IPTC tag extraction

8 files changed, 225 insertions, 1 deletions

diff --git a/bsie/extractor/image/iptc.py b/bsie/extractor/image/iptc.py
new file mode 100644
index 0000000..195eff7
--- /dev/null
+++ b/bsie/extractor/image/iptc.py
@@ -0,0 +1,70 @@
+
+# standard imports
+import typing
+
+# bsie imports
+from bsie.utils import bsfs, node, ns
+
+# inner-module imports
+from .. import base
+
+# exports
+__all__: typing.Sequence[str] = (
+    'Iptc',
+    )
+
+
+## code ##
+
+class Iptc(base.Extractor):
+    """Turn IPTC keywords into tags."""
+
+    CONTENT_READER = 'bsie.reader.exif.Iptc'
+
+    def __init__(self):
+        super().__init__(bsfs.schema.from_string(base.SCHEMA_PREAMBLE + '''
+            bsn:Tag rdfs:subClassOf bsfs:Node .
+
+            bse:tag rdfs:subClassOf bsfs:Predicate ;
+                rdfs:domain bsn:Entity ;
+                rdfs:range bsn:Tag .
+
+            <https://schema.bsfs.io/ie/Node/Tag#label> rdfs:subClassOf bsfs:Predicate ;
+                rdfs:domain bsn:Tag ;
+                rdfs:range xsd:string ;
+                bsfs:unique "true"^^xsd:boolean .
+
+            '''))
+        self._callmap = {
+            self.schema.predicate(ns.bse.tag): self._keywords,
+            }
+
+    def extract(
+            self,
+            subject: node.Node,
+            content: dict,
+            principals: typing.Iterable[bsfs.schema.Predicate],
+            ) -> typing.Iterator[typing.Tuple[node.Node, bsfs.schema.Predicate, typing.Any]]:
+        for pred in principals:
+            # find callback
+            clbk = self._callmap.get(pred)
+            if clbk is None:
+                continue
+            # produce triples
+            yield from clbk(subject, content)
+
+    def _keywords(
+            self,
+            subject: node.Node,
+            content: dict,
+            ) -> typing.Iterator[typing.Tuple[node.Node, bsfs.schema.Predicate, typing.Any]]:
+        if 'Iptc.Application2.Keywords' not in content:
+            return
+        for keyword in content['Iptc.Application2.Keywords']:
+            tag = node.Node(ns.bsn.Tag, label=keyword)
+            yield subject, self.schema.predicate(ns.bse.tag), tag
+            yield tag, self.schema.predicate(ns.bst.label), keyword
+
+
+
+## EOF ##
diff --git a/bsie/lib/naming_policy.py b/bsie/lib/naming_policy.py
index 9b9a45d..3e7c940 100644
--- a/bsie/lib/naming_policy.py
+++ b/bsie/lib/naming_policy.py
@@ -4,6 +4,9 @@ import abc
 import os
 import typing
 
+# external imports
+import urllib.parse
+
 # bsie imports
 from bsie.utils import bsfs, errors, ns
 from bsie.utils.node import Node
@@ -84,6 +87,8 @@ class DefaultNamingPolicy(NamingPolicy):
             return self.name_file(node)
         if node.node_type == ns.bsn.Preview:
             return self.name_preview(node)
+        if node.node_type == ns.bsn.Tag:
+            return self.name_tag(node)
         raise errors.ProgrammingError('no naming policy available for {node.node_type}')
 
     def name_file(self, node: Node) -> Node:
@@ -112,4 +117,14 @@ class DefaultNamingPolicy(NamingPolicy):
         node.uri = getattr(self._prefix.preview(), fragment)
         return node
 
+    def name_tag(self, node: Node) -> Node:
+        # NOTE: Must ensure to produce the same name for that tags with the same label.
+        if 'label' in node.hints: # tag label
+            fragment = urllib.parse.quote(node.hints['label'])
+        else: # random name
+            fragment = self._uuid()
+        # FIXME: match to existing tags in bsfs storage!
+        node.uri = getattr(self._prefix.tag(), fragment)
+        return node
+
 ## EOF ##
diff --git a/bsie/reader/exif.py b/bsie/reader/exif.py
index 2d0428b..7ec7574 100644
--- a/bsie/reader/exif.py
+++ b/bsie/reader/exif.py
@@ -17,6 +17,7 @@ MATCH_RULE = 'mime=image/jpeg'
 # exports
 __all__: typing.Sequence[str] = (
     'Exif',
+    'Iptc',
     )
 
