Mercurial > repos > peterjc > effectivet3

changeset 10:a46d7861c32c draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
"Update all the pico_galaxy tools on main Tool Shed"
author peterjc
date Fri, 16 Apr 2021 22:34:56 +0000
parents 512530020360
children ed8c1babc166
files tools/effectiveT3/README.rst tools/effectiveT3/effectiveT3.py tools/effectiveT3/effectiveT3.xml tools/effectiveT3/tool_dependencies.xml
diffstat 4 files changed, 161 insertions(+), 38 deletions(-) [+]
line wrap: on
line diff
--- a/tools/effectiveT3/README.rst	Tue May 16 09:17:17 2017 -0400
+++ b/tools/effectiveT3/README.rst	Fri Apr 16 22:34:56 2021 +0000
@@ -10,12 +10,12 @@
 Jehl, Arnold and Rattei.
 Effective - a database of predicted secreted bacterial proteins
 Nucleic Acids Research, 39(Database issue), D591-5, 2011.
-http://dx.doi.org/10.1093/nar/gkq1154
+https://doi.org/10.1093/nar/gkq1154
 
 Arnold, Brandmaier, Kleine, Tischler, Heinz, Behrens, Niinikoski, Mewes, Horn and Rattei.
 Sequence-based prediction of type III secreted proteins.
 PLoS Pathog. 5(4):e1000376, 2009.
-http://dx.doi.org/10.1371/journal.ppat.1000376
+https://doi.org/10.1371/journal.ppat.1000376
 
 http://effectors.org/
 
@@ -98,6 +98,9 @@
         - Minor internal changes to Python script for error reporting & style.
 v0.0.18 - Use ``<command detect_errors="aggressive">`` (internal change only).
         - Single quote command line arguments (internal change only).
+v0.0.19 - Python 3 compatible exception handling.
+v0.0.20 - Work with Effective T3 as installed by BioConda (which provides a
+          wrapper script which can be used to find the main JAR file folder).
 ======= ======================================================================
 
 
--- a/tools/effectiveT3/effectiveT3.py	Tue May 16 09:17:17 2017 -0400
+++ b/tools/effectiveT3/effectiveT3.py	Fri Apr 16 22:34:56 2021 +0000
@@ -12,20 +12,33 @@
 tab separated output for use in Galaxy.
 """
 import os
+
+# We want to be able to use shutil.which, but need Python 3.3+
+# import shutil
 import subprocess
 import sys
 
-# The Galaxy auto-install via tool_dependencies.xml will set this environment variable
-effective_t3_dir = os.environ.get("EFFECTIVET3", "/opt/EffectiveT3/")
-effective_t3_jar = os.path.join(effective_t3_dir, "TTSS_GUI-1.0.1.jar")
+# The Galaxy auto-install via tool_dependencies.xml will set the
+# environment variable $EFFECTIVET3 pointing at the folder with
+# the JAR file.
+#
+# The BioConda recipe will put a wrapper script on the $PATH,
+# which we can use to find the JAR file.
+#
+# We fall back on /opt/EffectiveT3/
+#
+effective_t3_jarname = "TTSS_GUI-1.0.1.jar"
 
 if "-v" in sys.argv or "--version" in sys.argv:
     # TODO - Get version of the JAR file dynamically?
-    print("Wrapper v0.0.17, TTSS_GUI-1.0.1.jar")
+    print("Wrapper v0.0.20, for %s" % effective_t3_jarname)
     sys.exit(0)
 
 if len(sys.argv) != 5:
-    sys.exit("Require four arguments: model, threshold, input protein FASTA file & output tabular file")
+    sys.exit(
+        "Require four arguments: model, threshold, input protein "
+        "FASTA file & output tabular file"
+    )
 
 model, threshold, fasta_file, tabular_file = sys.argv[1:]
 
