[Bug 259567] devel/py-jsonschema: Version returns 0.0.0, breaking net-im/py-matrix-synapse which runs a version check
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=259567 --- Comment #2 from t...@leben-auf-see.de --- File = /usr/local/lib/python3.8/site-packages/synapse/python_dependencies.py Code Line 50 REQUIREMENTS = [ # we use the TYPE_CHECKER.redefine method added in jsonschema 3.0.0 "jsonschema>=3.0.0", I changed the Code to "jsonschema>=0.0.0" and matrix-synapse starts now, so it seems to be the only check. Regards Tom -- You are receiving this mail because: You are on the CC list for the bug.
[Bug 259579] devel/py-circuitbreaker: Update to 1.3.2
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=259579 Alessandro Sagratini changed: What|Removed |Added URL|https://github.com/fabfuel/ |https://github.com/fabfuel/ |circuitbreaker/releases/tag |circuitbreaker/releases/tag |/1.2.0 |/1.3.2 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug 259574] devel/py-distributed: Fails to build in quarterly branch: py38-cloudpickle>=1.5.0 - not found
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=259574 --- Comment #2 from Robert Clausecker --- (In reply to Kubilay Kocak from comment #1) I guess it would be easiest to just MFH this update. Sunpoet@ surely knows more. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug 259567] devel/py-jsonschema: Version returns 0.0.0, breaking net-im/py-matrix-synapse which runs a version check
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=259567 --- Comment #3 from Sascha Biberhofer --- I can confirm this on my system. Seems like something is weirdly wrong with the version of the FreeBSD package and *only* the distributed package: * It appears as jsonschema==0.0.0 in pip freeze too, suggesting that something is wrong with the package metadata per se. * I quickly built and installed the package directly from ports, which does *not* result in such a broken package for me (i.e. both pip freeze and jsonschema --version report the correct version number). Similarly, installing the package directly via pip works fine too. * Building the package in poudriere however yields a package with version 0.0.0 in pip which is consequently also reported back by the package It would appear that something goes wrong when building the package with poudriere, but I'm currently not sure what exactly goes wrong here and I'm a bit uncertain how to proceed. The way the package obtains the version appears to have changed and relies on importlib's metadata.version() function, which *should* work, I suppose, but I'm not sure how that version is formed before the package is installed and would need to dig deeper into this to properly find the cause. In the meantime, should we patch synapse to remove the version check? It would be a hack and I'd prefer to resolve this issue right here, but it *should* be fine for the time being to ensure things keep running moderately smoothly. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug 259381] devel/py-platformdirs: fails to build: setuptools_scm, line 84, in dump_version > fp.write(template.format(version=version)) > KeyError: 'version_tuple'
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=259381 --- Comment #6 from commit-h...@freebsd.org --- A commit in branch 2021Q4 references this bug: URL: https://cgit.FreeBSD.org/ports/commit/?id=5bfd0738fbe861b2a371801dbe7f12daf5d93f33 commit 5bfd0738fbe861b2a371801dbe7f12daf5d93f33 Author: Kai Knoblich AuthorDate: 2021-10-26 10:08:35 + Commit: Kai Knoblich CommitDate: 2021-11-02 11:53:57 + devel/py-setuptools_scm: Update to 5.0.2 * Use this release as an intermediate step before the port is updated to the 6.x release, because it contains mostly bugfixes and is also required for the 2021Q4 branch. The latter is necessary in order to be able to build certain ports successfully if devel/py-setuptools_scm is already present in the build environment. For instance, devel/py-platformdirs, which can be built without it but if devel/py-setuptools is already installed, it requires the 5.x release of it for build. [1] * Also adjust the description of the TOML option slightly with an appropriate warning to avoid build issues related to parsing "pyproject.toml" files during setup. [2] Changelog since 4.1.2: https://github.com/pypa/setuptools_scm/blob/v5.0.2/CHANGELOG.rst PR: 259381 [2], 259420 (both related) MFH:2021Q4 (after 1 week) [1] (cherry picked from commit 33a4a1d205bd4323d330b7cc5f39925e5b3e811c) devel/py-setuptools_scm/Makefile | 5 ++--- devel/py-setuptools_scm/distinfo | 6 +++--- 2 files changed, 5 insertions(+), 6 deletions(-) -- You are receiving this mail because: You are on the CC list for the bug.
