Merge remote-tracking branch 'upstream/master'

[nominatim.git] / src / nominatim_api / search / icu_tokenizer.py

diff --git a/src/nominatim_api/search/icu_tokenizer.py b/src/nominatim_api/search/icu_tokenizer.py
index c2a265105a69d08eb3d7d8a75331e4a8c4d61dc9..c18dd8be62ed1190284e9c0751464b5e54091a47 100644 (file)
--- a/src/nominatim_api/search/icu_tokenizer.py
+++ b/src/nominatim_api/search/icu_tokenizer.py
@@ -48,6 +48,7 @@ class QueryPart(NamedTuple):
 QueryParts = List[QueryPart]
 WordDict = Dict[str, List[qmod.TokenRange]]
 
+
 def yield_words(terms: List[QueryPart], start: int) -> Iterator[Tuple[str, qmod.TokenRange]]:
     """ Return all combinations of words in the terms list after the
         given position.
@@ -72,7 +73,6 @@ class ICUToken(qmod.Token):
         assert self.info
         return self.info.get('class', ''), self.info.get('type', '')
 
-
     def rematch(self, norm: str) -> None:
         """ Check how well the token matches the given normalized string
             and add a penalty, if necessary.
@@ -91,7 +91,6 @@ class ICUToken(qmod.Token):
                 distance += abs((ato-afrom) - (bto-bfrom))
         self.penalty += (distance/len(self.lookup_word))
 
-
     @staticmethod
     def from_db_row(row: SaRow) -> 'ICUToken':
         """ Create a ICUToken from the row of the word table.
@@ -128,16 +127,13 @@ class ICUToken(qmod.Token):
                         addr_count=max(1, addr_count))
 
 
-
 class ICUQueryAnalyzer(AbstractQueryAnalyzer):
     """ Converter for query strings into a tokenized query
         using the tokens created by a ICU tokenizer.
     """
-
     def __init__(self, conn: SearchConnection) -> None:
         self.conn = conn
 
-
     async def setup(self) -> None:
         """ Set up static data structures needed for the analysis.
         """
@@ -163,7 +159,6 @@ class ICUQueryAnalyzer(AbstractQueryAnalyzer):
                      sa.Column('word', sa.Text),
                      sa.Column('info', Json))
 
-
     async def analyze_query(self, phrases: List[qmod.Phrase]) -> qmod.QueryStruct:
         """ Analyze the given list of phrases and return the
             tokenized query.
@@ -202,7 +197,6 @@ class ICUQueryAnalyzer(AbstractQueryAnalyzer):
 
         return query
 
-
     def normalize_text(self, text: str) -> str:
         """ Bring the given text into a normalized form. That is the
             standardized form search will work with. All information removed
@@ -215,7 +209,6 @@ class ICUQueryAnalyzer(AbstractQueryAnalyzer):
 
         return norm
 
-
     def split_query(self, query: qmod.QueryStruct) -> Tuple[QueryParts, WordDict]:
         """ Transliterate the phrases and split them into tokens.
 
@@ -248,7 +241,6 @@ class ICUQueryAnalyzer(AbstractQueryAnalyzer):
 
         return parts, words
 
-
     async def lookup_in_db(self, words: List[str]) -> 'sa.Result[Any]':
         """ Return the token information from the database for the
             given word tokens.
@@ -256,7 +248,6 @@ class ICUQueryAnalyzer(AbstractQueryAnalyzer):
         t = self.conn.t.meta.tables['word']
         return await self.conn.execute(t.select().where(t.c.word_token.in_(words)))
 
-
     def add_extra_tokens(self, query: qmod.QueryStruct, parts: QueryParts) -> None:
         """ Add tokens to query that are not saved in the database.
         """
@@ -268,7 +259,6 @@ class ICUQueryAnalyzer(AbstractQueryAnalyzer):
                                          count=1, addr_count=1, lookup_word=part.token,
                                          word_token=part.token, info=None))
 
-
     def rerank_tokens(self, query: qmod.QueryStruct, parts: QueryParts) -> None:
         """ Add penalties to tokens that depend on presence of other token.
         """
@@ -279,8 +269,8 @@ class ICUQueryAnalyzer(AbstractQueryAnalyzer):
                        and (repl.ttype != qmod.TokenType.HOUSENUMBER
                             or len(tlist.tokens[0].lookup_word) > 4):
                         repl.add_penalty(0.39)
-            elif tlist.ttype == qmod.TokenType.HOUSENUMBER \
-                 and len(tlist.tokens[0].lookup_word) <= 3:
+            elif (tlist.ttype == qmod.TokenType.HOUSENUMBER
+                  and len(tlist.tokens[0].lookup_word) <= 3):
                 if any(c.isdigit() for c in tlist.tokens[0].lookup_word):
                     for repl in node.starting:
                         if repl.end == tlist.end and repl.ttype != qmod.TokenType.HOUSENUMBER: