Re: [PATCH] show: new extension for displaying various repository data
On Sun, 12 Mar 2017 21:38:00 -0700, Gregory Szorc wrote:
> # HG changeset patch
> # User Gregory Szorc
> # Date 1489378362 25200
> # Sun Mar 12 21:12:42 2017 -0700
> # Node ID d30057d358076cbe7d632cd573095af97543f932
> # Parent 1c3352d7eaf24533ad52d4b8a024211e9189fb0b
> show: new extension for displaying various repository data
The idea sounds nice to me. I just checked minor implementation details
about formatter.
> +@command('show', commands.formatteropts, _('[VIEW]'))
> +def show(ui, repo, view=None, template=None):
> +"""show various repository information
> +
> +A requested view of repository data is displayed.
> +
> +If no view is requested, the list of available views is shown.
> +
> +.. note::
> +
> + The default output from this command is not covered under Mercurial's
> + default backwards-compatible mechanism (which puts an emphasis on
> + not changing behavior). This means output from this command may change
> + in any future release. However, the values fed to the formatter are
> + covered under the default backwards-compatible mechanism.
> +
> + Automated consumers of this command should specify an explicit
> template
> + via ``-T/--template`` to guarantee output is stable.
> +
> +List of available views:
> +
> +"""
> +views = showview._table
> +
> +if not view:
> +ui.write(_('available views:\n'))
> +ui.write('\n')
> +
> +for name, func in sorted(views.items()):
> +ui.write(_('%s\n') % func.__doc__)
> +
> +ui.write('\n')
> +raise error.Abort(_('no view requested'),
> + hint=_('use `hg show ` to choose a view'))
> +
> +if view not in views:
> +raise error.Abort(_('unknown view: %s') % view,
> + hint=_('run `hg show` to see available views'))
> +
> +template = template or 'show'
> +fmtopic = views[view]._formatter
> +formatter = ui.formatter(fmtopic, {'template': template})
> +
> +return views[view](ui, repo, formatter)
s/formatter/fm/ seems better for consistency. Also, I prefer calling fm.end()
here as it should be paired with the construction.
with ui.formatter(...) as fm:
return views[view](...)
> +@showview('bookmarks', formatter='bookmarks')
s/formatter=/topic=/ or /fmtopic=/ ?
If this 'bookmarks' template can't be compatible with the default template
used by "hg bookmarks" command, we'll need another topic name.
> +def showbookmarks(ui, repo, fm):
> +"""bookmarks and their associated changeset"""
> +marks = repo._bookmarks
> +if not len(marks):
> +ui.write_err('(no bookmarks set)\n')
> +return 0
If this is an error, it should error out. Otherwise, use fm.plain().
> +active = repo._activebookmark
> +longest = max(len(b) for b in marks)
> +
> +for bm, node in sorted(marks.items()):
> +fm.startitem()
> +fm.write('bookmark', '%s', bm)
> +fm.write('node', fm.hexfunc(node), fm.hexfunc(node))
> +fm.data(ctx=repo[node],
> +active=bm == active,
> +longestlen=longest)
As changectx can't be serialized, it shouldn't be passed to e.g. jsonformatter.
fm.context(ctx=repo[node]) can be used. It's unfortunate that 'longestlen'
appears in JSON output, but I have no better idea other than abusing
fm.context().
___
Mercurial-devel mailing list
Mercurial-devel@mercurial-scm.org
https://www.mercurial-scm.org/mailman/listinfo/mercurial-devel
Re: [PATCH] show: new extension for displaying various repository data
On Mon, Mar 13, 2017 at 7:21 PM, Yuya Nishihara wrote:
> On Sun, 12 Mar 2017 21:38:00 -0700, Gregory Szorc wrote:
> > # HG changeset patch
> > # User Gregory Szorc
> > # Date 1489378362 25200
> > # Sun Mar 12 21:12:42 2017 -0700
> > # Node ID d30057d358076cbe7d632cd573095af97543f932
> > # Parent 1c3352d7eaf24533ad52d4b8a024211e9189fb0b
> > show: new extension for displaying various repository data
>
> The idea sounds nice to me. I just checked minor implementation details
> about formatter.
