Version 6 (modified by xi, 11 years ago) (diff)


PyYAML Legacy Code

The following text was written by Tim Parkin and was hosted by Pollenation Internet. It's grabbed from  the old Feel free to add or modify the content here if you want to add some historic notes or attributions.


This is a home for the code developed by various authors (Steve Howell, Clark C Evans, John Pywtorak. Please correct or add if I'm mistaken. See below for historic notes). The code had no home and seemed to be forking. As we have used, and intend to continue using,  YAML under Python we felt a warm development environment may help. We'll try to unfork the code ( Meld should make this a less onerous task). It has been asked why we need a pure python version. Our intention is to provide both a pure python version with an option of using the same API into  Syck. I won't go into the advantages or disadvantages of either method, but there are individual circumstances where one or the other is essential.


Subversion Repository

We have an anonymous access subversion repository which you can access directly with the following (although trac allows you to view the source also).

svn co


SVN Tarball

Official Version

The last official version

Work that has been carried out since 0.32

Mike Orr had started to work on changes including an excellent USAGE page. I've included the USAGE in the docs folder

Steve Howell wanted to have all data loaded as strings (I think) so this departed from LegacyTestingSuite spec

John Pywtorak added some patches (the effects of which can be seen in LegacyPywtorakPatches?) for empty scalars and stripping trailing line modifiers for folded scalars (this tar is just the .patch files for load and stream)

This version was the one linked from (This is actually 0.30 as noted on the page linked to).

Older Archive

Old versions kept here for completeness

Commit/Timeline? Addition Notes

For the moment, additional/temporary notes about commits that can be added in LegacyChangesetNotes?.



  • March 2002 - Steve Howell starts work on parser only
  • April 2002 - no progress
  • May 2002 - no progress
  • June 2002 - Steve Howell begins work on the project again, gets a little help from Ryan King, and gets a basic emitter working, along with a few more cases for the parser
    • SlideShowell - first PyYAML application
  • July 2002 - Clark Evans and Steve Howell add major features:
    • distutils setup
    • catch up to flurry of spec changes from late Spring
    • better parser compliance
    • alias support for the emitter
    • YAML is fully bootstrapped – test cases are YAML-driven (see LegacyTestingSuite)
    • Clark Evans begins support for YPATH and queries (with some help from Steve Howell)
  • August 2002 - Dave Kuhlman joins the project
    • better support for serializing objects
    • XmlYaml (XML to/from YAML mappings)
    • dumpToFile method
    • started to spike some ideas for pull parser and schemas
  • September 2002 - way too much time on the mailing list
    • starting to clean up flow collections loading
    • reoranizing code a bit to prepare for more flexible implicit handling
    • giving talk to DC Python users group, using SlideShowell (9/21)
    • bug fixing