PyPI - tdfs4ds - Versions diffs - 0.2.4.19__py3-none-any.whl → 0.2.4.20__py3-none-any.whl - Mend

tdfs4ds 0.2.4.19py3-none-any.whl → 0.2.4.20py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (6) hide show

tdfs4ds/__init__.py CHANGED Viewed

@@ -1,4 +1,4 @@
-__version__ = '0.2.4.19'
+__version__ = '0.2.4.20'
 import logging
 # Setup the logger
 logging.basicConfig(

tdfs4ds/utils/lineage.py CHANGED Viewed

@@ -631,17 +631,27 @@ def get_ddl(view_name, schema_name, object_type='view'):
     # Replace carriage returns with newlines for consistent formatting
     return ddl.replace('\r', '\n')
-from datetime import datetime
-import sqlparse
-import re
 import os
-def generate_process_report(format="html", output_file=None, collapsible=False, sort_by="view_name"):
+import datetime
+import sqlparse
+import tdfs4ds
+import importlib.resources as pkg_resources
+def generate_process_report(
+    format="html",
+    output_file=None,
+    collapsible=False,
+    sort_by="view_name",
+    theme_mode="light",
+    company_name=None,
+    company_logo_url=None
+):
     """
-    Generate a process catalog report with sidebar index grouped by database,
-    with sub-items for Entity, Features, and DDL.
+    Generate a process catalog report styled after the Teradata website.
     """
+    # Retrieve processes
     processes = tdfs4ds.process_catalog()
     processes = processes[processes.DATA_DOMAIN == tdfs4ds.DATA_DOMAIN].to_pandas()
     processes['VIEW'] = processes['VIEW_NAME'].apply(lambda x: x.split('.')[1].replace('"', ""))
@@ -650,9 +660,9 @@ def generate_process_report(format="html", output_file=None, collapsible=False,
         db, vw = full_name.replace('"', '').split('.')
         return db, vw
-    processes["DB"], processes["VW"] = zip(*processes["VIEW_NAME"].map(split_view_name))
     if sort_by:
+        processes = processes.copy()
+        processes["DB"], processes["VW"] = zip(*processes["VIEW_NAME"].map(split_view_name))
         if sort_by == "database":
             processes = processes.sort_values(["DB", "VW"])
         elif sort_by == "view_name":
@@ -660,123 +670,187 @@ def generate_process_report(format="html", output_file=None, collapsible=False,
         elif sort_by == "database,view_name":
             processes = processes.sort_values(["DB", "VW"])
-    timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
-    readable_ts = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
-    data_domain = tdfs4ds.DATA_DOMAIN
-    if output_file is None:
-        output_file = f"report_{data_domain}_{timestamp}.{ 'html' if format=='html' else format }"
+    # Timestamp and output file
+    timestamp = datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S")
+    if not output_file:
+        output_file = f"process_report_{tdfs4ds.DATA_DOMAIN}_{datetime.datetime.now().strftime('%Y%m%d_%H%M%S')}.html"
+    # Theme
+    if theme_mode == "dark":
+        bg_color = "#121212"
+        text_color = "#f5f5f5"
+        sidebar_bg = "#1e1e1e"
+        border_color = "#333"
+        pre_bg = "#222"
+        teradata_logo_file = "teradata_sym_rgb_wht_rev.png"
+    else:
+        bg_color = "#ffffff"
+        text_color = "#000000"
+        sidebar_bg = "#f8f9fa"
+        border_color = "#e1e1e1"
+        pre_bg = "#f4f6f8"
+        teradata_logo_file = "teradata_sym_rgb_pos.png"
+    # Load Teradata logo
+    with pkg_resources.path("tdfs4ds.data.logo", teradata_logo_file) as p:
+        teradata_logo_path = str(p)
+    with pkg_resources.path("tdfs4ds.data.logo", "tdfs4ds_logo.png") as p:  # ensure you add this logo to package
+        tdfs4ds_logo_path = str(p)
+    # HTML Report
     report_parts = []
     if format == "html":
+        report_parts.append("<html><head>")
+        report_parts.append("<meta charset='utf-8'>")
+        report_parts.append("<title>Process Catalog Report</title>")
         report_parts.append(f"""
-        <html>
-        <head>
-            <title>Process Report - {data_domain}</title>
-            <style>
-                body {{
-                    margin: 0;
-                    font-family: Arial, sans-serif;
-                }}
-                .sidebar {{
-                    position: fixed;
-                    top: 0;
-                    left: 0;
-                    width: 280px;
-                    height: 100%;
-                    background: #f4f4f4;
-                    border-right: 1px solid #ccc;
-                    padding: 20px;
-                    overflow-y: auto;
-                }}
-                .sidebar h2 {{
-                    font-size: 18px;
-                    margin-top: 0;
