* bug#44614: 26.3; bibtex convert case undefined
@ 2020-11-13 11:05 Francesco Potortì
2020-11-14 16:33 ` Lars Ingebrigtsen
0 siblings, 1 reply; 4+ messages in thread
From: Francesco Potortì @ 2020-11-13 11:05 UTC (permalink / raw)
To: 44614
GNU Emacs 26.3 (build 1, x86_64-pc-linux-gnu, X toolkit, Xaw3d scroll bars) of 2020-05-17, modified by Debian
bibtex.el offers the opportunity of unifying the case of fields (see the
string unify-case). Unfortunately the docs and comments do not tell
what is the chosen case. In practice, I see that entries are downcased.
An old version of bibtex.el (last copyright 1997) uniquified cases based
on the bibtex-unify-case-convert variable, which is now removed.
This is unfortunate, and in my opinion a bug.
Looking at the old code, I see that the current behaviour is the same as
the old 'identity value for the mentioned variable. I am working on a
local patch for reverting to the old behaviour. After I finish, I can
provide it, if there is any hope of it to be accepted.
Here is the old definition:
(defcustom bibtex-unify-case-convert 'downcase
"*Function called when unifying case on entry and field names.
This variable is buffer-local."
:group 'bibtex
:type '(choice (const :tag "Same case as in `bibtex-field-alist'" identity)
(const :tag "Downcase" downcase)
(const :tag "Capitalize" capitalize)
(const :tag "Upcase" upcase)
(function :tag "Conversion function")))
(make-variable-buffer-local 'bibtex-unify-case-convert)
^ permalink raw reply [flat|nested] 4+ messages in thread
* bug#44614: 26.3; bibtex convert case undefined
2020-11-13 11:05 bug#44614: 26.3; bibtex convert case undefined Francesco Potortì
@ 2020-11-14 16:33 ` Lars Ingebrigtsen
2020-11-14 18:47 ` Francesco Potortì
0 siblings, 1 reply; 4+ messages in thread
From: Lars Ingebrigtsen @ 2020-11-14 16:33 UTC (permalink / raw)
To: Francesco Potortì; +Cc: 44614
Francesco Potortì <pot@gnu.org> writes:
> An old version of bibtex.el (last copyright 1997) uniquified cases based
> on the bibtex-unify-case-convert variable, which is now removed.
>
> This is unfortunate, and in my opinion a bug.
I tried to poke around to see whether there's any reason the variable
was removed, but I can't find any mention of it in the Emacs tree, so
it's possible that the in-tree version of bibtex.el never had the
variable?
> Looking at the old code, I see that the current behaviour is the same as
> the old 'identity value for the mentioned variable. I am working on a
> local patch for reverting to the old behaviour. After I finish, I can
> provide it, if there is any hope of it to be accepted.
>
> Here is the old definition:
>
> (defcustom bibtex-unify-case-convert 'downcase
> "*Function called when unifying case on entry and field names.
> This variable is buffer-local."
Sure; I think (re-)introducing such a variable would make sense.
--
(domestic pets only, the antidote for overdose, milk.)
bloggy blog: http://lars.ingebrigtsen.no
^ permalink raw reply [flat|nested] 4+ messages in thread
* bug#44614: 26.3; bibtex convert case undefined
2020-11-14 16:33 ` Lars Ingebrigtsen
@ 2020-11-14 18:47 ` Francesco Potortì
2020-11-16 21:36 ` Lars Ingebrigtsen
0 siblings, 1 reply; 4+ messages in thread
From: Francesco Potortì @ 2020-11-14 18:47 UTC (permalink / raw)
To: Lars Ingebrigtsen; +Cc: 44614
>> An old version of bibtex.el (last copyright 1997) uniquified cases based
>> on the bibtex-unify-case-convert variable, which is now removed.
>>
>> This is unfortunate, and in my opinion a bug.
>
>I tried to poke around to see whether there's any reason the variable
>was removed, but I can't find any mention of it in the Emacs tree, so
>it's possible that the in-tree version of bibtex.el never had the
>variable?
Mh. I found a very old version that i had set aside with the name
bibtex.el.orig, which does not contain that variable, so it may well be
that I had introduced that variable myself as a local change with the
aim of pushing it in the Emacs tree but then I forgot to do that. It
was a long time ago, so I may well have forgotten.
Anyway, here is my patch, which I think does not make any harm. It
changes the behaviour slightly in what I consider it being a more
consistent way. When unifying is requested, in the default case, which
is 'identity, if an unknown field is found it is not unified (rather
than being downcased as it is now): only known fields are unified to the
known case.
Note that the last hunk is only a change of 'if' with 'when', it is big
because of reindentation and just a matter of aesthetics (or clarity).
*** bibtex-2019.el 2020-11-13 11:54:34.000000000 +0100
--- bibtex.el 2020-11-14 19:05:48.000000000 +0100
***************
*** 89,92 ****
--- 89,103 ----
(put 'bibtex-include-OPTkey 'risky-local-variable t)
+ (defcustom bibtex-unify-case-convert 'identity
+ "*Function called when unifying case on entry and field names.
+ This variable is buffer-local."
+ :group 'bibtex
+ :type '(choice (const :tag "Same case as in `bibtex-field-alist'" identity)
+ (const :tag "Downcase" downcase)
+ (const :tag "Capitalize" capitalize)
+ (const :tag "Upcase" upcase)
+ (function :tag "Conversion function")))
+ (make-variable-buffer-local 'bibtex-unify-case-convert)
+
(defcustom bibtex-user-optional-fields
'(("annote" "Personal annotation (ignored)"))
***************
*** 123,127 ****
delimiters Change delimiters according to variables
`bibtex-field-delimiters' and `bibtex-entry-delimiters'.
