MW_thesaurus.py

@@ -1,6 +1,6 @@
 from itertools import chain
 
-from dict_dl import Queue, WordParser, ot, rb, uq, uqall, only_first_text
+from dict_dl import Queue, WordParser, ot, rb, uq, uqall
 
 
 class MWThesaurusParser(WordParser):
@@ -12,39 +12,64 @@ class MWThesaurusParser(WordParser):
     def thes(self):
         thes = {}
         for i in range(1, 10):
-            for j in range(1, 10):
+            for entry in self.root.findall(f".//div[@id='thesaurus-entry-{i}']"):
-                for entry in self.root.findall(
+                for se in chain(
-                    f".//div[@id='thesaurus-entry-{i}-{j}']"
+                    entry.findall(".//div[@class='sb no-sn']"),
+                    entry.findall(".//div[@class='sb has-num']"),
                 ):
-                    d = ""
+                    for e in se.findall(".//span[@class='dt']"):
-                    for e in entry.findall(".//span[@class='dt']"):
+                        examples = [ot(li) for li in e.findall(".//li")]
-                        d = only_first_text(e)
+                        for ul in e.findall(".//ul"):
-                    thes[d] = {}
+                            ul.clear()
-                    for relev in [4, 3]:
+                        d = ot(e)
-                        for e in entry.findall(
+                        thes[d] = {"examples": examples}
-                            f".//span[@class='thes-list sim-list-scored']//span[@class='lozenge color-{relev}']"
+                    thes[d]["synonyms"] = [
-                        ):
+                        ot(li)
-                            thes[d]["synonyms"] = thes[d].get("synonyms", []) + [ot(e)]
+                        for li in se.findall(
-                    for relev in [2, 1]:
+                            ".//span[@class='thes-list syn-list']/div[@class='thes-list-content synonyms_list']//li//a"
-                        for e in entry.findall(
+                        )
-                            f".//span[@class='thes-list sim-list-scored']//span[@class='lozenge color-{relev}']"
+                    ]
-                        ):
+                    thes[d]["synonyms"].extend([
-                            thes[d]["near synonyms"] = thes[d].get(
+                        ot(li)
-                                "near synonyms", []
+                        for li in se.findall(
-                            ) + [ot(e)]
+                            ".//span[@class='thes-list phrase-list']/div[@class='thes-list-content synonyms_list']//li//a"
+                        )
+                    ])
+                    thes[d]["near synonyms"] = [
+                        ot(li)
+                        for li in se.findall(
+                            ".//span[@class='thes-list rel-list']/div[@class='thes-list-content synonyms_list']//li//a"
+                        )
+                    ]
+                    thes[d]["near synonyms"].extend(
+                        [
+                            ot(li)
+                            for li in se.findall(
+                                ".//span[@class='thes-list sim-list']/div[@class='thes-list-content synonyms_list']//li//a"
+                            )
+                        ]
+                    )
+                    thes[d]["near antonyms"] = [
+                        ot(li)
+                        for li in se.findall(
+                            ".//span[@class='thes-list near-list']/div[@class='thes-list-content synonyms_list']//li//a"
+                        )
+                    ]
+                    thes[d]["near antonyms"].extend(
+                        [
+                            ot(li)
+                            for li in se.findall(
+                                ".//span[@class='thes-list opp-list']/div[@class='thes-list-content synonyms_list']//li//a"
+                            )
+                        ]
+                    )
+                    thes[d]["antonyms"] = [
+                        ot(li)
+                        for li in se.findall(
+                            ".//span[@class='thes-list ant-list']/div[@class='thes-list-content synonyms_list']//li//a"
+                        )
+                    ]
 
-                    for relev in [4, 3]:
-                        for e in entry.findall(
-                            f".//span[@class='thes-list opp-list-scored']//span[@class='lozenge color-{relev}']"
-                        ):
-                            thes[d]["antonyms"] = thes[d].get("antonyms", []) + [ot(e)]
-                    for relev in [2, 1]:
-                        for e in entry.findall(
-                            f".//span[@class='thes-list opp-list-scored']//span[@class='lozenge color-{relev}']"
-                        ):
-                            thes[d]["near antonyms"] = thes[d].get(
-                                "near antonyms", []
-                            ) + [ot(e)]
         return thes
 
     @property
@@ -67,15 +92,13 @@ class MWThesaurusParser(WordParser):
         assert (
             self.type or self.thes
         ), f"{self.time} {self.word}: type or definitions came back empty..."
-        return uqall(
+        return uqall({self.word: self.thes | {"type": self.type}})
-            {self.word: self.thes} | {"type": self.type, "time_of_retrieval": self.time}
-        )
 