>
> > +@command('show', commands.formatteropts, _('[VIEW]'))
> > +def show(ui, repo, view=None, template=None):
> > +"""show various repository information
> > +
> > +A requested view of repository data is displayed.
> > +
> > +If no view is requested, the list of available views is shown.
> > +
> > +.. note::
> > +
> > + The default output from this command is not covered under
> Mercurial's
> > + default backwards-compatible mechanism (which puts an emphasis on
> > + not changing behavior). This means output from this command may
> change
> > + in any future release. However, the values fed to the formatter
> are
> > + covered under the default backwards-compatible mechanism.
> > +
> > + Automated consumers of this command should specify an explicit
> template
> > + via ``-T/--template`` to guarantee output is stable.
> > +
> > +List of available views:
> > +
> > +"""
> > +views = showview._table
> > +
> > +if not view:
> > +ui.write(_('available views:\n'))
> > +ui.write('\n')
> > +
> > +for name, func in sorted(views.items()):
> > +ui.write(_('%s\n') % func.__doc__)
> > +
> > +ui.write('\n')
> > +raise error.Abort(_('no view requested'),
> > + hint=_('use `hg show ` to choose a
> view'))
> > +
> > +if view not in views:
> > +raise error.Abort(_('unknown view: %s') % view,
> > + hint=_('run `hg show` to see available
> views'))
> > +
> > +template = template or 'show'
> > +fmtopic = views[view]._formatter
> > +formatter = ui.formatter(fmtopic, {'template': template})
> > +
> > +return views[view](ui, repo, formatter)
>
> s/formatter/fm/ seems better for consistency. Also, I prefer calling
> fm.end()
> here as it should be paired with the construction.
>
> with ui.formatter(...) as fm:
> return views[view](...)
>
> > +@showview('bookmarks', formatter='bookmarks')
>
> s/formatter=/topic=/ or /fmtopic=/ ?
>
> If this 'bookmarks' template can't be compatible with the default template
> used by "hg bookmarks" command, we'll need another topic name.
>
> > +def showbookmarks(ui, repo, fm):
> > +"""bookmarks and their associated changeset"""
> > +marks = repo._bookmarks
> > +if not len(marks):
> > +ui.write_err('(no bookmarks set)\n')
> > +return 0
>
> If this is an error, it should error out. Otherwise, use fm.plain().
>
fm.plain() no-ops when using templateformatter though.
I'm not sure what to do here. We probably want a message printed to the
user without an error exit code. So I'm going to keep this as ui.write() in
V2 and you can suggest a workaround.
>
> > +active = repo._activebookmark
> > +longest = max(len(b) for b in marks)
> > +
> > +for bm, node in sorted(marks.items()):
> > +fm.startitem()
> > +fm.write('bookmark', '%s', bm)
> > +fm.write('node', fm.hexfunc(node), fm.hexfunc(node))
> > +fm.data(ctx=repo[node],
> > +active=bm == active,
> > +longestlen=longest)
>
> As changectx can't be serialized, it shouldn't be passed to e.g.
> jsonformatter.
> fm.context(ctx=repo[node]) can be used. It's unfortunate that 'longestlen'
> appears in JSON output, but I have no better idea other than abusing
> fm.context().
>
I have an idea. I may test the waters in V2.
