bsie extraction and info apps

3 files changed, 225 insertions, 0 deletions

diff --git a/bsie/apps/__init__.py b/bsie/apps/__init__.py
new file mode 100644
index 0000000..a548c3c
--- /dev/null
+++ b/bsie/apps/__init__.py
@@ -0,0 +1,20 @@
+"""
+
+Part of the bsie module.
+A copy of the license is provided with the project.
+Author: Matthias Baumgartner, 2022
+"""
+# imports
+import typing
+
+# inner-module imports
+from .index import main as index
+from .info import main as info
+
+# exports
+__all__: typing.Sequence[str] = (
+    'index',
+    'info',
+    )
+
+## EOF ##
diff --git a/bsie/apps/index.py b/bsie/apps/index.py
new file mode 100644
index 0000000..821aa4c
--- /dev/null
+++ b/bsie/apps/index.py
@@ -0,0 +1,131 @@
+"""
+
+Part of the bsie module.
+A copy of the license is provided with the project.
+Author: Matthias Baumgartner, 2022
+"""
+# imports
+import argparse
+import os
+import typing
+
+# bsfs imports
+import bsfs
+
+# bsie imports
+from bsie.base import errors
+from bsie.lib.bsie import BSIE
+from bsie.tools import builder
+from bsie.utils.bsfs import URI
+
+# exports
+__all__: typing.Sequence[str] = (
+    'main',
+    )
+
+
+## code ##
+
+def main(argv):
+    """Index files or directories into BSFS."""
+    parser = argparse.ArgumentParser(description=main.__doc__, prog='index')
+    parser.add_argument('--user', type=URI, default=URI('http://example.com/me'),
+        help='')
+    parser.add_argument('--collect', action='append', default=[],
+        help='')
+    parser.add_argument('--discard', action='append', default=[],
+        help='')
+    parser.add_argument('-r', '--recursive', action='store_true', default=False,
+        help='')
+    parser.add_argument('--follow', action='store_true', default=False,
+        help='')
+    parser.add_argument('--print', action='store_true', default=False,
+        help='')
+    parser.add_argument('input_file', nargs=argparse.REMAINDER,
+        help='')
+    args = parser.parse_args(argv)
+
+    # FIXME: Read reader/extractor configs from a config file
+    # reader builder
+    rbuild = builder.ReaderBuilder({})
+    # extractor builder
+    ebuild = builder.ExtractorBuilder([
+        {'bsie.extractor.generic.path.Path': {}},
+        {'bsie.extractor.generic.stat.Stat': {}},
+        {'bsie.extractor.generic.constant.Constant': dict(
+            tuples=[('http://bsfs.ai/schema/Entity#author', 'Me, myself, and I')],
+            schema='''
+                bse:author rdfs:subClassOf bsfs:Predicate ;
+                    rdfs:domain bsfs:Entity ;
+                    rdfs:range xsd:string ;
+                    bsfs:unique "true"^^xsd:boolean .
+                ''',
+            )},
+        ])
+    # pipeline builder
+    prefix = URI(args.user + ('file#' if args.user.endswith('/') else '/file#'))
+    pbuild = builder.PipelineBuilder(
+        prefix,
+        rbuild,
+        ebuild,
+        )
+
+    # build pipeline
+    pipeline = pbuild.build()
+    # build BSIE frontend
+    bsie = BSIE(pipeline, args.collect, args.discard)
+
+
+    def walk(handle):
+        """Walk through given input files."""
+        # FIXME: collect all triples by node, set all predicates at once
+        # FIXME: simplify code (below but maybe also above)
+        # FIXME: How to handle dependencies between data?
+        #        E.g. do I still want to link to a tag despite not being permitted to set its label?
+        # FIXME: node renaming?
+
+        # index input paths
+        for path in args.input_file:
+            if os.path.isdir(path) and args.recursive:
+                for dirpath, _, filenames in os.walk(path, topdown=True, followlinks=args.follow):
+                    for filename in filenames:
+                        for node, pred, value in bsie.from_file(os.path.join(dirpath, filename)):
+                            handle(node, pred, value)
+            elif os.path.isfile(path):
+                for node, pred, value in bsie.from_file(path):
+                    handle(node, pred, value)
+            else:
+                raise errors.UnreachableError()
+
+
+    if args.print:
+        walk(print)
+        return None
+
+    else:
+        # initialize bsfs
+        # NOTE: With presistent storages, the schema migration will be a seaparte operation.
+        # Here, we'd simply examine the schema and potentially discard more predicates.
+        store = bsfs.Open({
+            'Graph': {
+                'user': args.user,
+                'backend': {
+                    'SparqlStore': {}},
+                }})
+        store.migrate(bsie.schema)
+        # process files
+        def handle(node, pred, value):
+            store.node(node.node_type, node.uri).set(pred.uri, value)
+        walk(handle)
+        # return store
+        return store
+
+
+
+## main ##
+
+if __name__ == '__main__':
+    import sys
+    main(sys.argv[1:])
+
+## EOF ##
diff --git a/bsie/apps/info.py b/bsie/apps/info.py
new file mode 100644
index 0000000..8cc6dca
--- /dev/null
+++ b/bsie/apps/info.py
@@ -0,0 +1,74 @@
+"""
+
+Part of the bsie module.
+A copy of the license is provided with the project.
+Author: Matthias Baumgartner, 2022
+"""
+# imports
+import argparse
+import sys
+import typing
+
+# bsie imports
+from bsie.base import errors
+from bsie.tools import builder
+from bsie.utils.bsfs import URI
+
+# exports
+__all__: typing.Sequence[str] = (
+    'main',
+    )
+
+
+## code ##
+
+def main(argv):
+    """Show information from BSIE."""
+    parser = argparse.ArgumentParser(description=main.__doc__, prog='info')
+    parser.add_argument('what', choices=('predicates', ),
+        help='Select what information to show.')
+    args = parser.parse_args(argv)
+
+    # FIXME: Read reader/extractor configs from a config file
+    # reader builder
+    rbuild = builder.ReaderBuilder({})
+    # extractor builder
+    ebuild = builder.ExtractorBuilder([
+        {'bsie.extractor.generic.path.Path': {}},
+        {'bsie.extractor.generic.stat.Stat': {}},
+        {'bsie.extractor.generic.constant.Constant': dict(
+            tuples=[('http://bsfs.ai/schema/Entity#author', 'Me, myself, and I')],
+            schema='''
+                bse:author rdfs:subClassOf bsfs:Predicate ;
+                    rdfs:domain bsfs:Entity ;
+                    rdfs:range xsd:string ;
+                    bsfs:unique "true"^^xsd:boolean .
+                ''',
+            )},
+        ])
+    # pipeline builder
+    pbuild = builder.PipelineBuilder(
+        URI('http://example.com/me/file#'), # not actually used
+        rbuild,
+        ebuild,
+        )
+
+    # build pipeline
+    pipeline = pbuild.build()
+
+    # show info
+    if args.what == 'predicates':
+        # show predicates
+        for pred in pipeline.schema.predicates():
+            print(pred.uri)
+    else:
+        # args.what is already checked by argparse
+        raise errors.UnreachableError()
+
+
+## main ##
+
+if __name__ == '__main__':
+    main(sys.argv[1:])
+
+## EOF ##