SQL, Python, and Jupyter Notebook Search Tool

Owner: Brad Youles
Status: Active
Last reviewed: 2026-07-15

Purpose

This desktop application searches SQL files, Python files, and Jupyter notebooks for one or more case-insensitive terms. It recursively scans a selected folder, displays matching lines in the application, and exports all results to a CSV file.

Supported File Types

The application searches:

• SQL files (.sql)
• Python files (.py)
• Jupyter notebooks (.ipynb)

For Jupyter notebooks, the application searches both code cells and Markdown cells.

Requirements

• Python 3.8 or newer
• Tkinter

Tkinter is included with most standard Windows Python installations. No third-party Python packages are required.

Run The Application

Open PowerShell or Command Prompt in the folder containing the script.

Run:

python search-app.py

If the python command is not available, run:

py search-app.py

Replace search-app.py with the actual script name if it differs.

Use The Application

1. Select Choose Folder.
2. Select the top-level folder to search.
3. Enter one or more comma-separated search terms.
4. Select Search or press Enter while the search-term field is active.
5. Review the matches displayed in the application.
6. Open search_results.csv in the selected folder for the complete exported results.

Example search terms:

employee_id, release_version, lab_id

Searches are case-insensitive. For example, searching for employee_id also finds EMPLOYEE_ID, Employee_ID, and employee_id.

Search Behavior

The application recursively scans all supported files under the selected folder.

Each search term is evaluated separately. If the same term occurs more than once on a line, each occurrence is recorded as a separate match.

For SQL and Python files, the location is reported as a source-file line number:

line 42

For Jupyter notebooks, the location includes the cell number, cell type, and line within the cell:

cell 7 (code), line 4

The search includes matches in:

• Source code
• Comments
• String values
• SQL statements
• Jupyter code cells
• Jupyter Markdown cells

The application performs literal text matching. It does not currently support regular expressions or syntax-aware searching.

Output

The application writes the results to:

search_results.csv

The file is saved in the top-level folder selected for the search. A new search overwrites the existing file.

The output includes:

Column	Description
file_path	Full path to the matching file
file_kind	File type: sql, python, or ipynb
needle	Search term that matched
location	Source line or notebook cell location
start_index	Zero-based character position where the match begins
line_preview	Preview of the matching line

The CSV uses UTF-8 encoding with a byte-order mark so it opens correctly in Microsoft Excel.

Ignored Directories

The application does not search these directories:

.git
.venv
venv
__pycache__
.ipynb_checkpoints

The ignored directories are defined in the script:

IGNORE_DIRS = {
    ".git",
    ".venv",
    "venv",
    "__pycache__",
    ".ipynb_checkpoints",
}

Add additional directory names to this set when needed.

Searchable Extensions

The supported extensions are defined in the script:

SEARCHABLE_SUFFIXES = {".sql", ".py", ".ipynb"}

Additional text-based file types may be added to this set, but the search-routing logic must also send those files to the appropriate search function.

Search-Term Rules

Enter multiple terms as a comma-separated list:

release_version, employee number, final_cluster

The application:

• Removes leading and trailing spaces
• Ignores blank entries
• Removes duplicate terms case-insensitively
• Treats commas as separators

The current version cannot search for a literal phrase that contains a comma.

Error Handling

The application displays an error message when:

• No folder is selected
• The selected path is invalid
• No search terms are entered
• The results CSV cannot be written

If an individual file cannot be read or a notebook cannot be parsed, the application records the error in the results and continues scanning the remaining files.