Permalink
Commits on Dec 18, 2009
  1. Timestamped for 0.23 release.

    theory committed Dec 18, 2009
  2. Not includeing pg_tapgen in the release.

    theory committed Dec 18, 2009
    Not sure what to do with this script, but will likely move it to its own project.
  3. 8.0 fixes passing on 8.1 now.

    theory committed Dec 18, 2009
  4. All tests pass on 8.0.

    theory committed Dec 18, 2009
  5. Update testsed versions.

    theory committed Dec 18, 2009
  6. All tests pass on 8.1.

    theory committed Dec 18, 2009
  7. All tests oow pass on 8.2.

    theory committed Dec 18, 2009
  8. Build the .so only when < 8.4.

    theory committed Dec 18, 2009
  9. Add display_oper()`.

    theory committed Dec 18, 2009
Commits on Dec 17, 2009
  1. Finally all tests pass on 8.3.

    theory committed Dec 17, 2009
    Have to cast from records to text in `row_eq()` in 8.3 and earlier.
Commits on Dec 16, 2009
  1. Update patches for 8.3.

    theory committed Dec 16, 2009
  2. List 8.5 as supported.

    theory committed Dec 16, 2009
  3. Fixes to run on 8.5.

    theory committed Dec 16, 2009
Commits on Dec 11, 2009
  1. Added a note about being about to use descriptions without error mess…

    KevinField committed with theory Dec 10, 2009
    …ages.
Commits on Dec 9, 2009
  1. Add `columns_are()`.

    theory committed Dec 9, 2009
  2. Exclude `information_schema` as well as `pg_catalog`.

    theory committed Dec 9, 2009
    This is for all the `*_are()` functions.
  3. * Update `has_operator()` default descriptions.

    theory committed Dec 9, 2009
    Changed the default descriptions output by `has_operator()`, `has_leftop()`, and `has_rightop()` to format the operator names more like the display of `regoperator` types.
  4. Add `operators_are()`.

    theory committed Dec 9, 2009
    Also changed `casts_are()` to accept an array of strings rather than a
    multidimensional array of types. It's just easier, and close enough to the
    syntax of `CREATE CAST` to make sense. Did this after I decided that a similar
    approach to `operators_are()` made much more sense, given that `::regoperator`
    nicely formats operators.
    
    In the process, added `_areni()`, which is the same as `_are()` except that it
    doesn't `quote_ident()` extra or missing things. Used for the string syntax of
    `casts_are()` and `operators_are()`.
Commits on Dec 8, 2009
  1. Add `casts_are()`.

    theory committed Dec 8, 2009
Commits on Dec 7, 2009
  1. Add `triggers_are()`.

    theory committed Dec 7, 2009
Commits on Dec 4, 2009
  1. Add `row_eq()`.

    theory committed Dec 4, 2009
  2. * Add `display_type()`.

    theory committed Dec 4, 2009
    Replaced all internal use of `format_type()` with it. This is so that schemas are always displayed properly.
  3. Add type schema variants of `col_is_type()`

    theory committed Dec 4, 2009
    Also:
    
    * Add `col_type_is( schema, table, column, type )`
    * Fix `domain_type_is()` to compare domain names case-insensitvely and to restrict the search to domains.
Commits on Dec 2, 2009
  1. Add `domain_type_is()` and `domain_type_isnt()`.

    theory committed Dec 2, 2009
    Also fixed up the diagnostic output from `isnt()` while I was at it. It was just wrong.
Commits on Nov 21, 2009
  1. Note bugfix from dukeleto.

    theory committed Nov 21, 2009
Commits on Nov 20, 2009
  1. Add `types_are()`.

    theory committed Nov 20, 2009