Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
Commits on Mar 20, 2015
  1. Update multimarkdown.

    authored
Commits on Feb 2, 2015
  1. Fix label ordering in `enum_has_labels()`.

    authored
    On 9.0 and higher, where enums have an explcit order required by
    `ALTER TYPE ADD VALUE`. Resolves #60.
Commits on Feb 7, 2014
Commits on Jan 4, 2014
Commits on Sep 5, 2013
  1. Simplify DATA wildcard.

    authored
    Closes #48.
Commits on Apr 26, 2013
  1. @eneault
Commits on Apr 2, 2013
  1. Whitespace tweaks.

    authored
  2. Easier way of saying the same thing.

    Aaron W. Swenson authored
Commits on Mar 30, 2013
  1. Allow make to be called as 'PG_CONFIG="/path/to/pg_config" make'

    Aaron W. Swenson authored
    without having to export an environment variable.
Commits on Jan 28, 2013
  1. Remove CREATE EXTENSION support for -core and -schema.

    authored
    They were never really supported, anyway. The files are still created for folks to copy around.
Commits on Jan 15, 2013
  1. Get tests passing on 8.4.

    authored
Commits on Sep 11, 2012
  1. 8.1 or later, not 8.0.

    authored
  2. Update 8.3 patch.

    authored
    Turns out it applies cleanly for the core and schema files, too, so remove the stuff to work around those issues.
Commits on Sep 10, 2012
  1. Silence -core and -schema patch application.

    authored
    And delete the .orig files. No way to stop these, since the 8.3 patches have offsets that vary for these derived scripts.
Commits on Aug 31, 2012
  1. Set $(PERL) after loading PGXS.

    authored
    To ensure that it is the same value throughout. Resolves #26.
Commits on Dec 16, 2011
  1. Use `$(SHELL)` to execute `getos.sh`.

    authored
    Closes #16.
Commits on Nov 11, 2011
  1. Drop support for PostgreSQL 8.0.

    authored
    The version of pg_regress in 8.0 doesn't support the --inputdir option, so there is no longer any way to adequately test it. Besides, 8.0 has been EOLed for a long time now.
Commits on Nov 10, 2011
  1. Update build for 8.2.

    authored
  2. Fix version checks.

    authored
Commits on Aug 25, 2011
  1. Start simplifiying document headers.

    authored
    There are way too many `h3` headers. Some functions have many different signaures. Instead, have one header for each function name, and then list the parameters as a definition list after the synopsis. This keeps the signatures explicit while minimizing the number (and length!) of the `h3` headers.
    
    To make this work, the documentation must be Multimarkdown, as Markdown does not support definition lists.
Commits on Aug 24, 2011
  1. Get working on 9.0 again.

    authored
    Also, don't remove `test/setup.sql` during `make clean`, but do remove `sql/pgtap.sql` and `sql/uninstall_pgtap.sql`.
  2. Remove TAPSCHEMA option.

    authored
  3. Add PostgreSQL 9.1 CREATE EXTENSION support.

    authored
    While at it, rewrite the Makefile to be a bit easier to understand, with much better version determination (more future-proof, too). Might run into some issues when building for other versions; will have to test that soon.
Commits on Feb 26, 2011
  1. Update HTML generation for new verson of dicsount.

    authored
    Move TOC generation and cleanup to an external script that's easier to maintain.
Commits on Feb 3, 2011
  1. Generate the compat files.

    authored
    Instead of a patch, just use a simple Perl script with a list of core functions. It can also be inverted to generate a script with the schema functions.
  2. Add target to build portable core pgTAP.

    authored
    It will just be built, and can then be taken and embedded in other
    distributions for testing on PostgreSQL 8.3+. Still trying to create its
    complement with the schema-testing stuff in it. Once I get that figured out,
    I'll document it.
Commits on Feb 2, 2011
  1. Restore line to generate pg_prove docs

    authored
    But generate them from the CPAN-installed version.
  2. Fix HTML generation.

    authored
Something went wrong with that request. Please try again.