wiki:BittenRecipesExtractorScript

Version 8 (modified by anonymous, 12 years ago) (diff)

--

Bitten Recipe Extractor

Description

Two small scripts who's purpose is to retrieve a Trac's Bitten rule sets. Thus allowing you to make off-line changes quickly with your favorite text editor.
Once satisfied with your changes, a second script uploads said changes.

Bugs/Feature Requests

Existing bugs and feature requests for BittenRecipesExtractorScript are here.

If you have any issues, create a new ticket.

Known Issues

Pattern matching may include swap and backup specific files from editing your recipes. The only issue this would cause, may be a file with no changes from the original, or other non-desirable changes. Keep this in mind if your finding recipes not updating with correct data. Search for hidden files in your recipe directory, and delete them. I have attempted to solve this, but unfortunately I have found that Trac allows periods and dashes in the naming convention of bitten rules. So this rules out a lot of simple pattern matching I can perform. eg: if str.isalnum(file_name)

Super User access is needed when reading and writing to the Trac database directly.
Also, on some of my clusters that I work with, multiple version of python can exist. Meaning, if you receive errors like:

ImportError: No module named trac.env

You need to adjust your PATH environment to include the path to which ever python version was used to install and run Trac:

export PATH=/usr/local/python2.7/bin:$PATH

Note: This, being the one reason I do not include the common #!/usr/bin/python interrupter header in my scripts.

Download

Download the zipped source from [download:bittenrecipesextractorscript here].

Source

You can check out BittenRecipesExtractorScript from here using Subversion, or browse the source with Trac.

Example

I may combined these two scripts into one someday, but for now the usage follows:

Reading and Writing Recipes

Each Bitten Rule will be a separate <recipe_name>.recipe file.
Each python script uses the same single argument. The only argument that is needed, is the path to your Trac environment. Which is exactly what is necessary when ever you run:

trac-admin /some/trac/environment

To be absolutely clear on this, and to add additional text spam to this site:

sudo python recipe_get.py /some/trac/environment

and:

sudo python recipe_put.py /some/trac/environment

Accordingly.

When you perform a recipe_get.py, a directory will be created with a name equal to your Trac directory with '_recipes' appended to it (trac-directory-name_recipes). The recipes will be saved (or updated for recipes currently residing) in this directory. You can leave these files there and continue to run both scripts for eternity. No need to clean them up. And quite frankly, leaving them alone for backup reasons, is a darn good idea.

Editing the Recipe Files

The file syntax is fairly straight forward. The values are delimited by two carrot characters. Please, if there is a quicker better way, do submit a ticket.

From the first value to last:

  • The unique recipe name (NEVER CHANGE THIS)
  • The path in your repository
  • Is this recipe active or not? 1 being active, 0 being inactive
  • The bitten recipe itself
  • The minimum revision (if blank set to None)
  • The maximum revision (if blank set to None)
  • The recipe name as seen on the Trac Bitten Builds page
  • The recipe description

I have seen no harm in changing all but the first value. In fact, I have altered the other values with incorrect data, and was easily able to recover from it, by making my corrections, and running the recipe_put.py script again.

Recent Changes

11018 by mjasonmiller on 2011-12-05 16:08:09
I give up. Its too early to be coding. But this should fix everything.
11017 by mjasonmiller on 2011-12-05 16:01:49
Sigh. Ignore my last comment...
11016 by mjasonmiller on 2011-12-05 16:00:03
Woops. Fixing simple list issue. I swear to test my code before commiting...
(more)

Author/Contributors

Author: mjasonmiller
Maintainer: mjasonmiller
Contributors: