> On 24 Apr 2025, at 07:53, Ted Lemon <[email protected]> wrote: > > What we really want is to not gratuitously add newlines. That is, to not > reflow the text. When a change is being made anyway, a newline in the middle > of the change is fine. Lines should not be too long, because if they are then > git diff output becomes incomprehensible.
If you mean that a git diff output on a command line does not wrap then this config, from a quick search [1], should fix that: > $ GIT_PAGER='' git diff > > Without a pager, the lines will wrap. > So ideally you want to flow things to a reasonable line width, but then not > reflow to try to keep that length. You add newlines when adding text that > makes lines too long, but never reflow. > > Of course, this is a hard discipline to follow. Do you have another example where long lines cause an issue? [1] https://stackoverflow.com/questions/136178/how-should-i-use-git-diff-for-long-lines Jay > > On Wed, Apr 23, 2025, at 12:30 PM, Jay Daley wrote: >> >> >> I’m really confused with this. Isn’t it true that 100% of problems with git >> identifying changes where there weren’t actually changes, are caused by >> CRLFs? In other words, isn’t the best strategy to use as few of them as >> possible and allow the tools to soft wrap for display but otherwise work >> with very long lines that would not be suitable for display? >> >> Jay >> >>> On 24 Apr 2025, at 05:20, Michael Richardson <[email protected]> wrote: >>> >>> Marc Petit-Huguenin <[email protected]> wrote: >>>> >>>> >>>> Not sure I understand the difference. Can you give an example of both? >>>> Note that in the case of RFCXML, empty lines inside a <t> elements are >>>> removed. >>> >>> >>> >>> OLPS: >>> Not sure I understand the difference. >>> Can you give an example of both? >>> Note that in the case of RFCXML, empty lines inside a <t> elements are >>> removed. >>> >>> NSNL: >>> Not sure I understand the difference. >>> Can you give an example of both? >>> Note that in the case of RFCXML, >>> empty lines inside a <t> elements are removed. >>> >>> >>> (I could make your "empty lines..." sentence longer to more clearly make the >>> point, but 80% of readers' MUA will wrap it and the point will be lost) >>> >>> >>> -- >>> ] Never tell me the odds! | ipv6 mesh >>> networks [ >>> ] Michael Richardson, Sandelman Software Works | IoT architect >>> [ >>> ] [email protected] http://www.sandelman.ca/ | ruby on rails >>> [ >>> >>> _______________________________________________ >>> rfc-interest mailing list -- [email protected] >>> To unsubscribe send an email to [email protected] >> >> >> -- >> Jay Daley >> IETF Executive Director >> [email protected] >> _______________________________________________ >> rfc-interest mailing list -- [email protected] >> To unsubscribe send an email to [email protected] >> > -- Jay Daley IETF Executive Director [email protected] _______________________________________________ rfc-interest mailing list -- [email protected] To unsubscribe send an email to [email protected]
