PgsFile 0.3.5__py3-none-any.whl → 0.3.6__py3-none-any.whl

PgsFile/PgsFile.py

@@ -1237,6 +1237,259 @@ claws_c7_tags = {
     }
 }
 
+spacy_pos_tags = {
+    "$": {
+        "description": "Dollar sign",
+        "chinese_translation": "美元符号",
+        "examples": ["$"]
+    },
+    "''": {
+        "description": "Closing quotation mark",
+        "chinese_translation": "闭合引号",
+        "examples": ["'"]
+    },
+    ",": {
+        "description": "Comma",
+        "chinese_translation": "逗号",
+        "examples": [","]
+    },
+    "-LRB-": {
+        "description": "Left round bracket (i.e., '(')",
+        "chinese_translation": "左圆括号",
+        "examples": ["("]
+    },
+    "-RRB-": {
+        "description": "Right round bracket (i.e., ')')",
+        "chinese_translation": "右圆括号",
+        "examples": [")"]
+    },
+    ".": {
+        "description": "Sentence-final punctuation",
+        "chinese_translation": "句末标点",
+        "examples": ["."]
+    },
+    ":": {
+        "description": "Colon, semi-colon, or dash",
+        "chinese_translation": "冒号、分号或破折号",
+        "examples": [":", ";", "-"]
+    },
+    "ADD": {
+        "description": "Email address",
+        "chinese_translation": "电子邮件地址",
+        "examples": ["example@example.com"]
+    },
+    "AFX": {
+        "description": "Affix",
+        "chinese_translation": "词缀",
+        "examples": ["un-", "re-", "-ing"]
+    },
+    "CC": {
+        "description": "Coordinating conjunction",
+        "chinese_translation": "并列连词",
+        "examples": ["and", "but", "or"]
+    },
+    "CD": {
+        "description": "Cardinal number",
+        "chinese_translation": "基数",
+        "examples": ["one", "two", "three"]
+    },
+    "DT": {
+        "description": "Determiner",
+        "chinese_translation": "限定词",
+        "examples": ["the", "a", "an"]
+    },
+    "EX": {
+        "description": "Existential 'there'",
+        "chinese_translation": "存在句中的there",
+        "examples": ["there"]
+    },
+    "FW": {
+        "description": "Foreign word",
+        "chinese_translation": "外来词",
+        "examples": ["rendezvous", "schadenfreude"]
+    },
+    "HYPH": {
+        "description": "Hyphen",
+        "chinese_translation": "连字符",
+        "examples": ["-"]
+    },
+    "IN": {
+        "description": "Preposition or subordinating conjunction",
+        "chinese_translation": "介词或从属连词",
+        "examples": ["in", "on", "at", "if", "because"]
+    },
+    "JJ": {
+        "description": "Adjective",
+        "chinese_translation": "形容词",
+        "examples": ["happy", "sad", "big"]
+    },
+    "JJR": {
+        "description": "Adjective, comparative",
+        "chinese_translation": "形容词比较级",
+        "examples": ["happier", "sadder", "bigger"]
+    },
+    "JJS": {
+        "description": "Adjective, superlative",
+        "chinese_translation": "形容词最高级",
+        "examples": ["happiest", "saddest", "biggest"]
+    },
+    "LS": {
+        "description": "List item marker",
+        "chinese_translation": "列表项标记",
+        "examples": ["1.", "2.", "3."]
+    },
+    "MD": {
+        "description": "Modal",
+        "chinese_translation": "情态动词",
+        "examples": ["can", "could", "may"]
+    },
+    "NFP": {
+        "description": "Superfluous punctuation",
+        "chinese_translation": "多余的标点符号",
+        "examples": ["..."]
+    },
+    "NN": {
+        "description": "Noun, singular or mass",
+        "chinese_translation": "单数或质量名词",
+        "examples": ["cat", "water", "sand"]
+    },
+    "NNP": {
+        "description": "Proper noun, singular",
+        "chinese_translation": "单数专有名词",
+        "examples": ["John", "London", "Everest"]
+    },
+    "NNPS": {
+        "description": "Proper noun, plural",
+        "chinese_translation": "复数专有名词",
+        "examples": ["Smiths", "Alps"]
+    },
+    "NNS": {
+        "description": "Noun, plural",
+        "chinese_translation": "复数名词",
+        "examples": ["cats", "dogs", "houses"]
+    },
+    "PDT": {
+        "description": "Predeterminer",
+        "chinese_translation": "前位限定词",
+        "examples": ["all", "both", "half"]
+    },
+    "POS": {
+        "description": "Possessive ending",
+        "chinese_translation": "所有格结尾",
+        "examples": ["'s"]
+    },
+    "PRP": {
+        "description": "Personal pronoun",
+        "chinese_translation": "人称代词",
+        "examples": ["I", "you", "he"]
+    },
+    "PRP$": {
+        "description": "Possessive pronoun",
+        "chinese_translation": "所有格代词",
+        "examples": ["my", "your", "his"]
+    },
+    "RB": {
+        "description": "Adverb",
+        "chinese_translation": "副词",
+        "examples": ["quickly", "happily", "sadly"]
+    },
+    "RBR": {
+        "description": "Adverb, comparative",
+        "chinese_translation": "副词比较级",
+        "examples": ["faster", "happier", "more quickly"]
+    },
+    "RBS": {
+        "description": "Adverb, superlative",
+        "chinese_translation": "副词最高级",
+        "examples": ["fastest", "happiest", "most quickly"]
+    },
+    "RP": {
+        "description": "Particle",
+        "chinese_translation": "小品词",
+        "examples": ["up", "down", "off"]
+    },
+    "SYM": {
+        "description": "Symbol",
+        "chinese_translation": "符号",
+        "examples": ["+", "=", "<"]
+    },
+    "TO": {
+        "description": "'to'",
+        "chinese_translation": "'to'",
+        "examples": ["to"]
+    },
+    "UH": {
+        "description": "Interjection",
+        "chinese_translation": "感叹词",
+        "examples": ["oh", "ah", "wow"]
+    },
+    "VB": {
+        "description": "Verb, base form",
+        "chinese_translation": "动词原形",
+        "examples": ["run", "jump", "eat"]
+    },
+    "VBD": {
+        "description": "Verb, past tense",
+        "chinese_translation": "动词过去式",
+        "examples": ["ran", "jumped", "ate"]
+    },
+    "VBG": {
+        "description": "Verb, gerund or present participle",
+        "chinese_translation": "动词动名词或现在分词",
+        "examples": ["running", "jumping", "eating"]
+    },
+    "VBN": {
+        "description": "Verb, past participle",
+        "chinese_translation": "动词过去分词",
+        "examples": ["run", "jumped", "eaten"]
+    },
+    "VBP": {
+        "description": "Verb, non-3rd person singular present",
+        "chinese_translation": "动词非第三人称单数现在式",
+        "examples": ["run", "jump", "eat"]
+    },
+    "VBZ": {
+        "description": "Verb, 3rd person singular present",
+        "chinese_translation": "动词第三人称单数现在式",
+        "examples": ["runs", "jumps", "eats"]
+    },
+    "WDT": {
+        "description": "Wh-determiner",
+        "chinese_translation": "Wh限定词",
+        "examples": ["which", "that", "what"]
+    },
+    "WP": {
+        "description": "Wh-pronoun",
+        "chinese_translation": "Wh代词",
+        "examples": ["who", "whom", "what"]
+    },
+    "WP$": {
+        "description": "Possessive wh-pronoun",
+        "chinese_translation": "所有格Wh代词",
+        "examples": ["whose"]
+    },
+    "WRB": {
+        "description": "Wh-adverb",
+        "chinese_translation": "Wh副词",
+        "examples": ["where", "when", "why"]
+    },
+    "XX": {
+        "description": "Unknown",
+        "chinese_translation": "未知",
+        "examples": []
+    },
+    "_SP": {
+        "description": "Space",
+        "chinese_translation": "空格",
+        "examples": [" "]
+    },
+    "``": {
+        "description": "Opening quotation mark",
+        "chinese_translation": "开放引号",
+        "examples": ["`"]
+    }
+}
+
 
