rename privacy_policy_analyzer package to poligrapher

.gitignore

@@ -157,3 +157,8 @@ stats/
 figs/
 # cProfile data
 profile*.bin
+
+# Extra data, distributed separately
+*/extra-data/entity_info.json
+*/extra-data/named_entity_recognition/
+*/extra-data/purpose_classification/

MANIFEST.in

@@ -0,0 +1 @@
+recursive-include poligrapher/extra-data/ *

environment.yml

@@ -302,6 +302,7 @@ dependencies:
   - spacy-legacy=3.0.12=pyhd8ed1ab_0
   - spacy-loggers=1.0.1=pyhd3eb1b0_0
   - spacy-model-en_core_web_lg=3.5.0=pyhd8ed1ab_0
+  - spacy-model-en_core_web_md=3.5.0=pyhd8ed1ab_0
   - spacy-model-en_core_web_sm=3.5.0=pyhd8ed1ab_0
   - spacy-model-en_core_web_trf=3.5.0=pyh55bf090_0
   - spacy-transformers=1.2.4=py310hfb6f7a9_0

privacy_policy_analyzer/__init__.py → poligrapher/__init__.py

@@ -1,5 +1,5 @@
 from spacy.tokens import Token
-from privacy_policy_analyzer import utils
+from . import utils
 
 Token.set_extension("src", getter=utils.token_to_source)
 Token.set_extension("ent", getter=utils.token_to_ent)

privacy_policy_analyzer/annotators/collection_annotator.py → poligrapher/annotators/collection_annotator.py

@@ -6,7 +6,7 @@
 import yaml
 from spacy.tokens import Token
 
-import privacy_policy_analyzer
+import poligrapher
 
 from .base import BaseAnnotator
 
@@ -341,7 +341,7 @@ class CollectionAnnotator(BaseAnnotator):
     def __init__(self, nlp):
         super().__init__(nlp)
 
-        with pkg_resources.open_text(privacy_policy_analyzer, "verb_patterns.yml") as fin:
+        with pkg_resources.open_text(poligrapher, "verb_patterns.yml") as fin:
             config = yaml.safe_load(fin)
 
         token_map = config["token_map"]

privacy_policy_analyzer/phrase_normalization.py → poligrapher/phrase_normalization.py

@@ -6,7 +6,7 @@
 
 import regex
 
-from privacy_policy_analyzer.utils import TRIVIAL_WORDS
+from .utils import TRIVIAL_WORDS
 
 
 TRIM_TRANSITIONS = {

privacy_policy_analyzer/scripts/build_graph.py → poligrapher/scripts/build_graph.py

@@ -1,20 +1,22 @@
 #!/usr/bin/env python3
 
 import argparse
+from collections import defaultdict, deque
+import importlib.resources as pkg_resources
 import itertools
 import logging
 import os
-from collections import defaultdict, deque
 
 import networkx as nx
 import yaml
 
-from privacy_policy_analyzer.annotators import CollectionAnnotator
-from privacy_policy_analyzer.document import PolicyDocument, SegmentType
-from privacy_policy_analyzer.graph_utils import yaml_dump_graph, contracted_nodes
-from privacy_policy_analyzer.phrase_normalization import EntityMatcher, RuleBasedPhraseNormalizer
-from privacy_policy_analyzer.purpose_classification import PurposeClassifier
-from privacy_policy_analyzer.utils import setup_nlp_pipeline
+import poligrapher
+from poligrapher.annotators import CollectionAnnotator
+from poligrapher.document import PolicyDocument, SegmentType
+from poligrapher.graph_utils import contracted_nodes, yaml_dump_graph
+from poligrapher.phrase_normalization import EntityMatcher, RuleBasedPhraseNormalizer
+from poligrapher.purpose_classification import PurposeClassifier
+from poligrapher.utils import setup_nlp_pipeline
 
 
 def dag_add_edge(G, n1, n2, *args, **kwargs):
@@ -447,14 +449,21 @@ def main():
     parser = argparse.ArgumentParser()
     parser.add_argument("--nlp", required=True, help="NLP model directory")
     parser.add_argument("--purpose-classification", required=True, help="Purpose classification model directory")
-    parser.add_argument("-p", "--phrase-map", required=True, help="Path to phrase_map.yml")
-    parser.add_argument("-e", "--entity-info", required=True, help="Path to entity_info.json")
+    parser.add_argument("-p", "--phrase-map", default="", help="Path to phrase_map.yml")
+    parser.add_argument("-e", "--entity-info", default="", help="Path to entity_info.json")
     parser.add_argument("-v", "--variant", choices=["default", "original", "policylint", "per_section"],
                         default="default", help="Variant of the graph")
     parser.add_argument("--pretty", action="store_true", help="Generate pretty GraphML graph for visualization")
     parser.add_argument("workdirs", nargs="+", help="Input directories")
     args = parser.parse_args()
 
+    with pkg_resources.path(poligrapher, "extra-data") as extra_data:
+        if not args.phrase_map:
+            args.phrase_map = extra_data / "phrase_map.yml"
+
+        if not args.entity_info:
+            args.entity_info = extra_data / "entity_info.json"
+
     nlp = setup_nlp_pipeline(args.nlp)
     graph_builder = GraphBuilder(args.phrase_map, args.entity_info, args.purpose_classification, args.variant)
 

privacy_policy_analyzer/scripts/init_document.py → poligrapher/scripts/init_document.py

@@ -14,8 +14,8 @@
 import spacy
 import torch
 
-from privacy_policy_analyzer.document import PolicyDocument
-from privacy_policy_analyzer.utils import setup_nlp_pipeline
+from poligrapher.document import PolicyDocument
+from poligrapher.utils import setup_nlp_pipeline
 
 
 def main():

privacy_policy_analyzer/scripts/run_annotators.py → poligrapher/scripts/run_annotators.py

@@ -3,16 +3,16 @@
 import argparse
 import logging
 
-from privacy_policy_analyzer.annotators import (
+from poligrapher.annotators import (
     CollectionAnnotator,
     CoreferenceAnnotator,
     ListAnnotator,
     PurposeAnnotator,
     SubsumptionAnnotator,
     SubjectAnnotator,
 )
-from privacy_policy_analyzer.document import PolicyDocument
-from privacy_policy_analyzer.utils import setup_nlp_pipeline
+from poligrapher.document import PolicyDocument
+from poligrapher.utils import setup_nlp_pipeline
 
 
 def main():

setup.py

@@ -3,7 +3,8 @@
 from setuptools import setup
 
 setup(
-    name='Privacy Policy Analyzer',
+    name='poligrapher',
     author='UCI Networking Group',
-    packages=['privacy_policy_analyzer'],
+    include_package_data=True,
+    packages=['poligrapher', 'poligrapher.annotators', 'poligrapher.scripts'],
 )