Move existing preprocessors to entry_points

This also allows us to remove some logic from regparser.plugins

regparser/plugins.py

@@ -1,23 +1,9 @@
-from collections import OrderedDict, defaultdict
-
-from importlib import import_module
+from collections import defaultdict
 
 from stevedore import extension
 from stevedore.exception import NoMatches
 
 
-def extend_list(namespace, original_list):
-    """Use the plugin manager to tack on extra entries to a list of strings"""
-    try:
-        results = list(original_list)   # shallow copy
-        mgr = extension.ExtensionManager(namespace=namespace,
-                                         invoke_on_load=False)
-        mgr.map(lambda ext: results.append(ext.entry_point_target))
-        return results
-    except NoMatches:
-        return original_list
-
-
 def update_dictionary(namespace, original):
     """
         Use the extension manager to update a dictionary.
@@ -39,23 +25,3 @@ def handle_plugin(ext):
         return dict(original)
     except NoMatches:
         return dict(original)
-
-
-def class_paths_to_classes(class_paths):
-    """We use an idiom of a list of class paths relatively often. This
-    function will convert that list into the appropriate classes"""
-    results = []
-    for class_path in class_paths:
-        split_char = ':' if ':' in class_path else '.'
-        mod_string, class_name = class_path.rsplit(split_char, 1)
-        mod = import_module(mod_string)
-        results.append(getattr(mod, class_name))
-    return results
-
-
-def classes_by_shorthand(class_paths):
-    """We often give our plugin entities specific names, indicated by their
-    "shorthand" field. This creates an (ordered) dictionary, mapping the
-    constructed classes by their shorthand name"""
-    return OrderedDict([(cls.shorthand, cls)
-                        for cls in class_paths_to_classes(class_paths)])

regparser/tree/xml_parser/xml_wrapper.py

@@ -2,10 +2,9 @@
 
 from lxml import etree
 import six
+from stevedore.extension import ExtensionManager
 
-from regparser.plugins import class_paths_to_classes
 from regparser.tree.xml_parser.preprocessors import replace_html_entities
-import settings
 
 
 class XMLWrapper(object):
@@ -29,8 +28,9 @@ def preprocess(self):
         attempts to fix some of those (general) flaws. For specific issues, we
         tend to instead use the files in settings.LOCAL_XML_PATHS"""
 
-        for preprocessor in class_paths_to_classes(settings.PREPROCESSORS):
-            preprocessor().transform(self.xml)
+        for extension in ExtensionManager('eregs_ns.parser.preprocessors',
+                                          invoke_on_load=True):
+            extension.plugin.transform(self.xml)
 
         return self
 

settings.py

@@ -106,20 +106,6 @@
     "ATF I 5300.1": "https://atf-eregs.apps.cloud.gov/static/atf_eregs/5300_1.pdf",
     "ATF I 5300.2": "https://www.atf.gov/file/58806/download"}
 
-PREPROCESSORS = plugins.extend_list('eregs_ns.parser.preprocessors', [
-    "regparser.tree.xml_parser.preprocessors.MoveLastAMDPar",
-    "regparser.tree.xml_parser.preprocessors.SupplementAMDPar",
-    "regparser.tree.xml_parser.preprocessors.ParenthesesCleanup",
-    "regparser.tree.xml_parser.preprocessors.MoveAdjoiningChars",
-    "regparser.tree.xml_parser.preprocessors.ApprovalsFP",
-    "regparser.tree.xml_parser.preprocessors.ExtractTags",
-    "regparser.tree.xml_parser.preprocessors.Footnotes",
-    "regparser.tree.xml_parser.preprocessors.ParseAMDPARs",
-    "regparser.tree.xml_parser.preprocessors.AtfI50032",
-    "regparser.tree.xml_parser.preprocessors.AtfI50031",
-    "regparser.tree.xml_parser.preprocessors.ImportCategories",
-])
-
 # Regulations.gov settings. The demo key is rate limited by IP; sign up for
 # your own key at
 # http://regulationsgov.github.io/developers/key/

setup.py

@@ -58,6 +58,27 @@
              "ExternalCitationParser"),
             "formatting = regparser.layer.formatting:Formatting",
             "graphics = regparser.layer.graphics:Graphics",
+        ],
+        "eregs_ns.parser.preprocessors": [
+            ("move-last-amdpar = regparser.tree.xml_parser.preprocessors:"
+             "MoveLastAMDPar"),
+            ("supplement-amd-par = regparser.tree.xml_parser.preprocessors:"
+             "SupplementAMDPar"),
+            ("parenthesis-cleanup = regparser.tree.xml_parser.preprocessors:"
+             "ParenthesesCleanup"),
+            ("move-adjoining-chars = regparser.tree.xml_parser.preprocessors:"
+             "MoveAdjoiningChars"),
+            ("approvals-fp = regparser.tree.xml_parser.preprocessors:"
+             "ApprovalsFP"),
+            ("extract-tags = regparser.tree.xml_parser.preprocessors:"
+             "ExtractTags"),
+            "footnotes = regparser.tree.xml_parser.preprocessors:Footnotes",
+            ("parse-amdpars = regparser.tree.xml_parser.preprocessors:"
+             "ParseAMDPARs"),
+            "atf-i-50032 = regparser.tree.xml_parser.preprocessors:AtfI50032",
+            "atf-i-50031 = regparser.tree.xml_parser.preprocessors:AtfI50031",
+            ("atf-import-categories = regparser.tree.xml_parser.preprocessors:"
+             "ImportCategories"),
         ]
     }
 )