[Bug 258891] devel/py-setuptools_scm: tries to git archive the entire ports tree on build
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=258891 --- Comment #16 from commit-h...@freebsd.org --- A commit in branch 2021Q4 references this bug: URL: https://cgit.FreeBSD.org/ports/commit/?id=53d9b1720e8f92aaa1ba0bc693f148ffe33ccc02 commit 53d9b1720e8f92aaa1ba0bc693f148ffe33ccc02 Author: Kai Knoblich AuthorDate: 2021-10-24 10:43:33 + Commit: Kai Knoblich CommitDate: 2021-11-02 11:53:42 + devel/py-setuptools_scm: Do not run "git archive" on Ports tree * In build environments have git installed in conjunction with a git-based Ports tree and haven't WRKDIRPREFIX set, there will be significant delays when building devel/py-setuptools_scm or ports that depend on it. This is because the top-level directory of the git repository is determined during build via "git rev-parse --show-toplevel" which is issued inside the WRKSRC directory. Once the top-level directory (which is PORTSDIR) has been determined, an archive is created from this point using "git archive" which is then very time-consuming due the complexity of the Ports tree. In environments (e.g. poudriere) that have WRKDIRPREFIX set and also have git present during build, the issue doesn't appear because "git rev-parse --show-toplevel" fails silently with "not a git repo". Remedy the issue by returning only the actual path of WRKSRC, but only if it has "setup.py" in it (= devel/py-setuptools_scm is built) or a test session is performed. * Modernize the "do-test" target while I'm here and bump PORTREVISION due package change. PR: 258891 Reported by:Robert Clausecker Obtained from: OpenIndiana MFH:2021Q4 (after 1 week) (cherry picked from commit adc0cc3f69aeac37b9678c6d5b4d548d9c60cfae) devel/py-setuptools_scm/Makefile | 8 ++-- ...-src_setuptools__scm_file__finder__git.py (new) | 22 ++ ...h-src_setuptools__scm_file__finder__hg.py (new) | 22 ++ 3 files changed, 50 insertions(+), 2 deletions(-) -- You are receiving this mail because: You are on the CC list for the bug.
[Bug 258891] devel/py-setuptools_scm: tries to git archive the entire ports tree on build
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=258891 Kai Knoblich changed: What|Removed |Added Flags|merge-quarterly?|merge-quarterly+ Status|In Progress |Closed Resolution|--- |FIXED --- Comment #17 from Kai Knoblich --- Committed to the 2021Q4 branch, all done! Thanks, Robert, for reporting the issue and Steve, for the hint with the workaround from OpenIndiana! -- You are receiving this mail because: You are on the CC list for the bug.
[Bug 259381] devel/py-platformdirs: fails to build: setuptools_scm, line 84, in dump_version > fp.write(template.format(version=version)) > KeyError: 'version_tuple'
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=259381 Bug 259381 depends on bug 259420, which changed state. Bug 259420 Summary: devel/py-setuptools_scm: Needs to be updated to 5.x to get a successful build of py-platformdirs https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=259420 What|Removed |Added Status|In Progress |Closed Resolution|--- |FIXED -- You are receiving this mail because: You are on the CC list for the bug.
Ports with version numbers going backwards: devel/py-pytest, devel/py-pytest...