@@ -33,7 +46,9 @@
     sys.exit("Input FASTA file not found: %s" % fasta_file)
 
 if threshold not in ["selective", "sensitive"] and not threshold.startswith("cutoff="):
-    sys.exit("Threshold should be selective, sensitive, or cutoff=..., not %r" % threshold)
+    sys.exit(
+        "Threshold should be selective, sensitive, or cutoff=..., not %r" % threshold
+    )
 
 
 def clean_tabular(raw_handle, out_handle):
@@ -42,7 +57,11 @@
     positive = 0
     errors = 0
     for line in raw_handle:
-        if not line or line.startswith("#") or line.startswith("Id; Description; Score;"):
+        if (
+            not line
+            or line.startswith("#")
+            or line.startswith("Id; Description; Score;")
+        ):
             continue
         assert line.count(";") >= 3, repr(line)
         # Normally there will just be three semi-colons, however the
@@ -73,8 +92,10 @@
     # Avoid using shell=True when we call subprocess to ensure if the Python
     # script is killed, so too is the child process.
     try:
-        child = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
-    except Exception, err:
+        child = subprocess.Popen(
+            cmd, universal_newlines=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE
+        )
+    except Exception as err:
         sys.exit("Error invoking command:\n%s\n\n%s\n" % (" ".join(cmd), err))
     # Use .communicate as can get deadlocks with .wait(),
     stdout, stderr = child.communicate()
@@ -82,41 +103,141 @@
     if return_code or stderr.startswith("Exception in thread"):
         cmd_str = " ".join(cmd)  # doesn't quote spaces etc
         if stderr and stdout:
-            sys.exit("Return code %i from command:\n%s\n\n%s\n\n%s" % (return_code, cmd_str, stdout, stderr))
+            sys.exit(
+                "Return code %i from command:\n%s\n\n%s\n\n%s"
+                % (return_code, cmd_str, stdout, stderr)
+            )
         else:
-            sys.exit("Return code %i from command:\n%s\n%s" % (return_code, cmd_str, stderr))
+            sys.exit(
+                "Return code %i from command:\n%s\n%s" % (return_code, cmd_str, stderr)
+            )
 
 
-if not os.path.isdir(effective_t3_dir):
-    sys.exit("Effective T3 folder not found: %r" % effective_t3_dir)
+try:
+    from shutil import which
+except ImportError:
+    # Likely running on Python 2, use backport:
+    def which(cmd, mode=os.F_OK | os.X_OK, path=None):
+        """Python implementation of command line tool which.
+
+        Given a command, mode, and a PATH string, return the path which
+        conforms to the given mode on the PATH, or None if there is no such
+        file.
+
+        `mode` defaults to os.F_OK | os.X_OK. `path` defaults to the result
+        of os.environ.get("PATH"), or can be overridden with a custom search
+        path.
+        """
+        # Check that a given file can be accessed with the correct mode.
+        # Additionally check that `file` is not a directory, as on Windows
+        # directories pass the os.access check.
+        def _access_check(fn, mode):
+            return os.path.exists(fn) and os.access(fn, mode) and not os.path.isdir(fn)
+
+        # Short circuit. If we're given a full path which matches the mode
+        # and it exists, we're done here.
+        if _access_check(cmd, mode):
+            return cmd
+
+        path = (path or os.environ.get("PATH", os.defpath)).split(os.pathsep)
+
+        if sys.platform == "win32":
+            # The current directory takes precedence on Windows.
+            if os.curdir not in path:
+                path.insert(0, os.curdir)
 