 
@@ -41,4 +42,24 @@ class Exif(base.Reader):
         except (TypeError, OSError, RuntimeError) as err:
             raise errors.ReaderError(path) from err
 
+
+class Iptc(base.Reader):
+    """Use pyexiv2 to read iptc metadata from image files."""
+
+    def __init__(self):
+        self._match = filematcher.parse(MATCH_RULE)
+
+    def __call__(self, path: str) -> dict:
+        # perform quick checks first
+        if not self._match(path):
+            raise errors.UnsupportedFileFormatError(path)
+
+        try:
+            # open the file
+            img = pyexiv2.Image(path)
+            # read metadata
+            return img.read_iptc()
+        except (TypeError, OSError, RuntimeError) as err:
+            raise errors.ReaderError(path) from err
+
 ## EOF ##
diff --git a/bsie/utils/namespaces.py b/bsie/utils/namespaces.py
index 4a66048..9357253 100644
--- a/bsie/utils/namespaces.py
+++ b/bsie/utils/namespaces.py
@@ -20,6 +20,7 @@ bsf = bsie.Literal.Array.Feature
 bsl = bsfs.Literal
 bsn = bsie.Node
 bsp = bsie.Node.Preview()
+bst = bsie.Node.Tag()
 
 # export
 __all__: typing.Sequence[str] = (
@@ -32,6 +33,7 @@ __all__: typing.Sequence[str] = (
     'bsl',
     'bsn',
     'bsp',
+    'bst',
     'xsd',
     )
 
diff --git a/test/extractor/image/test_iptc.py b/test/extractor/image/test_iptc.py
new file mode 100644
index 0000000..5fa763d
--- /dev/null
+++ b/test/extractor/image/test_iptc.py
@@ -0,0 +1,69 @@
+
+# standard imports
+import unittest
+
+# bsie imports
+from bsie.extractor import base
+from bsie.utils import bsfs, node as _node, ns
+
+# objects to test
+from bsie.extractor.image.iptc import Iptc
+
+
+## code ##
+
+class TestIptc(unittest.TestCase):
+
+    def test_eq(self):
+        # identical instances are equal
+        self.assertEqual(Iptc(), Iptc())
+        self.assertEqual(hash(Iptc()), hash(Iptc()))
+        # comparison respects type
+        class Foo(): pass
+        self.assertNotEqual(Iptc(), Foo())
+        self.assertNotEqual(hash(Iptc()), hash(Foo()))
+        self.assertNotEqual(Iptc(), 1234)
+        self.assertNotEqual(hash(Iptc()), hash(1234))
+        self.assertNotEqual(Iptc(), None)
+        self.assertNotEqual(hash(Iptc()), hash(None))
+
+    def test_schema(self):
+        self.assertSetEqual({pred.uri for pred in Iptc().schema.predicates()}, {
+            ns.bsfs.Predicate,
+            ns.bse.tag,
+            ns.bst.label,
+            })
+
+    def test_extract(self):
+        ext = Iptc()
+        node = _node.Node(ns.bsfs.File, '') # Blank node
+        content = {
+            'Iptc.Application2.Keywords': ['hello', 'world'],
+            'Iptc.Application2.RecordVersion': '4',
+            }
+        # target tags
+        t_hello = _node.Node(ns.bsn.Tag, label='hello')
+        t_world = _node.Node(ns.bsn.Tag, label='world')
+
+        # invalid principals are ignored
+        self.assertSetEqual(set(ext.extract(node, content, {ns.bse.filename})), set())
+        # extract finds all relevant information
+        self.assertSetEqual(set(ext.extract(node, content, {ext.schema.predicate(ns.bse.tag)})), {
+            (node, ext.schema.predicate(ns.bse.tag), t_hello),
+            (node, ext.schema.predicate(ns.bse.tag), t_world),
+            (t_hello, ext.schema.predicate(ns.bst.label), 'hello'),
+            (t_world, ext.schema.predicate(ns.bst.label), 'world'),
+            })
+
+        # empty content is acceptable
+        self.assertSetEqual(set(ext.extract(node, {}, set(ext.principals))), set())
+        # no principals is acceptable
+        self.assertSetEqual(set(ext.extract(node, content, set())), set())
+
+
+## main ##
+
+if __name__ == '__main__':
+    unittest.main()
+
+## EOF ##
diff --git a/test/lib/test_naming_policy.py b/test/lib/test_naming_policy.py
index c9b0cd2..b284fc0 100644
--- a/test/lib/test_naming_policy.py
+++ b/test/lib/test_naming_policy.py
@@ -31,6 +31,10 @@ class TestDefaultNamingPolicy(unittest.TestCase):
         self.assertEqual(policy.handle_node(
             Node(ns.bsn.Preview, ucid='abc123cba', size=123)).uri,
             URI('http://example.com/me/preview#abc123cba_s123'))
+        # processes bsn:Tag
+        self.assertEqual(policy.handle_node(
+            Node(ns.bsn.Tag, label='hello')).uri,
+            URI('http://example.com/me/tag#hello'))
         # raises an exception on unknown types
         self.assertRaises(errors.ProgrammingError, policy.handle_node,
             Node(ns.bsn.Invalid, ucid='abc123cba', size=123))
@@ -71,6 +75,29 @@ class TestDefaultNamingPolicy(unittest.TestCase):
         self.assertTrue(policy.name_preview(
             Node(ns.bsn.Preview, size=200)).uri.endswith('_s200'))
 