-                }}
-                .sidebar ul {{
-                    list-style: none;
-                    padding-left: 15px;
-                }}
-                .sidebar li {{
-                    margin-bottom: 4px;
-                }}
-                .sidebar a {{
-                    text-decoration: none;
-                    color: #007BFF;
-                    font-size: 14px;
-                }}
-                .sidebar a:hover {{
-                    text-decoration: underline;
-                }}
-                .content {{
-                    margin-left: 300px;
-                    padding: 20px;
-                }}
-                pre {{
-                    background: #f8f8f8;
-                    border: 1px solid #ddd;
-                    padding: 10px;
-                    overflow-x: auto;
-                }}
-            </style>
-        </head>
-        <body>
-            <div class="sidebar">
-                <h2>Teradata Feature Store</h2>
+        <style>
+        body {{
+            font-family: "Segoe UI", Arial, sans-serif;
+            margin: 0;
+            display: flex;
+            background: {bg_color};
+            color: {text_color};
+            height: 100vh;
+            overflow: hidden;
+        }}
+        .sidebar {{
+            width: 260px;
+            background: {sidebar_bg};
+            display: flex;
+            flex-direction: column;
+            border-right: 1px solid {border_color};
+            box-shadow: 2px 0 8px rgba(0,0,0,0.05);
+        }}
+        .sidebar-header {{
+            padding: 20px;
+            text-align: center;
+            border-bottom: 1px solid {border_color};
+        }}
+        .sidebar-header img {{
+            height: 35px;
+            max-width: 90%;
+            margin: 10px 0;
+        }}
+        .sidebar-content {{
+            flex: 1;
+            overflow-y: auto;
+            padding: 20px;
+        }}
+        .content-wrapper {{
+            flex: 1;
+            display: flex;
+            flex-direction: column;
+            height: 100vh;
+        }}
+        .header {{
+            flex: 0 0 auto;
+            background: #007CBA;
+            color: white;
+            padding: 20px;
+        }}
+        .header h1 {{
+            margin: 0 0 10px;
+        }}
+        .header p {{
+            margin: 5px 0;
+            font-size: 0.95em;
+        }}
+        .content {{
+            flex: 1;
+            padding: 30px;
+            overflow-y: auto;
+        }}
+        .footer {{
+            flex: 0 0 auto;
+            padding: 15px;
+            border-top: 1px solid {border_color};
+            font-size: 0.9em;
+            text-align: center;
+            color: #666;
+            display: flex;
+            justify-content: center;
+            align-items: center;
+            gap: 20px;
+        }}
+        .footer img {{
+            height: 25px;
+        }}
+        pre {{
+            background: {pre_bg};
+            padding: 12px;
+            border-radius: 6px;
+            overflow-x: auto;
+            font-size: 0.9em;
+        }}
+        ul {{ list-style-type: none; padding-left: 15px; }}
+        a {{ text-decoration: none; color: #007CBA; }}
+        a:hover {{ text-decoration: underline; }}
+        h2, h3 {{ color: #007CBA; }}
+        </style>
         """)
-        # Build structured index: group by database, with sub-items
-        grouped = processes.groupby("DB")
-        for db, group in grouped:
-            report_parts.append(f"<h3>DB: {db}</h3><ul>")
-            for _, row in group.iterrows():
-                view_name = row['VIEW_NAME']
-                vw = row["VW"]
-                safe_id = re.sub(r'[^A-Za-z0-9_]+', '_', view_name)
-                # Main view link
-                report_parts.append(f'<li><a href="#{safe_id}">{vw}</a>')
-                # Sub-links: entity, features, ddl
-                report_parts.append("<ul>")
-                report_parts.append(f'<li><a href="#{safe_id}_entity">Entity</a></li>')
-                report_parts.append(f'<li><a href="#{safe_id}_features">Features</a></li>')
-                report_parts.append(f'<li><a href="#{safe_id}_ddl">DDL</a></li>')
+        report_parts.append("</head><body>")
+        # Sidebar
+        report_parts.append("<div class='sidebar'>")
+        report_parts.append("<div class='sidebar-header'>")
+        if company_logo_url:
+            report_parts.append(f'<img src="{company_logo_url}" alt="Company logo"><br>')
+        report_parts.append(f'<img src="{teradata_logo_path}" alt="Teradata logo">')
