Spaces:

SamitF
/

TextExp

Sleeping

App Files Files Community

SamitF commited on 18 days ago

Commit

e47737d

verified ·

1 Parent(s): 9e8dd52

Update app.py

Browse files

Files changed (1) hide show

app.py +223 -3

app.py CHANGED Viewed

@@ -812,6 +812,226 @@ SAMPLES = {
     "Simple English": "Hello World! Feel free to paste any text here to test the NLP Text Explorer system. It works immediately without external libraries.",
     "Multilingual & Emoji": "Multilingual test: Bonjour, こんにちは, 안녕하세요! Here is an emoji breakdown to inspect UTF-8 bytes: 😊 🚀 🍕. What is their representation in binary memory space?",
     "HTML document sample": '<!DOCTYPE html>\n<html>\n<head>\n  <title>Sample Sandbox</title>\n</head>\n<body>\n  <div class="content">\n    <h1>Welcome to Text Explorer!</h1>\n    <p>This is standard markup designed to verify our pure HTML-Parser extraction utilities.</p>\n    <a href="https://example.com">Visit our project</a>\n  </div>\n</body>\n</html>',
-    "Tabular CSV format": "username,email,role,joined_date\nsmith_john,john.smith@gmail.com,Administrator,2024-03-24\nelizabeth_k,k.elizabeth@yahoo.com,Contributor,2025-05-15\ntech_support,support@domain.org,User,2026-01-10",
-    "Markdown document": "# Markdown Document Guide\n\nWelcome to this structural text parsing review. Here are some key attributes:\n\n- Support lists\n- Supports link items: [Hugging Face Space](https://huggingface.co/spaces)\n\n## Sub-headers\n",
-}

