Fix version.dtsi is reset after local firmware build #385
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
In #376 a new step was introduced for local builds to undo changes to the version.dtsi file after a build in order to reduce noise to the repo. Unfortunately the way used to execute the step causes the version.dtsi file to be reset too early and therefore causes an incorrect version number to be used for the version macro when run locally. This went unfortunately undiscovered as the checked in version.dtsi was the same on the day the change in #376 was tested and was not noticed until I build a new change to my keymap locally a few days ago. The git command introduced in #376 is wrapped into a shell function. However what was missed is that commands run by the shell function are run when the function calls are expanded by make. This causes the version.dtsi file to be reset before the firmware build process is even started which resultes in the version.dtsi currently checked in to the repo to be used for local builds instead of the newly generated file when make starts. This change updates how the git command to reset the version.dtsi is being called. It is no longer called within a shell function to ensure it does run in the order it is defined after the firmware build is complete. Builds run through GitHub actions are not impacted and always used the correct version.dtsi For more information and context see: - https://www.gnu.org/software/make/manual/html_node/Shell-Function.html - https://www.gnu.org/software/make/manual/html_node/Reading-Makefiles.html
|
Thanks for spotting and fixing this! |
cr0rc
pushed a commit
to cr0rc/Adv360-Pro-ZMK
that referenced
this pull request
Feb 25, 2024
…ion#385) In KinesisCorporation#376 a new step was introduced for local builds to undo changes to the version.dtsi file after a build in order to reduce noise to the repo. Unfortunately the way used to execute the step causes the version.dtsi file to be reset too early and therefore causes an incorrect version number to be used for the version macro when run locally. This went unfortunately undiscovered as the checked in version.dtsi was the same on the day the change in KinesisCorporation#376 was tested and was not noticed until I build a new change to my keymap locally a few days ago. The git command introduced in KinesisCorporation#376 is wrapped into a shell function. However what was missed is that commands run by the shell function are run when the function calls are expanded by make. This causes the version.dtsi file to be reset before the firmware build process is even started which resultes in the version.dtsi currently checked in to the repo to be used for local builds instead of the newly generated file when make starts. This change updates how the git command to reset the version.dtsi is being called. It is no longer called within a shell function to ensure it does run in the order it is defined after the firmware build is complete. Builds run through GitHub actions are not impacted and always used the correct version.dtsi For more information and context see: - https://www.gnu.org/software/make/manual/html_node/Shell-Function.html - https://www.gnu.org/software/make/manual/html_node/Reading-Makefiles.html
arogulin
pushed a commit
to arogulin/Adv360-Pro-ZMK
that referenced
this pull request
Feb 26, 2024
…ion#385) In KinesisCorporation#376 a new step was introduced for local builds to undo changes to the version.dtsi file after a build in order to reduce noise to the repo. Unfortunately the way used to execute the step causes the version.dtsi file to be reset too early and therefore causes an incorrect version number to be used for the version macro when run locally. This went unfortunately undiscovered as the checked in version.dtsi was the same on the day the change in KinesisCorporation#376 was tested and was not noticed until I build a new change to my keymap locally a few days ago. The git command introduced in KinesisCorporation#376 is wrapped into a shell function. However what was missed is that commands run by the shell function are run when the function calls are expanded by make. This causes the version.dtsi file to be reset before the firmware build process is even started which resultes in the version.dtsi currently checked in to the repo to be used for local builds instead of the newly generated file when make starts. This change updates how the git command to reset the version.dtsi is being called. It is no longer called within a shell function to ensure it does run in the order it is defined after the firmware build is complete. Builds run through GitHub actions are not impacted and always used the correct version.dtsi For more information and context see: - https://www.gnu.org/software/make/manual/html_node/Shell-Function.html - https://www.gnu.org/software/make/manual/html_node/Reading-Makefiles.html
OAGr
added a commit
to OAGr/Adv360-Pro-ZMK
that referenced
this pull request
Mar 2, 2024
* V3.0: (57 commits) Fix version.dtsi is reset after local firmware build (KinesisCorporation#385) Update build workflows (KinesisCorporation#376) Makefile enhancements to optimize local workflows (KinesisCorporation#363) Update Makefile variables (KinesisCorporation#335) Base ZMK update (KinesisCorporation#326) Prefer `tr` to ${char^^}, which does not work on older bash versions (KinesisCorporation#303) Add version macro (KinesisCorporation#300) Add pull request template (KinesisCorporation#293) Revert "Updated keymap" Revert "Updated keymap" Updated keymap Updated keymap Make get_version use bash from $PATH (KinesisCorporation#287) Update bluetooth settings in light of user feedback (KinesisCorporation#289) Revert "Add version macro to keymap.json (KinesisCorporation#269)" Add version macro to keymap.json (KinesisCorporation#269) Changelog new base ZMK update (KinesisCorporation#268) Version compiled FW with automatic macro (KinesisCorporation#267) Document new NKRO settings (KinesisCorporation#264) Improve documentation based on feedback (KinesisCorporation#260) ...
