Revision history for PostgreSQL extension semver. 0.32.0 2022-10-23T21:53:54Z - Add support for binary input (receive) and output (send) functions. Thanks to Anna Clemens for the pull request (#61)! 0.31.2 2021-09-28T02:03:35Z - Add an overflow check and properly compare the max size of INT32 rather than INT. Thanks to Felix Lechner for the report and Tom Lane for the C lesson (#58). 0.31.1 2021-04-27T00:10:07Z - Updated the C code to pass the correct number of arguments to `hashint2()`. Thanks to Andrew Gierth for the spot. - Fixed an error in processing the prerelease where it could sometimes incorrectly report throw an error saying "prerelease numbers can't start with 0" for prereleases with no such leading zero. 0.31.0 2020-10-17T22:30:00Z - Added a workaround for an LLVM bitcode compile error. Thanks to @mark-s-a for the report (#40). - Removed `--load-language` from the options for running the tests, as it has not been needed since 9.1, we support 9.2 and higher, and it has been removed from Postgres 13. - Fixed an a collation error on Postgres 12 and higher. Thanks to Andrew for Marc Munro for the report and to Andrew Gierth for the fix (pgxn/pgxn-manager#67). - Prerelease parts are no longer compared compared case-insensitively, but in ASCII sort order, as required by the spec. This is a breaking change in the sense that `1.0.0-rc1` will now be considered greater than `1.0.0-RC1`, rather than equivalent, but they're both still valid. See https://github.com/semver/semver/issues/176 for the relevant discussion. Thanks to Andrew Gierth for the spot! 0.30.0 2020-05-16T19:28:36Z - ***** WARNING: This release breaks compatibility with previous versions! ****** Previous versions of the semver extension incorrectly allowed some invalid prerelease and build metadata values. Details below, but *BEFORE YOU UPGRADE* we strongly recommend that you check for and repair any invalid semvers. You can find them using the official [SemVer regular expression](https://regex101.com/r/vkijKf/1/) like so: SELECT name, version FROM packages WHERE version::text !~ '^(0|[1-9]\d*)\.(0|[1-9]\d*)\.(0|[1-9]\d*)(?:-((?:0|[1-9]\d*|\d*[a-zA-Z-][0-9a-zA-Z-]*)(?:\.(?:0|[1-9]\d*|\d*[a-zA-Z-][0-9a-zA-Z-]*))*))?(?:\+([0-9a-zA-Z-]+(?:\.[0-9a-zA-Z-]+)*))?$'; If no rows are returned, you should be good to go. If there are results, here are Examples of invalid semantic versions and how they should be repaired. 1.0.0-02799 -> 1.0.0-2799 1.0.0-0.02 -> 1.0.0-0.2 1.0.0-.20 -> 1.0.0-0.20 1.0.0+0+20 -> 1.0.0+0-20 or 1.0.0+0.20 1.0.0+.af -> 1.0.0+0.af or 1.0.0+af And now, back to your regularly-scheduled changes. - Fixed an error formatting prerelease parts for semvers longer than 32 bytes. Thanks to @Nemo157 for the report and suggested fix (#48). - Removed code that converted dashes to dots in prerelease parts. It had been doing so on the assumption that dashes were invalid in SemVer 1.0 and 2.0 prerelease parts, but that does not turn out to be the case. Thanks to @Nemo157 for the report (#46). - Fixed the parsing of prerelease and metadata parts to allow leading zeros for parts with non-numeric characters, e.g., `1.0.0-alpha.0a`, and to disallow parts with leading zeros and only numeric characters, e.g., `1.0.0-02799`. Thanks to @Nemo157 for the bug report, and to Joseph Donahue for the SemVer spec expertise (#45). - The metadata part may no longer contain plus signs other than the one used to start the metadata part. - The prerelease and metadata parts may no longer start with a dot. 0.22.0 2020-04-02T14:13:55Z - Fixed `get_semver_prerelease()` so that it returns only the prerelease without any build metadata. For example, for `1.0.0-1.2+1.02`, it will now return `1.2`, not `1.2+1.02`. Thanks to Artur Troian for the report (#41) and fix (#42)! 0.21.0 2020-03-21T21:56:20Z - Fixed build metadata part to allow leading zeros. In other words, `1.0.0-1.02` is not a legitimate semver, because it has a leading zero in a numeric part of the prerelease version. `1.0.0-1.2+1.02`, however, is valid, because the leading zero numeric expression is in the build metadata part, which does not evaluate as numeric, and so is allowed. This fix allows source code hashes, for example, to be used in build metadata. Thanks to Artur Troian for the report (#38) and fix (#39)! 0.20.3 2018-11-10T19:32:26Z - Fixed test failure on Postgres 11 due to a renamed column that this extension doesn't even use for testing. 0.20.2 2018-10-19T02:05:11Z - Fix Makefile bug that installed the version script with the release version instead of the extension version. Thanks again to @ninjaslikecheese for the report (#33). 0.20.1 2018-10-18T00:30:04Z - Properly incremented version to v0.20.0 in the spec file and docs. Thanks to @ninjaslikecheese for the report (#32). 0.20.0 2018-10-17T12:59:22Z - Fixed file name in the v0.17.0 upgrade script. - Added the semverrange type. - Increased the minimum required version of PostgreSQL to 9.2 for range type support. - Fixed the `Makefile` so it properly installs `semver.sql`. Thanks to Matej Marušák for the report (Issue #31). 0.17.0 2017-05-07T07:15:25Z - Fixed potential memory leak in is_semver function. Thanks to Arun Tejasvi Chaganty (Issue #29). 0.16.0 2016-12-04T00:51:19Z - Added `get_semver_*` accessor functions to retrieve the individual parts of a semver, thanks to Geoff Montee (PR #27). 