[PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

[-- Attachment #1: Type: text/plain, Size: 1738 bytes --]

Hi all,

I'm attaching a patch with a proposal to unify in a single constant
(named `org-latex-language-alist')
`org-latex-polyglossia-language-alist' and
`org-latex-babel-language-alist', along with some necessary (minor)
modifications in `org-latex-guess-polyglossia-language' and
`org-latex-guess-babel-language'

The new list, which is not exhaustive, is built taking as a reference
the documentation of Babel and Polyglossia in their latest versions
within TeX live 2021. It also assumes the latest improvements in the
babel package (on the current state of the art regarding the babel and
polyglossia packages, see this previous thread:
https://list.orgmode.org/87wnmv4s87.fsf@posteo.net/). I have also
corrected some minor inconsistencies in the previous two lists.

This new alist supports three types of members:

- Members with two elements: CODE BABEL/POLYGLOSSIA-OPTION.

i.e.: ("ar" "arabic")

- Members with three elements: CODE BABEL/POLYGLOSSIA-OPTION ASTERISK
(the presence of the asterisk indicates that this language is not loaded
in Babel using the old method of ldf files but using ini files. If Babel
is loaded in an Org document with these languages, the \"AUTO \"
argument is just removed, to avoid compilation errors. The new babel
method with ini files  is not supported, for backward
compatibility with 'old' ldf method).

i.e. ("bo" "tibetan" "*")

- Members with four elements (for variants of languages): CODE
BABEL-OPTION POLYGLOSSIA-OPTION POLYGLOSSIA-VARIANT

i.e. ("es" "spanishmx" "spanish" "mexican")

==> babel:

\usepackage[mexican]{babel}

==> polyglossia:

\usepackage{polyglossia}
\setmainlanguage[variant=mexican]{spanish}

I also attach an Org document for testing.

Best regards,

Juan Manuel


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-ox-latex.el-Unify-in-one-list-babel-and-polyglossia-.patch --]
[-- Type: text/x-patch, Size: 8021 bytes --]

From 389a4e43756a7c195c2c1f751b7dc9c03447526d Mon Sep 17 00:00:00 2001
From: Juan Manuel Macias <maciaschain@posteo.net>
Date: Sun, 3 Oct 2021 16:55:31 +0200
Subject: [PATCH] ox-latex.el: Unify in one list babel and polyglossia language
 alists

* lisp/ox-latex.el (org-latex-language-alist): Unify in a single list `org-latex-polyglossia-language-alist' and `org-latex-babel-language-alist'
---
 lisp/ox-latex.el | 167 +++++++++++++++++++----------------------------
 1 file changed, 68 insertions(+), 99 deletions(-)

diff --git a/lisp/ox-latex.el b/lisp/ox-latex.el
index 3e3967033..de03470fa 100644
--- a/lisp/ox-latex.el
+++ b/lisp/ox-latex.el
@@ -160,144 +160,109 @@
 \f
 ;;; Internal Variables
 