+        report_parts.append("</div>")  # end sidebar-header
+        # Sidebar content (index)
+        report_parts.append("<div class='sidebar-content'>")
+        report_parts.append("<h2>Process Catalog Index</h2><ul id='indexList'>")
+        for db, db_group in processes.groupby("DB"):
+            report_parts.append(f"<li><h3>DB: {db}</h3><ul>")
+            for _, row in db_group.iterrows():
+                view_id = f"{row['DB']}_{row['VW']}"
+                report_parts.append(f'<li><a href="#{view_id}">{row["VW"]}</a><ul>')
+                report_parts.append(f'<li><a href="#{view_id}_entity">Entity</a></li>')
+                report_parts.append(f'<li><a href="#{view_id}_features">Features</a></li>')
+                report_parts.append(f'<li><a href="#{view_id}_ddl">DDL</a></li>')
                 report_parts.append("</ul></li>")
-            report_parts.append("</ul>")
+            report_parts.append("</ul></li>")
+        report_parts.append("</ul></div></div>")  # close sidebar-content & sidebar
-        report_parts.append("""
-            </div>
-            <div class="content">
-        """)
+        # Main content wrapper
+        report_parts.append("<div class='content-wrapper'>")
         # Header
-        report_parts.append(f"<h1>Process Catalog Report - {data_domain}</h1>")
-        report_parts.append(f"<p><em>Generated on {readable_ts}</em></p>")
+        report_parts.append("<div class='header'>")
+        report_parts.append("<h1>Process Catalog Report</h1>")
+        report_parts.append("<p>Data Domain: Customer Transaction Analytics Time Management | "
+                            f"Generated on {timestamp}</p>")
+        report_parts.append("<p>Powered by <strong>Teradata</strong> and <strong>tdfs4ds</strong></p>")
+        report_parts.append("</div>")  # end header
+        # Content
+        report_parts.append("<div class='content'>")
+    # Sections per process
     for _, row in processes.iterrows():
         view_name = row['VIEW_NAME']
-        db, vw = row["DB"], row["VW"]
+        db, vw = split_view_name(view_name)
+        view_id = f"{db}_{vw}"
         entity_list = row['ENTITY_ID'].split(',')
         features_list = row['FEATURE_NAMES'].split(',')
-        ddl_raw = tdml.execute_sql(f"SHOW VIEW {view_name}").fetchall()[0][0]
+        process_id = row['PROCESS_ID']
+        ddl_raw = tdfs4ds.tdml.execute_sql(f"SHOW VIEW {view_name}").fetchall()[0][0]
         ddl = sqlparse.format(ddl_raw, reindent=True, keyword_case="upper")
-        safe_id = re.sub(r'[^A-Za-z0-9_]+', '_', view_name)
         if format == "html":
-            section = [f'<h2 id="{safe_id}">{db}.{vw} ({data_domain})</h2>']
-            section.append(f'<h3 id="{safe_id}_entity">Entity</h3><ul>')
+            section = [f'<h2 id="{view_id}">{view_name}</h2>']
+            section.append(f"<p><strong>PROCESS_ID:</strong> {process_id}</p>")
+            section.append(f'<h3 id="{view_id}_entity">Entity</h3><ul>')
             section.extend([f"<li>{t}</li>" for t in entity_list])
             section.append("</ul>")
-            section.append(f'<h3 id="{safe_id}_features">Features ({len(features_list)} total)</h3><ul>')
+            section.append(f'<h3 id="{view_id}_features">Features ({len(features_list)} total)</h3><ul>')
             section.extend([f"<li>{t}</li>" for t in features_list])
             section.append("</ul>")
+            section.append(f'<h3 id="{view_id}_ddl">DDL</h3>')
-            section.append(f'<h3 id="{safe_id}_ddl">DDL</h3>')
             if collapsible:
                 section.append("<details><summary>Show/Hide DDL</summary><pre><code>")
                 section.append(ddl)
@@ -787,9 +861,20 @@ def generate_process_report(format="html", output_file=None, collapsible=False,
             report_parts.append("\n".join(section))
     if format == "html":
+        report_parts.append("</div>")  # end content
+        report_parts.append(
+            f"<div class='footer'>"
+            f'<a href="https://www.teradata.com" target="_blank">'
+            f'<img src="{teradata_logo_path}" alt="Teradata logo"></a>'
+            f'<a href="https://pypi.org/project/tdfs4ds/" target="_blank">'
+            f'<img src="{tdfs4ds_logo_path}" alt="tdfs4ds logo"></a>'
+            f"<span>© 2025 . Generated using tdfs4ds on Teradata.</span>"
+            f"</div>"
+        )
         report_parts.append("</div></body></html>")
+    # Write output
     with open(output_file, "w", encoding="utf-8") as f:
         f.write("\n".join(report_parts))
-    print(f"✅ Report generated: {os.path.abspath(output_file)}")
+    return output_file

