Skip to content

Latest commit

 

History

History
116 lines (78 loc) · 3.75 KB

README.rst

File metadata and controls

116 lines (78 loc) · 3.75 KB
Raw

yamlloader

This module provides loaders and dumpers for PyYAML. Currently, an OrderedDict loader/dumper is implemented, allowing to keep items order when loading resp. dumping a file from/to an OrderedDict (Python 3.7+: Also regular dicts are supported and are the default items to be loaded to. As of Python 3.7 preservation of insertion order is a language feature of regular dicts.)

This project was originally mirrored from yamlordereddict Many thanks to the original author François Ménabé! The library contains several improvements including automated testing and the much faster C-versions of the Loaders/Dumpers.

API Documentation

Install

There is a pip and a conda version available

$ pip install yamlloader

or

$ conda install yamlloader -c conda-forge

But does [your special case here] also work?

Tests are run continuously using randomly generated yaml files. Also, there are no fails to be expected.

Still, if you are concerned that your special case may breaks in the future, please add your own tests as test_ext_anyname.py under tests/ or let us know about your needs. This guarantees that no code will be added that breaks your case.

C vs non-C version

A significant speedup can be reached by replacing the Loader* and Dumper* classes by CLoader* and CDumper*. The package hereby relies on the implementations from PyYAML. If they have not been compiled, yamlloader automatically falls back to the non-C versions.

Therefore using the C-version is safe: if it is not available, the pure Python version is automatically used.

Usage examples

Loader usage

import yaml
import yamlloader

with open('myfile.yml') as yaml_file:
    data = yaml.load(yaml_file,
                     Loader=yamlloader.ordereddict.CLoader)
                     # CLoader is faster than Loader

Note: For using the safe loader (which takes standard YAML tags and does not construct arbitrary Python objects), replace yamlloader.ordereddict.CLoader by yamlloader.ordereddict.CSafeLoader.

Dumper usage

import yaml
import yamlloader
from collections import OrderedDict

data = OrderedDict([('key1', 'val1'),
                    ('key2', OrderedDict([('key21', 'val21'),
                                          ('key22', 'val22')]))])

with open('myfile.yaml', 'w') as yaml_file:
    yaml.dump(data, yaml_file,
              Dumper=yamlloader.ordereddict.CDumper)

Note: For using the safe dumper (which produce standard YAML tags and does not represent arbitrary Python objects), replace yamlloader.ordereddict.CDumper by yamlloader.ordereddict.CSafeDumper.

FAQ

C version not working

If the C version is not working (it falls back by default to a non-C version), check if yaml.cyaml exists. If not, the cyaml module was not compiled during the installation of yaml (pyyaml). Make sure that cython is installed (pip install Cython) and the yaml.h file is there (apt: libyaml-dev).