From 0f984d3bfda9efaa6869286b13ff6842a6df67fb Mon Sep 17 00:00:00 2001
From: Nicola Soranzo <nicola.soranzo@tgac.ac.uk>
Date: Thu, 5 May 2016 11:49:05 +0100
Subject: [PATCH 1/2] Unicodify pretty datetimes which can contain non-ASCII
 chars

---
 templates/show_params.mako                          | 4 ++--
 templates/webapps/galaxy/data_manager/view_job.mako | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/templates/show_params.mako b/templates/show_params.mako
index 7c99075407f..8d807dfd2db 100644
--- a/templates/show_params.mako
+++ b/templates/show_params.mako
@@ -1,6 +1,6 @@
 <%inherit file="/base.mako"/>
 <%namespace file="/message.mako" import="render_msg" />
-<% from galaxy.util import nice_size %>
+<% from galaxy.util import nice_size, unicodify %>
 
 <style>
     .inherit {
@@ -117,7 +117,7 @@
         encoded_history_id = trans.security.encode_id( hda.history_id )
         %>
         <tr><td>Name:</td><td>${hda.name | h}</td></tr>
-        <tr><td>Created:</td><td>${hda.create_time.strftime(trans.app.config.pretty_datetime_format)}</td></tr>
+        <tr><td>Created:</td><td>${unicodify(hda.create_time.strftime(trans.app.config.pretty_datetime_format))}</td></tr>
         ##      <tr><td>Copied from another history?</td><td>${hda.source_library_dataset}</td></tr>
         <tr><td>Filesize:</td><td>${nice_size(hda.dataset.file_size)}</td></tr>
         <tr><td>Dbkey:</td><td>${hda.dbkey | h}</td></tr>
diff --git a/templates/webapps/galaxy/data_manager/view_job.mako b/templates/webapps/galaxy/data_manager/view_job.mako
index 0bb1398a3cb..d1a5a6d4b65 100644
--- a/templates/webapps/galaxy/data_manager/view_job.mako
+++ b/templates/webapps/galaxy/data_manager/view_job.mako
@@ -1,6 +1,6 @@
 <%inherit file="/base.mako"/>
 <%namespace file="/message.mako" import="render_msg" />
-<% from galaxy.util import nice_size %>
+<% from galaxy.util import nice_size, unicodify %>
 
 <%def name="title()">Data Manager: ${ data_manager.name | h } - ${ data_manager.description | h }</%def>
 
@@ -26,7 +26,7 @@
     </thead>
     <tbody>
         <tr><td>Name:</td><td>${hda.name | h}</td></tr>
-        <tr><td>Created:</td><td>${hda.create_time.strftime(trans.app.config.pretty_datetime_format) | h}</td></tr>
+        <tr><td>Created:</td><td>${unicodify(hda.create_time.strftime(trans.app.config.pretty_datetime_format)) | h}</td></tr>
         <tr><td>Filesize:</td><td>${nice_size(hda.dataset.file_size) | h}</td></tr>
         <tr><td>Tool Exit Code:</td><td>${job.exit_code | h}</td></tr>
         <tr><td>Full Path:</td><td>${hda.file_name | h}</td></tr>

From 427804ff823e4b4277e768e1822bd1642c549a1f Mon Sep 17 00:00:00 2001
From: Nicola Soranzo <nicola.soranzo@tgac.ac.uk>
Date: Thu, 5 May 2016 17:59:59 +0100
Subject: [PATCH 2/2] Do not sniff one-line files as CSV/TSV

---
 lib/galaxy/datatypes/sniff.py   |  4 ++++
 lib/galaxy/datatypes/tabular.py | 27 ++++++++++++++++-----------
 2 files changed, 20 insertions(+), 11 deletions(-)

diff --git a/lib/galaxy/datatypes/sniff.py b/lib/galaxy/datatypes/sniff.py
index ebbbec01ea7..0e4f76bd551 100644
--- a/lib/galaxy/datatypes/sniff.py
+++ b/lib/galaxy/datatypes/sniff.py
@@ -297,6 +297,10 @@ def guess_ext( fname, sniff_order, is_multi_byte=False ):
     >>> guess_ext(fname, sniff_order)
     'gff3'
     >>> fname = get_test_fname('temp.txt')
+    >>> file(fname, 'wt').write("a\\t2")
+    >>> guess_ext(fname, sniff_order)
+    'txt'
+    >>> fname = get_test_fname('temp.txt')
     >>> file(fname, 'wt').write("a\\t2\\nc\\t1\\nd\\t0")
     >>> guess_ext(fname, sniff_order)
     'tabular'
diff --git a/lib/galaxy/datatypes/tabular.py b/lib/galaxy/datatypes/tabular.py
index 4c5038c17ed..0ac9085f4ea 100644
--- a/lib/galaxy/datatypes/tabular.py
+++ b/lib/galaxy/datatypes/tabular.py
@@ -923,15 +923,19 @@ class BaseCSV( TabularData ):
                 # No columns so not separated by this dialect.
                 return False
 
-            # check all rows can be read as otherwise set_meta throws an exception
+            # Check that there is a second row as it is used by set_meta and
+            # that all rows can be read
             if self.strict_width:
                 num_columns = len(header_row)
+                found_second_line = False
                 for data_row in reader:
+                    found_second_line = True
                     # All columns must be the same length
                     if num_columns != len(data_row):
                         return False
+                if not found_second_line:
+                    return False
             else:
-                # Check the next row as it is used by set_meta
                 data_row = reader.next()
                 if len(data_row) < 2:
                     # No columns so not separated by this dialect.
@@ -958,7 +962,6 @@ class BaseCSV( TabularData ):
             """
             if not csv.Sniffer().has_header(open(filename, 'r').read(self.big_peek_size)):
                 return False
-
             return True
         except:
             # Not readable by Python's csv using this dialect
@@ -995,8 +998,8 @@ class BaseCSV( TabularData ):
 @dataproviders.decorators.has_dataproviders
 class CSV( BaseCSV ):
     """
-    Comma separated table data.
-    Only sniffs comma separated files with at least 2 columns
+    Comma-separated table data.
+    Only sniffs comma-separated files with at least 2 rows and 2 columns.
     """
 
     def __init__(self, **kwd):
@@ -1009,14 +1012,16 @@ class CSV( BaseCSV ):
 @dataproviders.decorators.has_dataproviders
 class TSV( BaseCSV ):
     """
-    Comma separated table data.
-    Only sniff tab separated files with at least two columns
+    Tab-separated table data.
+    Only sniff tab-separated files with at least 2 rows and 2 columns.
 
-    Note: Use of this datatype is optional as the general tabular format will handle most tab separated files.
-    This datatype would only be required for dataset with tabs INSIDE double quotes.
+    Note: Use of this datatype is optional as the general tabular datatype will
+    handle most tab-separated files. This datatype is only required for datasets
+    with tabs INSIDE double quotes.
 
-    This datatype currently does not support tsv files where the header has one column less to indicate first column is row names
-    This kind of file is handled fine by tabular.
+    This datatype currently does not support TSV files where the header has one
+    column less to indicate first column is row names. This kind of file is
+    handled fine by the tabular datatype.
     """
 
     def __init__(self, **kwd):