+    def test_name_tag(self):
+        # setup
+        policy = DefaultNamingPolicy('http://example.com', 'me')
+        # name_tag uses label
+        self.assertEqual(policy.name_tag(
+            Node(ns.bsn.Tag, label='hello')).uri,
+            URI('http://example.com/me/tag#hello'))
+        # name_tag matches the label
+        self.assertEqual(
+            policy.name_tag(Node(ns.bsn.Tag, label='world')),
+            policy.name_tag(Node(ns.bsn.Tag, label='world')),
+            )
+        self.assertNotEqual(
+            policy.name_tag(Node(ns.bsn.Tag, label='hello')),
+            policy.name_tag(Node(ns.bsn.Tag, label='world')),
+            )
+        # label can include characters that are not valid for an uri
+        self.assertEqual(policy.name_tag(
+            Node(ns.bsn.Preview, label='hello world { foo bar ] ')).uri,
+            URI('http://example.com/me/tag#hello%20world%20%7B%20foo%20bar%20%5D%20'))
+        # name_tag falls back to a random guid
+        self.assertTrue(policy.name_tag(
+            Node(ns.bsn.Tag,)).uri.startswith('http://example.com/me/tag#'))
 
 class TestNamingPolicyIterator(unittest.TestCase):
 
diff --git a/test/reader/test_exif.py b/test/reader/test_exif.py
index de6e801..1767f12 100644
--- a/test/reader/test_exif.py
+++ b/test/reader/test_exif.py
@@ -10,7 +10,7 @@ import pyexiv2
 from bsie.utils import errors
 
 # objects to test
-from bsie.reader.exif import Exif
+from bsie.reader.exif import Exif, Iptc
 
 
 ## code ##
@@ -44,6 +44,26 @@ class TestExif(unittest.TestCase):
             })
 
 
+class TestIptc(unittest.TestCase):
+    def test_call(self):
+        rdr = Iptc()
+        # discards non-image files
+        self.assertRaises(errors.UnsupportedFileFormatError, rdr,
+            os.path.join(os.path.dirname(__file__), 'invalid.doc'))
+        # raises on invalid image files
+        self.assertRaises(errors.UnsupportedFileFormatError, rdr,
+            os.path.join(os.path.dirname(__file__), 'invalid.jpg'))
+        # raises on invalid image files
+        pyexiv2.set_log_level(3) # suppress log message
+        self.assertRaises(errors.ReaderError, rdr,
+            os.path.join(os.path.dirname(__file__), 'testimage_exif_corrupted.jpg'))
+        # returns dict with exif info
+        self.assertDictEqual(rdr(os.path.join(os.path.dirname(__file__), 'testimage_exif.jpg')), {
+            'Iptc.Application2.Keywords': ['hello', 'world'],
+            'Iptc.Application2.RecordVersion': '4',
+            })
+
+
 ## main ##
 
 if __name__ == '__main__':
diff --git a/test/reader/testimage_exif.jpg b/test/reader/testimage_exif.jpg
index a774bc2..bc331ac 100644
--- a/test/reader/testimage_exif.jpg
+++ b/test/reader/testimage_exif.jpg