PgsFile 0.5.0__py3-none-any.whl → 0.5.2__py3-none-any.whl

PgsFile/PgsFile.py

@@ -4144,9 +4144,21 @@ def tfidf_keyword_extraction(documents, top_percent=(0.0, 0.10)):
 
 
 from xml.dom.minidom import Document
-def maketmx(tmx_path, source_list, target_list, source_lang="zh-CN", target_lang="en-US"):
+from datetime import datetime
+def maketmx(
+    tmx_path,
+    source_list,
+    target_list,
+    source_lang="zh-CN",
+    target_lang="en-US",
+    author="Petercusin",
+    client_name=None,
+    project_id=None,
+    domain=None,
+    status="Final"
+):
     """
-    Generate a TMX (Translation Memory eXchange) file for any two language pairs.
+    Generate a TMX (Translation Memory eXchange) file for any two language pairs, with optional metadata.
 
     Parameters
     ----------
@@ -4161,34 +4173,86 @@ def maketmx(tmx_path, source_list, target_list, source_lang="zh-CN", target_lang
         Source language code, e.g., "zh-CN" (default), "fr-FR", "de-DE".
     target_lang : str, optional
         Target language code, e.g., "en-US" (default), "es-ES", "ja-JP".
+    author : str, optional
+        Author of the TMX file, e.g., "Petercusin" (default).
+    client_name : str, optional
+        Name of the client or company, e.g., "Acme Corp".
+    project_id : str, optional
+        Project identifier, e.g., "Project_XYZ_2025".
+    domain : str, optional
+        Domain or subject field, e.g., "Medical", "Legal", "Technical".
+    status : str, optional
+        Translation status, e.g., "Draft", "Reviewed", "Final" (default).
 
     Returns
     -------
     None
         Writes the TMX file to the specified path.
 
+    Raises
+    ------
+    ValueError
+        If `source_list` and `target_list` have different lengths.
+
     Example
     -------
-    # Chinese to English
-    maketmx("zh_en.tmx", ["你好", "再见"], ["Hello", "Goodbye"], "zh-CN", "en-US")
-
-    # French to German
-    maketmx("fr_de.tmx", ["Bonjour", "Au revoir"], ["Hallo", "Auf Wiedersehen"], "fr-FR", "de-DE")
+    # Chinese to English, with metadata
+    maketmx(
+        "zh_en.tmx",
+        ["你好", "再见"],
+        ["Hello", "Goodbye"],
+        "zh-CN",
+        "en-US",
+        author="Dr. Guisheng PAN",
+        client_name="Acme Corp",
+        project_id="Project_XYZ_2025",
+        domain="Technical",
+        status="Final"
+    )
     """
+    if len(source_list) != len(target_list):
+        raise ValueError("source_list and target_list must have the same number of elements.")
+
     doc = Document()
     tmx = doc.createElement("tmx")
-    tmx.setAttribute("version", "1.0")
+    tmx.setAttribute("version", "1.4")
     doc.appendChild(tmx)
 
     header = doc.createElement("header")
     header.setAttribute("creationtool", "PgsFile")
-    header.setAttribute("creationtoolversion", "1.1.0")
-    header.setAttribute("creationtooldeveloper", "Dr. Guisheng PAN, panguisheng@sufe.edu.cn")
+    header.setAttribute("creationtoolversion", "0.5.1")
+    header.setAttribute("creationtooldeveloper", author)
+    header.setAttribute("creationdate", datetime.now().strftime("%Y-%m-%dT%H:%M:%SZ"))
     header.setAttribute("srclang", source_lang)
     header.setAttribute("trtlang", target_lang)
     header.setAttribute("datatype", "plaintext")
+    header.setAttribute("segtype", "sentence")
+    header.setAttribute("adminlang", "en-US")
+    header.setAttribute("o-tmf", "PgsFile_TM")
     tmx.appendChild(header)
 
+    # Add optional metadata as <prop> elements
+    if client_name:
+        prop = doc.createElement("prop")
+        prop.setAttribute("type", "x-Client")
+        prop.appendChild(doc.createTextNode(client_name))
+        header.appendChild(prop)
+    if project_id:
+        prop = doc.createElement("prop")
+        prop.setAttribute("type", "x-Project")
+        prop.appendChild(doc.createTextNode(project_id))
+        header.appendChild(prop)
+    if domain:
+        prop = doc.createElement("prop")
+        prop.setAttribute("type", "x-Domain")
+        prop.appendChild(doc.createTextNode(domain))
+        header.appendChild(prop)
+    if status:
+        prop = doc.createElement("prop")
+        prop.setAttribute("type", "x-Status")
+        prop.appendChild(doc.createTextNode(status))
+        header.appendChild(prop)
+
     body = doc.createElement("body")
     tmx.appendChild(body)
 