     "Simple English": "Hello World! Feel free to paste any text here to test the NLP Text Explorer system. It works immediately without external libraries.",
     "Multilingual & Emoji": "Multilingual test: Bonjour, こんにちは, 안녕하세요! Here is an emoji breakdown to inspect UTF-8 bytes: 😊 🚀 🍕. What is their representation in binary memory space?",
     "HTML document sample": '<!DOCTYPE html>\n<html>\n<head>\n  <title>Sample Sandbox</title>\n</head>\n<body>\n  <div class="content">\n    <h1>Welcome to Text Explorer!</h1>\n    <p>This is standard markup designed to verify our pure HTML-Parser extraction utilities.</p>\n    <a href="https://example.com">Visit our project</a>\n  </div>\n</body>\n</html>',
+    "Tabular CSV format": "username,email,role,joined_date\nsmith_john,john.smith@gmail.com,Administrator,2024-03-24\elizabeth_k,k.elizabeth@yahoo.com,Contributor,2025-05-15\ntech_support,support@domain.org,User,2026-01-10",
+    "Markdown document": (
+        "# Markdown Document Guide\n\n"
+        "Welcome to this structural text parsing review. Here are some key attributes:\n\n"
+        "- Support lists\n- Supports link items: [Hugging Face Space](https://huggingface.co/spaces)\n\n"
+        "## Sub-headers\n"
+        "```python\n"
+        "def test():\n"
+        "    return 'Hello, Markdown analyzer!'\n"
+        "```"
+    )
+}
+# =====================================================================
+# GRADIO BLOCKS APPLICATION VIEW
+# =====================================================================
+with gr.Blocks(css=CSS, title="Text Explorer - Educational Pre-NLP Inspector") as demo:
+    # Hero Title Banner
+    gr.HTML("""
+    <div class="header-hero">
+        <h1>🔍 Text Explorer</h1>
+        <p>Analyze and demystify raw character representation, encodings, byte streams, and structural text layouts before high-level NLP tokenization.</p>
+    </div>
+    """)
+    with gr.Row():
+        # Input column
+        with gr.Column(scale=4):
+            gr.Markdown("### 📥 Document Source Input")
+            input_text = gr.Textbox(
+                label="Source Text Area",
+                placeholder="Paste your paragraphs, CSV structure, HTML content, or multilingual strings here...",
+                lines=10,
+                max_lines=25,
+                elem_id="input-textbox"
+            )
+            with gr.Row():
+                file_upload = gr.File(
+                    label="Or Upload Text File (.txt, .csv, .json, .html, .md)",
+                    file_types=[".txt", ".csv", ".json", ".html", ".md"],
+                    type="filepath",
+                )
+            gr.Markdown("💡 **Educational Presets**")
+            with gr.Row():
+                # Loop samples to create buttons
+                for name, content in SAMPLES.items():
+                    gr.Button(name, size="sm").click(
+                        fn=lambda val=content: val,
+                        outputs=input_text
+                    )
+        # Output / Analysis display tab matrix
+        with gr.Column(scale=6):
+            gr.Markdown("### ⚡ Real-Time Analytical Breakdown")
+            with gr.Tabs():
+                # Overview
+                with gr.TabItem("📊 Overview"):
+                    with gr.Accordion("📚 Educational Spotlight: The Lexical Units", open=True):
+                        gr.HTML("""
+                        <div class="educational-card">
+                            <h4>How Computers Partition Text</h4>
+                            <p>Before an NLP model can comprehend context, the document must be mathematically measured:</p>
+                            <ul style="list-style-type: disc; margin-left: 1.25rem; margin-top: 0.5rem;">
+                                <li><b>Bytes</b>: The actual digital footprint in physical memory (UTF-8 allocates 1 to 4 bytes per character).</li>
+                                <li><b>Words & Sentences</b>: Fundamental logic nodes. Without standard tools like NLTK/spaCy, computers extract tokens using punctuation borders and boundary regex.</li>
+                            </ul>
+                        </div>
+                        """)
+                    overview_html = gr.HTML("<div class='text-center text-slate-400 py-6'>Analyze text to generate document insights.</div>")
+                    overview_chart = gr.HTML()
+                # Character Encodings
+                with gr.TabItem("🔢 Encodings Table"):
+                    with gr.Accordion("🏫 Educational Spotlight: Encodings Simplified", open=False):
+                        gr.HTML("""
+                        <div class="educational-card">
+                            <h4>Encoding Architecture Explained</h4>
+                            <p>Every character is assigned a unique tracking coordinate:</p>
+                            <ul style="list-style-type: disc; margin-left: 1.25rem; margin-top: 0.5rem;">
+                                <li><b>ASCII</b>: Traditional 7-bit system spanning decimal positions 0 through 127. Standard English keys only.</li>
+                                <li><b>Unicode</b>: A universal catalog assigning identical codepoint indices (e.g., U+1F60A) for all global languages and emojis.</li>
+                                <li><b>UTF-8</b>: An optimized encoding translating those codepoints into binary arrays, using smaller slices (1 byte) for English and wider slices for eastern languages or emojis.</li>
+                            </ul>
+                        </div>
+                        """)
+                    encoding_table = gr.DataFrame(
+                        headers=["Char", "ASCII Representation", "Unicode Point", "Decimal", "Hexadecimal", "Binary Value", "UTF-8 Length", "UTF-8 Bytes"],
+                        datatype=["str", "str", "str", "str", "str", "str", "str", "str"],
+                        wrap=True
+                    )
+                # Unicode Explorer
+                with gr.TabItem("🌌 Unicode Explorer"):
+                    with gr.Accordion("🎓 Educational Spotlight: Bit Patterns of UTF-8", open=False):
+                        gr.HTML("""
+                        <div class="educational-card">
+                            <h4>Deciphering the UTF-8 Multi-byte Rules</h4>
+                            <p>UTF-8 dynamically sizes bytes based on prefix triggers. Inspect the binary tags below:</p>