 def word_list(split_words):
     """
@@ -1359,6 +1612,14 @@ def remove_empty_folders(folder_path):
     print(delet_root)
     print("Folders removed: ",len(delet_root))
 
+def remove_file(file_path):
+    import os
+    if os.path.exists(file_path):
+        os.remove(file_path)
+        print(f'{file_path} removed!')
+    else:
+        print(f"{file_path} doesn't exist")
+            
 def concatenate_excel_files(directory_path, output_file):
     # List to hold DataFrames
     dataframes = []

PgsFile/__init__.py

@@ -25,7 +25,8 @@ from .PgsFile import FilePath, FileName, DirList
 from .PgsFile import get_subfolder_path, get_full_path
 from .PgsFile import makedirec, makefile
 from .PgsFile import source_path, next_folder_names, get_directory_tree_with_meta, find_txt_files_with_keyword
-from .PgsFile import remove_empty_folders, remove_empty_txts, remove_empty_lines, remove_empty_last_line, move_file, copy_file
+from .PgsFile import remove_empty_folders, remove_empty_txts, remove_empty_lines, remove_empty_last_line
+from .PgsFile import move_file, copy_file, remove_file
 from .PgsFile import concatenate_excel_files
 from .PgsFile import set_permanent_environment_variable
 from .PgsFile import delete_permanent_environment_variable
@@ -35,7 +36,7 @@ from .PgsFile import get_system_info
 # 6. Data cleaning
 from .PgsFile import BigPunctuation, StopTags, Special, yhd
 from .PgsFile import ZhStopWords, EnPunctuation, get_stopwords, get_CET_dics, get_BNC_dic
-from .PgsFile import nltk_en_tags, nltk_tag_mapping, thulac_tags, ICTCLAS2008, LangCodes, pgs_abbres_words, usua_tag_set, claws_c7_tags
+from .PgsFile import nltk_en_tags, nltk_tag_mapping, thulac_tags, ICTCLAS2008, LangCodes, pgs_abbres_words, usua_tag_set, claws_c7_tags, spacy_pos_tags
 from .PgsFile import check_contain_chinese, check_contain_number
 from .PgsFile import replace_chinese_punctuation_with_english
 from .PgsFile import replace_english_punctuation_with_chinese

