From d0768bac8bd69711f095e2c24b0416e045a64f98 Mon Sep 17 00:00:00 2001
From: Jeremy Goecks <jeremy.goecks@emory.edu>
Date: Wed, 30 Jan 2013 12:22:59 -0500
Subject: [PATCH] Collect and output comments when reading unordered GTF.
 Handle comments when converting from GTF to FLI.

---
 .../datatypes/converters/interval_to_fli.py       |  5 +++++
 lib/galaxy/datatypes/util/gff_util.py             | 15 +++++++++++++--
 2 files changed, 18 insertions(+), 2 deletions(-)

diff --git a/lib/galaxy/datatypes/converters/interval_to_fli.py b/lib/galaxy/datatypes/converters/interval_to_fli.py
index dd27e04119f..8448ee62d29 100644
--- a/lib/galaxy/datatypes/converters/interval_to_fli.py
+++ b/lib/galaxy/datatypes/converters/interval_to_fli.py
@@ -16,6 +16,8 @@ and symbols are sorted in lexigraphical order.
 
 import sys, optparse
 from galaxy import eggs
+import pkg_resources; pkg_resources.require( "bx-python" )
+from bx.tabular.io import Comment
 from galaxy.datatypes.util.gff_util import GFFReaderWrapper, read_unordered_gtf, convert_gff_coords_to_bed
 
 def main():
@@ -38,6 +40,9 @@ def main():
             in_reader = read_unordered_gtf( open( in_fname, 'r' ) )
 
         for feature in in_reader:
+            if isinstance( feature, Comment ):
+                continue
+
             for name in feature.attributes:
                 val = feature.attributes[ name ]
                 try:
diff --git a/lib/galaxy/datatypes/util/gff_util.py b/lib/galaxy/datatypes/util/gff_util.py
index 5eee0124d63..98f4a571f2d 100644
--- a/lib/galaxy/datatypes/util/gff_util.py
+++ b/lib/galaxy/datatypes/util/gff_util.py
@@ -384,9 +384,14 @@ def read_unordered_gtf( iterator, strict=False ):
         key_fn = lambda fields: fields[0] + '_' + get_transcript_id( fields )
 
     
-    # Aggregate intervals by transcript_id.
+    # Aggregate intervals by transcript_id and collect comments.
     feature_intervals = odict()
+    comments = []
     for count, line in enumerate( iterator ):
+        if line.startswith( '#' ):
+            comments.append( Comment( line ) )
+            continue
+
         line_key = key_fn( line.split('\t') )
         if line_key in feature_intervals:
             feature = feature_intervals[ line_key ]
@@ -413,7 +418,13 @@ def read_unordered_gtf( iterator, strict=False ):
     for features in chroms_features_sorted:
         features.sort( lambda a,b: cmp( a.start, b.start ) )
         
-    # Yield.
+    # Yield comments first, then features.
+    # FIXME: comments can appear anywhere in file, not just the beginning. 
+    # Ideally, then comments would be associated with features and output 
+    # just before feature/line.
+    for comment in comments:
+        yield comment
+
     for chrom_features in chroms_features_sorted:
         for feature in chrom_features:
             yield feature