Merge pull request #5 from ekorkut/ekorkut1_hashsign

Hash sign in urls is not encoded and more strict error check for order and number of metadata columns

Author: ekorkut

pycsvw/csvw.py

@@ -28,7 +28,8 @@
 
 from . import nt_serializer
 from .csvw_exceptions import NoDefaultOrValueUrlError, \
-    BothDefaultAndValueUrlError, BothLangAndDatatypeError, RiotWarning, RiotError
+    BothDefaultAndValueUrlError, BothLangAndDatatypeError, \
+    VirtualColumnPrecedesNonVirtualColumn, RiotWarning, RiotError
 
 READ_PERMISSIONS = stat.S_IRUSR | stat.S_IRGRP | stat.S_IROTH
 
@@ -92,6 +93,16 @@ def _read_metadata(handle):
 
                 col["aboutUrl"] = col.get("aboutUrl", None)
                 col["suppressOutput"] = col.get("suppressOutput", False)
+            # Non-virtual columns should precede virtual columns
+            virtual_seen_yet = False
+            for col in table_schema["columns"]:
+                if not virtual_seen_yet and col["virtual"]:
+                    virtual_seen_yet = True
+                    continue
+                if virtual_seen_yet and not col["virtual"]:
+                    raise VirtualColumnPrecedesNonVirtualColumn(
+                        "Non-virtual column {} comes after a virtual column. "
+                        "All virtual columns should come after non-virtual columns.".format(col))
 
         return out
 

pycsvw/csvw_exceptions.py

@@ -20,6 +20,22 @@ class NoDefaultOrValueUrlError(Exception):
     pass
 
 
+class NumberOfNonVirtualColumnsMismatch(Exception):
+    """
+    The exception throw when number of non-virtual columns in metadata
+    does not match number of columns in CSV file.
+    """
+    pass
+
+
+class VirtualColumnPrecedesNonVirtualColumn(Exception):
+    """
+    The exception thrown when a virtual column precedes a non-virtual column
+    in the specified metadata.
+    """
+    pass
+
+
 class BothDefaultAndValueUrlError(Exception):
     """
     The exception thrown when a virtual column specifies both

pycsvw/nt_serializer.py

@@ -16,7 +16,8 @@
 
 from .generator_utils import process_dates_times, DATATYPE_MAP, read_csv
 from .csvw_exceptions import NullValueException, BothValueAndLiteralError, \
-    BothValueAndDatatypeError, NoValueOrLiteralError, InvalidItemError
+    BothValueAndDatatypeError, NoValueOrLiteralError, InvalidItemError, \
+    NumberOfNonVirtualColumnsMismatch
 from .rdf_utils import is_null_value, get_column_map, get_subject_for_cell, \
     get_predicate_for_cell, apply_all_subs
 
@@ -271,6 +272,7 @@ def serialize(tables, md_tables, custom_prefixes, output_obj):
                 'prefixes': custom_prefixes
             }
         }
+        num_nonvirtual_columns = sum([1 for x in metadata["tableSchema"]["columns"] if not x["virtual"]])
         # Read the csv file fresh after rewinding the file
         table_file_obj = tables[table_url]
         table_file_obj.seek(0)
@@ -279,4 +281,10 @@ def serialize(tables, md_tables, custom_prefixes, output_obj):
         next(table_csv_reader)  # Ignore header
 
         for row_num, row in enumerate(table_csv_reader):
+            if len(row) != num_nonvirtual_columns:
+                raise NumberOfNonVirtualColumnsMismatch(
+                    "The number of non-virtual columns in metadata, {}, "
+                    "do not match with the number of columns in row {}, {}, "
+                    "of the csv file '{}'.".format(
+                        num_nonvirtual_columns, row_num + 1, len(row), table_url))
             write_row(output_obj, str(row_num + 1), row, table_info)

pycsvw/rdf_utils.py

@@ -17,7 +17,7 @@
 from .csvw_exceptions import NullValueException, MissingColumnError, FailedSubstitutionError
 