{PgsFile-0.3.5.dist-info → PgsFile-0.3.6.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: PgsFile
-Version: 0.3.5
+Version: 0.3.6
 Summary: This module simplifies Python package management, script execution, file handling, web scraping, and multimedia downloads. The module supports LLM-based NLP tasks such as tokenization, lemmatization, POS tagging, NER, dependency parsing, MDD, WSD, and MIP analysis. It also generates word lists and plots data, aiding literary students. Ideal for scraping data, cleaning text, and analyzing language, it offers user-friendly tools to streamline workflows.
 Home-page: https://mp.weixin.qq.com/s/12-KVLfaPszoZkCxuRd-nQ?token=1589547443&lang=zh_CN
 Author: Pan Guisheng

{PgsFile-0.3.5.dist-info → PgsFile-0.3.6.dist-info}/RECORD

@@ -1,5 +1,5 @@
-PgsFile/PgsFile.py,sha256=qKo8b6fgwBwhuPi4sEDKh4inr_1SoietOUIGbVLoToM,141592
-PgsFile/__init__.py,sha256=cS9OjLOCwguaFasoazpcqsCvBvJFCZijxFfzQACMJjE,3358
+PgsFile/PgsFile.py,sha256=52Uxj1gii1F1J9rvWnR1cFffIumeqBDsJRaN_uLoZUg,149704
+PgsFile/__init__.py,sha256=9vTeHtnxXaf_Qo36pes9o5_MU_M5M7MSMUddDWvkoDA,3408
 PgsFile/Corpora/Idioms/English_Idioms_8774.txt,sha256=qlsP0yI_XGECBRiPZuLkGZpdasc77sWSKexANu7v8_M,175905
 PgsFile/Corpora/Monolingual/Chinese/People's Daily 20130605/Raw/00000000.txt,sha256=SLGGSMSb7Ff1RoBstsTW3yX2wNZpqEUchFNpcI-mrR4,1513
 PgsFile/Corpora/Monolingual/Chinese/People's Daily 20130605/Raw/00000001.txt,sha256=imOa6UoCOIZoPXT4_HNHgCUJtd4FTIdk2FZNHNBgJyg,3372
@@ -2585,8 +2585,8 @@ PgsFile/models/fonts/博洋行书3500.TTF,sha256=VrgeHr8cgOL6JD05QyuD9ZSyw4J2aIV
 PgsFile/models/fonts/陆柬之行书字体.ttf,sha256=Zpd4Z7E9w-Qy74yklXHk4vM7HOtHuQgllvygxZZ1Hvs,1247288
 PgsFile/models/prompts/1. MIP prompt.txt,sha256=4lHlHmleayRytqr1n9jtt6vn1rQvyf4BKeThpbwI8o8,1638
 PgsFile/models/prompts/2. WSD prompt.txt,sha256=o-ZFtCRUCDrXgm040WTQch9v2Y_r2SIlrZaquilJjgQ,2348
-PgsFile-0.3.5.dist-info/LICENSE,sha256=cE5c-QToSkG1KTUsU8drQXz1vG0EbJWuU4ybHTRb5SE,1138
-PgsFile-0.3.5.dist-info/METADATA,sha256=Hi45rI7as2xrOLm4nKIIbnGuGfmQEnxgGDHYqHPDqrI,2892
-PgsFile-0.3.5.dist-info/WHEEL,sha256=eOLhNAGa2EW3wWl_TU484h7q1UNgy0JXjjoqKoxAAQc,92
-PgsFile-0.3.5.dist-info/top_level.txt,sha256=028hCfwhF3UpfD6X0rwtWpXI1RKSTeZ1ALwagWaSmX8,8
-PgsFile-0.3.5.dist-info/RECORD,,
+PgsFile-0.3.6.dist-info/LICENSE,sha256=cE5c-QToSkG1KTUsU8drQXz1vG0EbJWuU4ybHTRb5SE,1138
+PgsFile-0.3.6.dist-info/METADATA,sha256=HkRDJ8CCZoGrkKIuwlSpju61tF5bdn-2Hbrwt4B6zd8,2892
+PgsFile-0.3.6.dist-info/WHEEL,sha256=eOLhNAGa2EW3wWl_TU484h7q1UNgy0JXjjoqKoxAAQc,92
+PgsFile-0.3.6.dist-info/top_level.txt,sha256=028hCfwhF3UpfD6X0rwtWpXI1RKSTeZ1ALwagWaSmX8,8
+PgsFile-0.3.6.dist-info/RECORD,,