 
-# w = MWThesaurusParser("content")
+w = MWThesaurusParser("augur")
 # print(w.neighbours)
-# print(w.todict())
+print(w.todict())
-# exit()
+exit()
 
 q = Queue(MWThesaurusParser, "en_MWThesaurus/", "_MWT.json")
 q.loadDB()

analysis.py

@@ -1,25 +1,12 @@
-from dict_dl import FullDictionary
-
 # import matplotlib.pyplot as plt
 # from PIL import Image
 # from wordcloud import STOPWORDS, WordCloud
 
-d = FullDictionary("en_MerriamWebster/", "_MW.json")
+d = FullDictionary("en_merriam_webster/", "_mw.json")
 # d = Dictionary("en_MW_thesaurus/", "_mwt.json")
 # d = Dictionary("de_duden/", "_duden.json")
 print(f"{d.readtime:.06f}")
 
-print(
-    sorted(
-        [
-            k
-            for k in d
-            if not any([c in ["a", "e", "i", "o", "u", "_"] for c in k.lower()])
-            and len(k) > 2
-            and k[-1] not in string.ascii_uppercase
-        ]
-    )
-)
 # print([k for k in d if not all([c in string.ascii_letters for c in k])])
 print([k for k in d if "?" in k])
 exit()

d.py

@@ -18,7 +18,7 @@ d = DictFile(os.path.expandvars(f"$DICT_DL/en_MerriamWebster/{prefix}_MW.json"))
 
 print(f"||||||||{query}||||||||")
 for k, v in d[query].items():
-    print(k, v)
+    print(k,v)
 #     if k != "type":
 #         table = Table(title=k)
 #         table.add_column("synonyms", justify="center", style="cyan", no_wrap=True)

dict_dl.py

@@ -122,8 +122,8 @@ def url2str(url: str, clean=True) -> str:
     xml_str = re.sub(r"<>", "-->", xml_str)
     xml_str = remove_tag(xml_str, "head")
     # xml_str = remove_tag(xml_str)
-    # with open("test.html", "w") as f:
+    with open("test.html", "w") as f:
-    #     f.write(xml_str)
+        f.write(xml_str)
     return xml_str
 
 

duden.py

@@ -96,10 +96,10 @@ class DudenParser(WordParser):
         )
 
 
-d = DudenParser("hineintauchen")
+# d = DudenParser("hinfallen")
-print(d.neighbours)
+# print(d.neighbours)
-print(d.todict())
+# print(d.todict())
-exit()
+# exit()
 
 q = Queue(DudenParser, "de_Duden/", "_D.json")
 q.loadDB()

main.py

@@ -80,7 +80,7 @@ def phrases(n: int = 4, nouns: int = 1, adjs: int = 2, pw: bool = False):
         if pw:
             # ps = [ "".join(p)[:-1] for p in [ [word + char for word, char in zip(p, [random_char() for w in p])] for p in phrases ] ]
             ps = [
-                "".join([w.capitalize() if i > 0 else w for i, w in enumerate(p)])
+                "".join([w.capitalize() for i,w in enumerate(p) if i > 0])
                 + random_char()
                 + f"{random.randint(0,999):03d}"
                 for p in phrases

merriam_webster.py

@@ -21,7 +21,7 @@ class MerriamWebsterParser(WordParser):
             for entry in self.root.findall(f".//div[@id='dictionary-entry-1']"):
                 for e in entry.findall(".//span[@class='cxl']"):
                     words = [ot(d) for d in entry.findall(".//a")]
-                    definitions[f'{ot(e)} {", ".join(words)}'] = []
+                    definitions[f'{ot(e)} {", ".join(words)}'] = [ ]
         return cw(definitions)
 
     @property

mw_scraper.py

@@ -1,12 +1,11 @@
 import random
-from typing import List
 import re
 from pathlib import Path
 from time import sleep
 from urllib.parse import urlparse
 
 from requests_html import HTMLSession
-from sqlmodel import Field, Relationship, Session, SQLModel, create_engine, select, Column, ARRAY, String
+from sqlmodel import Field, Relationship, Session, SQLModel, create_engine, select
 
 with open("db.secrets", "r") as f:
     db_pass = f.readline().strip()
@@ -29,7 +28,7 @@ class Description(SQLModel, table=True):
     word: str | None = Field(default=None, foreign_key="word.word")
     sense_id: int | None = Field(default=None, foreign_key="sense.id")
     description: str
-    examples: List[str] | None = Field(sa_column=Column(ARRAY(String)))
+    examples: list[str] | None
 
