From mboxrd@z Thu Jan 1 00:00:00 1970 From: Piyush Srivastava Subject: Re: Bug: Bug in handling of named fields in the org spreadsheet [9.1.13 (release_9.1.13)] Date: Wed, 23 May 2018 20:05:22 +0530 Message-ID: <87po1msamt.fsf@tifr.res.in> References: <87y3gf4gso.fsf@tifr.res.in> <87po1oxdac.fsf@nicolasgoaziou.fr> Mime-Version: 1.0 Content-Type: text/plain Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:47523) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1fLUsX-0008D5-4Q for emacs-orgmode@gnu.org; Wed, 23 May 2018 10:36:38 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1fLUsT-0007Pc-Vi for emacs-orgmode@gnu.org; Wed, 23 May 2018 10:36:37 -0400 Received: from tifr.res.in ([2406:f00:1:1::6c]:58962 helo=mailhost.tifr.res.in) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1fLUsS-00073c-BN for emacs-orgmode@gnu.org; Wed, 23 May 2018 10:36:33 -0400 Received: from localhost (localhost.localdomain [127.0.0.1]) by mailhost.tifr.res.in (Postfix) with ESMTP id 93BBA4181 for ; Wed, 23 May 2018 20:05:29 +0530 (IST) Received: from mailhost.tifr.res.in ([127.0.0.1]) by localhost (mailhost.tifr.res.in [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id LtyG0jHSokwa for ; Wed, 23 May 2018 20:05:24 +0530 (IST) Received: from vyom (unknown [158.144.106.131]) by mailhost.tifr.res.in (Postfix) with ESMTPSA id 3E3BF4184 for ; Wed, 23 May 2018 20:05:24 +0530 (IST) In-reply-to: <87po1oxdac.fsf@nicolasgoaziou.fr> List-Id: "General discussions about Org-mode." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: emacs-orgmode-bounces+geo-emacs-orgmode=m.gmane.org@gnu.org Sender: "Emacs-orgmode" Cc: emacs-orgmode@gnu.org Thank you. I will get back to creating the patch this weekend and will include your suggested change. Best regards, -- Piyush. Nicolas Goaziou writes: > Hello, > > Piyush Srivastava writes: > >> In the current org mode version (release_9.1.13), when a named cell >> formula is stored using C-u C-c =, the #+TBLFM line stores the name of >> the cell *without* a prefix $ sign, as in the following table (notice >> that in the #+TBLFM row, there is no $ sign before the 'sum' variable): >> >> | | c | >> |---+-----| >> | | 3 | >> | | 10 | >> |---+-----| >> | # | 14 | >> | ^ | sum | >> |---+-----| >> >> #+TBLFM: sum=vsum(@-II$0..@-I$0) >> >> >> With this convention, pressing TAB in the row after making a change in >> the table does not update the field referred to by $sum, contrary to >> what is said in the manual. >> >> >> However, if I manually add a prefix $ to the variable sum in the #+TBLFM >> row (as in the table below), then TAB update work as described in the >> manual (notice the $sum variable in the #+TBLFM row): >> >> >> | | c | >> |---+-----| >> | | 3 | >> | | 12 | >> |---+-----| >> | # | 15 | >> | ^ | sum | >> |---+-----| >> >> #+TBLFM: $sum=vsum(@-II$0..@-I$0) >> >> >> A fix is to ensure that if the LHS of a field formula being inserted is >> a field name rather than a numerical reference, then a '$' should be >> prefixed to it. I have the following patch that implements this: >> >> >> diff --git a/lisp/org-table.el b/lisp/org-table.el >> --- a/lisp/org-table.el >> +++ b/lisp/org-table.el >> @@ -2301,7 +2301,13 @@ LOCATION instead." >> (org-indent-line) >> (insert (or (match-string 2) "#+TBLFM:"))) >> (insert " " >> - (mapconcat (lambda (x) (concat (car x) "=" (cdr x))) >> + (mapconcat (lambda (x) >> + (let* ((name (car x)) >> + (first (substring-no-properties name 0 1)) >> + (test (or (string= first "@") (string= first "$"))) >> + ) >> + (concat (if test name (concat "$" name)) >> + "=" (cdr x)))) >> (sort alist #'org-table-formula-less-p) >> "::") >> "\n")))) > > You're right. > > I suggest the following, though: > > (mapconcat (lambda (x) > (pcase-let ((`(,lhs . ,rhs) x)) > (format "%s=%s" > (if (string-match-p "\\`[$@]" lhs) lhs > ;; Named fields are stored > ;; without the "$" prefix. > (concat "$" lhs)) > rhs))) > (sort alist #'org-table-formula-less-p) > "::") > >> I am happy to assign copyright for the patch, if that is needed and if >> the patch if found suitable, for it to be included in org-mode code. > > This is not needed because this patch is a TINYCHANGE. However, starting > the copyright assignment process is still valuable if you intend to > provide more patches. > > In any case, would you mind providing a patch incorporating the changes > above with a proper commit message? Also, it might be interesting to add > a test for it in "test-org-table.el". > > Thank you. > > Regards,