{tdfs4ds-0.2.4.19.dist-info → tdfs4ds-0.2.4.20.dist-info}/METADATA RENAMED Viewed

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: tdfs4ds
-Version: 0.2.4.19
+Version: 0.2.4.20
 Summary: A python package to simplify the usage of feature store using Teradata Vantage ...
 Author: Denis Molin
 Requires-Python: >=3.6
@@ -13,6 +13,10 @@ Requires-Dist: tqdm
 Requires-Dist: networkx
 Requires-Dist: sqlparse
+<p align="center">
+  <img src="https://github.com/denismolin/tdfs4ds/blob/main/tdfs4ds_logo.png" alt="tdfs4ds logo" width="200"/>
+</p>
 # tdfs4ds : A Feature Store Library for Data Scientists working with Clearscape Analytics
 The tdfs library is a Python package designed for managing and utilizing Feature Stores in a Teradata Database. With a set of easy-to-use functions, tdfs enables the efficient creation, registration, and storage of features. It also simplifies the process of preparing feature data for ingestion, building datasets for data analysis, and obtaining already existing features.

{tdfs4ds-0.2.4.19.dist-info → tdfs4ds-0.2.4.20.dist-info}/RECORD RENAMED Viewed

@@ -2,7 +2,7 @@ tdfs/__init__.py,sha256=7AcO7uB1opRCt7t2JOHworKimfAaDeO3boRW7u9Geo8,23
 tdfs/datasets.py,sha256=-b2MPEKGki2V1M8iUcoDR9uc2krIK7u1CK-EhChvihs,985
 tdfs/feature_store.py,sha256=Honu7eOAXxP4Ivz0mRlhuNkfTDzgZl5HB1WlQUwzcZ0,31354
 tdfs/data/curves.csv,sha256=q0Tm-0yu7VMK4lHvHpgi1LMeRq0lO5gJy2Q17brKbEM,112488
-tdfs4ds/__init__.py,sha256=_H_VK1ezxvUSLKuIq9WKdNwZuu_iWjNcBJ9IwXYkpDo,64168
+tdfs4ds/__init__.py,sha256=oGMGP13DYQowGbWRIP8xJ2RnhHqnVk5AD-t33mHrx0o,64168
 tdfs4ds/datasets.py,sha256=LE4Gn0muwdyrIrCrbkE92cnafUML63z1lj5bFIIVzmc,3524
 tdfs4ds/feature_engineering.py,sha256=oVnZ2V_XNGE12LKC_fNfkrWSQZLgtYRmaf8Dispi6S4,7081
 tdfs4ds/feature_store.py,sha256=y-oItPZw6nBkBcGAceaATZbkLPTsvpk0OnpzTxYofDs,68576
