Thanks for testing out Depfu, we hope it has been useful and saved you some time! Your free trial for private repos has now expired. We're not just going to cut you off right away, but if you want to continue using Depfu, please upgrade your account as soon as possible.
Any questions or feedback, please let us know. Thanks for using Depfu!
Here is everything you need to know about this update. Please take a good look at what changed and the test results before merging this pull request.
Now when the Gettext locale is being changed on runtime with:
Gettext.put_locale("nl")
The Gettext backend for Timex will follow suit. If for some reason you want Timex, and just Timex, to change its locale to something else you should target the Timex.Gettext backend with:
Timex.Gettext.put_locale("de")
Potentially Breaking
Fixed handling of Etc/GMT vs GMT timezones. The former must be POSIX compatible, which inverts the meaning of the sign, whereas the latter have no such restriction and are equivalent to UTC(+/-)HH:MM,
while this is fixing incorrect behavior, it could potentially break users relying on the incorrect behavior.
Incorrect stringification of error values returned in some circumstances
Let Timex follow the locale set by the global Gettext. See #501
Added/Changed
Removed Timex.Timezone.diff, as it is no longer used in Timex
Added support for obtaining the Julian day of year via Calendar.Julian.date_for_day_of_year/3
Fixed
3.7.1 (from changelog)
Fixed
local/0 and local/1 were incorrectly returning TimezoneInfo structs due to a bad match
3.7.0 (from changelog)
Potentially Breaking
Elixir 1.8+ is now required
Tzdata 1.0+ is now required
If you were previously relying on ? suffixed functions to return
{:error, reason} if given invalid date/time inputs (other than
is_valid?), these functions now always return booleans and raise if an
error with the input is encountered
Added/Changed
Refactored much of the library to delegate to the Calendar API where
appropriate, we now make more of an effort to avoid duplication of the
standard library functionality
Functions with the ? suffix now correctly raise on invalid inputs,
and always return booleans. This was implicitly broken before, we need
to follow convention here.
Local timezone handling no longer requires parsing the zoneinfo files,
instead we attempt to observe the timezone name that is active and feed
that into the timezone database directly. We were just using the
abbreviations before, but that wasn't correct behavior at all. In the
future we may want to support the system timezone database as a proper
implementation of Calendar.TimeZoneDatabase, but for now we've just
removed the unnecessary parsing work that was going on here.
Timex.today/1 which returns today's date in the provided timezone
Fixed
Handling of timezones across DST. More generally we now handle gaps/ambiguity
much more consistently
ZoneInfo parser was refactored, now properly supports version 2/3
files, addresses some small bugs in previous code
Some incorrect/redundant typespecs were removed/fixed
We now support alternative timezone databases for API operations that
do not need to interact with the Timex.Timezone module directly. That
module is still tied to tzdata for now, but in the future may be
modified to remove the direct dependency.
Depfu will automatically keep this PR conflict-free, as long as you don't add any commits to this branch yourself. You can also trigger a rebase manually by commenting with @depfu rebase.
All Depfu comment commands
@depfu rebase
Rebases against your default branch and redoes this update
@depfu recreate
Recreates this PR, overwriting any edits that you've made to it
@depfu merge
Merges this PR once your tests are passing and conflicts are resolved
@depfu close
Closes this PR and deletes the branch
@depfu reopen
Restores the branch and reopens this PR (if it's closed)
@depfu pause
Ignores all future updates for this dependency and closes this PR
@depfu pause [minor|major]
Ignores all future minor/major updates for this dependency and closes this PR
@depfu resume
Future versions of this dependency will create PRs again (leaves this PR as is)
Go to the Depfu Dashboard to see the state of your dependencies and to customize how Depfu works.
⌛Your trial has expired ⌛
Thanks for testing out Depfu, we hope it has been useful and saved you some time! Your free trial for private repos has now expired. We're not just going to cut you off right away, but if you want to continue using Depfu, please upgrade your account as soon as possible.
Any questions or feedback, please let us know. Thanks for using Depfu!
Here is everything you need to know about this update. Please take a good look at what changed and the test results before merging this pull request.
What changed?
✳️ timex (3.6.3 → 3.7.5) · Repo · Changelog
Release Notes
3.7.5 (from changelog)
3.7.3 (from changelog)
3.7.1 (from changelog)
3.7.0 (from changelog)
Does any of this look wrong? Please let us know.