-if not os.path.isfile(effective_t3_jar):
-    sys.exit("Effective T3 JAR file not found: %r" % effective_t3_jar)
+            # PATHEXT is necessary to check on Windows.
+            pathext = os.environ.get("PATHEXT", "").split(os.pathsep)
+            # See if the given file matches any of the expected path extensions.
+            # This will allow us to short circuit when given "python.exe".
+            matches = [cmd for ext in pathext if cmd.lower().endswith(ext.lower())]
+            # If it does match, only test that one, otherwise we have to try
+            # others.
+            files = [cmd] if matches else [cmd + ext.lower() for ext in pathext]
+        else:
+            # On other platforms you don't have things like PATHEXT to tell you
+            # what file suffixes are executable, so just pass on cmd as-is.
+            files = [cmd]
+
+        seen = set()
+        for dir in path:
+            dir = os.path.normcase(dir)
+            if dir not in seen:
+                seen.add(dir)
+                for thefile in files:
+                    name = os.path.join(dir, thefile)
+                    if _access_check(name, mode):
+                        return name
+        return None
+
+
+# Try in order the following to find the JAR file:
+# - Location of any wrapper script, e.g. from BioConda installation
+# - The $EFFECTIVET3 env var, e.g. old-style Galaxy tool installation
+# - The /opt/EffectiveT3/ folder.
+effective_t3_jar = None
+effective_t3_dir = None
+dirs = ["/opt/EffectiveT3/"]
+if "EFFECTIVET3" in os.environ:
+    dirs.insert(0, os.environ.get("EFFECTIVET3"))
+if which("effectivet3"):
+    # Assuming this is a BioConda installed wrapper for effective T3,
+    # this will get the directory of the wrapper script which is where
+    # the JAR file will be:
+    dirs.insert(0, os.path.split(os.path.realpath(which("effectivet3")))[0])
+for effective_t3_dir in dirs:
+    effective_t3_jar = os.path.join(effective_t3_dir, effective_t3_jarname)
+    if os.path.isfile(effective_t3_jar):
+        # Good
+        break
+    effective_t3_jar = None
+if not effective_t3_dir or not effective_t3_jar:
+    sys.exit("Effective T3 JAR file %r not found in %r" % (effective_t3_jarname, dirs))
 
 if not os.path.isdir(os.path.join(effective_t3_dir, "module")):
-    sys.exit("Effective T3 module folder not found: %r" % os.path.join(effective_t3_dir, "module"))
+    sys.exit(
+        "Effective T3 module folder not found: %r"
+        % os.path.join(effective_t3_dir, "module")
+    )
 
 effective_t3_model = os.path.join(effective_t3_dir, "module", model)
 if not os.path.isfile(effective_t3_model):
-    sys.stderr.write("Contents of %r is %s\n"
-                     % (os.path.join(effective_t3_dir, "module"),
-                        ", ".join(repr(p) for p in os.listdir(os.path.join(effective_t3_dir, "module")))))
+    sys.stderr.write(
+        "Contents of %r is %s\n"
+        % (
+            os.path.join(effective_t3_dir, "module"),
+            ", ".join(
+                repr(p) for p in os.listdir(os.path.join(effective_t3_dir, "module"))
+            ),
+        )
+    )
     sys.stderr.write("Main JAR was found: %r\n" % effective_t3_jar)
     sys.exit("Effective T3 model JAR file not found: %r" % effective_t3_model)
 
-# We will have write access whereever the output should be,
-temp_file = os.path.abspath(tabular_file + ".tmp")
+# We will have write access wherever the output should be,
+if tabular_file == "/dev/stdout":
+    temp_file = os.path.abspath("effectivet3_tabular_output.tmp")
+else:
+    temp_file = os.path.abspath(tabular_file + ".tmp")
 
 # Use absolute paths since will change current directory...
 tabular_file = os.path.abspath(tabular_file)
 fasta_file = os.path.abspath(fasta_file)
 
-cmd = ["java", "-jar", effective_t3_jar,
-       "-f", fasta_file,
-       "-m", model,
-       "-t", threshold,
-       "-o", temp_file,
-       "-q"]
+cmd = [
+    "java",
+    "-jar",
+    effective_t3_jar,
+    "-f",
+    fasta_file,
+    "-m",
+    model,
+    "-t",
+    threshold,
+    "-o",
+    temp_file,
+    "-q",
+]
 
 try:
     # Must run from directory above the module subfolder:
@@ -139,8 +260,7 @@
 os.remove(temp_file)
 
 if errors:
-    print("%i sequences, %i positive, %i errors"
-          % (count, positive, errors))
+    print("%i sequences, %i positive, %i errors" % (count, positive, errors))
 else:
     print("%i/%i sequences positive" % (positive, count))
 
