From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Schulte Subject: Re: [babel] Expand code is failing on a #+call line Date: Thu, 20 Oct 2011 11:52:48 -0600 Message-ID: <87aa8vk0sv.fsf@gmail.com> References: <8039eny9vy.fsf@somewhere.org> <87fwinll4y.fsf@gmail.com> <80y5wfwrko.fsf@somewhere.org> Mime-Version: 1.0 Content-Type: text/plain Return-path: Received: from eggs.gnu.org ([140.186.70.92]:53424) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1RGwna-0003RU-JP for emacs-orgmode@gnu.org; Thu, 20 Oct 2011 13:53:00 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1RGwnX-0006m1-Ri for emacs-orgmode@gnu.org; Thu, 20 Oct 2011 13:52:58 -0400 Received: from mail-yx0-f169.google.com ([209.85.213.169]:46696) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1RGwnX-0006lv-Il for emacs-orgmode@gnu.org; Thu, 20 Oct 2011 13:52:55 -0400 Received: by yxs7 with SMTP id 7so3646225yxs.0 for ; Thu, 20 Oct 2011 10:52:54 -0700 (PDT) In-Reply-To: <80y5wfwrko.fsf@somewhere.org> (Sebastien Vauban's message of "Thu, 20 Oct 2011 18:33:43 +0200") 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-bounces+geo-emacs-orgmode=m.gmane.org@gnu.org To: Sebastien Vauban Cc: emacs-orgmode@gnu.org > >> As to this specific question... I am not sure where you are trying to >> expand a code block. It seems as if you are trying to expand a call >> line, but I do not believe that is supposed to be possible. > > As for any other code block, all I want is to see its expanded (instantiated) > version before I run it. > > Is this not doable when you call the code block remotely (I mean: from > a #+call line, instead of from the code block itself)? > That is correct. There is no reason (that I can think of at the moment) why this functionality could not be added, but currently the code block preview functionality expects to be called from within a code block. > > If not, does that mean we better not put code block into the lob, when we're > interested by having a look at their expanded form? > Yes, given the current state of affairs that is correct. Best -- Eric > > Best regards, > Seb > >> "Sebastien Vauban" writes: >> >>> Hi Eric, >>> >>> * Config >>> >>> ** Some file to be ingested >>> >>> Let's say I have this code in a =my-lob.org= file: >>> >>> #+srcname: add-column-in-table(table="", column="", type="", nullability="") >>> #+begin_src sql >>> -- add column `$column' (if column does not exist yet) >>> IF NOT EXISTS (SELECT * >>> FROM INFORMATION_SCHEMA.COLUMNS >>> WHERE TABLE_NAME = '$table' >>> AND COLUMN_NAME = '$column') >>> BEGIN >>> ALTER TABLE $table >>> ADD $column $type $nullability >>> END >>> #+end_src >>> >>> ** In .emacs >>> >>> ... which I ingest at Emacs startup through the lines: >>> >>> #+begin_src emacs-lisp >>> (require 'ob-lob) >>> (org-babel-lob-ingest "~/org/my-lob.org") >>> #+end_src >>> >>> * Some other document >>> >>> ** Source params >>> >>> In some other document, I have a table with columns I'd like to add in a >>> database: >>> >>> #+results: params >>> | table | column | type | nullability | >>> |---------+----------------+-------------+-------------| >>> | dossier | pfiResetDate | date | NULL | >>> | dossier | pfiResetOprID | tinyint | NULL | >>> | dossier | pfiResetOprNom | varchar(64) | NULL | >>> >>> Normally, I could call statements like this: >>> >>> #+call: add-column-in-table(table=params[2,0], column=params[2,1], type=params[2,2], nullability=params[2,3]) >>> >>> But... >>> >>> ** Expand code is failing >>> >>> =C-c C-v C-v= does generate an error: >>> >>> setf: Wrong type argument: consp, nil >>> >>> #+begin_src text >>> Debugger entered--Lisp error: (wrong-type-argument consp nil) >>> setcar(nil ((:cache . "") (:comments . "") (:exports . "") (:noweb . "") (:padline . "") (:results . "") (:shebang . "") (:tangle . ""))) >>> (setf (nth 2 info) (sort (org-babel-merge-params ... params) (lambda ... ...))) >>> (let* ((info ...) (lang ...) (params ...) (body ...) (expand-cmd ...) (assignments-cmd ...) (expanded ...)) (org-edit-src-code nil expanded (concat "*Org-Babel Preview " ... "[ " lang " ]*"))) >>> org-babel-expand-src-block() >>> call-interactively(org-babel-expand-src-block nil nil) >>> #+end_src >>> >>> This stays very unclear to me... >>> >>> ** Execute code is failing >>> >>> By the way, =C-c C-v C-e= returns as well an error: >>> >>> let: Wrong type argument: stringp, nil >>> >>> This is explainable: that's because the SQL =engine= is not known (not given >>> neither in the header of the code block, neither anywhere in this file). >>> >>> #+begin_src text >>> Debugger entered--Lisp error: (wrong-type-argument stringp nil) >>> intern(nil) >>> (let ((--cl-var-- ...)) (cond (... ...) (... ...) (... ...) (t ...))) >>> (case (intern engine) ((quote msosql) (format "osql %s -s \" \" -i %s -o %s" ... ... ...)) ((quote mysql) (format "mysql %s < %s > %s" ... ... ...)) ((quote postgresql) (format "psql -A -P footer=off -F \" \" -f %s -o %s %s" ... ... ...)) (t (error "no support for the %s sql engine" engine))) >>> (let* ((result-params ...) (cmdline ...) (engine ...) (in-file ...) (out-file ...) (header-delim "") (command ...)) (with-temp-file in-file (insert ...)) (message command) (shell-command command) (if (or ... ... ... ... ...) (with-temp-buffer ...) (with-temp-buffer ... ... ...))) >>> org-babel-execute:sql("-- add column `$column' (if column does not exist yet)\nIF NOT EXISTS (SELECT *\n FROM INFORMATION_SCHEMA.COLUMNS\n WHERE TABLE_NAME = '$table'\n AND COLUMN_NAME = '$column')\nBEGIN\n ALTER TABLE $table\n ADD $column $type $nullability\nEND" ((:var table . "dossier") (:var column . "pfiResetDate") (:var type . "date") (:var nullability . "NULL") (:colname-names) (:rowname-names) (:result-params "silent" "replace") (:result-type . value) (:comments . "") (:shebang . "") (:cache . "no") (:padline . "") (:noweb . "no") (:tangle . "no") (:exports . "code") (:results . "silent") (:padnewline . "yes") (:hlines . "no") (:session . "none") (:result-type . value) (:result-params "replace") (:rowname-names) (:colname-names) )) >> >>> #+end_src >>> >>> Though, I guess we should have a proper manner to report that some necessary >>> arguments are missing, instead of failing with a unclear message. >>> >>> ** Other weirdnesses >>> >>> While the variable =org-babel-library-of-babel= contains the ingested code >>> (here, of =add-column-in-table=), the variable =org-babel-lob-files= is =nil=!? >>> >>> ** Speed commands >>> >>> Speed commands don't work on the =#+call= lines. If I press =v= or =e=, they're >>> inserted verbatim. >>> >>> Can you help me understand the 1^st point of these 4? Thanks a lot!! -- Eric Schulte http://cs.unm.edu/~eschulte/