 
 class Pronunciation(SQLModel, table=True):
@@ -40,6 +39,11 @@ class Pronunciation(SQLModel, table=True):
 
 
 engine = create_engine(db_pass)
+SQLModel.metadata.create_all(engine)
+
+html_session = HTMLSession()
+
+QUEUE = {line.strip() for line in open("queue.db", "rt")}
 
 
 def add_word(word):
@@ -74,7 +78,6 @@ def add_word(word):
                 .where(Sense.word == _word)
                 .where(Sense.word_class == _class)
             ).one_or_none()
-
             if results:
                 sense = results
             else:
@@ -94,26 +97,14 @@ def add_word(word):
                     for sents in dt.find("span.sents"):
                         _example.append(sents.text)
                     _examples.append("; ".join(_example))
-
-                _final_description = "; ".join(_desc)
-                results = session.exec(
-                    select(Description).where(
-                        Description.word == word.word,
-                        Description.description == _final_description,
-                    )
-                ).one_or_none()
-                if results:
-                    continue
-                else:
                 session.add(
                     Description(
                         word=word.word,
                         sense_id=sense.id,
-                            description=_final_description,
+                        description="; ".join(_desc),
                         examples=_examples,
                     )
                 )
-
             for pron in c.find(
                 "span.prons-entries-list-inline div.prons-entry-list-item,a.prons-entry-list-item"
             ):
@@ -126,26 +117,16 @@ def add_word(word):
     return links
 
 
-def presently_available():
+def present():
     with Session(engine) as session:
         return session.exec(select(Word.word)).unique()
 
 
-if __name__ == "__main__":
+while True:
-    SQLModel.metadata.create_all(engine)
-    html_session = HTMLSession()
-
-    QUEUE = {line.strip() for line in open("queue.db", "rt")}
-
-    while True:
     try:
-            if len(QUEUE) < 20:
+        QUEUE |= add_word(random.choice(list(QUEUE)))
-                exit()
+        QUEUE -= set(present())
-            next_word = random.choice(list(QUEUE))
+        print(len(QUEUE))
-            already_present = set(presently_available())
-            print(next_word, len(QUEUE), len(already_present))
-            QUEUE |= add_word(next_word)
-            QUEUE -= already_present | {next_word}
         sleep(random.random() * 5)
     except KeyboardInterrupt:
         with open("queue.db", "wt") as f:

t.py

@@ -1,60 +1,25 @@
 #!/bin/python
-"""search the Merriam Webster Thesaurus with ease"""
-import argparse
 import os
+import sys
 from itertools import zip_longest
 
 from rich.console import Console
 from rich.table import Table
 
 from dict_dl import DictFile
-import string
 
-letters = string.ascii_lowercase
+if len(sys.argv) < 2:
-prefix_length = 3
+    query = next(sys.stdin).strip()
-unusual = (
+else:
-    lambda prefix: not all([c in letters for c in prefix.lower()])
+    query = sys.argv[1].strip()
-    or len(prefix) < prefix_length
+prefix = query[:3]
-)
 
-parser = argparse.ArgumentParser(description="Merriam Webster Thesaurus")
-parser.add_argument("-p", "--preview", action="store_true", help="FZF preview")
-parser.add_argument("query", type=str, help="query")
-
-args = parser.parse_args()
-prefix = args.query[:prefix_length].lower()
-
-if unusual(prefix):
-    prefix = "_" * prefix_length
 d = DictFile(os.path.expandvars(f"$DICT_DL/en_MWThesaurus/{prefix}_MWT.json"))
 
-if args.preview:
+print(f"||||||||{query}||||||||")
-    for k, v in d[args.query].items():
-        if k == "type":
-            word_type = k
-        else:
-            syns = v["synonyms"]
-            nsyns = v["related" if "related" in v else "near synonyms"]
-            nants = v["near antonyms"]
-            ants = v["antonyms"]
-            print(f"> {k}")
-            if syns:
-                print("  SYNONYMS\n  ", ", ".join(syns))
-            if nsyns:
-                print("  NEAR SYNONYMS\n  ", ", ".join(nsyns))
-            if nants:
-                print("  NEAR ANTONYMS\n  ", ", ".join(nants))
-            if ants:
-                print("  ANTONYMS\n  ", ", ".join(ants))
-        print()
-    exit()
-
-print(f"||||||||{args.query}||||||||")
 print()
-for k, v in d[args.query].items():
+for k, v in d[query].items():
-    if k == "type":
+    if k != "type":
-        word_type = k
-    else:
         table = Table(title=k)
         table.add_column("synonyms", justify="center", style="cyan", no_wrap=True)
         table.add_column("near synonyms", justify="center", style="cyan", no_wrap=True)