docs: update installation instructions and dependencies

- Add `uv` installation method as recommended for ease of use
- Clarify installation options for various features using `uv` and `pip`
- Update feature-specific installation instructions in FAQ and advanced formats
- Add DuckDB backend documentation to mkdocs navigation
- Include `click` and `lxml` in `pyproject.toml` for interactive and all features

These changes improve clarity and accessibility of installation instructions, making it easier for users to set up SQLstream with desired features.

Author: subhayu99

README.md

@@ -42,12 +42,31 @@ $ sqlstream shell data.csv
 
 ## Installation
 
+### Using [`uv`](https://docs.astral.sh/uv/#installation) (recommended)
+
+**Basic (CSV only)**:
+```bash
+uv tool install sqlstream
+```
+
+**All features**:
+```bash
+uv tool install "sqlstream[all]"
+```
+
+**Multiple Sub Dependencies**:
+```bash
+uv tool install "sqlstream[interactive,pandas,s3,http,html,duckdb]"
+```
+
+### Using `pip`
+
 **Basic (CSV only)**:
 ```bash
 pip install sqlstream
 ```
 
-**All features** (recommended):
+**All features**:
 ```bash
 pip install "sqlstream[all]"
 ```

docs/cli/interactive-mode.md

@@ -30,6 +30,9 @@ The interactive shell requires the `textual` library:
 # Install with CLI support
 pip install "sqlstream[cli]"
 
+# Install full TUI
+pip install "sqlstream[interactive]"
+
 # Or install all features
 pip install "sqlstream[all]"
 ```

docs/faq.md

@@ -212,9 +212,12 @@ pip install "sqlstream[all]"
 **By feature**:
 - `sqlstream[parquet]` - Parquet support (pyarrow)
 - `sqlstream[pandas]` - Pandas backend (pandas)
+- `sqlstream[duckdb]` - DuckDB backend (duckdb)
 - `sqlstream[s3]` - S3 support (s3fs)
-- `sqlstream[cli]` - Interactive shell (textual)
+- `sqlstream[cli]` - Basic CLI shell (click)
+- `sqlstream[interactive]` - Full blown TUI (textual)
 - `sqlstream[http]` - HTTP data sources (httpx)
+- `sqlstream[html]` - HTML parsing (lxml)
 
 ---
 
@@ -371,7 +374,7 @@ SQLStream supports three execution backends:
 
 ### How do I use the DuckDB backend?
 
-Install with `pip install duckdb` or `pip install "sqlstream[all]"`.
+Install with `pip install duckdb` or `pip install "sqlstream[duckdb]"`.
 
 Then use it in your code:
 ```python

docs/features/advanced-formats.md

@@ -324,10 +324,9 @@ reader.set_columns(['col1', 'col2'])
 
 ## Requirements
 
-- **HTML Support**: Requires `pandas` and `lxml` or `html5lib`
+- **HTML Support**: Requires `pandas`, `bs4` and `lxml` or `html5lib`
   ```bash
-  pip install sqlstream[pandas]
-  pip install lxml  # or html5lib
+  pip install sqlstream[html]
   ```
 
 - **Markdown Support**: No additional dependencies (built-in parser)

docs/index.md

@@ -86,22 +86,38 @@ $ sqlstream query "SELECT * FROM 's3://my-bucket/data.parquet' WHERE date > '202
 
 ## Installation
 
+### Using [`uv`](https://docs.astral.sh/uv/#installation) (recommended)
+
 === "Basic (CSV only)"
 
     ```bash
-    pip install sqlstream
+    uv tool install sqlstream
+    ```
+
+=== "With Parquet and Pandas support"
+
+    ```bash
+    uv tool install "sqlstream[parquet,pandas]"
+    ```
+
+=== "All Features"
+
+    ```bash
+    uv tool install "sqlstream[all]"
     ```
 
-=== "With Parquet"
+=== "Multiple Sub Dependencies"
 
     ```bash
-    pip install "sqlstream[parquet]"
+    uv tool install "sqlstream[interactive,pandas,s3,http,html,duckdb]"
     ```
 
-=== "With Pandas (10-100x faster)"
+### Using `pip`
+
+=== "Basic (CSV only)"
 
     ```bash
-    pip install "sqlstream[pandas]"
+    pip install sqlstream
     ```
 
 === "All Features"

mkdocs.yml

@@ -91,6 +91,7 @@ nav:
       - JOIN Operations: features/joins.md
       - Aggregations: features/aggregations.md
       - Pandas Backend: features/pandas-backend.md
+      - DuckDB Backend (Full SQL): features/duckdb-backend.md
       - Inline File Paths: features/inline-paths.md
   - CLI Reference:
       - Overview: cli/overview.md
@@ -110,6 +111,8 @@ nav:
       - Design Overview: architecture/design.md
       - Volcano Model: architecture/volcano-model.md
       - Optimizations: architecture/optimizations.md
+  - FAQ: faq.md
+  - Limitations: limitations.md
   - Contributing: contributing.md
 
 extra:

pyproject.toml

@@ -34,10 +34,11 @@ cli = [
     "textual>=0.40.0",
 ]
 interactive = [
+    "click>=8.0.0",
+    "rich>=13.0.0",
     "textual>=0.40.0",
     "tree-sitter>=0.21.3",
     "tree-sitter-sql>=0.3.5",
-    "rich>=13.0.0",
 ]
 parquet = ["pyarrow>=10.0.0"]
 pandas = [
@@ -57,6 +58,7 @@ duckdb = [
     "duckdb>=1.0.0",
 ]
 all = [
+    "lxml>=4.8.0",
     "pandas>=2.0.0",
     "pyarrow>=14.0.0",
     "httpx>=0.27.0",