___
Mercurial-devel mailing list
Mercurial-devel@mercurial-scm.org
https://www.mercurial-scm.org/mailman/listinfo/mercurial-devel
Re: [PATCH] show: new extension for displaying various repository data
Yuya Nishihara writes: > On Sun, 12 Mar 2017 21:38:00 -0700, Gregory Szorc wrote: >> # HG changeset patch >> # User Gregory Szorc >> # Date 1489378362 25200 >> # Sun Mar 12 21:12:42 2017 -0700 >> # Node ID d30057d358076cbe7d632cd573095af97543f932 >> # Parent 1c3352d7eaf24533ad52d4b8a024211e9189fb0b >> show: new extension for displaying various repository data > > The idea sounds nice to me. I just checked minor implementation details > about formatter. Just a quick reply (as I whittle down my backlog), but a lot of people (including myself) have a 'show' alias (inspired by 'git show'). That may or may not be a factor in this. ___ Mercurial-devel mailing list Mercurial-devel@mercurial-scm.org https://www.mercurial-scm.org/mailman/listinfo/mercurial-devel
Re: [PATCH] show: new extension for displaying various repository data
On 3/22/17 7:29 AM, Sean Farley wrote: Yuya Nishihara writes: On Sun, 12 Mar 2017 21:38:00 -0700, Gregory Szorc wrote: # HG changeset patch # User Gregory Szorc # Date 1489378362 25200 # Sun Mar 12 21:12:42 2017 -0700 # Node ID d30057d358076cbe7d632cd573095af97543f932 # Parent 1c3352d7eaf24533ad52d4b8a024211e9189fb0b show: new extension for displaying various repository data The idea sounds nice to me. I just checked minor implementation details about formatter. Just a quick reply (as I whittle down my backlog), but a lot of people (including myself) have a 'show' alias (inspired by 'git show'). That may or may not be a factor in this. Greg called this out specifically in his excellent summary. FB also has a "show" extension that replaced our "show" alias: https://bitbucket.org/facebook/hg-experimental/src/default/hgext3rd/show.py Therefore, I would also slightly prefer "view", but I admit I don't care about hgk (even though we have it on at FB and it doesn't seem to work at all...) I'll respond to the original as well so I can respond inline to the code and summary. ___ Mercurial-devel mailing list Mercurial-devel@mercurial-scm.org https://www.mercurial-scm.org/mailman/listinfo/mercurial-devel
Re: [PATCH] show: new extension for displaying various repository data
> On Mar 22, 2017, at 04:51, Ryan McElroy wrote: > > > >> On 3/22/17 7:29 AM, Sean Farley wrote: >> Yuya Nishihara writes: >> On Sun, 12 Mar 2017 21:38:00 -0700, Gregory Szorc wrote: # HG changeset patch # User Gregory Szorc # Date 1489378362 25200 # Sun Mar 12 21:12:42 2017 -0700 # Node ID d30057d358076cbe7d632cd573095af97543f932 # Parent 1c3352d7eaf24533ad52d4b8a024211e9189fb0b show: new extension for displaying various repository data >>> The idea sounds nice to me. I just checked minor implementation details >>> about formatter. >> Just a quick reply (as I whittle down my backlog), but a lot of people >> (including myself) have a 'show' alias (inspired by 'git show'). >> >> That may or may not be a factor in this. > > Greg called this out specifically in his excellent summary. > > FB also has a "show" extension that replaced our "show" alias: > https://bitbucket.org/facebook/hg-experimental/src/default/hgext3rd/show.py At Paris, people from Facebook mildly objected to "show" because of the conflict. At Mountain View, Durham said to just use "show" and FB would deal with it. I myself was convinced to use "display" after Paris and that's what I initially implemented. Justification for "display" was summarized in that RFC patch's commit message. However, enough people said the reasons weren't strong enough and Durham gave me a green light, so I changed to my preference: show. Anyway, the command and extension name is a bikeshed and can be changed before 4.2. I'd prefer review focus on the behavior so we can get something landed and iterate on the feature. I *really* want a smartlog/wip feature in the core distribution and this extension/command is where it will live. > > Therefore, I would also slightly prefer "view", but I admit I don't care > about hgk (even though we have it on at FB and it doesn't seem to work at > all...) > > I'll respond to the original as well so I can respond inline to the code and > summary. ___ Mercurial-devel mailing list Mercurial-devel@mercurial-scm.org https://www.mercurial-scm.org/mailman/listinfo/mercurial-devel