yukitaka
pushed a commit
to yukitaka/Adv360-Pro-ZMK
that referenced
this pull request
Mar 21, 2024
…ion#385) In KinesisCorporation#376 a new step was introduced for local builds to undo changes to the version.dtsi file after a build in order to reduce noise to the repo. Unfortunately the way used to execute the step causes the version.dtsi file to be reset too early and therefore causes an incorrect version number to be used for the version macro when run locally. This went unfortunately undiscovered as the checked in version.dtsi was the same on the day the change in KinesisCorporation#376 was tested and was not noticed until I build a new change to my keymap locally a few days ago. The git command introduced in KinesisCorporation#376 is wrapped into a shell function. However what was missed is that commands run by the shell function are run when the function calls are expanded by make. This causes the version.dtsi file to be reset before the firmware build process is even started which resultes in the version.dtsi currently checked in to the repo to be used for local builds instead of the newly generated file when make starts. This change updates how the git command to reset the version.dtsi is being called. It is no longer called within a shell function to ensure it does run in the order it is defined after the firmware build is complete. Builds run through GitHub actions are not impacted and always used the correct version.dtsi For more information and context see: - https://www.gnu.org/software/make/manual/html_node/Shell-Function.html - https://www.gnu.org/software/make/manual/html_node/Reading-Makefiles.html
luna-flowers
pushed a commit
to luna-flowers/Adv360-Pro-ZMK
that referenced
this pull request
Mar 22, 2024
…ion#385) In KinesisCorporation#376 a new step was introduced for local builds to undo changes to the version.dtsi file after a build in order to reduce noise to the repo. Unfortunately the way used to execute the step causes the version.dtsi file to be reset too early and therefore causes an incorrect version number to be used for the version macro when run locally. This went unfortunately undiscovered as the checked in version.dtsi was the same on the day the change in KinesisCorporation#376 was tested and was not noticed until I build a new change to my keymap locally a few days ago. The git command introduced in KinesisCorporation#376 is wrapped into a shell function. However what was missed is that commands run by the shell function are run when the function calls are expanded by make. This causes the version.dtsi file to be reset before the firmware build process is even started which resultes in the version.dtsi currently checked in to the repo to be used for local builds instead of the newly generated file when make starts. This change updates how the git command to reset the version.dtsi is being called. It is no longer called within a shell function to ensure it does run in the order it is defined after the firmware build is complete. Builds run through GitHub actions are not impacted and always used the correct version.dtsi For more information and context see: - https://www.gnu.org/software/make/manual/html_node/Shell-Function.html - https://www.gnu.org/software/make/manual/html_node/Reading-Makefiles.html
alok
added a commit
to alok/Adv360-Pro-ZMK
that referenced
this pull request
Apr 8, 2024
…ro-ZMK * 'V3.0' of https://github.com/KinesisCorporation/Adv360-Pro-ZMK: Fix Makefile errors that prevent builds on macOS (KinesisCorporation#409) Fix version.dtsi is reset after local firmware build (KinesisCorporation#385) Update build workflows (KinesisCorporation#376)
ztomer
pushed a commit
to ztomer/Adv360-Pro-ZMK
that referenced
this pull request
Apr 16, 2024
…ion#385) In KinesisCorporation#376 a new step was introduced for local builds to undo changes to the version.dtsi file after a build in order to reduce noise to the repo. Unfortunately the way used to execute the step causes the version.dtsi file to be reset too early and therefore causes an incorrect version number to be used for the version macro when run locally. This went unfortunately undiscovered as the checked in version.dtsi was the same on the day the change in KinesisCorporation#376 was tested and was not noticed until I build a new change to my keymap locally a few days ago. The git command introduced in KinesisCorporation#376 is wrapped into a shell function. However what was missed is that commands run by the shell function are run when the function calls are expanded by make. This causes the version.dtsi file to be reset before the firmware build process is even started which resultes in the version.dtsi currently checked in to the repo to be used for local builds instead of the newly generated file when make starts. This change updates how the git command to reset the version.dtsi is being called. It is no longer called within a shell function to ensure it does run in the order it is defined after the firmware build is complete. Builds run through GitHub actions are not impacted and always used the correct version.dtsi For more information and context see: - https://www.gnu.org/software/make/manual/html_node/Shell-Function.html - https://www.gnu.org/software/make/manual/html_node/Reading-Makefiles.html
