Re: Unify subcommand structure; introduce double dashes for all subcommands?
On 23.07.2014 19:52, Junio C Hamano wrote: Stefan Beller stefanbel...@gmail.com writes: A git command is generally setup as: git command [subcommand] [options] ... The subcommands vary wildly by the nature of the command. However all subcommands could at least follow one style. The commands bundle, notes, stash and submodule have subcommands without two leading dashes (i.e. git stash list) as opposed to all other commands (i.e. git tag --list). Sounds familiar. E.g. here is a similar thread about a year ago. http://thread.gmane.org/gmane.comp.version-control.git/231376/focus=231478 Further discussions to make the plan more concrete is very much welcomed. Thanks. So I'd want to add have the subcommands without double dashes ideally. It's just less typing and more english like, not cryptic with dashes. This may come handy for newcomers/beginners using git. It also benefits the powerusers of git as you spare the typing of 2 dashes on the subcommand. I'm currently trying to understand the functions to parse options and I wonder if the following is possible: If there is an option set by OPT_CMDMODE (such as git tag --delete / --verify) we want to add an internal option, (such as PARSE_OPT_NODASH ?), that you can deliver this CMD_MODE option without leading 2 dashes. The current behavior with leading 2 dashes is still supported, but maybe a warning is printed about deprecating the option with 2 dashes. Having this change in place will switch over revert, replace, tag and merge-base to having sane subcommands without dashes possible. Once this change is in, we can rewrite the other commands, which as of now don't require the dashes. Coincidentally these commands heavily rely on option parsing themselves, such as git-notes having this code in place: if (argc 1 || !strcmp(argv[0], list)) result = list(argc, argv, prefix); else if (!strcmp(argv[0], add)) result = add(argc, argv, prefix); else if (!strcmp(argv[0], copy)) result = copy(argc, argv, prefix); ... Does that make sense? Stefan -- To unsubscribe from this list: send the line unsubscribe git in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Unify subcommand structure; introduce double dashes for all subcommands?
Stefan Beller stefanbel...@gmail.com writes: On 23.07.2014 19:52, Junio C Hamano wrote: Sounds familiar. E.g. here is a similar thread about a year ago. http://thread.gmane.org/gmane.comp.version-control.git/231376/focus=231478 Further discussions to make the plan more concrete is very much welcomed. Thanks. So I'd want to add have the subcommands without double dashes ideally. That is not ideal at all, I am afraid. A command that started only with its primary operating mode, e.g. git tag [-s|-a] tagname [object], may have to gain I do not want to create, I just want to list and the way to signal that has to be an option that cannot be mistaken as its valid first argument (to avoid git tag list that cannot create a tag called list, we use git tag --list). You could add an entirely new command git foo that always takes the command-mode word, i.e. git foo mode$n args, but you will be typing the operating mode name all the time only to save --mode$n for 2=$n, which may not be a good economy in the end. Please do not go there. -- To unsubscribe from this list: send the line unsubscribe git in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Unify subcommand structure; introduce double dashes for all subcommands?
On 10.08.2014 20:13, Junio C Hamano wrote: Stefan Beller stefanbel...@gmail.com writes: On 23.07.2014 19:52, Junio C Hamano wrote: Sounds familiar. E.g. here is a similar thread about a year ago. http://thread.gmane.org/gmane.comp.version-control.git/231376/focus=231478 Further discussions to make the plan more concrete is very much welcomed. Thanks. So I'd want to add have the subcommands without double dashes ideally. That is not ideal at all, I am afraid. A command that started only with its primary operating mode, e.g. git tag [-s|-a] tagname [object], may have to gain I do not want to create, I just want to list and the way to signal that has to be an option that cannot be mistaken as its valid first argument (to avoid git tag list that cannot create a tag called list, we use git tag --list). You could add an entirely new command git foo that always takes the command-mode word, i.e. git foo mode$n args, but you will be typing the operating mode name all the time only to save --mode$n for 2=$n, which may not be a good economy in the end. Please do not go there. I see your point. However how often do you really want to create a tag called list? As of now it's easy: git tag list and for listing all tags you'd need to type: git tag --list and if you want to create a tag called --list, I'd assume you'd go git tag -- --list # However: fatal: '--list' is not a valid tag name. So even as of now certain tag names cannot be done easily. Also you have to type two more dashes for an action you'd probably want to perform more often (as opposed to creating a tag 'list') In my (ideal) world we'd rather have this behavior: git tag list # behaves the same as git tag Now creating a tag called 'list' is not as easy, because 'list' is a primary operating mode name, so we need to tell git we're actually meaning the name as opposed to the operating mode: git tag create -- list # or even git tag create -- --list Anyways despite my arguing, it seems you rather want to rather have the leading double dashes everywhere for the operating modes? So the plan is to not touch the parsing, but to adjust notes and stash ? Thanks, Stefan -- To unsubscribe from this list: send the line unsubscribe git in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Unify subcommand structure; introduce double dashes for all subcommands?
In the user survey 2012 question 23 (In your opinion, which areas in Git need improvement?), the most crucial point identified was the user interface. I wonder if there are any more recent surveys, showing if this has changed. Now when we want to improve the user interface, we're likely talking about the porcelain commands only, as that's what most users perceive as the commandline user interface. A git command is generally setup as: git command [subcommand] [options] ... The subcommands vary wildly by the nature of the command. However all subcommands could at least follow one style. The commands bundle, notes, stash and submodule have subcommands without two leading dashes (i.e. git stash list) as opposed to all other commands (i.e. git tag --list). So my proposal is to unify the structure of the subcommands to either have always leading dashes or never. This would need a longterm thinking of course (e.g. introduce new options with(out) dashes, but support existing commands until git 3.0 or such, then drop them.) Was there a discussion about this topic already? I'd like to read on these discussions, if any. I could think about the following points being interesting * user interface (what is more appealing to a user?) * ease of transition (Is it really worth it? How long does it take to pay off?) * ease of implementation (Could we reuse the option parser already in place for the double-dashed subcommands, i.e. have less LoC) * error-proneness of the transition Thanks, Stefan -- To unsubscribe from this list: send the line unsubscribe git in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Unify subcommand structure; introduce double dashes for all subcommands?
Stefan Beller stefanbel...@gmail.com writes: A git command is generally setup as: git command [subcommand] [options] ... The subcommands vary wildly by the nature of the command. However all subcommands could at least follow one style. The commands bundle, notes, stash and submodule have subcommands without two leading dashes (i.e. git stash list) as opposed to all other commands (i.e. git tag --list). Sounds familiar. E.g. here is a similar thread about a year ago. http://thread.gmane.org/gmane.comp.version-control.git/231376/focus=231478 Further discussions to make the plan more concrete is very much welcomed. Thanks. -- To unsubscribe from this list: send the line unsubscribe git in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
