Update specification for config.ini

The full validator isn't included in the configobj module, so it
has to be downloaded:

```
wget https://raw.githubusercontent.com/DiffSK/configobj/master/src/configobj/validate.py
python -c 'import configobj; from validate import Validator; print( configobj.ConfigObj("tral/tral_configuration/config.ini", 
configspec="tral/tral_configuration/spec.ini").validate(Validator()))'
```

Author: sbliven

.gitignore

@@ -123,3 +123,6 @@ ENV/
 
 # easy install
 easy_setup/install_ext_software/phyml.sh
+
+# configobj code
+validate.py

tral/repeat/repeat_align.py

@@ -43,7 +43,7 @@ def realign_repeat(my_msa, realignment='mafft', sequence_type='AA', rate_distrib
     Returns:
         my_msa realigned (list of strings)
 
-    .. todo:: 
+    .. todo::
         - decide what happens if branch length of tree is getting zero
     """
 

tral/sequence/repeat_detection_run.py

@@ -199,7 +199,7 @@ def __init__(self):
             self.valopts = {
                 # <file> input query alignment  (fasta/a2m/a3m) or HMM file (.hhm)
                 "-i": None,
-                "-d": REPEAT_DETECTOR_PATH['HHrepID_dummyhmm'],   # <path> dummy hmm database file
+                "-d": os.path.expanduser(REPEAT_DETECTOR_PATH['HHrepID_dummyhmm']),   # <path> dummy hmm database file
                 "-o": 'hhrepID.o',    # <file> write results and multiple sequence alignment to file (default=none)
                 "-v": 0,           # -v: verbose mode (default: show only warnings)  ;  -v 0: suppress all screen outpu
                 "-P": None,        # <float> max p-value of suboptimal alignments in all search rounds but the last one (def=0.1)

tral/tral_configuration/config.ini

@@ -10,7 +10,7 @@ sequence_type = AA
         # If the executable is in the system path, supply its name. Otherwise, supply the full path to the executable. Details are explained in TRAL's online docs.
         PHOBOS = phobos
         HHrepID = hhrepid_64
-        HHrepID_dummyhmm = /path/to/home/.tral/data/hhrepid/dummyHMM.hmm
+        HHrepID_dummyhmm = ~/.tral/data/hhrepid/dummyHMM.hmm
         T-REKS = T-REKS
         TRED = tred
         TRF = trf
@@ -36,13 +36,14 @@ sequence_type = AA
                 type = min
                 threshold = 1.9
 
+
 [repeat]
     scoreslist = phylo_gap01, # score (the comma in the end is needed for TRAL)
     calc_score = False # is the score calculated?
     calc_pvalue = False # is the pvalue calculated?
     precision = 10
     ginsi = ginsi # integrated in MAFFT
-	Castor = Castor
+    Castor = Castor
     [[castor_parameter]]
         rate_distribution = constant # either constant or gamma
     alfsim = alfsim

tral/tral_configuration/spec.ini

@@ -1,26 +1,53 @@
+# Config specification
+# To validate, download:
+# https://raw.githubusercontent.com/DiffSK/configobj/master/src/configobj/validate.py
+#
+# Then run:
+#    python -c 'import configobj; from validate import Validator;
+#        print( configobj.ConfigObj("tral/tral_configuration/config.ini",
+#                                   configspec="tral/tral_configuration/spec.ini"
+#                                  ).validate(Validator()))'
+
 sequence_type = option('AA', 'DNA', default="AA")
 
 [sequence]
     [[repeat_detection]]
         AA = string_list(min=0, max=100, default=list(''))
         DNA = string_list(min=0, max=100, default=list(''))
+    [[repeat_detector_path]]
+        PHOBOS = string(default="phobos")
+        HHrepID = string(default="hhrepid_64")
+        HHrepID_dummyhmm = string(default="~/.tral/data/hhrepid/dummyHMM.hmm")
+        T-REKS = string(default="T-REKS")
+        TRED = string(default="tred")
+        TRF = string(default="trf")
+        TRUST = string(default="TRUST")
+        TRUST_substitutionmatrix = string(default="/usr/bin/tral_tools/TRUST/Align/BLOSUM50")
+        XSTREAM = string(default="XSTREAM")
+
+[hmm]
+    hmmbuild = string(default='hmmbuild')
+    l_effective_max = integer(default=50)
+
+[filter]
+    [[basic]]
+        tag = string(default='generic_filter_tag')
+        [[[dict]]]
+            # Each filter requires a func_name (to look up the function)
+            # Other values are passed as arguments to the function.
+            [[[[__many__]]]]
+                func_name = string()  # pvalue, divergence, n_effective, etc.
+
+        [[[tags]]]
 
 [repeat]
     scoreslist = string_list(min=1, max=100, default=list('phylo_gap01'))
     calc_score = boolean(default=False)
     calc_pvalue = boolean(default=False)
     precision = float(min=1, max=1000, default=10)
+    ginsi = string(default='ginsi')
+    Castor = string(default='Castor')
+    [[castor_parameter]]
+        rate_distribution = option('constant', 'gamma', default='constant')
+    alfsim = string(default='alfsim')
 
-[filter]
-    [[basic]]
-        tag = string_list(min=1, max=100, default=list('generic_filter_tag'))
-        [[[dict]]]
-            [[[[1]]]]
-                func_name = string_list(min=1, max=100, default=list('pvalue'))
-                score = string_list(min=1, max=100, default=list('phylo_gap01'))
-                threshold = float(min=0.0, max=1.0, default=0.1)
-            [[[[2]]]]
-                func_name = string_list(min=1, max=100, default=list('pvalue'))
-                attribute = string_list(min=1, max=100, default=list('n_effective'))
-                type = option('min', 'max', default="min")
-                threshold = float(min=0.1, max=10000, default=2.5)