ztomer
pushed a commit
to ztomer/Adv360-Pro-ZMK
that referenced
this pull request
Apr 16, 2024
…ion#385) In KinesisCorporation#376 a new step was introduced for local builds to undo changes to the version.dtsi file after a build in order to reduce noise to the repo. Unfortunately the way used to execute the step causes the version.dtsi file to be reset too early and therefore causes an incorrect version number to be used for the version macro when run locally. This went unfortunately undiscovered as the checked in version.dtsi was the same on the day the change in KinesisCorporation#376 was tested and was not noticed until I build a new change to my keymap locally a few days ago. The git command introduced in KinesisCorporation#376 is wrapped into a shell function. However what was missed is that commands run by the shell function are run when the function calls are expanded by make. This causes the version.dtsi file to be reset before the firmware build process is even started which resultes in the version.dtsi currently checked in to the repo to be used for local builds instead of the newly generated file when make starts. This change updates how the git command to reset the version.dtsi is being called. It is no longer called within a shell function to ensure it does run in the order it is defined after the firmware build is complete. Builds run through GitHub actions are not impacted and always used the correct version.dtsi For more information and context see: - https://www.gnu.org/software/make/manual/html_node/Shell-Function.html - https://www.gnu.org/software/make/manual/html_node/Reading-Makefiles.html
gabrielfreiberg
added a commit
to gabrielfreiberg/Adv360-Pro-ZMK
that referenced
this pull request
May 5, 2024
* upstream_V3.0: Changelog update (KinesisCorporation#448) Minor docs update (KinesisCorporation#445) Document new layer colors and modifier color configuration option (KinesisCorporation#431) Zephyr 3.5 Update (KinesisCorporation#426) Fix Makefile errors that prevent builds on macOS (KinesisCorporation#409) Fix version.dtsi is reset after local firmware build (KinesisCorporation#385) Update build workflows (KinesisCorporation#376)
gabrielfreiberg
added a commit
to gabrielfreiberg/Adv360-Pro-ZMK
that referenced
this pull request
May 5, 2024
* main: Changelog update (KinesisCorporation#448) Minor docs update (KinesisCorporation#445) Document new layer colors and modifier color configuration option (KinesisCorporation#431) Zephyr 3.5 Update (KinesisCorporation#426) Fix Makefile errors that prevent builds on macOS (KinesisCorporation#409) Fix version.dtsi is reset after local firmware build (KinesisCorporation#385) Update build workflows (KinesisCorporation#376)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Advantage 360 Pro PR template
I discovered yesterday that the command I introduced in #376 to reset the version.dtsi file after a local build to reduce noise on code changes to repositories is not behaving as intended and went undetected during my testing given that my checked in version.dtsi was from the same day and branch. Apologies for that.
What's changed:
This change updates how the git command to reset the version.dtsi is being called. It is no longer called within a shell function to ensure it does run in the order it is defined after the firmware build is complete.
Why has this change been implemented:
In #376 a new step was introduced for local builds to undo changes to the version.dtsi file after a build in order to reduce noise to the repo.
Unfortunately the way used to execute the step causes the version.dtsi file to be reset too early and therefore causes an incorrect version number to be used for the version macro when run locally. This went unfortunately undiscovered as the checked in version.dtsi was the same on the day the change in #376 was tested and was not noticed until I build a new change to my keymap locally a few days ago.
The git command introduced in #376 is wrapped into a shell function.
However what was missed is that commands run by the shell function are run when the function calls are expanded by make. This causes the version.dtsi file to be reset before the firmware build process is even started which resultes in the version.dtsi currently checked in to the repo to be used for local builds instead of the newly generated file when make starts.
Builds run through GitHub actions are not impacted and always used the correct version.dtsi
For more information and context see:
What (if any) actions must a user take after this change:
No further actions must be taken by users. Local builds will start using the correct version number generated by
bin/get_version.shagain while maintaining the reset of the version.dtsi file after a local firmware build is complete.