diff --git a/benchmark.py b/benchmark.py
index cf7c578..60cfea3 100644
--- a/benchmark.py
+++ b/benchmark.py
@@ -1,7 +1,6 @@
import json
import shutil
from statistics import mean, stdev
-from typing import Tuple
from alive_progress import alive_bar
from peewee import SelectQuery
@@ -11,7 +10,7 @@ from app import db
from server import search, suggest, exact_search
-def benchmark_query(query: SelectQuery, filename: str = None) -> Tuple[float, float]:
+def benchmark_query(query: SelectQuery, filename: str = None) -> tuple[float, float]:
query, params = query.sql()
query = "EXPLAIN (ANALYZE, COSTS, VERBOSE, BUFFERS, FORMAT JSON) " + query
diff --git a/data.py b/data.py
index f15907c..694d655 100644
--- a/data.py
+++ b/data.py
@@ -1,5 +1,4 @@
from dataclasses import dataclass
-from typing import Optional, List
colors = {
"campaign1": {
@@ -94,10 +93,10 @@ assert set(single_speaker["Handbooker Helper"].keys()) == set(range(1, 44 + 1))
class SeriesData:
name: str
slug: str
- playlist_id: Optional[str] = None
- videos: Optional[List[str]] = None
+ playlist_id: str | None = None
+ videos: list[str] | None = None
single_speaker: bool = False
- initial_speaker: Optional[str] = None
+ initial_speaker: str | None = None
series_data = [
diff --git a/fetch.py b/fetch.py
index cc44fe2..bcba184 100644
--- a/fetch.py
+++ b/fetch.py
@@ -64,7 +64,7 @@ def main(args: argparse.Namespace) -> None:
f.write(r.content)
changed = False
try:
- e = Episode.select().where((Episode.youtube_id == url)).get()
+ e = Episode.select().where(Episode.youtube_id == url).get()
if args.skip_existing and e.downloaded:
continue
except DoesNotExist:
diff --git a/import.py b/import.py
index a92532b..8a66008 100644
--- a/import.py
+++ b/import.py
@@ -1,7 +1,6 @@
import os
import re
from html import unescape
-from typing import List, Optional, Set, Union
from alive_progress import alive_bar
from peewee import fn, chunked
@@ -29,13 +28,13 @@ def add_to_text(text: str, add: str) -> str:
return add
-def line_key(line: Line) -> Union[str, Line]:
+def line_key(line: Line) -> str | Line:
if line.ismeta or line.isnote:
return line
return line.person
-def group_lines(dblines: List[Line]) -> List[Line]:
+def group_lines(dblines: list[Line]) -> list[Line]:
final_lines = []
order = 0
@@ -74,7 +73,7 @@ def group_lines(dblines: List[Line]) -> List[Line]:
return final_lines
-def insert_subtitle(text: str, person: Optional[Person], subline: Subtitle, episode: Episode, order: int,
+def insert_subtitle(text: str, person: Person | None, subline: Subtitle, episode: Episode, order: int,
isnote: bool = False, ismeta: bool = False) -> Line:
dbline = Line()
if not text:
@@ -94,7 +93,7 @@ def insert_subtitle(text: str, person: Optional[Person], subline: Subtitle, epis
def main() -> None:
os.nice(15)
- all_people: Set[str] = set()
+ all_people: set[str] = set()
for series in Series.select().order_by(Series.id):
for episode in Episode.select().where(
(Episode.text_imported == False) & (Episode.series == series) & (Episode.downloaded)
@@ -103,9 +102,9 @@ def main() -> None:
f.write("\n".join(sorted(p for p in all_people if "\n" not in p)))
file = srtdir / f"{episode.id}.srt"
strtext = file.read_text()
