 3]
+        ):
+            return False
+        else:
+            return True
 
 
 @build_sniff_from_prefix
@@ -161,34 +152,21 @@ def sniff_prefix(self, file_prefix: FilePrefix) -> bool:
         allowed_classes = ["fill", "gap"]
         strands = ["+", "-"]
 
-        fh = file_prefix.string_io()
-        for line in fh:
-            line = line.strip()
-            if line:  # first non-empty line
-                if line.startswith("net"):
-                    tokens = line.split()
-                    if not (len(tokens) == 3 and tokens[2].isdigit()):
-                        return False
-                    for line in fh:
-                        if line[0] != " ":  # children are indented one space
-                            return False
-                        line = line.strip()
-                        if line == "":
-                            break
-                        tokens = line.split()
-                        if not (
-                            len(tokens) >= 7  # seven fixed fields
-                            and len(tokens) <= 41  # plus seventeen optional name/value pairs
-                            and tokens[0] in allowed_classes
-                            and tokens[1].isdigit()
-                            and tokens[2].isdigit()
-                            and tokens[4] in strands
-                            and tokens[5].isdigit()
-                            and tokens[6].isdigit()
-                        ):
-                            return False
-                        else:
-                            return True
-                else:
-                    return False
-        return False
+        headers = get_headers(file_prefix, None, count=2, comment_designator="#")
+        if not (
+            len(headers) == 2
+            and len(headers[0]) == 3
+            and headers[0][0] == "net"
+            and headers[0][2].isdecimal()
+            and len(headers[1]) >= 7  # seven fixed fields
+            and len(headers[1]) <= 41  # plus seventeen optional name/value pairs
+            and headers[1][0] in allowed_classes
+            and headers[1][1].isdecimal()
+            and headers[1][2].isdecimal()
+            and headers[1][4] in strands
+            and headers[1][5].isdecimal()
+            and headers[1][6].isdecimal()
+        ):
+            return False
+        else:
+            return True

lib/galaxy/datatypes/sequence.py

@@ -1205,26 +1205,30 @@ def sniff_prefix(self, file_prefix: FilePrefix) -> bool:
         >>> fname = get_test_fname( 'alignment.lav' )
         >>> Axt().sniff( fname )
         False
+        >>> fname = get_test_fname( '2.chain' )
+        >>> Axt().sniff( fname )
+        False
         """
-        headers = get_headers(file_prefix, None)
-        if len(headers) < 4:
+        headers = get_headers(file_prefix, None, count=4, comment_designator="#")
+        if not (
+            len(headers) >= 3
+            and len(headers[0]) == 9
+            and headers[0][0] == "0"
+            and headers[0][2].isdecimal()
+            and headers[0][3].isdecimal()
+            and headers[0][5].isdecimal()
+            and headers[0][6].isdecimal()
+            and headers[0][7] in data.valid_strand
+            and headers[0][8].isdecimal()
+            and len(headers[1]) == 1
+            and len(headers[2]) == 1
+        ):
             return False
-        for hdr in headers:
-            if len(hdr) > 0 and hdr[0].startswith("##matrix=axt"):
-                return True
-            if len(hdr) > 0 and not hdr[0].startswith("#"):
-                if len(hdr) != 9:
-                    return False
-                try:
-                    for _ in (hdr[0], hdr[2], hdr[3], hdr[5], hdr[6], hdr[8]):
-                        int(_)
-                except ValueError:
-                    return False
-                if hdr[7] not in data.valid_strand:
-                    return False
-                else:
-                    return True
-        return False
+        # the optional fourth non-comment line has to be empty
+        if len(headers) == 4 and not headers[3] == []:
+            return False
+        else:
+            return True
 
 
 @build_sniff_from_prefix

lib/galaxy/datatypes/test/2.chain

@@ -0,0 +1,10 @@
+##matrix=axtChain 16 91,-114,-31,-123,-114,100,-125,-31,-31,-125,100,-114,-123,-31,-114,91
+##gapPenalties=axtChain O=400 E=30
+chain 67224 chr22 50818468 + 26560645 26561468 chr19 61431566 - 54838449 54839272 1
+823
+
+chain 48985 chr22 50818468 + 26560497 26561116 chr19 61431566 + 29160089 29160708 2
+619
+
+chain 46902 chr22 50818468 + 19792341 19793000 chr19 61431566 + 59180700 59181359 3
+659