Render PDF pages as images

Author: utkarsh0493

Computations/Code/__init__.py

@@ -24,16 +24,19 @@
 import pandas as pd
 import streamlit as st
 
+import streamlit.components.v1 as components
+from streamlit_tree_select import tree_select
+
 
 CURRENT_FILE = Path(__file__).resolve()
 CODE_DIR = CURRENT_FILE.parents[1]
 sys.path.append(str(CODE_DIR))
 
 from core.indexer import scan_project  # noqa: E402
 from core.validator import validate_project_structure, calculate_structure_score  # noqa: E402
-from core.previewer import get_file_type, read_text_file, read_table_file  # noqa: E402
+from core.previewer import get_file_type, read_text_file, read_table_file, render_notebook_html, render_pdf_pages  # noqa: E402
 from core.exporter import file_index_to_csv, file_index_to_json, file_index_to_html  # noqa: E402
-from core.tree_builder import build_tree, tree_to_markdown  # noqa: E402
+from core.tree_builder import build_tree, tree_to_markdown, tree_to_select_nodes  # noqa: E402
 from core.reporter import generate_markdown_report  # noqa: E402
 from core.output_writer import create_output_directory, save_text_outputs  # noqa: E402
 
@@ -116,6 +119,7 @@
 
     project_tree = build_tree(str(path), include_hidden=include_hidden)
     tree_markdown = tree_to_markdown(project_tree)
+    tree_nodes = tree_to_select_nodes(project_tree)
 
 except Exception as error:
     st.error(f"Could not scan project: {error}")
@@ -152,18 +156,37 @@
     st.write("### Files by extension")
     st.bar_chart(df["extension"].replace("", "[no extension]").value_counts())
 
+
 elif page == "Directory tree":
 
-    st.subheader("Directory tree")
+    st.subheader("Interactive directory tree")
 
     st.write(
         """
-        This view shows the selected project as a nested directory tree.
-        Hidden files and hidden folders are excluded from this view.
+        Select files or folders from the project tree.
+        Hidden files and ignored system folders follow the scan settings from the sidebar.
         """
     )
 
-    st.code(tree_markdown, language="markdown")
+    selected_nodes = tree_select(
+        tree_nodes,
+        checked=[],
+        expanded=[str(path)],
+    )
+
+    st.write("### Selected paths")
+
+    checked_paths = selected_nodes.get("checked", [])
+
+    if checked_paths:
+        for selected_path in checked_paths:
+            st.code(selected_path)
+    else:
+        st.info("No file or folder selected.")
+
+    with st.expander("Show markdown tree"):
+        st.code(tree_markdown, language="markdown")
+
 
 elif page == "Structure validation":
 
@@ -275,7 +298,29 @@
                 )
 
         elif preview_type == "pdf":
-            st.info("PDF preview will be added in a later step.")
+            try:
+                max_pages = st.slider(
+                    "Number of PDF pages to preview",
+                    min_value=1,
+                    max_value=50,
+                    value=5,
+                )
+
+                pdf_pages = render_pdf_pages(selected_path, max_pages=max_pages)
+
+                for page_number, image in pdf_pages:
+                    st.write(f"Page {page_number}")
+                    st.image(image, use_container_width=True)
+
+            except Exception as error:
+                st.error(f"Could not render PDF: {error}")
+
+        elif preview_type == "notebook":
+            try:
+                notebook_html = render_notebook_html(selected_path)
+                components.html(notebook_html, height=900, scrolling=True)
+            except Exception as error:
+                st.error(f"Could not render notebook: {error}")
 
         else:
             st.warning("Preview is not supported for this file type yet.")