- subtitlelines: List[Subtitle] = list(parse(strtext))
+ subtitlelines: list[Subtitle] = list(parse(strtext))
print(episode.video_number, episode.pretty_title)
- person: Optional[Person] = None
+ person: Person | None = None
with db.atomic():
dblines = []
i = 0
diff --git a/phrases.py b/phrases.py
index 0d1f5ad..1fa941b 100644
--- a/phrases.py
+++ b/phrases.py
@@ -1,6 +1,5 @@
import os
from dataclasses import dataclass
-from typing import Dict
import en_core_web_md
from alive_progress import alive_bar
@@ -23,7 +22,7 @@ class Noun:
count: int = 1
-lemma_cache: Dict[str, str] = {}
+lemma_cache: dict[str, str] = {}
nlp: Language = en_core_web_md.load(disable=["ner", "textcat"])
nlp.Defaults.stop_words = STOP_WORDS
@@ -48,7 +47,7 @@ for episode in Episode.select().where((Episode.phrases_imported == False) & (Epi
print("run nlp")
doc = nlp(text)
print("nlp finished")
- nouns: Dict[str, Noun] = {}
+ nouns: dict[str, Noun] = {}
chunk: Span
noun_chunks = list(doc.noun_chunks)
with alive_bar(len(noun_chunks), title='lemmatizing and counting') as bar:
diff --git a/server.py b/server.py
index f00b158..4360904 100644
--- a/server.py
+++ b/server.py
@@ -1,6 +1,5 @@
import random
import time
-from typing import List
from flask import request, jsonify, Response, abort, g
from peewee import fn, Alias, SQL, DoesNotExist, Expression, ModelSelect, JOIN
@@ -19,6 +18,7 @@ from suggestions import suggestions
app.register_blueprint(ssr_routes)
+
def add_cors(response: Response) -> Response:
header = response.headers
header['Access-Control-Allow-Origin'] = '*'
@@ -34,7 +34,7 @@ def before_request():
def after_request(response: Response):
diff = time.perf_counter() - g.start
if response.response:
- response.headers.set("Server-Timing", f"server;dur={diff *1000 :.5f}")
+ response.headers.set("Server-Timing", f"server;dur={diff * 1000 :.5f}")
return response
@@ -194,11 +194,11 @@ def series():
@app.route("/api/episodes")
@cache.cached(timeout=60 * 60 * 24)
def api_episodes():
- all_series: List[Series] = Series.select().order_by(Series.order)
+ all_series: list[Series] = Series.select().order_by(Series.order)
data = []
for series in all_series:
- episodes: List[Episode] = Episode.select().where(Episode.series == series).order_by(Episode.video_number)
+ episodes: list[Episode] = Episode.select().where(Episode.series == series).order_by(Episode.video_number)
series_data = []
for episode in episodes:
@@ -241,7 +241,7 @@ def transcript():
(Episode.series.slug == series)
).join(Series).get()
- lines: List[Line] = Line.select(Line, Person).where(
+ lines: list[Line] = Line.select(Line, Person).where(
(Episode.episode_number == episode_number)
&
(Episode.series.slug == series)
diff --git a/stats.py b/stats.py
index 05a4967..0b68f4e 100644
--- a/stats.py
+++ b/stats.py
@@ -27,7 +27,7 @@ class Stats(ABC):
class MultiColumnStats(Stats):
- def as_data(self) -> List[Dict[str, Any]]:
+ def as_data(self) -> list[dict[str, Any]]:
data = []
cur = self.execute()
column_names = [d.name for d in cur.description]
diff --git a/suggestions.py b/suggestions.py
index d3877fa..54cfc43 100644
--- a/suggestions.py
+++ b/suggestions.py
@@ -28,14 +28,13 @@ That said, I try to only use phrases that don't contain spoilers themselves.
