I think when you've merged with master a few times already it gets fiddly to do anything other than what Chris suggests. The rebase can be worth it if there was a clean commit history but if you're squashing then you can just do it the direct way.
I would do this like: $ git checkout mybranch $ git checkout -b mybranch_backup # make a backup! $ git checkout mybranch $ git fetch upstream $ git merge upstream/master $ git diff upstream/master > mybranch.diff $ git reset --hard HEAD~1000 # <-- wipe 1000 commits, more might be needed... $ git pull upstream/master $ git apply mybranch.diff Then you add, commit etc, use Co-authored-by for other authors and finally $ git push --force Oscar On Thu, 13 May 2021 at 17:47, gu...@uwosh.edu <gu...@uwosh.edu> wrote: > > Having tried various versions of what Matthias suggests, I think the solution > for my case is probably what Chris suggests. > > Thanks, > Jonathan > > On Thursday, May 13, 2021 at 10:47:23 AM UTC-5 matthia...@gmail.com wrote: >> >> Hi Jonathan. >> >> On Thu, May 13, 2021 at 3:02 PM gu...@uwosh.edu wrote: >> > >> > Jisoo, >> > >> > If you can get it to work that would be great. I tried to squash >> > everything into one commit in PR #21333, but I could not get GIT to do it. >> > I'm not sure why. If you do get it to work, please let me know how. >> >> You could do that: >> >> git rebase master --interactive >> >> Your favorite editor will be opened and you can replace the "pick" >> with "squash" or "fixup" as you like. >> After saving the file and closing the editor, your new commit(s) will >> be available. >> >> This will keep the author information. >> >> HTH, >> Matthias >> >> > >> > Jonathan >> > >> > On Wednesday, May 12, 2021 at 10:53:05 PM UTC-5 JSS95 wrote: >> >> >> >> >> >> Jonathan, may I squash the commits when the PR is merged? This means that >> >> your 80 commit logs will be lost, but you will still have the credits as >> >> a co-author. >> >> >> >> Jisoo Song >> >> 2021년 5월 12일 수요일 오전 9시 24분 39초 UTC+9에 gu...@uwosh.edu님이 작성: >> >>> >> >>> > I called myself naive, in that I suppose I think it would ideally know >> >>> > that SymPy would not generate ambiguous results. One simple answer here >> >>> > might be not to supply a simple rendering of Equation(a,b) except to >> >>> > for >> >>> > use with TeX, where I suppose it would be possible to render the '=' in >> >>> > a larger size, or different colour. >> >>> >> >>> > Imagine what would happen if someone cut and pasted an Equation object >> >>> > rendered using '=' to another place in the code. >> >>> >> >>> Yes, this is something I have struggled with what might work best. >> >>> Presently, SymPy latex output in a Jupyter notebook converts `*` and >> >>> `**` to more standard representations, which cannot be copied and pasted >> >>> into code. The programmer solution is to assign the expression to a name >> >>> and use that name where you want the code version. This works equally >> >>> well for the Eqn object. I would still like to be able to copy and paste >> >>> from the output, which means we may want something like what Sagemath >> >>> used to do, which allowed you to toggle between latex and code view. I >> >>> think that capability went away in the Jupyter compatible version, but >> >>> have not tested it recently. >> >>> >> >>> I agree that when Latex output is not used the output should probably be >> >>> in a representation that can be directly copies into code. That is an >> >>> easy change. After I grade my exams I will incorporate it into the >> >>> various versions. >> >>> >> >>> Jonathan >> >>> >> >>> On Monday, May 10, 2021 at 8:47:02 AM UTC-5 da...wrote: >> >>>> >> >>>> On 09/05/2021 23:52, wrote: >> >>>> > David, >> >>>> > >> >>>> > I do not think you are being naive. The choice of representation is to >> >>>> > keep things as close to standard mathematics as possible. However, >> >>>> > your suggestions are approaches taken by others. For example Sagemath >> >>>> > uses a==4 as the way to input and display something similar to the >> >>>> > proposed Equation type. My problem with this is that it looks like the >> >>>> > logical comparison operator in most computer languages that should >> >>>> > yield True or False. I am not sure that is very important to most >> >>>> > people doing math, but since I do both coding and math it bothers me. >> >>>> >> >>>> Well of course, even people who don't do coding will understand the >> >>>> other meaning of '=' within SymPy work. >> >>>> >> >>>> I called myself naive, in that I suppose I think it would ideally know >> >>>> that SymPy would not generate ambiguous results. One simple answer here >> >>>> might be not to supply a simple rendering of Equation(a,b) except to for >> >>>> use with TeX, where I suppose it would be possible to render the '=' in >> >>>> a larger size, or different colour. >> >>>> >> >>>> Imagine what would happen if someone cut and pasted an Equation object >> >>>> rendered using '=' to another place in the code. >> >>>> >> >>>> David >> >>>> >> >>>> >> > -- >> > You received this message because you are subscribed to the Google Groups >> > "sympy" group. >> > To unsubscribe from this group and stop receiving emails from it, send an >> > email to sympy+un...@googlegroups.com. >> > To view this discussion on the web visit >> > https://groups.google.com/d/msgid/sympy/a2aebc5c-0d80-4fe6-a906-703d42d8d128n%40googlegroups.com. > > -- > You received this message because you are subscribed to the Google Groups > "sympy" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to sympy+unsubscr...@googlegroups.com. > To view this discussion on the web visit > https://groups.google.com/d/msgid/sympy/e5781162-6a5b-4742-9192-f67ca7d9f409n%40googlegroups.com. -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to sympy+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/CAHVvXxREjsXz9dr%2BEPL6i7qLppm74ewVE8EseNyoQ2wim-eFAQ%40mail.gmail.com.