@@ -4216,3 +4280,4 @@ def maketmx(tmx_path, source_list, target_list, source_lang="zh-CN", target_lang
 
     with open(tmx_path, 'w', encoding='utf-8') as f:
         doc.writexml(f, indent='\t', newl='\n', addindent='\t')
+

{PgsFile-0.5.0.dist-info → PgsFile-0.5.2.dist-info}/METADATA

@@ -1,8 +1,8 @@
 Metadata-Version: 2.1
 Name: PgsFile
-Version: 0.5.0
+Version: 0.5.2
 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 OCR, tokenization, lemmatization, POS tagging, NER, ATE, dependency parsing, MDD, WSD, LIWC, MIP analysis and Chinese-English sentence alignment. 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/lWMkYDWQMjBJNKY2vMYTpw
+Home-page: https://github.com/Petercusin/PgsFile
 Author: Pan Guisheng
 Author-email: panguisheng@sufe.edu.cn
 License: Educational free
@@ -40,3 +40,4 @@ Key Features:
 
 Author: Pan Guisheng, a PhD student at the Graduate Institute of Interpretation and Translation of Shanghai International Studies University
 Email: 895284504@qq.com
+Homepage: https://mp.weixin.qq.com/s/lWMkYDWQMjBJNKY2vMYTpw

{PgsFile-0.5.0.dist-info → PgsFile-0.5.2.dist-info}/RECORD

@@ -1,4 +1,4 @@
-PgsFile/PgsFile.py,sha256=szt4jGqE_q2FYN399mZzdtsWUNK678Sy7y41xEvdL3M,172583
+PgsFile/PgsFile.py,sha256=YzdCEPFvT1jisKcihRUNyi2hqAWhStkP-yIcj_xqiLI,174646
 PgsFile/__init__.py,sha256=YkDTLWtveSeN4I5ZXSmdp7YZFOHb4yEzxQB82DEHk9s,3704
 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
@@ -2593,8 +2593,8 @@ PgsFile/models/prompts/5. ATE prompt.txt,sha256=5wu0gGlsV7DI0LruYM3-uAC6brppyYD0
 PgsFile/models/prompts/6. ATE3 prompt.txt,sha256=VnaXpPa6BgZHUcm8PxmP_qgU-8xEoTB3XcBqjwCUy_g,1254
 PgsFile/models/prompts/7. SentAlign prompt.txt,sha256=hXpqqC-CAgo8EytkJ0MaLhevLefALazWriY-ew39jxs,1537
 PgsFile/models/prompts/8. TitleCase prompt.txt,sha256=4p-LfGy0xAj2uPi9amyMm41T6Z17VNpFFsGZOgWhROs,1136
-PgsFile-0.5.0.dist-info/LICENSE,sha256=cE5c-QToSkG1KTUsU8drQXz1vG0EbJWuU4ybHTRb5SE,1138
-PgsFile-0.5.0.dist-info/METADATA,sha256=PFoIaOZCwuxKpZtCcvblbO2Xs-i6DQnUvn9s9uIuW_A,3070
-PgsFile-0.5.0.dist-info/WHEEL,sha256=eOLhNAGa2EW3wWl_TU484h7q1UNgy0JXjjoqKoxAAQc,92
-PgsFile-0.5.0.dist-info/top_level.txt,sha256=028hCfwhF3UpfD6X0rwtWpXI1RKSTeZ1ALwagWaSmX8,8
-PgsFile-0.5.0.dist-info/RECORD,,
+PgsFile-0.5.2.dist-info/LICENSE,sha256=cE5c-QToSkG1KTUsU8drQXz1vG0EbJWuU4ybHTRb5SE,1138
+PgsFile-0.5.2.dist-info/METADATA,sha256=fEpBEg0EhJCZOZdSTzLdD5Yn1eRlvdv1YFmWKmM4F2E,3119
+PgsFile-0.5.2.dist-info/WHEEL,sha256=eOLhNAGa2EW3wWl_TU484h7q1UNgy0JXjjoqKoxAAQc,92
+PgsFile-0.5.2.dist-info/top_level.txt,sha256=028hCfwhF3UpfD6X0rwtWpXI1RKSTeZ1ALwagWaSmX8,8
+PgsFile-0.5.2.dist-info/RECORD,,