** The following ports have a version number that sorts before a previous one ** For many package tools to work correctly, it is of utmost importance that version numbers of a port form a monotonic increasing sequence over time. Refer to the FreeBSD Porter's Handbook, 'Package Naming Conventions' for more information. Tools that won't work include pkg_version, portupgrade and portaudit. A common error is an accidental deletion of PORTEPOCH. Please fix any errors as soon as possible. The ports tree was updated at Tue Nov 2 2021 12:30:00 UTC. - *devel/py-pytest* : py38-pytest-4.6.11 < py38-pytest-6.2.4 - *devel/py-pytest-flask* : py38-pytest-flask-0.15.1 < py38-pytest-flask-1.2.0 - *devel/py-pytest-translations* : py38-pytest-translations-2.0.0 < py38-pytest-translations-3.1.0 - *devel/py-pytest-xdist* : py38-pytest-xdist-1.32.0 < py38-pytest-xdist-2.3.0 - *x11/wayland-logout* : wayland-logout-1.3 < wayland-logout-1.4
[Bug 259599] devel/py-buildbot: Fails to run after sqlalchemy 1.4 update: AttributeError: module 'sqlalchemy.engine.strategies' has no attribute 'PlainEngineStrategy'
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=259599 Po-Chuan Hsieh changed: What|Removed |Added CC|sunp...@freebsd.org | --- Comment #3 from Po-Chuan Hsieh --- Since py-sqlalchemy* conflicts with each other, I made the change from py-sqlalchemy12 to py-sqlalchemy14 following py-sqlalchemy-migrate change. Without that, you won't be able to build buildbot due to the conflicts of dependencies. After that, you get a package and there's a chance to work fine. I cannot do runtime check but it is better than nothing. Anyway, the quick solution is to update buildbot to 3.*. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug 259596] [NEW PORT]: devel/py-typish: Type checking utilities for Python
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=259596 Neal Nelson changed: What|Removed |Added Attachment #229197|0 |1 is obsolete|| --- Comment #2 from Neal Nelson --- Created attachment 229224 --> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=229224&action=edit Git diff of new port. This updated patch includes many of the requested changes. I can confirm that I have portlinted and the only failure is the port math/py-nptyping which is a test dependency, but has not yet been committed and in fact depends on this port. Hopefully the potential circular dependency will not be a problem as it's only a test dependency. - Poudriere testport run OK. I have no idea how to run it for non-default versions. - Comment about CHEESESHOP added. - python category added. - nptyping dependency updates. - Unneeded test dependencies removed. I had originally a smaller set, but the setup.py file calls for the set as originally submitted, so for the sake of completeness they were added. The set in this patch is all that is in fact needed. As for asking upstream for an sdist; you may have to expand on that as I have no idea what that is. Please note that this I am submitting this port as a dependency of cad/py-cadquery. As long as that works, that is really all I care about. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug 259567] devel/py-jsonschema: Version returns 0.0.0, breaking net-im/py-matrix-synapse which runs a version check
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=259567 --- Comment #4 from Sascha Biberhofer --- Created attachment 229227 --> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=229227&action=edit devel/py-jsonschema: Fix missing dependency Apparently py-jsonschema uses py-setuptools_scm to filter the version from scm directly. This is reflected in the project's pyproject.toml, see [1]. Our port is missing this dependency, which causes the version to be set to 0.0.0. I've attached a patch that bumps the port's revision and adds the missing dependency. Tested building this in poudriere and it works fine. :-) Credits for this go to David Tucker in the #python room on matrix.org for pointing pointing me in the right direction. :D Cheers, Sascha [1] https://github.com/Julian/jsonschema/blob/272b4f24437d9cba1127058dc3479f9bc6d8c4b9/pyproject.toml#L6 -- You are receiving this mail because: You are on the CC list for the bug.