 
-SUB_PATTERN = re.compile(r'{(\w+)}')
+SUB_PATTERN = re.compile(r'{([A-Za-z0-9_\-# /:]+)}')
 
 
 def is_null_value(val, null_values):
@@ -60,7 +60,7 @@ def apply_sub(url, row, column_name_to_sub, column_info, quote_sub=True):
 
     rep_before = "{" + column_name_to_sub + "}"
     if quote_sub:
-        return url.replace(rep_before, quote(rep_after.encode('utf-8'), safe=':/'))
+        return url.replace(rep_before, quote(rep_after.encode('utf-8'), safe=':/#'))
     else:
         return url.replace(rep_before, rep_after)
 

setup.py

@@ -13,7 +13,7 @@
 
 setup(
     name='pycsvw',
-    version="1.0.1",
+    version="1.0.2",
     description='Generate JSON and RDF from csv files with metadata',
     url='https://github.com/bloomberg/pycsvw',
     author='Dev Ramudit, Erman Korkut',

tests/negative.NumberOfNonVirtualColumnsMismatch1.csv-metadata.json

@@ -0,0 +1,11 @@
+{
+  "@context": "http://www.w3.org/ns/csvw",
+  "url": "http://example.org/simple.csv",
+  "tableSchema": {
+    "columns": [{
+      "titles": "t1"
+    },{
+      "titles": "t2"
+    }]
+  }
+}

tests/negative.NumberOfNonVirtualColumnsMismatch2.csv-metadata.json

@@ -0,0 +1,16 @@
+{
+  "@context": "http://www.w3.org/ns/csvw",
+  "url": "http://example.org/simple.csv",
+  "tableSchema": {
+    "columns": [{
+      "titles": "t1"
+    },{
+      "titles": "t2"
+    },{
+      "titles": "t3"
+    }, {
+      "titles": "t4"
+    }
+    ]
+  }
+}

tests/negative.VirtualColumnPrecedesNonVirtualColumn.csv-metadata.json

@@ -0,0 +1,18 @@
+{
+  "@context": "http://www.w3.org/ns/csvw",
+  "url": "http://example.org/simple.csv",
+  "tableSchema": {
+    "columns": [{
+      "titles": "t1"
+    },
+      {
+      "name": "v1",
+      "virtual": true,
+      "aboutUrl": "owl:sub-{_row}",
+      "propertyUrl": "owl:obj-{_row}",
+      "valueUrl": "owl:pred-{_row}"
+    }, {
+      "titles": "t2"
+    }]
+  }
+}

tests/negative.metadata_mismatch.csv

@@ -0,0 +1,3 @@
+expense,description,amount
+taxi,from conference to hotel,20
+fee,conference registration fee,50

