update for pg 0.11.0 and a few minor changes #28
Conversation
…_may_ be single - depending on precision - currently taking the safe route and migrate all to double)
all tests passing
… uninitialized class variable errors: https://gist.github.com/4c76c2894b51ac911fba
The value of a boolean column may actually be the string containing the 0 or 1 byte (note: not character) rather than the integer 0 or 1. The row processing should respect that and still output a 't' or 'f' rather than the original value string.
… uninitialized class variable errors: https://gist.github.com/4c76c2894b51ac911fba
Refactor string escaping & add with/without timezone option
Commits to increase test coverage
* refactor config file template options to be more sane * re-align test options and expectations for test_truncate * update config_all_options fixture to match currently available options
* update sql used to check for index presence * index migration to db was missing index name * added note on limitation concerning index names == table_name
Conflicts: lib/mysql2psql/config.rb lib/mysql2psql/converter.rb test/fixtures/config_all_options.yml test/lib/test_helper.rb
* green tests
* some major changes have been accumulated * including breaking change in config options
Conflicts: .gitignore Gemfile Gemfile.lock README.rdoc lib/mysql2psql/mysql_reader.rb lib/mysql2psql/postgres_writer.rb
Conflicts: .gitignore README.rdoc lib/mysql2psql/mysql_reader.rb lib/mysql2psql/postgres_writer.rb
Conflicts: mysql2psql.gemspec
* merges intent of ef9951e by Aaron Peckham with preious technique for duplicate detection for PostgreSQL < 9
|
Hi Max, are you still maintaining this gem? I just spent a bit of time today merging the changes from a bunch of people: James Coleman, Aaron Peckham, James Tippett, Tim Morgan", dakhota, Matthew Soldo. Changes are all on this pull request now (which has also merged you own master, so it should be an easy fast-forward merge). Tests are all green - hopefully that means I got everyone's changes in OK(!) |
|
Hi, can you, please, rebase your updates so, that I merge them? |
|
Hi Max, sure I'll take another look now. Hopefully we can get the number of independent forks back to something a little more reasonable.. |
|
hmm, so I took a look at the latest master and its not a very clean merge. various people have been heading in different directions. a few things that need to be resolved, and they require an opinion: I'm not so sure about the changes made to use config/database.yml by default, and drop the ability of the tool to generate a config template for the user. Fine if you are running the tool from a cloned repo, but not consistent with using the tool as a gem. There's also an unresolved issue of who's managing the gemspec. Current master is along the lines of manually managed, but jeweler is still in the project and Rakefile. Also, version in the gemspec is not consistent with the version that the gem reports. Why have we removed all of the good info in the readme? Max, you have any thoughts on which way to go on these issues? |
|
I think you can do anything you consider right =) |
|
ok;-) I'll get a fully resolved merge first, then look at some changes on top to clean up a bit |
|
Any updates on these changes? |
|
|
Ok, well let's see if I could sort these out... |
Hi Max, I've integrated a few contributions to the project, updated some of the docs and stepped the gem version to 0.1.1. All tests green at my end!