#
"""
from dataclasses import dataclass
-from typing import Optional
@dataclass
class Suggestion:
text: str
# only show this suggestion to people who have watched at least this episode
- episode: Optional[int] = None
+ episode: int | None = None
suggestions = {
diff --git a/tests.sql b/tests.sql
index 5b65d14..344fca8 100644
--- a/tests.sql
+++ b/tests.sql
@@ -1,4 +1,7 @@
-select e.pretty_title, text,char_length(line.text) as len from line join episode e on e.id = line.episode_id order by len desc;
+select e.pretty_title, text, char_length(line.text) as len
+from line
+ join episode e on e.id = line.episode_id
+order by len desc;
SELECT pg_size_pretty(pg_relation_size('phrase'));
@@ -6,12 +9,15 @@ SELECT pg_size_pretty(pg_relation_size('phrase'));
delete
from phrase;
-delete from line;
+delete
+from line;
update episode
-set text_imported= False, phrases_imported=False;
+set text_imported= False,
+ phrases_imported= False;
-update person set color=null;
+update person
+set color=null;
EXPLAIN analyse
SELECT text, sum(count) as total_count
@@ -76,3 +82,20 @@ SELECT *
FROM ts_stat('SELECT search_text from line')
order by nentry desc
limit 500;
+
+SELECT *, ts_rank("search_text", websearch_to_tsquery('english', 'I cast regret')) AS "rank"
+FROM line
+ INNER JOIN person ON (line.person_id = person.id)
+ INNER JOIN episode ON (line.episode_id = episode.id)
+WHERE (
+ (line.search_text @@ websearch_to_tsquery('english', 'I cast regret')) AND
+ (episode.episode_number <= 1000) AND
+ (episode.series_id = 2)
+ )
+ORDER BY rank DESC
+LIMIT 20;
+
+select websearch_to_tsquery('english', 'I cast regret');
+
+INSERT INTO line (text, search_text, ...) values ('This is a longer example text', to_tsvector('english', 'This is a longer example text'));
+select to_tsvector('english', 'This is a longer example text');
diff --git a/utils.py b/utils.py
index acda9f9..c021c79 100644
--- a/utils.py
+++ b/utils.py
@@ -1,7 +1,6 @@
import re
from datetime import timedelta
from pathlib import Path
-from typing import Optional
from app import cache
from data import single_speaker
@@ -17,7 +16,7 @@ def milliseconds_to_td(ms: int) -> timedelta:
return timedelta(milliseconds=ms)
-def episode_speaker(series_title: str, episode: int) -> Optional[str]:
+def episode_speaker(series_title: str, episode: int) -> str | None:
try:
series = single_speaker[series_title]
except KeyError:
diff --git a/web/package-lock.json b/web/package-lock.json
index a110005..4639093 100644
--- a/web/package-lock.json
+++ b/web/package-lock.json
@@ -3976,9 +3976,9 @@
}
},
"node_modules/caniuse-lite": {
- "version": "1.0.30001363",
- "resolved": "https://registry.npmjs.org/caniuse-lite/-/caniuse-lite-1.0.30001363.tgz",
- "integrity": "sha512-HpQhpzTGGPVMnCjIomjt+jvyUu8vNFo3TaDiZ/RcoTrlOq/5+tC8zHdsbgFB6MxmaY+jCpsH09aD80Bb4Ow3Sg==",
+ "version": "1.0.30001458",
+ "resolved": "https://registry.npmjs.org/caniuse-lite/-/caniuse-lite-1.0.30001458.tgz",
+ "integrity": "sha512-lQ1VlUUq5q9ro9X+5gOEyH7i3vm+AYVT1WDCVB69XOZ17KZRhnZ9J0Sqz7wTHQaLBJccNCHq8/Ww5LlOIZbB0w==",
"dev": true,
"funding": [
{
@@ -15459,9 +15459,9 @@
}
},
"caniuse-lite": {
- "version": "1.0.30001363",
- "resolved": "https://registry.npmjs.org/caniuse-lite/-/caniuse-lite-1.0.30001363.tgz",
- "integrity": "sha512-HpQhpzTGGPVMnCjIomjt+jvyUu8vNFo3TaDiZ/RcoTrlOq/5+tC8zHdsbgFB6MxmaY+jCpsH09aD80Bb4Ow3Sg==",
+ "version": "1.0.30001458",
+ "resolved": "https://registry.npmjs.org/caniuse-lite/-/caniuse-lite-1.0.30001458.tgz",
+ "integrity": "sha512-lQ1VlUUq5q9ro9X+5gOEyH7i3vm+AYVT1WDCVB69XOZ17KZRhnZ9J0Sqz7wTHQaLBJccNCHq8/Ww5LlOIZbB0w==",
"dev": true
},
"case-sensitive-paths-webpack-plugin": {
diff --git a/web/src/components/Intro.vue b/web/src/components/Intro.vue
index e17fcd4..94e5285 100644
--- a/web/src/components/Intro.vue
+++ b/web/src/components/Intro.vue
@@ -11,7 +11,7 @@
experimental transcript view.
If you have any feedback, ideas for improvements or bugs, feel free to contact me at
cr@lw1.at or on Twitter.
-
+ You can learn more about this website here.