Computations/Code/app/app.py

@@ -8,18 +8,20 @@
 - Text/code files.
 - CSV/TSV tables.
 - Images.
-
-Unsupported or future previews
-------------------------------
-- PDF rendering.
-- Binary scientific objects such as H5AD/RDS.
+- PDFs.
+- Jupyter notebooks.
 """
 
 from pathlib import Path
 from typing import Optional
 
+import nbformat
 import pandas as pd
+from nbconvert import HTMLExporter
 
+import fitz
+from PIL import Image
+import io
 
 TEXT_EXTENSIONS = {
     ".txt", ".md", ".py", ".r", ".rmd", ".sh",
@@ -37,10 +39,6 @@
 
 
 def get_file_type(file_path: str) -> str:
-    """
-    Return broad preview type for a file.
-    """
-
     suffix = Path(file_path).suffix.lower()
 
     if suffix in TABLE_EXTENSIONS:
@@ -55,14 +53,13 @@ def get_file_type(file_path: str) -> str:
     if suffix == ".pdf":
         return "pdf"
 
+    if suffix == ".ipynb":
+        return "notebook"
+
     return "unsupported"
 
 
 def read_text_file(file_path: str, max_chars: int = 10000) -> str:
-    """
-    Read a text-like file safely.
-    """
-
     path = Path(file_path)
 
     try:
@@ -77,10 +74,6 @@ def read_text_file(file_path: str, max_chars: int = 10000) -> str:
 
 
 def read_table_file(file_path: str, max_rows: Optional[int] = 500) -> pd.DataFrame:
-    """
-    Read CSV/TSV file for preview.
-    """
-
     path = Path(file_path)
     suffix = path.suffix.lower()
 
@@ -90,4 +83,45 @@ def read_table_file(file_path: str, max_rows: Optional[int] = 500) -> pd.DataFra
     if suffix == ".tsv":
         return pd.read_csv(path, sep="\t", nrows=max_rows)
 
-    raise ValueError(f"Unsupported table format: {suffix}")
+    raise ValueError(f"Unsupported table format: {suffix}")
+
+
+def render_notebook_html(file_path: str) -> str:
+    """
+    Convert a Jupyter notebook to HTML for preview.
+    """
+
+    path = Path(file_path)
+
+    notebook = nbformat.read(path, as_version=4)
+
+    exporter = HTMLExporter()
+    exporter.exclude_input_prompt = True
+    exporter.exclude_output_prompt = True
+
+    body, _ = exporter.from_notebook_node(notebook)
+
+    return body
+
+
+def render_pdf_pages(file_path: str, max_pages: int = 5, zoom: float = 2.0):
+    """
+    Render PDF pages as PIL images for Streamlit preview.
+    """
+
+    path = Path(file_path)
+    document = fitz.open(path)
+
+    images = []
+
+    for page_number in range(min(len(document), max_pages)):
+        page = document.load_page(page_number)
+        matrix = fitz.Matrix(zoom, zoom)
+        pixmap = page.get_pixmap(matrix=matrix)
+
+        image = Image.open(io.BytesIO(pixmap.tobytes("png")))
+        images.append((page_number + 1, image))
+
+    document.close()
+
+    return images

Computations/Code/core/__init__.py

@@ -2,27 +2,22 @@
 Directory Tree Builder
 
 Builds a nested representation of a local project directory and converts it
-to a markdown-style tree view.
+to Streamlit tree-select compatible nodes.
 """
 
 from pathlib import Path
-from typing import Dict, Any
+from typing import Dict, Any, List
 
 from core.indexer import should_ignore
 
 
 def build_tree(project_path: str, include_hidden: bool = False) -> Dict[str, Any]:
-    """
-    Build a nested dictionary representation of a project directory.
-    """
-
     root = Path(project_path).expanduser().resolve()
 
     def walk(path: Path) -> Dict[str, Any]:
         children = []
 
         for item in sorted(path.iterdir(), key=lambda x: (x.is_file(), x.name.lower())):
-
             if should_ignore(item, include_hidden=include_hidden):
                 continue
 