--- a/tools/effectiveT3/effectiveT3.xml	Tue May 16 09:17:17 2017 -0400
+++ b/tools/effectiveT3/effectiveT3.xml	Fri Apr 16 22:34:56 2021 +0000
@@ -1,4 +1,4 @@
-<tool id="effectiveT3" name="Effective T3" version="0.0.18">
+<tool id="effectiveT3" name="Effective T3" version="0.0.20">
     <description>Find bacterial effectors in protein sequences</description>
     <requirements>
         <requirement type="package" version="1.0.1">effectiveT3</requirement>
@@ -16,7 +16,7 @@
 '$fasta_file' '$tabular_file'
     </command>
     <inputs>
-        <param name="fasta_file" type="data" format="fasta" label="FASTA file of protein sequences"/> 
+        <param name="fasta_file" type="data" format="fasta" label="FASTA file of protein sequences"/>
         <param name="module" type="select" display="radio" label="Classification module">
             <options from_file="effectiveT3.loc">
                 <column name="value" index="0"/>
@@ -62,7 +62,7 @@
         </test>
     </tests>
     <help>
-    
+
 **What it does**
 
 This calls the command line Effective T3 v1.0.1 tool for prediction of bacterial effector proteins.
@@ -87,17 +87,17 @@
 Peter J.A. Cock, Björn A. Grüning, Konrad Paszkiewicz and Leighton Pritchard (2013).
 Galaxy tools and workflows for sequence analysis with applications
 in molecular plant pathology. PeerJ 1:e167
-http://dx.doi.org/10.7717/peerj.167
+https://doi.org/10.7717/peerj.167
 
 Jehl, Arnold and Rattei (2011).
 Effective - a database of predicted secreted bacterial proteins
 Nucleic Acids Research, 39(Database issue), D591-5.
-http://dx.doi.org/10.1093/nar/gkq1154
+https://doi.org/10.1093/nar/gkq1154
 
 Arnold, Brandmaier, Kleine, Tischler, Heinz, Behrens, Niinikoski, Mewes, Horn and Rattei (2009).
 Sequence-based prediction of type III secreted proteins.
 PLoS Pathog. 5(4):e1000376.
-http://dx.doi.org/10.1371/journal.ppat.1000376
+https://doi.org/10.1371/journal.ppat.1000376
 
 See also http://effectors.org/
 
--- a/tools/effectiveT3/tool_dependencies.xml	Tue May 16 09:17:17 2017 -0400
+++ b/tools/effectiveT3/tool_dependencies.xml	Fri Apr 16 22:34:56 2021 +0000
@@ -11,7 +11,7 @@
                 <action type="make_directory">$INSTALL_DIR/module</action>
                 <!-- Original URL http://effectors.csb.univie.ac.at/sites/eff/files/others/TTSS_ANIMAL-1.0.1.jar -->
                 <action type="download_file" sha256sum="3d9cd8e805387d2dfa855076b3d5f7f97334aa612288075111329fb036c94e34" target_filename="TTSS_ANIMAL-1.0.1.jar">https://depot.galaxyproject.org/software/TTSS_ANIMAL/TTSS_ANIMAL_1.0.1_src_all.jar</action>
-                <action type="move_file"><source>TTSS_ANIMAL-1.0.1.jar</source><destination>$INSTALL_DIR/module/</destination></action>        
+                <action type="move_file"><source>TTSS_ANIMAL-1.0.1.jar</source><destination>$INSTALL_DIR/module/</destination></action>
                 <!-- Original URL http://effectors.csb.univie.ac.at/sites/eff/files/others/TTSS_PLANT-1.0.1.jar -->
                 <action type="download_file" sha256sum="593f0052ace030c2fa16cf336f3916a21bc2addbaefdfa149b084b1425e42a13" target_filename="TTSS_PLANT-1.0.1.jar">https://depot.galaxyproject.org/software/TTSS_PLANT/TTSS_PLANT_1.0.1_src_all.jar</action>
                 <action type="move_file"><source>TTSS_PLANT-1.0.1.jar</source><destination>$INSTALL_DIR/module/</destination></action>