Commits
See the full diff on Github. The new version differs by 63 commits:
fix: address non-alphabetic characters in some month abbrev translations
version 3.7.5
feat: add timezone.get/3
fix: empty timezone edge case in posix parser
fix: address some unhandled parser edge cases for ISO8601
fix: support set/2 :microsecond option without precision
fix: avoid delegating to tzdata for non-existent timezones
fix: converting with empty time zone (#655)
fix: parser could fail on timezones when parsed date is invalid
version 3.7.4
fix: validate date/time after parsing completes
fix: incorrect documentation in changelog
fix: bug in handling of etc/gmt vs gmt
fix: improper return value from local/1
doc: update readme
test: remove usage of ~U macro for backwards compat
version: 3.7.3
fix: issue where non-olson timezones could not be operated on via standard apis
version 3.7.2
fix: incorrect use of backported api
ci: fix tests that are sensitive to environment
fix: backport day_of_week
Fix Timex.Format.DateTime.Formatters.Default Docs (#547)
feat: support single-digit hour when parsing timezone offsets
Bugfix: make Timex follow the guidelines set by Gettext so it will follow the global Gettext locale instead of needing to be set independently. It's gettext backend can still be switched independently if necessary. (#501)
fix: multiple fixes/improvements to timezone handling internally
fix: improve error handling in formatter/parser
fix: parsing of week numbers
fix: fractional seconds should be present in asn1 generalized format
feat: add weekday/2 function
fix: return value of format, closes #604
feat: allow parsing of localized date/time strings using current locale
fix: port last remaining private gettext function to public api
fix: duration formatting with a locale, closes #500
fix: alias translator.with_locale to gettext
fix: typespecs in from_* duration apis, closes #606
fix: handle negative values when normalizing durations, normalize when converting representations
fix: use of private gettext apis, closes #569
fix: do not start dialyxir in dev
chore: update translations files
fix: finnish translations in wrong location
fix: use 'yesterday'/'tomorrow' in relative formatter
doc: added changelog entries for 3.6.2-3
fix: incorrect specs in after/before/between, closes #622
test: validate to_datetime/convert correspond with diff/shift
chore: remove unused timezone.diff function
version 3.7.1
fix: local/0,1 incorrectly matched on timezone value
fix: backport missing calendar apis needed after 1.8
ci: use erlef/setup-elixir action
ci: fix workflow
doc: update changelog
Add Timex.today(timezone)
doc: fix layout
Fix Markdown formatting in moduledoc
Fix zero behavior for default duration formatter
typespec for Timezone.convert/2 fixed
version 3.7.0
fix: handling of timezones over DST, a number of other fixes
chore: update deps
release: v3.6.4
Loosen tzdata constraint to allow minor versions
fix: next isn't future
↗️ certifi (indirect, 2.5.3 → 2.6.1) · Repo
Commits
See the full diff on Github. The new version differs by 21 commits:
certifi_res doesn't exist
make file relative to privdir in parse transform
simplify CI
fix CI
bump to 2.6.0
test more versions in CI
remove dependency to parse_trans
Merge pull request #44 from g-andrade/fix/silent-failure-upon-failing-to-read-certificates-file
Merge pull request #43 from enidgjoleka/add-more-entries-on-gitignore
bump certificates
Add ebin/ on .gitignore
Improve quality of GHA runs
Prevent silent failure upon error when reading certificates file
Merge pull request #42 from paulo-ferraz-oliveira/feature/gha
Try to fix for specific shell
Try to fix for git expectations
Move file to where it's expected
Actually run on different Erlang versions
Implement GitHub Actions
add empty actiion file
remove compatibility with rebar2
↗️ hackney (indirect, 1.17.0 → 1.17.4) · Repo · Changelog
Release Notes
1.17.4 (from changelog)
1.17.3 (from changelog)
1.17.2 (from changelog)
1.17.1 (from changelog)
Does any of this look wrong? Please let us know.
Commits
See the full diff on Github. The new version differs by 19 commits:
fix changelog date
add rebar.lock
bump to 1.17.4
fix use_count
bump to 1.17.3
fix date
Merge pull request #681 from benoitc/fix/gh680
ensuure we release a socket in the pool
Revert "Link checkout process to fix dangling aborted request"
bump to 1.17.2
fix parse_trans
bump version of certifi
Merge pull request #655 from spydon/spydon.expose-ssl-default-opts
regen doc
bump to 1.17.1
Merge pull request #674 from ronaldwind/master
Merge pull request #676 from paulo-ferraz-oliveira/bump/parse_trans
Avoid parse_trans warning when using hackney as a dependency
Link checkout process to fix dangling aborted request
↗️ tzdata (indirect, 1.0.5 → 1.1.0) · Repo · Changelog
Release Notes
1.1.0 (from changelog)
Does any of this look wrong? Please let us know.
Commits
See the full diff on Github. The new version differs by 8 commits:
Merge pull request #118 from lau/v-1-1
Update mix.lock
Bump version to 1.1.0
Bump hackney requirement to ~> 1.17
Update included tzdata .ets file from 2020d to 2020e
Merge pull request #106 from hoyon/fix-warnings
Merge pull request #113 from TheFirstAvenger/patch-1
Fix bad spec for zone_alias_list
Depfu will automatically keep this PR conflict-free, as long as you don't add any commits to this branch yourself. You can also trigger a rebase manually by commenting with
@depfu rebase
.All Depfu comment commands
Go to the Depfu Dashboard to see the state of your dependencies and to customize how Depfu works.