@@ -49,16 +44,32 @@ def walk(path: Path) -> Dict[str, Any]:
 
 
 def tree_to_markdown(tree: Dict[str, Any], indent: int = 0) -> str:
-    """
-    Convert nested tree dictionary to markdown text.
-    """
-
     prefix = "    " * indent
-
     icon = "📁" if tree["type"] == "directory" else "📄"
     line = f"{prefix}- {icon} {tree['name']}\n"
 
     for child in tree.get("children", []):
         line += tree_to_markdown(child, indent + 1)
 
-    return line
+    return line
+
+
+def tree_to_select_nodes(tree: Dict[str, Any]) -> List[Dict[str, Any]]:
+    """
+    Convert tree dictionary to streamlit-tree-select node format.
+    """
+
+    def convert_node(node: Dict[str, Any]) -> Dict[str, Any]:
+        icon = "📁" if node["type"] == "directory" else "📄"
+
+        converted = {
+            "label": f"{icon} {node['name']}",
+            "value": node["path"],
+        }
+
+        if node.get("children"):
+            converted["children"] = [convert_node(child) for child in node["children"]]
+
+        return converted
+
+    return [convert_node(tree)]

Computations/Code/core/previewer.py

@@ -28,11 +28,15 @@ dependencies:
       - annotated-types==0.7.0
       - anyio==4.13.0
       - attrs==26.1.0
+      - beautifulsoup4==4.14.3
+      - bleach==6.3.0
       - blinker==1.9.0
       - cachetools==7.1.1
       - certifi==2026.4.22
       - charset-normalizer==3.4.7
       - click==8.3.3
+      - defusedxml==0.7.1
+      - fastjsonschema==2.21.2
       - gitdb==4.0.12
       - gitpython==3.1.50
       - h11==0.16.0
@@ -43,39 +47,57 @@ dependencies:
       - jinja2==3.1.6
       - jsonschema==4.26.0
       - jsonschema-specifications==2025.9.1
+      - jupyter-client==8.8.0
+      - jupyter-core==5.9.1
+      - jupyterlab-pygments==0.3.0
       - markdown==3.10.2
       - markdown-it-py==4.1.0
       - markupsafe==3.0.3
       - mdurl==0.1.2
+      - mistune==3.2.1
       - narwhals==2.20.0
+      - nbclient==0.10.4
+      - nbconvert==7.17.1
+      - nbformat==5.10.4
       - numpy==2.4.4
       - pandas==3.0.2
+      - pandocfilters==1.5.1
       - pillow==12.2.0
+      - platformdirs==4.9.6
       - pluggy==1.6.0
       - protobuf==7.34.1
       - pyarrow==24.0.0
       - pydantic==2.13.4
       - pydantic-core==2.46.4
       - pydeck==0.9.2
       - pygments==2.20.0
+      - pymupdf==1.27.2.3
       - pytest==9.0.3
       - python-dateutil==2.9.0.post0
       - python-multipart==0.0.27
       - pyyaml==6.0.3
+      - pyzmq==27.1.0
       - referencing==0.37.0
       - requests==2.33.1
       - rich==15.0.0
       - rpds-py==0.30.0
       - six==1.17.0
       - smmap==5.0.3
+      - soupsieve==2.8.3
       - starlette==1.0.0
       - streamlit==1.57.0
+      - streamlit-pdf==2.0.1
+      - streamlit-tree-select==0.0.5
       - tenacity==9.1.4
+      - tinycss2==1.4.0
       - toml==0.10.2
+      - tornado==6.5.5
+      - traitlets==5.15.0
       - typing-extensions==4.15.0
       - typing-inspection==0.4.2
       - urllib3==2.6.3
       - uvicorn==0.46.0
       - watchdog==6.0.0
+      - webencodings==0.5.1
       - websockets==16.0
 prefix: /Users/utkarsh/anaconda3/envs/fssnav

Computations/Code/core/tree_builder.py

@@ -7,4 +7,9 @@ pygments
 rich
 watchdog
 pytest
-pyyaml
+pyyaml
+streamlit-tree-select
+nbconvert
+nbformat
+pymupdf
+pillow