#215 Done.

nicolay-r · Nov 29, 2021 · c199921 · c199921
1 parent c1c9ea4
commit c199921
Show file tree

Hide file tree

Showing 4 changed files with 16 additions and 15 deletions.
diff --git a/arekit/processing/frames/__init__.py b/arekit/processing/frames/__init__.py
diff --git a/arekit/common/frame_variants/search.py → arekit/processing/frames/annot.py b/arekit/common/frame_variants/search.py → arekit/processing/frames/annot.py
@@ -2,11 +2,9 @@
 from arekit.common.languages.ru.mods import RussianLanguageMods
 from arekit.common.text_frame_variant import TextFrameVariant
 from arekit.common.languages.mods import BaseLanguageMods
-from arekit.processing.text.enums import TermFormat
-from arekit.processing.text.parsed import ParsedText
 
 
-class FrameVariantsSearcher(object):
+class FrameVariantsAnnotationHelper(object):
 
     # region private methods
 
@@ -24,14 +22,14 @@ def __get_preposition(terms, index):
     # endregion
 
     @staticmethod
-    def iter_frames_from_parsed_text(frame_variants, parsed_text, locale_mods=RussianLanguageMods):
+    def iter_frames_from_lemmas(frame_variants, lemmas, locale_mods=RussianLanguageMods):
+        """
+        Considered to perform frames annotation across lemmatized terms.
+        """
         assert(isinstance(frame_variants, FrameVariantsCollection))
-        assert(isinstance(parsed_text, ParsedText))
+        assert(isinstance(lemmas, list))
         assert(issubclass(locale_mods, BaseLanguageMods))
 
-        lemmas = [locale_mods.replace_specific_word_chars(lemma) if isinstance(lemma, str) else lemma
-                  for lemma in parsed_text.iter_terms(term_format=TermFormat.Lemma)]
-
         start_ind = 0
         last_ind = 0
         max_variant_len = max([len(variant) for _, variant in frame_variants.iter_variants()])
@@ -43,7 +41,7 @@ def iter_frames_from_parsed_text(frame_variants, parsed_text, locale_mods=Russia
                 if not(last_ind < len(lemmas)):
                     continue
 
-                is_all_words_within = FrameVariantsSearcher.__check_all_words_within(
+                is_all_words_within = FrameVariantsAnnotationHelper.__check_all_words_within(
                     terms=lemmas,
                     start_index=start_ind,
                     last_index=last_ind)
@@ -56,8 +54,8 @@ def iter_frames_from_parsed_text(frame_variants, parsed_text, locale_mods=Russia
                 if not frame_variants.has_variant(ctx_value):
                     continue
 
-                prep_term = FrameVariantsSearcher.__get_preposition(terms=lemmas,
-                                                                    index=start_ind)
+                prep_term = FrameVariantsAnnotationHelper.__get_preposition(terms=lemmas,
+                                                                            index=start_ind)
 
                 yield TextFrameVariant(
                     variant=frame_variants.get_variant_by_value(ctx_value),

diff --git a/arekit/common/frame_variants/parse.py → arekit/processing/frames/parser.py b/arekit/common/frame_variants/parse.py → arekit/processing/frames/parser.py
@@ -4,7 +4,7 @@
 from arekit.common.languages.ru.mods import RussianLanguageMods
 from arekit.common.text_frame_variant import TextFrameVariant
 from arekit.common.languages.mods import BaseLanguageMods
-from arekit.common.frame_variants.search import FrameVariantsSearcher
+from arekit.processing.frames.annot import FrameVariantsAnnotationHelper
 from arekit.processing.text.enums import TermFormat
 from arekit.processing.text.parsed import ParsedText
 
@@ -41,9 +41,12 @@ def parse_frames_in_parsed_text(frame_variants_collection, parsed_text, locale_m
         assert(isinstance(parsed_text, ParsedText))
         assert(issubclass(locale_mods, BaseLanguageMods))
 
-        frame_variants_iter = FrameVariantsSearcher.iter_frames_from_parsed_text(
+        lemmas = [locale_mods.replace_specific_word_chars(lemma) if isinstance(lemma, str) else lemma
+                  for lemma in parsed_text.iter_terms(term_format=TermFormat.Lemma)]
+
+        frame_variants_iter = FrameVariantsAnnotationHelper.iter_frames_from_lemmas(
             frame_variants=frame_variants_collection,
-            parsed_text=parsed_text,
+            lemmas=lemmas,
             locale_mods=locale_mods)
 
         if frame_variants_iter is None:

diff --git a/arekit/processing/text/parser.py b/arekit/processing/text/parser.py
@@ -3,11 +3,11 @@
 
 from arekit.common.entities.base import Entity
 from arekit.common.frame_variants.collection import FrameVariantsCollection
-from arekit.common.frame_variants.parse import FrameVariantsParser
 from arekit.common.news.base import News
 from arekit.common.news.parse_options import NewsParseOptions
 from arekit.common.news.parsed.base import ParsedNews
 from arekit.common.utils import split_by_whitespaces
+from arekit.processing.frames.parser import FrameVariantsParser
 from arekit.processing.text.parsed import ParsedText
 from arekit.processing.text.tokens import Tokens
 from arekit.processing.text.token import Token