Re: elpa.gnu.org packages requiring external packages

[Eric Abrahamsen]

Stefan Monnier <address@hidden> writes:

>>>>>     > ebdb-i18n-chn    # needs "pyim"
>>> BTW, I think Emacs comes with 99% of what's needed already (it does
>>> have a pinyin table and looking at pyim-hanzi2pinyin, there doesn't
>>> seem to be much more to it), so maybe we could add to Emacs a function
>>> equivalent to pyim-hanzi2pinyin and get rid of this dependency.
>> I'll take a look and see how close it already is. I'd be happy to add a
>> function to Emacs to do this or, if it's a real no-brainer, just add it
>> to EBDB and remove this dependency (and probably the whole package).
>
> AFAICT the pinyin table is only kept in the quail table: quail tables
> are somewhat similar to keymaps, so they work well to incrementally map
> a sequence of ASCII chars to the possible chinese characters, but for
> pyim-hanzi2pinyin we'll need to traverse this table to build a "reverse
> table" (most likely represented as a char-table).

Okay, I'll take a look. I suppose we can defer building it until first
usage, etc.

>>> IOW if we could get rid of the second dependency, then I think it would
>>> make sense to remove `helm` as a dependency (and probably just merge
>>> helm-ebdb into ebdb).  So my question here is: why do we need to call
>>> helm-marked-candidates?
>>
>> I don't quite understand -- the package is useless without the call to
>> `helm-other-buffer' (that's the whole point), so how could I remove
>> that, or fold this into ebdb?
>
> The helm-ebdb function is useless without Helm, yes.  But that doesn't
> mean that it has to appear as a dependency on the containing package.
>
> Imagine if font-lock, outline, newcomment, and indent were packaged as
> ELPA packages.  Would you want separate python-font-lock,
> python-outline, python-comment, and python-indent packages each with its
> corresponding dependency?
>
> I think it makes a lot of sense to fold helm-ebdb into the ebdb package,
> without adding `helm` as a dependency.  This way, users of Helm and EBDB
> won't have to additionally install helm-ebdb to enjoy the combination of
> the two: just by installing `ebdb` and `helm` they'll get `helm-ebdb`.

Oh I'm all in favor, but I assumed it would be a no-no from a
dependency-graph point of view. If it doesn't matter, I can fold this
back into ebdb and just use `declare-function' with "ext:" to quiet the
compiler? In which case, I might as well just do the same for
`helm-marked-candidates' and leave it as-is?

I've also got these other silly little packages -- counsel-ebdb,
company-ebdb -- etc, should those all be reabsorbed? There are an awful
lot of counsel-* packages in there already.

Sorry, I'm just trying to figure out best policy for managing this kind
of stuff.

>> I think the only way around that would be to ask the helm maintainers to
>> allow some sort of flag to be set in the alist definition, saying "these
>> actions should accept all the marked candidates".
>
> Something along these lines sounds good, yes.
>
>> I'm not too optimistic they'd do that.
>
> Why not?

My observation is that the helm maintainers are not enthusiastic about
adding complexity without a good reason, and I would guess this wouldn't
count. I've opened an issue, though.

Eric