tests/rdf/test_url_special_chars.py

@@ -0,0 +1,109 @@
+# Copyright 2017 Bloomberg Finance L.P.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""
+This test covers the cases where csv file and metadata works with url safe characters like #, /, :
+"""
+
+from rdflib import ConjunctiveGraph, Literal, URIRef, Namespace
+
+from pycsvw import CSVW
+
+PRE_NS = Namespace('http://www.example.org/')
+
+
+def test_url_safe_chars():
+
+    csvw = CSVW(csv_path="tests/url_special_chars.csv",
+                metadata_path="tests/url_special_chars.csv-metadata.json")
+    rdf_output = csvw.to_rdf()
+
+    g = ConjunctiveGraph()
+    g.parse(data=rdf_output, format="turtle")
+
+    # Check subjects
+    sub1 = URIRef('http://www.example.org/c#1/chash2/chash3/chash4/chash5/chash6')
+    literals = [Literal('c#1'), Literal('chash2'), Literal('chash3'),
+                Literal('chash4'), Literal('chash6'), Literal('chash5')]
+    verify_non_virtual_columns(sub1, g, literals)
+    verify_virtual_columns(sub1, g, '#/:- _r1', '#/:-%20_r1')
+
+    sub2 = URIRef('http://www.example.org/c/1/c/2/c/3/c/4/c/5/c/6')
+    literals = [Literal('c/1'), Literal('c/2'), Literal('c/3'),
+                Literal('c/4'), Literal('c/6'), Literal('c/5')]
+    verify_non_virtual_columns(sub2, g, literals)
+    verify_virtual_columns(sub2, g, '/#:- _r2','/#:-%20_r2')
+
+    sub3 = URIRef('http://www.example.org/c:1/c:2/c:3/c:4/c:5/c:6')
+    literals = [Literal('c:1'), Literal('c:2'), Literal('c:3'),
+                Literal('c:4'), Literal('c:6'), Literal('c:5')]
+    verify_non_virtual_columns(sub3, g, literals)
+    verify_virtual_columns(sub3, g, ':#/-_ r3', ':#/-_%20r3')
+
+    sub4 = URIRef('http://www.example.org/c-1/c-2/c-3/c-4/c-5/c-6')
+    literals = [Literal('c-1'), Literal('c-2'), Literal('c-3'),
+                Literal('c-4'), Literal('c-6'), Literal('c-5')]
+    verify_non_virtual_columns(sub4, g, literals)
+    verify_virtual_columns(sub4, g, '-/#_ :r4', '-/#_%20:r4')
+
+    sub5 = URIRef('http://www.example.org/c%201/c%202/c%203/c%204/c%205/c%206')
+    literals = [Literal('c 1'), Literal('c 2'), Literal('c 3'),
+                Literal('c 4'), Literal('c 6'), Literal('c 5')]
+    verify_non_virtual_columns(sub5, g, literals)
+    verify_virtual_columns(sub5, g, ' -/#:_r5', '%20-/#:_r5')
+
+    sub6 = URIRef('http://www.example.org/c_1/c_2/c_3/c_4/c_5/c_6')
+    literals = [Literal('c_1'), Literal('c_2'), Literal('c_3'),
+                Literal('c_4'), Literal('c_6'), Literal('c_5')]
+    verify_non_virtual_columns(sub6, g, literals)
+    verify_virtual_columns(sub6, g, '_ /:#r6', '_%20/:#r6')
+
+
+def verify_virtual_columns(sub, g, orig_value_str, encoded_value_str):
+    v1_triples = list(g.triples((sub, PRE_NS['v1p{}'.format(encoded_value_str)], None)))
+    assert len(v1_triples) == 1
+    assert "v1p{}".format(encoded_value_str) in str(v1_triples[0][1])
+    assert orig_value_str == str(v1_triples[0][2])
+    v2_triples = list(g.triples((sub, PRE_NS['v2p{}'.format(encoded_value_str)], None)))
+    assert len(v2_triples) == 1
+    assert "v2p{}".format(encoded_value_str) in str(v2_triples[0][1])
+    assert 'v2v{}'.format(encoded_value_str) in str(v2_triples[0][2])
+
+    # Standalone virtual column
+    standalone_sub = URIRef('http://www.example.org/v3s{}'.format(encoded_value_str))
+    v3_triples = list(g.triples((standalone_sub, None, None)))
+    assert len(v3_triples) == 1
+    assert "v3p{}".format(encoded_value_str) in str(v3_triples[0][1])
+    assert 'v3v{}'.format(encoded_value_str) in str(v3_triples[0][2])
+
+
+def verify_non_virtual_columns(sub, g, literals):
+    all_triples = list(g.triples((sub, None, None)))
+    assert len(all_triples) == 9
+    assert (sub, PRE_NS['t#p'], literals[0]) in all_triples
+    assert (sub, PRE_NS['t/p'], literals[1]) in all_triples
+    assert (sub, PRE_NS['t:p'], literals[2]) in all_triples
+    assert (sub, PRE_NS['t-p'], literals[3]) in all_triples
+    assert (sub, PRE_NS['t_p'], literals[4]) in all_triples
+    # Space value
+    space_triple = list(g.triples((sub, PRE_NS['t%20p'], None)))
+    assert len(space_triple) == 1
+    assert "%20" in str(space_triple[0][1])
+    assert " " not in str(space_triple[0][1])
+    assert literals[5] == space_triple[0][2]
+
+
+
+
+
+
+
+