Commit graph

3900 commits

Author SHA1 Message Date
165e0e743d bump: version 0.21.1 → 0.21.2 2026-05-28 10:45:23 -05:00
37271dc577 fix: remove reference to click package
typer now "vendors" click so the true package may not be present
2026-05-28 10:44:16 -05:00
656f8fce8d bump: version 0.21.0 → 0.21.1 2025-10-19 16:42:40 -05:00
65c3714d9f fix: add Documentation link in pyproject.toml 2025-10-19 16:41:09 -05:00
10ebde55b1 bump: version 0.20.6 → 0.21.0 2025-10-19 13:42:10 -05:00
d2498091c5 feat: require latest wuttjamaican 2025-10-19 13:40:31 -05:00
6d8a3f40bc bump: version 0.20.5 → 0.20.6 2025-10-04 10:07:34 -05:00
326187cea1 fix: add dependency for SQLAlchemy-Utils
SQLAlchemy-Continuum used to depend on this directly, but no longer
does, so now we must

https://github.com/sqlalchemy-continuum/sqlalchemy-continuum/blob/main/CHANGES.rst#150-2025-08-30
2025-10-04 09:57:44 -05:00
be4688e676 bump: version 0.20.4 → 0.20.5 2025-09-20 14:25:25 -05:00
efbc7a39ab fix: fix config extension entry point
per upstream changes
2025-09-20 14:24:07 -05:00
fb3650d739 docs: update intersphinx doc links per server migration 2025-02-18 12:13:14 -06:00
061eaa44e9 bump: version 0.20.3 → 0.20.4 2025-02-15 17:40:41 -06:00
5439c7ba3b fix: add rattail purge-reports command 2025-02-15 17:39:55 -06:00
d23b539439 bump: version 0.20.2 → 0.20.3 2025-01-07 10:53:58 -06:00
f617ff29d5 fix: *do* keep exit code for run-n-mail overnight luigi task
This reverts commit ede2e049f1.

turns out the overnight luigi task logic needs to know whether the
command fails, so it will *not* mark the task as completed..  should
figure out a more elegant way to go about this at some point..
2024-12-28 10:49:04 -06:00
4b086c2e89 todo: add comment reminder re: naming trainwreck constraints 2024-12-27 19:32:25 -06:00
7ee9bf8b9c docs: tweak readme (test commit) 2024-12-16 18:36:35 -06:00
370af2bc45 bump: version 0.20.1 → 0.20.2 2024-12-16 12:45:45 -06:00
214e83b1cb fix: force sorting of (sub)commands when displaying with --help 2024-12-12 21:27:57 -06:00
ede2e049f1 fix: do not keep exit code for run-n-mail overnight luigi task
pretty sure run-n-mail is handling the sending of email if task fails,
so no need to consider the parent command a failure and get 2nd email
2024-12-10 13:46:22 -06:00
dc8539990c bump: version 0.20.0 → 0.20.1 2024-12-10 12:51:27 -06:00
57262c9ff2 fix: use simple string instead of UUID for special role getters
upstream now uses proper UUID but we still can't do that here
2024-12-08 08:10:41 -06:00
68be73b256 fix: move some logic to avoid error when no sqlalchemy 2024-12-05 08:22:28 -06:00
Lance Edgar
31a0f92365 fix: add custom make_uuid() function, app handler method
more overrides so upstream can start to experiment without us
2024-11-30 16:40:20 -06:00
Lance Edgar
4ef111592a fix: add custom uuid_column() and uuid_fk_column() functions
must override wuttjamaican explicitly here, so that it can start to
explore using true UUID data type for column in PG etc.  but for now
and possibly always, rattail will stick with string data type
2024-11-30 15:47:07 -06:00
Lance Edgar
eacc8417ab fix: cleanup some code for wutta project template
even though cookiecutter-wuttaweb is the preferred approach now
2024-11-30 15:29:04 -06:00
Lance Edgar
6bd9f17ba6 bump: version 0.19.2 → 0.20.0 2024-11-24 10:31:39 -06:00
Lance Edgar
7d30be68ec feat: add project generator for 'wutta'
really it's for 'wuttaweb' - just the basics, hopefully can add
generator logic to wutta project soon instead