! unify-case Change case of entry types and field names.
braces Enclose parts of field entries by braces according to
`bibtex-field-braces-alist'.
--- 134,139 ----
delimiters Change delimiters according to variables
`bibtex-field-delimiters' and `bibtex-entry-delimiters'.
! unify-case Change case of entry and field names according to
! `bibtex-unify-case-convert'.
braces Enclose parts of field entries by braces according to
`bibtex-field-braces-alist'.
***************
*** 2309,2313 ****
(when (memq 'unify-case format)
(delete-region beg-type end-type)
! (insert (car entry-list)))
;; update left entry delimiter
--- 2321,2325 ----
(when (memq 'unify-case format)
(delete-region beg-type end-type)
! (insert (funcall bibtex-unify-case-convert (car entry-list))))
;; update left entry delimiter
***************
*** 2510,2523 ****
;; unify case of field name
! (if (memq 'unify-case format)
! (let ((fname (car (assoc-string field-name
! default-field-list t))))
! (if fname
! (progn
! (delete-region beg-name end-name)
! (goto-char beg-name)
! (insert fname))
! ;; there are no rules we could follow
! (downcase-region beg-name end-name))))
;; update point
--- 2522,2533 ----
;; unify case of field name
! (when (memq 'unify-case format)
! (let ((fname (car (assoc-string field-name
! default-field-list t)))
! (curname (buffer-substring beg-name end-name)))
! (delete-region beg-name end-name)
! (goto-char beg-name)
! (insert (funcall bibtex-unify-case-convert
! (or fname curname)))))
;; update point
***************
*** 2525,2554 ****
;; check whether all required fields are present
! (if (memq 'required-fields format)
! (let ((alt-expect (make-vector num-alt nil))
! (alt-found (make-vector num-alt 0)))
! (dolist (fname req-field-list)
! (cond ((setq idx (nth 3 fname))
! ;; t if field has alternative flag
! (bibtex-vec-push alt-expect idx (car fname))
! (if (member-ignore-case (car fname) field-list)
! (bibtex-vec-incr alt-found idx)))
! ((not (member-ignore-case (car fname) field-list))
! ;; If we use the crossref field, a required field
! ;; can have the OPT prefix. So if it was empty,
! ;; we have deleted by now. Nonetheless we can
! ;; move point on this empty field.
! (setq error-field-name (car fname))
! (error "Mandatory field `%s' is missing" (car fname)))))
! (dotimes (idx num-alt)
! (cond ((= 0 (aref alt-found idx))
! (setq error-field-name (car (last (aref alt-fields idx))))
! (error "Alternative mandatory field `%s' is missing"
! (aref alt-expect idx)))
! ((< 1 (aref alt-found idx))
! (setq error-field-name (car (last (aref alt-fields idx))))
! (error "Alternative fields `%s' are defined %s times"
! (aref alt-expect idx)
! (length (aref alt-fields idx))))))))
;; update comma after last field
--- 2535,2564 ----
;; check whether all required fields are present
! (when (memq 'required-fields format)
! (let ((alt-expect (make-vector num-alt nil))
! (alt-found (make-vector num-alt 0)))
! (dolist (fname req-field-list)
! (cond ((setq idx (nth 3 fname))
! ;; t if field has alternative flag
! (bibtex-vec-push alt-expect idx (car fname))
! (if (member-ignore-case (car fname) field-list)
! (bibtex-vec-incr alt-found idx)))
! ((not (member-ignore-case (car fname) field-list))
! ;; If we use the crossref field, a required field
! ;; can have the OPT prefix. So if it was empty,
! ;; we have deleted by now. Nonetheless we can
! ;; move point on this empty field.
! (setq error-field-name (car fname))
! (error "Mandatory field `%s' is missing" (car fname)))))
! (dotimes (idx num-alt)
! (cond ((= 0 (aref alt-found idx))
! (setq error-field-name (car (last (aref alt-fields idx))))
! (error "Alternative mandatory field `%s' is missing"
! (aref alt-expect idx)))
! ((< 1 (aref alt-found idx))
! (setq error-field-name (car (last (aref alt-fields idx))))
! (error "Alternative fields `%s' are defined %s times"
! (aref alt-expect idx)
! (length (aref alt-fields idx))))))))
;; update comma after last field
^ permalink raw reply [flat|nested] 4+ messages in thread
* bug#44614: 26.3; bibtex convert case undefined
2020-11-14 18:47 ` Francesco Potortì
@ 2020-11-16 21:36 ` Lars Ingebrigtsen
0 siblings, 0 replies; 4+ messages in thread
From: Lars Ingebrigtsen @ 2020-11-16 21:36 UTC (permalink / raw)
To: Francesco Potortì; +Cc: 44614
Francesco Potortì <pot@gnu.org> writes:
> Anyway, here is my patch, which I think does not make any harm. It
> changes the behaviour slightly in what I consider it being a more
> consistent way. When unifying is requested, in the default case, which
> is 'identity, if an unknown field is found it is not unified (rather
> than being downcased as it is now): only known fields are unified to the
> known case.
Makes sense to me. I've now applied this to Emacs 28 (and added a NEWS
entry).
--
(domestic pets only, the antidote for overdose, milk.)
bloggy blog: http://lars.ingebrigtsen.no
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2020-11-16 21:36 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-11-13 11:05 bug#44614: 26.3; bibtex convert case undefined Francesco Potortì
2020-11-14 16:33 ` Lars Ingebrigtsen
2020-11-14 18:47 ` Francesco Potortì
2020-11-16 21:36 ` Lars Ingebrigtsen
Code repositories for project(s) associated with this public inbox
https://git.savannah.gnu.org/cgit/emacs.git
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).