Re: [PATCH] show: new extension for displaying various repository data
Ryan McElroy a écrit : On 3/22/17 7:29 AM, Sean Farley wrote: Yuya Nishihara writes: On Sun, 12 Mar 2017 21:38:00 -0700, Gregory Szorc wrote: # HG changeset patch # User Gregory Szorc # Date 1489378362 25200 # Sun Mar 12 21:12:42 2017 -0700 # Node ID d30057d358076cbe7d632cd573095af97543f932 # Parent 1c3352d7eaf24533ad52d4b8a024211e9189fb0b show: new extension for displaying various repository data The idea sounds nice to me. I just checked minor implementation details about formatter. Just a quick reply (as I whittle down my backlog), but a lot of people (including myself) have a 'show' alias (inspired by 'git show'). That may or may not be a factor in this. Greg called this out specifically in his excellent summary. FB also has a "show" extension that replaced our "show" alias: https://bitbucket.org/facebook/hg-experimental/src/default/hgext3rd/show.py It seems to me that all "show" extensions/aliases can be replaced by the proposed extension. Also if we look to the future, it seems to me that we'd have hard time to explain that this command is not called "show" because there used to be extensions/aliases with this name and that got replaced by said command. ___ Mercurial-devel mailing list Mercurial-devel@mercurial-scm.org https://www.mercurial-scm.org/mailman/listinfo/mercurial-devel
Re: [PATCH] show: new extension for displaying various repository data
On 3/22/17 8:29 AM, Gregory Szorc wrote: On Mar 22, 2017, at 04:51, Ryan McElroy wrote: On 3/22/17 7:29 AM, Sean Farley wrote: Yuya Nishihara writes: On Sun, 12 Mar 2017 21:38:00 -0700, Gregory Szorc wrote: # HG changeset patch # User Gregory Szorc # Date 1489378362 25200 # Sun Mar 12 21:12:42 2017 -0700 # Node ID d30057d358076cbe7d632cd573095af97543f932 # Parent 1c3352d7eaf24533ad52d4b8a024211e9189fb0b show: new extension for displaying various repository data The idea sounds nice to me. I just checked minor implementation details about formatter. Just a quick reply (as I whittle down my backlog), but a lot of people (including myself) have a 'show' alias (inspired by 'git show'). That may or may not be a factor in this. Greg called this out specifically in his excellent summary. FB also has a "show" extension that replaced our "show" alias: https://urldefense.proofpoint.com/v2/url?u=https-3A__bitbucket.org_facebook_hg-2Dexperimental_src_default_hgext3rd_show.py&d=DwIGaQ&c=5VD0RTtNlTh3ycd41b3MUw&r=nuarHzhP1wi1T9iURRCj1A&m=KZNXs3QMg2M7YtCvXYwgouTxwTN2WFhsYLOLHtLfE2Y&s=qWD0RFF73TrSukNubnCJl-EdsGcgjpl5olyQNX88XQ4&e= At Paris, people from Facebook mildly objected to "show" because of the conflict. At Mountain View, Durham said to just use "show" and FB would deal with it. I myself was convinced to use "display" after Paris and that's what I initially implemented. Justification for "display" was summarized in that RFC patch's commit message. However, enough people said the reasons weren't strong enough and Durham gave me a green light, so I changed to my preference: show. Anyway, the command and extension name is a bikeshed and can be changed before 4.2. I'd prefer review focus on the behavior so we can get something landed and iterate on the feature. I *really* want a smartlog/wip feature in the core distribution and this extension/command is where it will live. I'd just pick what makes the most sense for Mercurial users, and not worry about what people have aliased, etc. The conflict with 'git show' is a valid concern though. For 'hg view', you could argue that hgk is just one implementation of this new 'view' paradigm, and maybe it could change it's command line to accept the same bookmark/smartlog/etc inputs as the proposed 'hg view/show'. That might let us justify using 'hg view' ___ Mercurial-devel mailing list Mercurial-devel@mercurial-scm.org https://www.mercurial-scm.org/mailman/listinfo/mercurial-devel
