Re: Merging guix.el

[Alex Kost]

Ludovic Courtès (2014-08-28 16:41 +0400) wrote:

> Alex Kost <address@hidden> skribis:
>
>> Ludovic Courtès (2014-08-23 16:17 +0400) wrote:
>
> [...]
>
>>>   • Have them appropriately listed in the top-level Makefile.am (I can
>>>     help with that, if you’re not familiar.)
>>
>> Along with the small changes to top-level "Makefile.am", I made
>> "Makefile.am" in "emacs" dir and...
>
> I think it’s better to avoid recursive makefiles, which is why I
> suggested adding changes to the top-level Makefile.am.
>
> Could you make it this way?

No problem.

> More precisely, the Emacs-specific things could be kept in emacs.am, and
> that file would be included from the top-level Makefile.am.

OK, I think it would be good to make "emacs-ui" branch temporary, so
that after I'll fix everything that needs to be fixed, it may be
recreated with a nice and clean commit(s) for merging “guix.el”.  This
way I could push commits there without a fear that I mess it all up.
WDYT?

>> I imagine there may be... for example vi users, who wouldn't want to
>> install this feature, so I made some changes in "configure.ac" to add
>> “--disable-emacs-ui” option.
>
> It seems that the only things that cannot be done when Emacs is not
> available is the generation of the autoloads file, right?

Yes.

> Then, what about adding $(AUTOLOADS) to the distribution?  It would just
> need to be appended to dist_lisp_DATA, and not added to CLEANFILES.

OK, will be done.  And what about "configure.ac"?  I thought a new
configure option is good.  Should I delete it?

> Nitpick: could you use makefile-backslash-region for the $(AUTOLOADS)
> recipe?

Yes, sorry.

>> Also I use almost the same code in "guix-helper.scm.in" that is used in
>> "scripts/guix.in", so I think it will be good to have some little
>> additional module with ‘config-lookup’ function.  WDYT?
>
> It cannot be in a module, because at this point the module location
> isn’t known yet.  I don’t really know how to factorize it, so I propose
> to leave it for later, with a FIXME.  Maybe Mark has an idea?
>
> +(define %current-manifest)
> +(define current-manifest-entries-table)
> +(define packages)
> +(define packages-table)
>
> I didn’t know this was possible, but we shouldn’t rely on it.

Do you mean definitions without initial values?

> +(define name+version->key cons)
> +(define (key->name+version key)
> +  (values (car key) (cdr key)))
>
> I would find it easier to read if it ‘cons’ and ‘car+cdr’ (from SRFI-1)
> were used directly.

Thanks, I didn't know about that.

> +(define* (set-current-manifest-maybe! #:optional manifest)
> +  (define (manifest-entries->hash-table entries)
> +    (let ((entries-table (make-hash-table (length entries))))
> +      (map (lambda (entry)
> +             (let* ((key (name+version->key
> +                          (manifest-entry-name entry)
> +                          (manifest-entry-version entry)))
> +                    (ref (hash-ref entries-table key)))
> +               (hash-set! entries-table key
> +                          (if ref (cons entry ref) (list entry)))))
> +           entries)
> +      entries-table))
> +
> +  (let ((manifest (or manifest (profile-manifest %user-profile))))
> +    (unless (and (manifest? %current-manifest)
> +                 (equal? manifest %current-manifest))
> +      (set! %current-manifest manifest)
> +      (set! current-manifest-entries-table
> +            (manifest-entries->hash-table
> +             (manifest-entries manifest))))))
>
> Wouldn’t it be enough to pass the current manifest as an argument to the
> various functions, instead of defining a global variable?

Most likely, I'll fix it.

> Also, my understanding is that ‘current-manifest-entries-table’ is here
> to speed up lookups in ‘manifest-entries-by-name+version’, right?

Yes, ‘current-manifest-entries-table’ and ‘packages-table’ are there to
speed up the process of finding “manifest entries”/“packages” by
name+version (it is a very general need).  Also
‘current-manifest-entries-table’ is used in ‘fold-manifest-entries’.

> Then, I think this optimization should go into (guix profiles):
> <manifest> objects would carry that vhash, and ‘manifest-installed?’
> etc. would make use of it.  The constructor would be changed along these
> lines:
>
> diff --git a/guix/profiles.scm b/guix/profiles.scm
> index a2c73fd..98eb814 100644
> --- a/guix/profiles.scm
> +++ b/guix/profiles.scm
> @@ -84,9 +84,17 @@
>  ;;;
>  
>  (define-record-type <manifest>
> -  (manifest entries)
> +  (%manifest entries name->entry)
>    manifest?
> -  (entries manifest-entries))                     ; list of <manifest-entry>
> +  (entries     manifest-entries)         ; list of <manifest-entry>
> +  (name->entry manifest-name->entry))    ; vhash [string -> <manifest-entry>]
> +
> +(define (manifest entries)
> +  (%manifest entries
> +             (fold (lambda (entry result)
> +                     (vhash-cons ... result))
> +                   vlist-null
> +                   entries)))
>  
>  ;; Convenient alias, to avoid name clashes.
>  (define make-manifest manifest)
>
> WDYT?

I think it's a good idea, but if that "name" is just a package name,
I can't use this optimization: I need to define entries by
"name+version".  Also I think it should be ‘name->entries’ as there can
be several manifest entries with the same name (or name/version).

> +(define (set-packages!)
> +  (let ((count 0))
> +    (set! packages
> +          (fold-packages (lambda (pkg res)
> +                           (set! count (+ 1 count))
> +                           (vhash-consq (object-address pkg) pkg res))
> +                         vlist-null))
> +    (set! packages-table (make-hash-table count))
> +    (vlist-for-each (lambda (elem)
> +                      (let* ((pkg (cdr elem))
> +                             (key (name+version->key
> +                                   (package-name pkg)
> +                                   (package-version pkg)))
> +                             (ref (hash-ref packages-table key)))
> +                        (hash-set! packages-table key
> +                                   (if ref (cons pkg ref) (list pkg)))))
> +                    packages)))
> +
> +(set-packages!)
>
> Given that ‘set-packages!’ has only on call site, what about removing
> it, and instead writing directly:
>
>   (define %packages
>     (fold-packages ... vlist-null))
>
>   (define %package-count
>     (length %packages))
>
>   (define %package-table
>     (vlist-fold ...))
>
> It’s also best to prefix global variable names with ‘%’.

Yes, it's definitely better, except I don't know how to fill a table
with ‘vlist-fold’ and I don't see a better variant than this:

(define %package-table
  (let ((table (make-hash-table %package-count)))
    (vlist-for-each (lambda (elem)
                      (let* ((pkg (cdr elem))
                             (key (name+version->key
                                   (package-name pkg)
                                   (package-version pkg)))
                             (ref (hash-ref table key)))
                        (hash-set! table key
                                   (if ref (cons pkg ref) (list pkg)))))
                    packages)
    table))

> I should point out that we try to stick to functional style (in
> host-side code at least.)  Thus any identifier with an exclamation mark
> gives you a -1 during review.  ;-)  See “Coding Style” in HACKING.

My bad, I'll fix it, thanks.


P.S.  OMG!  Thank you very much for reviewing all that crap.  Elisp
code is much better (I hope :-)).

--
Alex