0.15.0 2016-10-28T03:38:59Z - Fixed incorrect parsing of semver build metadata. - Fixed text foratting of build metadata-only semver values. - Implemented pre-release precedence scheme as per semantic versioning 2.0.0 section 11. Thanks to Sergey Kozlov for the report (Issue #23). - Added Xavier Caron and Tom Davis as maintainers. 0.14.0 2016-08-11T05:38:27Z - Updated the extension version in `semver.control` to 0.13.0, which evidently I forgot to do for that release. Thanks to Stanislav Lorenc for the report (Issue #21). 0.13.0 2016-07-30T21:06:15Z - Fixed bug where a dash (-) was disallowed within the metadata part of a semver. Thanks to @TerraTech for the report (Issue #19) and Tom Davis for the fix (PR #20). 0.12.0 2016-06-11T06:00:19Z - Fixed bug where some items described in the SemVer 2.0.0 Spec, section 10, were not properly recognized as semvers. Thanks to Xavier Caron for the fix (with tests!) (PR #17). 0.11.0 2016-05-23T20:29:49Z - Added `is_semver()`, thanks to a pull request from Xavier Caron (PR #14). - Fixed bug where version parts with non-leading 0 were considered invalid. Thanks to @JeremySTX for the report (Issue #15) and Tom Davis for the fix (PR #16). 0.10.0 2016-05-12T00:28:03Z - Updated extension to support Semantic Versioning 2.0.0, thanks to Tom Davis (PR #13). - Improved error output to mark specific bad character, where possible, thanks to Tom Davis. - BREAKING CHANGE: `semver()` and cast constructors no longer support the semver 1.0.0-beta format format "X.Y.Zpre" (that is, without the dash), since it's not compatible with the semver 2.0.0 spec. Use `to_semver()` to convert it to X.Y.Z-pre", or specify "X.Y.Z-pre" directly. 0.5.0 2014-12-06T00:29:17Z - Fixed issue where the release file could be listed twice when installing, which causes an installation failure on some versions of PostgreSQL. - Fixed a typo in the README: `make installcheck` should be run *after* `make install`. - Fixed an issue where the installer would read the distribution version rather than the extension version. The latter is requierd to tell PostgreSQL the correct version number (since the distribution version is quite different from the extension version). - Added `LICENSE` file to simplify packaging, thanks to Richard Marko, who is packaging semver for Fedora. 0.4.0 2013-06-12T06:48:33Z - Updated the `Makefile` to reflect the recommended patterns from pgsql-hackers. Thanks to Cédric Villemain for the details. - Changed the Makefile to read the distribution name and version from META.json. - Fixed the Makefile to allow `PG_CONFIG=/path/to/pg_config` to actually work. - Fixed an off-by-one bug in the parser that could add garbage characters to the end of certain semvers with patch versions. Thanks to Andrew "RhodiumToad" Gierth for pointing out the bug. - Add a macro to fetch a semver from C function arguments, and cast to `semver*` rather than `void*`. Suggested by Andrew "RhodiumToad" Gierth. - Removed unnecessary calls to `PG_FREE_IF_COPY()`. Only really needed for toastable values, and semvers are not toastable. 0.3.0 2012-11-20T19:04:30Z - Updated the parser to support an optional dash before the prerelease version, and the formatter to always emit a dash for the prerelease version. This brings it in line with the final semver 1.0.0 specification. Thanks to Pieter van de Bruggen for making it happen. 0.2.4 2012-11-02T22:32:30Z - Fixed a memory allocation bug that could cause semvers to be displayed with missing or garbage characters. Thanks to Andrew "RhodiumToad" Gierth for help tracking down and fixing the issue. 0.2.3 2011-11-11T06:55:10Z - Fixed the `Makefile` so that the documentation file should properly be found and installed by `PGXS`. 0.2.2 2011-05-12T19:01:41 - Tweaked MultiMarkdown table layout in the docuemntation so that the header row is always processed as a header row, rather than a `pre` block. - Simplified the `CREATE EXTENSION` support in the `Makefile`. 0.2.1 2011-04-20T20:37:05 - Fixed the metadata file to reflect that the "pair" extension is not included in the semver distribution. - Added abstract and doc file to the `provides` section of `META.json`. - Removed the `NO_PGXS` stuff from `Makefile`, as the PostgreSQL core team does not recommend its use outside of the core contrib extensions. - Added PostgreSQL 9.1 `CREATE EXTENSION` support, including migration from an unpackaged install via `CREATE EXTENSION semver FROM unpackaged`. - Renamed `doc/semver.md` to `doc/semver.mdd`, so that PGXN will parse it as MultiMarkdown. This will allow the tables to be properly formatted. - Removed documentation that semver is implemented as a domain. It hasn't been since 0.2.0. - Removed Unicode characters `psql` output in the documentation. 0.2.0 2011-02-05T19:32:49 - Converted to a native type implemented in C by Sam Vilain. While David was at lunch, no less. - As a consequence, `USING` is no longer required in an `ORDER BY` clause to get proper semver sort ordering. - Added casts from nuermic types. - Renamed `clean_semver()` to `to_semver()`. 0.1.0 2010-10-07 18:31:43 - Initial version. - Implementation in pure PL/pgSQL. - Included in [PGXN Manager](https://github.com/theory/pgxn-manager). - Not otherwise released.