Re: [PATCH] show: new extension for displaying various repository data
Excerpts from Denis Laxalde's message of 2017-03-23 10:14:14 +0100: > Ryan McElroy a écrit : > > On 3/22/17 7:29 AM, Sean Farley wrote: > >> Yuya Nishihara writes: > >> > >>> On Sun, 12 Mar 2017 21:38:00 -0700, Gregory Szorc wrote: > # HG changeset patch > # User Gregory Szorc > # Date 1489378362 25200 > # Sun Mar 12 21:12:42 2017 -0700 > # Node ID d30057d358076cbe7d632cd573095af97543f932 > # Parent 1c3352d7eaf24533ad52d4b8a024211e9189fb0b > show: new extension for displaying various repository data > >>> The idea sounds nice to me. I just checked minor implementation details > >>> about formatter. > >> Just a quick reply (as I whittle down my backlog), but a lot of people > >> (including myself) have a 'show' alias (inspired by 'git show'). > >> > >> That may or may not be a factor in this. > > > > Greg called this out specifically in his excellent summary. > > > > FB also has a "show" extension that replaced our "show" alias: > > https://bitbucket.org/facebook/hg-experimental/src/default/hgext3rd/show.py > > It seems to me that all "show" extensions/aliases can be replaced by the > proposed extension. Also if we look to the future, it seems to me that The problem is top-level command names are rare resources. It's really hard to name things. I think subcommands is a great choice. > we'd have hard time to explain that this command is not called "show" > because there used to be extensions/aliases with this name and that got > replaced by said command. Another benefit of the "show" approach is it's supposed to be read-only. It feels cleaner than for example "hg bookmark" which could do either read or write. ___ Mercurial-devel mailing list Mercurial-devel@mercurial-scm.org https://www.mercurial-scm.org/mailman/listinfo/mercurial-devel
Re: [PATCH] show: new extension for displaying various repository data
Excerpts from Ryan McElroy's message of 2017-03-22 11:51:16 +: > > On 3/22/17 7:29 AM, Sean Farley wrote: > > Yuya Nishihara writes: > > > >> On Sun, 12 Mar 2017 21:38:00 -0700, Gregory Szorc wrote: > >>> # HG changeset patch > >>> # User Gregory Szorc > >>> # Date 1489378362 25200 > >>> # Sun Mar 12 21:12:42 2017 -0700 > >>> # Node ID d30057d358076cbe7d632cd573095af97543f932 > >>> # Parent 1c3352d7eaf24533ad52d4b8a024211e9189fb0b > >>> show: new extension for displaying various repository data > >> The idea sounds nice to me. I just checked minor implementation details > >> about formatter. > > Just a quick reply (as I whittle down my backlog), but a lot of people > > (including myself) have a 'show' alias (inspired by 'git show'). > > > > That may or may not be a factor in this. > > Greg called this out specifically in his excellent summary. > > FB also has a "show" extension that replaced our "show" alias: > https://bitbucket.org/facebook/hg-experimental/src/default/hgext3rd/show.py > > Therefore, I would also slightly prefer "view", but I admit I don't care > about hgk (even though we have it on at FB and it doesn't seem to work > at all...) "show" is not a problem because it supports subcommands. It we make the default subcommand to be "show commit" then there is no BC. > > I'll respond to the original as well so I can respond inline to the code > and summary. ___ Mercurial-devel mailing list Mercurial-devel@mercurial-scm.org https://www.mercurial-scm.org/mailman/listinfo/mercurial-devel