+                            <ul style="list-style-type: disc; margin-left: 1.25rem; margin-top: 0.5rem;">
+                                <li><b>0xxxxxxx</b> (1 Byte): Matches normal English ASCII indices seamlessly.</li>
+                                <li><b>110xxxxx 10xxxxxx</b> (2 Bytes): Directs unicode sequences (e.g. Greek / Cyrillic).</li>
+                                <li><b>1110xxxx 10xxxxxx 10xxxxxx</b> (3 Bytes): Covers general Asian characters (CJK blocks).</li>
+                                <li><b>11110xxx 10xxxxxx 10xxxxxx 10xxxxxx</b> (4 Bytes): Represents deep emoticons and rare historical glyphs.</li>
+                            </ul>
+                        </div>
+                        """)
+                    unicode_explorer_html = gr.HTML("<div class='text-center text-slate-400 py-6'>Displaying individual unicode bit structures...</div>")
+                # Frequencies
+                with gr.TabItem("📈 Frequencies"):
+                    with gr.Row():
+                        with gr.Column():
+                            gr.Markdown("#### Character Frequency Analysis")
+                            char_freq_table = gr.DataFrame(
+                                headers=["Character Key", "Occurrence", "Ratio"],
+                                datatype=["str", "number", "str"]
+                            )
+                        with gr.Column():
+                            gr.Markdown("#### Word Frequency Analysis (Vocabulary)")
+                            word_freq_table = gr.DataFrame(
+                                headers=["Lowercased Token", "Occurrence"],
+                                datatype=["str", "number"]
+                            )
+                # Advanced Statistics
+                with gr.TabItem("🧮 Detailed Statistics"):
+                    stats_table = gr.DataFrame(
+                        headers=["Property Metric", "Value", "Educational Context"],
+                        datatype=["str", "str", "str"],
+                        wrap=True
+                    )
+                # Regex Explorer
+                with gr.TabItem("🔎 Regex Explorer"):
+                    with gr.Accordion("🧠 Educational Spotlight: Structured Pattern Matching", open=False):
+                        gr.HTML("""
+                        <div class="educational-card">
+                            <h4>Regex: The Structural Extraction Language</h4>
+                            <p>Regular expressions look for matching text layouts. They are extremely effective before models compile lexical vectors:</p>
+                            <ul style="list-style-type: disc; margin-left: 1.25rem; margin-top: 0.5rem;">
+                                <li><b>Emails</b>: Matches user accounts bounded by domains (<code>username@host.ext</code>).</li>
+                                <li><b>IP Addresses</b>: Recognizes IPv4 address streams (digits separated by decimal boundaries).</li>
+                            </ul>
+                        </div>
+                        """)
+                    regex_explorer_html = gr.HTML("<div class='text-center text-slate-400 py-6'>Matches category analysis lists...</div>")
+                # Format Detectors
+                with gr.TabItem("🧱 Layout Formats"):
+                    with gr.Accordion("📦 Educational Spotlight: Layout Analysis Systems", open=False):
+                        gr.HTML("""
+                        <div class="educational-card">
+                            <h4>Markup and Matrix Structure Detection Rules</h4>
+                            <p>Computers parse document systems using distinctive delimiters and structural tags:</p>
+                            <ul style="list-style-type: disc; margin-left: 1.25rem; margin-top: 0.5rem;">
+                                <li><b>HTML/XML</b>: Uses bracket boundaries (<code>&lt;tag&gt;</code>) to represent nested elements.</li>
+                                <li><b>CSV Grid Structure</b>: Uses consistent grid markers (such as commas) containing identical segments per line row.</li>
+                            </ul>
+                        </div>
+                        """)
+                    format_overview_html = gr.HTML()
+                    with gr.Accordion("CSV Preview Engine (Non-Library Machine)", open=True):
+                        csv_preview_html = gr.HTML("<p class='text-xs text-slate-400 italic'>No CSV parsed grid display loaded yet.</p>")
+                    with gr.Accordion("HTML Tree Parser Detail", open=True):
+                        html_preview_html = gr.HTML("<p class='text-xs text-slate-400 italic'>No HTML tag analysis details loaded yet.</p>")
+                # Download Export
+                with gr.TabItem("💾 Export Report"):
+                    gr.Markdown("### Create Inspection Summary File")
+                    gr.Markdown("Download a complete, beautifully structured Markdown text report mapping the calculations of the explorer for external study.")
+                    report_btn = gr.Button("🔨 Compile Markdown Analysis Report", variant="primary")
+                    report_file = gr.File(label="Downloadable Analysis Report", type="filepath")
+    # =====================================================================
+    # CONTROLLER WIREFRAMES & ACTIONS
+    # =====================================================================
+    # Upload handler updating input text
+    file_upload.change(
+        fn=handle_file_upload,
+        inputs=file_upload,
+        outputs=input_text
+    )
+    # Core reactive interface processing all panels together
+    ui_outputs = [
+        overview_html,
+        encoding_table,
+        unicode_explorer_html,
+        char_freq_table,
+        word_freq_table,
+        stats_table,
+        regex_explorer_html,
+        format_overview_html,
+        csv_preview_html,
+        html_preview_html,
+        overview_chart
+    ]
+    # Input changes trigger analytical updates instantly
+    input_text.change(
+        fn=analyze_all_inputs,
+        inputs=input_text,
+        outputs=ui_outputs
+    )
+    # Export report trigger
+    report_btn.click(
+        fn=generate_report,
+        inputs=input_text,
+        outputs=report_file
+    )
+# Run the Gradio Space application
+if __name__ == "__main__":
+    demo.launch()