Posted by jpluimers on 2021/07/28
[WayBack] Frank A. Krueger on Twitter: “I made a build status IoT thing! Automatically polls @bitrise and is even Alexa controlled (for brightness and to turn off). Now I want to add github issue and PR counts. Just a $12 ESP32 and a $30 led matrix.”
[WayBack]
–jeroen
Posted in Development, DVCS - Distributed Version Control, ESP32, GitHub, Hardware Development, Software Development, Source Code Management | Leave a Comment »
Posted by jpluimers on 2021/07/21
If you get the below error when running hg.exe, then you are mixing a 64-bit Mercurial with 32-bit dependencies:
C:\>hg --version Traceback (most recent call last): File "hg", line 43, in File "hgdemandimport\demandimportpy2.pyc", line 150, in __getattr__ File "hgdemandimport\demandimportpy2.pyc", line 94, in _load File "hgdemandimport\demandimportpy2.pyc", line 43, in _hgextimport File "mercurial\dispatch.pyc", line 22, in File "hgdemandimport\demandimportpy2.pyc", line 248, in _demandimport File "hgdemandimport\demandimportpy2.pyc", line 43, in _hgextimport File "mercurial\i18n.pyc", line 28, in File "hgdemandimport\demandimportpy2.pyc", line 150, in __getattr__ File "hgdemandimport\demandimportpy2.pyc", line 94, in _load File "hgdemandimport\demandimportpy2.pyc", line 43, in _hgextimport File "mercurial\encoding.pyc", line 24, in File "mercurial\policy.pyc", line 101, in importmod File "mercurial\policy.pyc", line 63, in _importfrom File "hgdemandimport\demandimportpy2.pyc", line 164, in __doc__ File "hgdemandimport\demandimportpy2.pyc", line 94, in _load File "hgdemandimport\demandimportpy2.pyc", line 43, in _hgextimport File "mercurial\cext\parsers.pyc", line 12, in File "mercurial\cext\parsers.pyc", line 10, in __load ImportError: DLL load failed: %1 is geen geldige Win32-toepassing.
The equivalent English error is [WayBack] “ImportError: DLL load failed: %1 is not a valid Win32 application.” – Google Search.
The problem is the bitness of hg.exe: [WayBack] python – Error while installing Mercurial on IIS7 64bit: “DLL Load Failed: %1 is not a valid Win32 application” – Stack Overflow
You can quickly figure out the bitness of hg.exe:
C:\>where hg C:\Program Files\Mercurial\hg.exe C:\>sigcheck "C:\Program Files\Mercurial\hg.exe" Sigcheck v2.72 - File version and signature viewer Copyright (C) 2004-2019 Mark Russinovich Sysinternals - www.sysinternals.com c:\program files\mercurial\hg.exe: Verified: Unsigned Link date: 17:49 9-7-2019 Publisher: n/a Company: n/a Description: Fast scalable distributed SCM (revision control, version control) system Product: mercurial Prod version: 5.0.2 File version: 5.0.2 MachineType: 64-bit
Since I use chocolatey for most my installs, I forced x86 the Chocolatey way:
I was missing the
forcex86switch:
--x86, --forcex86 ForceX86 - Force x86 (32bit) installation on 64 bit systems. Defaults to false.
So after these:
choco uninstall --yes hg choco install --yes --force86 hg
I got this signature check:
C:\>sigcheck "C:\Program Files (x86)\Mercurial\hg.exe" Sigcheck v2.72 - File version and signature viewer Copyright (C) 2004-2019 Mark Russinovich Sysinternals - www.sysinternals.com c:\program files (x86)\mercurial\hg.exe: Verified: Unsigned Link date: 17:50 9-7-2019 Publisher: n/a Company: n/a Description: Fast scalable distributed SCM (revision control, version control) system Product: mercurial Prod version: 5.0.2 File version: 5.0.2 MachineType: 32-bit
–jeroen
Posted in Development, DVCS - Distributed Version Control, Encoding, Mercurial/Hg, Python, Scripting, Software Development, Source Code Management | Leave a Comment »
Posted by jpluimers on 2021/06/15
Every now and then you really goof up the final changeset in a series of changesets that are in a branch.
I have yet to figure out how to delete that changeset, other than keeping the branch stale and merge from the last good changeset to a new branch.
This seems hard to do given these from [Archive.is] plasticscm delete changeset – Google Search:
Delete empty branches only
Branches can only be deleted when they are empty. This means that if you want to delete a branch, you have to delete its changesets first. A changeset can only be deleted if it doesn’t have any descendants (children or linked by merges).
Consider the following scenario: If you want to delete
main/t124, you’ll need to delete its two changesets first. But, to delete changesetD, you’ll need to first removeA,BandC, to get rid of all its descendants.Yes, Plastic branches are admittedly hard to kill.
But, in our opinion, the question is: why would you want to delete a branch?
The answer is obvious for users coming from Git: deleting branches is a pretty common technique, you need to clean out old branches to keep the repo usable. But, that’s not the case in Plastic. If you want to learn more about why we don’t delete branches, please check “We don’t delete task branches”.
Changesets can be moved to a different branch
Changesets belong to a branch, but they can be moved under certain circumstances.
A common example is that you mistakenly checkin changes to
maininstead of the right task branch because you forgot to create the branch and switch to it.It is possible to fix the issue and move the changesets to the new branch:
Find the details about how to move changesets here.
Deleting the branch fails because there is still a changeset on it:
Deleting the changeset fails because the option is greyed out:
It does not happen that often, though there are now about a few % of the branches stale because of this.
–jeroen
Posted in Development, PlasticSCM, Source Code Management | Leave a Comment »
Posted by jpluimers on 2021/05/19
I needed this for a project there 3rdParty libraries could have .dcu and .obj files (to either speed up building, or because not all source code was supplied), but those files should not be present nowhere else in the repository.
This is an ignore.conf template for that:
*.obj *.dcu !/3rdParty/**/*.dcu !/3rdParty/**/*.obj
The ! exclamation mark undoes the ignore according to the pattern following it.
Note the starting slashes after the !: they indicate the root directory of the repository and are mandatory.
Based on (warning, the JavaScript on these pages are extremely CPU intensive!):
–jeroen
Posted in Development, PlasticSCM, Software Development, Source Code Management | Leave a Comment »
Posted by jpluimers on 2021/05/18
Since I always forget the command (I do not use it often enough) [WayBack] Plastic SCM – Branch per task guide:
cm add -R .
That adds the current directory (.) recursively, taking into account the ignore.conf file.
–jeroen
Posted in Development, PlasticSCM, Software Development, Source Code Management | Leave a Comment »