@@ -25,11 +25,11 @@ tdfs4ds/process_store/process_store_catalog_management.py,sha256=H135RRTYn-pyWIq
 tdfs4ds/utils/__init__.py,sha256=-yTMfDLZbQnIRQ64s_bczzT21tDW2A8FZeq9PX5SgFU,168
 tdfs4ds/utils/filter_management.py,sha256=7D47N_hnTSUVOkaV2XuKrlUFMxzWjDsCBvRYsH4lXdU,11073
 tdfs4ds/utils/info.py,sha256=sShnUxXMlvCtQ6xtShDhqdpTr6sMG0dZQhNBFgUENDY,12058
-tdfs4ds/utils/lineage.py,sha256=XvoiNyrVrsVhuSZTAJrCNjEZAQ4YVsoe61aIl2fBKzk,34757
+tdfs4ds/utils/lineage.py,sha256=gy5M42qy5fvdWmlohAY3WPYoqAyp5VakeEmeT1YjrJQ,37839
 tdfs4ds/utils/query_management.py,sha256=nAcE8QY1GWAKgOtb-ubSfDVcnYbU7Ge8CruVRLoPtmY,6356
 tdfs4ds/utils/time_management.py,sha256=1eqGs7rT3SGag0F30R3PzwiC7Aa7DKia2Ud0aSNKcPg,10593
 tdfs4ds/utils/visualization.py,sha256=5S528KoKzzkrAdCxfy7ecyqKvAXBoibNvHwz_u5ISMs,23167
-tdfs4ds-0.2.4.19.dist-info/METADATA,sha256=5ri5CNAhzx8igDa4PpsfwMQPeJSxhNsaM_c7_B8IYIs,11944
-tdfs4ds-0.2.4.19.dist-info/WHEEL,sha256=R0nc6qTxuoLk7ShA2_Y-UWkN8ZdfDBG2B6Eqpz2WXbs,91
-tdfs4ds-0.2.4.19.dist-info/top_level.txt,sha256=wMyVkMvnBn8RRt1xBveGQxOpWFijPMPkMiE7G2mi8zo,8
-tdfs4ds-0.2.4.19.dist-info/RECORD,,
+tdfs4ds-0.2.4.20.dist-info/METADATA,sha256=niFkaOR8VVaVVYY318GZ9PwX4wjgf53Awzt_kCUWY84,12084
+tdfs4ds-0.2.4.20.dist-info/WHEEL,sha256=R0nc6qTxuoLk7ShA2_Y-UWkN8ZdfDBG2B6Eqpz2WXbs,91
+tdfs4ds-0.2.4.20.dist-info/top_level.txt,sha256=wMyVkMvnBn8RRt1xBveGQxOpWFijPMPkMiE7G2mi8zo,8
+tdfs4ds-0.2.4.20.dist-info/RECORD,,

{tdfs4ds-0.2.4.19.dist-info → tdfs4ds-0.2.4.20.dist-info}/WHEEL RENAMED Viewed

File without changes

{tdfs4ds-0.2.4.19.dist-info → tdfs4ds-0.2.4.20.dist-info}/top_level.txt RENAMED Viewed

File without changes

tdfs4ds 0.2.4.19__py3-none-any.whl → 0.2.4.20__py3-none-any.whl

tdfs4ds 0.2.4.19py3-none-any.whl → 0.2.4.20py3-none-any.whl