but for now the wuttjamaican quick start docs will reference the
online Rattail Demo site for generating new project code
2024-11-24 10:25:51 -06:00
Lance Edgar
9093c53aeb fix: tweak generated output for new python/rattail projects 2024-11-22 19:43:55 -06:00
Lance Edgar
87f3555059 fix: avoid error in product autocomplete for duplicated key 2024-11-20 20:15:44 -06:00
Lance Edgar
8205a6ca14 fix: add problem report for duplicated product keys 2024-11-20 19:49:31 -06:00
Lance Edgar
8127e6173f bump: version 0.19.1 → 0.19.2 2024-11-18 14:07:22 -06:00
Lance Edgar
893050a8fc fix: remove enum for sqlalchemy-continuum operation types
This reverts commit bd79108ac4.

we can't include this b/c sqlalchemy is not a general requirement
2024-11-12 19:43:41 -06:00
Lance Edgar
02d9f48bd3 fix: define str() behavior for ProductCost 2024-11-12 18:23:49 -06:00
Lance Edgar
bd79108ac4 fix: add convenience enum for sqlalchemy-continuum operation types 2024-11-12 18:22:40 -06:00
Lance Edgar
f2bb06d19b bump: version 0.19.0 → 0.19.1 2024-11-12 10:17:53 -06:00
Lance Edgar
5004a87a38 fix: make vendor optional, for Vendor Sample Files
sometimes a sample file pertains to "any" vendor
2024-11-02 19:39:12 -05:00
Lance Edgar
93889b9694 bump: version 0.18.12 → 0.19.0 2024-10-22 17:08:04 -05:00
Lance Edgar
f3422cfd97 fix: fix method call signature 2024-10-22 15:05:05 -05:00
Lance Edgar
f7c03fd618 feat: add support for new ordering batch from parsed file 2024-10-22 14:25:55 -05:00
Lance Edgar
c19d359c2e bump: version 0.18.11 → 0.18.12 2024-09-15 10:54:57 -05:00
Lance Edgar
7705fb94a8 fix: update project links, kallithea -> forgejo 2024-09-14 10:47:02 -05:00
Lance Edgar
4937d76680 docs: use markdown for readme file 2024-09-13 18:47:39 -05:00
Lance Edgar
c2141c776b docs: update comment for startup versioning workaround
now that i know a *little* more about why it is needed
2024-09-08 19:40:34 -05:00
Lance Edgar
5dca0b8507 bump: version 0.18.10 → 0.18.11 2024-09-03 11:19:30 -05:00
Lance Edgar
a3df76ea98 fix: move startup workaround for trainwreck query bug
apparently this workaround will break alembic commands since
versioning must be handled a bit differently in that scenario.  so
must only do the workaround if versioning is enabled during the
startup sequence..
2024-09-03 11:13:15 -05:00
Lance Edgar
2a46b2226b bump: version 0.18.9 → 0.18.10 2024-09-03 11:00:04 -05:00
Lance Edgar
1641e0ea14 fix: add startup workaround for trainwreck query bug
not at all clear why, but c621e21cd4 has
introduced some strange bug where the "first" trainwreck query will
raise an error and yet subsequent queries work fine.

so now in the startup sequence we simply force a query, which for
unknown reasons does *not* raise any error, and will somehow magically
prevent the error when a "normal" trainwreck query happens later.

AFAIK this is a "benign" operation and causes no real side effects
other than preventing the strange error.  fingers crossed, this is
fine and will not cause further problems...??!
2024-09-03 10:57:59 -05:00
Lance Edgar
37ea063a6b bump: version 0.18.8 → 0.18.9 2024-08-30 20:36:47 -05:00
Lance Edgar
c621e21cd4 fix: change import for wuttjamaican base model
must avoid importing the actual model subpkg from wuttjamaican because
it can interfere with sqlalchemy-continuum versioning
2024-08-30 20:35:58 -05:00