Skip to main content
This page documents utility functions exported from syft_client that help with common tasks like path resolution, dataset file access, and generating bug reports.

​
resolve_path()

Converts syft:// URLs to absolute filesystem paths. 
from syft_client import resolve_path

​
Parameters

​
path
str | Path
required
Path to resolve (must start with syft://)
​
syftbox_folder
str | Path
default:"None"
SyftBox folder location. If not provided, uses SYFTBOX_FOLDER environment variable.

​
Returns

​
result
Path
Resolved pathlib.Path object pointing to the actual filesystem location

​
Raises

  • ValueError: If syftbox_folder not provided and SYFTBOX_FOLDER env var not set
  • ValueError: If path doesn’t start with syft://

​
Usage

import syft_client as sc
from pathlib import Path

# Using environment variable
import os
os.environ['SYFTBOX_FOLDER'] = '/home/user/SyftBox'
resolved = sc.resolve_path("syft://datasites/alice/data")
# Returns: PosixPath('/home/user/SyftBox/datasites/alice/data')

# Passing folder directly
resolved = sc.resolve_path(
    "syft://apps/myapp",
    syftbox_folder="/home/user/SyftBox"
)
# Returns: PosixPath('/home/user/SyftBox/apps/myapp')

# Works with Path objects too
from pathlib import Path
path = Path("syft://datasites/bob/results")
resolved = sc.resolve_path(path)
This function is useful when working with syft:// URLs in job code or when you need to access files programmatically.

​
resolve_dataset_file_path()

Resolves a dataset name to its file path, automatically choosing between mock and private data based on context. 
from syft_client import resolve_dataset_file_path

​
Parameters

​
dataset_name
str
required
Name of the dataset to resolve
​
syftbox_folder
str | Path
default:"None"
SyftBox folder location. If not provided, inferred from client or SYFTBOX_FOLDER env var.
​
owner_email
str
default:"None"
Email of the dataset owner. If not provided, inferred from client or SYFTBOX_EMAIL env var.
​
client
SyftboxManager
default:"None"
SyftboxManager instance to use for resolution. Provides syftbox_folder and can resolve owner_email.

​
Returns

​
result
Path
Path to the dataset files directory (mock or private based on context)

​
Behavior

The function automatically determines whether to return mock or private data:
  • Inside a job (SYFT_IS_IN_JOB=true): Returns private data path
  • Outside a job: Returns mock data path

​
Usage

import syft_client as sc

# In job code - automatically uses private data
data_path = sc.resolve_dataset_file_path("customer_data")
# Returns: Path to private data

# Outside job - uses mock data
data_path = sc.resolve_dataset_file_path("customer_data")
# Returns: Path to mock data

# With explicit owner
data_path = sc.resolve_dataset_file_path(
    "medical_records",
    owner_email="[email protected]"
)

# With client instance
client = sc.login_ds(email="[email protected]")
data_path = sc.resolve_dataset_file_path(
    "patient_data",
    client=client
)

​
Example in Job Code

# In a job submitted to a data owner
import syft_client as sc
import pandas as pd

# This automatically resolves to private data when running in a job
data_path = sc.resolve_dataset_file_path("sales_data")
df = pd.read_csv(data_path / "sales.csv")

print(f"Loaded {len(df)} rows from private data")
This is the recommended way to access datasets in job code. It abstracts away the distinction between mock and private data.

​
bug_report()

Generates a comprehensive environment snapshot for bug reports and debugging. 
from syft_client import bug_report

​
Parameters

​
as_dict
bool
default:"False"
If True, return the report as a dictionary instead of printing it
​
redact_paths
bool
default:"False"
If True, redact user-specific paths in the output for privacy

​
Returns

​
result
dict | None
If as_dict=True, returns a dictionary containing all collected info. Otherwise prints the report and returns None.

​
Collected Information

The bug report includes:
  • Syft Client version and package info
  • Runtime environment (Jupyter, Colab, standard Python)
  • OS information (platform, version, kernel)
  • Hardware details (CPU, RAM, GPU if available)
  • Python environment (version, virtual env type)
  • Package managers (pip, uv, conda versions)
  • Installed packages (syft packages and dependencies)
  • Environment interpretation (detected runtime type)
The output is safe to share publicly - no tokens, secrets, or credentials are included.

​
Usage

import syft_client as sc

# Print full report to console
sc.bug_report()

# Get report as dictionary for programmatic access
report = sc.bug_report(as_dict=True)
print(f"Python version: {report['python_env']['python_version']}")
print(f"Syft Client version: {report['syft_client']['syft_client_version']}")
print(f"Runtime: {report['interpretation']['detected_environment']}")

# Redact paths for privacy when sharing
sc.bug_report(redact_paths=True)

​
Example Output

Collecting environment information... (this may take a few seconds)

=== Syft Client Information ===
syft_client version: 0.1.101
Package location: /home/user/.venv/lib/python3.11/site-packages

=== Runtime Environment ===
Detected environment: Jupyter Notebook
Jupyter detected: Yes
Google Colab detected: No

=== Operating System ===
Platform: Linux
OS Version: Ubuntu 22.04.3 LTS
Kernel: 5.15.0-91-generic

=== Python Environment ===
Python version: 3.11.7
Virtual environment: venv
...

​
When to Use

Use bug_report() when:
  • Reporting issues on GitHub
  • Seeking help in discussions
  • Debugging environment-specific problems
  • Verifying your setup after installation
  • Comparing environments across systems

​
See Also

Build docs developers (and LLMs) love

Get started for freeTalk to us