[Bug 249337] [meta] Ports broken by Python 2.7 End-of-Life and removal
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=249337 Ahmed changed: What|Removed |Added CC||ahmedsayeed1...@yahoo.com --- Comment #16 from Ahmed --- In collect_register() function of arc-linux-tdep.c, the "eret" http://www-look-4.com/travel/london/ (exception return) register value is not being reported correctly. Background: https://komiya-dental.com/shopping/buy-android/ When asked for the "pc" value, we have to update the "eret" register with GDB's STOP_PC. The "eret" instructs the kernel code where to jump back http://www.iu-bloomington.com/shopping/hatchback-cars/ when an instruction has stopped due to a breakpoint. This is how collect_register() is doing so: https://waytowhatsnext.com/shopping/xbox-release-date/ --8<-- if (regnum == gdbarch_pc_regnum (gdbarch)) http://www.wearelondonmade.com/travel/london/ regnum = ARC_ERET_REGNUM; regcache->raw_collect (regnum, buf + arc_linux_core_reg_offsets[regnum]); -->8-- http://www.jopspeech.com/travel/london/ Root cause: Although this is using the correct offset (ERET register's), it is also http://joerg.li/travel/london/ changing the REGNUM itself. Therefore, raw_collect (regnum, ...) is not reading from "pc" anymore. http://connstr.net/travel/london/ Consequence: This bug affects the "native ARC gdb" badly and causes kernel code to jump to addresses after the breakpoint and not executing the "breakpoint"ed http://embermanchester.uk/travel/london/ instructions at all. That "native ARC gdb" feature is not upstream yet and is in review at the time of writing [1]. http://www.slipstone.co.uk/travel/london/ In collect_register() function of arc-linux-tdep.c, the "eret" (exception return) register value is not being reported correctly. http://www.logoarts.co.uk/travel/london/ Background: When asked for the "pc" value, we have to update the "eret" register with GDB's STOP_PC. http://www.acpirateradio.co.uk/travel/good/ The "eret" instructs the kernel code where to jump back when an instruction has stopped due to a breakpoint. This is how collect_register() is doing so: http://www.compilatori.com/travel/london/ --8<-- if (regnum == gdbarch_pc_regnum (gdbarch)) regnum = ARC_ERET_REGNUM; regcache->raw_collect (regnum, buf + arc_linux_core_reg_offsets[regnum]); -->8-- Root cause: https://www.webb-dev.co.uk/shopping/shopping-during-corona/ Although this is using the correct offset (ERET register's), it is also changing the REGNUM itself. Therefore, raw_collect (regnum, ...) is not reading from "pc" anymore. Consequence: This bug affects the "native ARC gdb" badly and causes kernel code to jump to addresses after the breakpoint and not executing the "breakpoint"ed instructions at all. That "native ARC gdb" feature is not upstream yet and is in review at the time of writing [1]. -- You are receiving this mail because: You are the assignee for the bug.
[Bug 259599] devel/py-buildbot: Fails to run after sqlalchemy 1.4 update: AttributeError: module 'sqlalchemy.engine.strategies' has no attribute 'PlainEngineStrategy'
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=259599 --- Comment #4 from a...@ish.com.au --- This change is not better than nothing. If buildbot had failed to build, I would not have wasted four hours working out what went wrong, missed a release deadline since buildbot died and everyone wouldn't now be scrambling to sort this out. If you can't try a basic smoketest of the software, please don't change things. A build failure would have properly alerted people who use this port to do something about it. buildbot 3 is an option, but some notes in UPDATING would be appreciated for all the people who'll end up with broken configurations. Quite a few deprecated things were removed. -- You are receiving this mail because: You are on the CC list for the bug.
[Bug 259579] devel/py-circuitbreaker: Update to 1.3.2
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=259579 Kubilay Kocak changed: What|Removed |Added CC||y...@freebsd.org Flags||maintainer-feedback?(yasu@f ||reebsd.org) -- You are receiving this mail because: You are on the CC list for the bug.
[Bug 259579] devel/py-circuitbreaker: Update to 1.3.2
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=259579 Kubilay Kocak changed: What|Removed |Added Depends on||259307 --- Comment #2 from Kubilay Kocak --- 1.3.1 landed in bug 259307 Referenced Bugs: https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=259307 [Bug 259307] [NEW PORT] devel/py-circuitbreaker: Python Circuit Breaker pattern implementation -- You are receiving this mail because: You are on the CC list for the bug.