-(defconst org-latex-babel-language-alist
-  '(("af" . "afrikaans")
-    ("bg" . "bulgarian")
-    ("ca" . "catalan")
-    ("cs" . "czech")
-    ("cy" . "welsh")
-    ("da" . "danish")
-    ("de" . "germanb")
-    ("de-at" . "naustrian")
-    ("de-de" . "ngerman")
-    ("el" . "greek")
-    ("en" . "english")
-    ("en-au" . "australian")
-    ("en-ca" . "canadian")
-    ("en-gb" . "british")
-    ("en-ie" . "irish")
-    ("en-nz" . "newzealand")
-    ("en-us" . "american")
-    ("es" . "spanish")
-    ("et" . "estonian")
-    ("eu" . "basque")
-    ("fi" . "finnish")
-    ("fr" . "french")
-    ("fr-ca" . "canadien")
-    ("gl" . "galician")
-    ("hr" . "croatian")
-    ("hu" . "hungarian")
-    ("id" . "indonesian")
-    ("is" . "icelandic")
-    ("it" . "italian")
-    ("la" . "latin")
-    ("ms" . "malay")
-    ("nl" . "dutch")
-    ("nb" . "norsk")
-    ("nn" . "nynorsk")
-    ("no" . "norsk")
-    ("pl" . "polish")
-    ("pt" . "portuguese")
-    ("pt-br" . "brazilian")
-    ("ro" . "romanian")
-    ("ru" . "russian")
-    ("sa" . "sanskrit")
-    ("sb" . "uppersorbian")
-    ("sk" . "slovak")
-    ("sl" . "slovene")
-    ("sq" . "albanian")
-    ("sr" . "serbian")
-    ("sv" . "swedish")
-    ("ta" . "tamil")
-    ("tr" . "turkish")
-    ("uk" . "ukrainian"))
-  "Alist between language code and corresponding Babel option.")
-
-(defconst org-latex-polyglossia-language-alist
-  '(("am" "amharic")
+(defconst org-latex-language-alist
+  '(("am" "amharic" "*")
     ("ar" "arabic")
-    ("ast" "asturian")
+    ("ast" "asturian" "*")
     ("bg" "bulgarian")
-    ("bn" "bengali")
-    ("bo" "tibetan")
+    ("bn" "bengali" "*")
+    ("bo" "tibetan" "*")
     ("br" "breton")
     ("ca" "catalan")
-    ("cop" "coptic")
+    ("cop" "coptic" "*")
     ("cs" "czech")
     ("cy" "welsh")
     ("da" "danish")
-    ("de" "german" "german")
-    ("de-at" "german" "austrian")
-    ("de-de" "german" "german")
-    ("dsb" "lsorbian")
-    ("dv" "divehi")
+    ("de" "ngerman" "german" "german")
+    ("de-at" "naustrian" "german" "austrian")
+    ("dsb" "lsorbian" "*")
+    ("dv" "divehi" "*")
     ("el" "greek")
-    ("en" "english" "usmax")
-    ("en-au" "english" "australian")
-    ("en-gb" "english" "uk")
-    ("en-nz" "english" "newzealand")
-    ("en-us" "english" "usmax")
+    ("el-polyton" "polutonikogreek" "greek" "polytonic")
+    ("en" "american" "english" "usmax")
+    ("en-au" "australian" "english" "australian")
+    ("en-gb" "british" "english" "uk")
+    ("en-nz" "newzealand" "english" "newzealand")
+    ("en-us" "american" "english" "usmax")
     ("eo" "esperanto")
     ("es" "spanish")
+    ("es" "spanishmx" "spanish" "mexican")
     ("et" "estonian")
     ("eu" "basque")
     ("fa" "farsi")
     ("fi" "finnish")
     ("fr" "french")
-    ("fu" "friulan")
+    ("fr-ca" "canadien" "french" "canadian")
+    ("fur" "friulan")
     ("ga" "irish")
     ("gd" "scottish")
     ("gl" "galician")
     ("he" "hebrew")
     ("hi" "hindi")
     ("hr" "croatian")
-    ("hsb" "usorbian")
+    ("hsb" "uppersorbian" "sorbian" "upper")
     ("hu" "magyar")
-    ("hy" "armenian")
+    ("hy" "armenian" "*")
     ("ia" "interlingua")
-    ("id" "bahasai")
+    ("id" "bahasai" "*")
     ("is" "icelandic")
     ("it" "italian")
-    ("kn" "kannada")
-    ("la" "latin" "modern")
-    ("la-classic" "latin" "classic")
-    ("la-medieval" "latin" "medieval")
-    ("la-modern" "latin" "modern")
-    ("lo" "lao")
+    ("kn" "kannada" "*")
+    ("la" "latin")
+    ("la-classic" "classiclatin" "latin" "classic")
+    ("la-medieval" "medievallatin" "latin" "medieval")
+    ("la-ecclesiastic" "ecclesiasticlatin" "latin" "ecclesiastic")
+    ("lo" "lao" "*")
     ("lt" "lithuanian")
     ("lv" "latvian")
-    ("ml" "malayalam")
-    ("mr" "maranthi")
-    ("nb" "norsk")
-    ("nko" "nko")
+    ("ml" "malayalam" "*")
+    ("mr" "maranthi" "*")
+    ("nb" "norsk" "norwegian" "bokmal")
     ("nl" "dutch")
-    ("nn" "nynorsk")
+    ("nn" "nynorsk" "norwegian" "nynorsk")
     ("no" "norsk")
     ("oc" "occitan")
     ("pl" "polish")
     ("pms" "piedmontese")
     ("pt" "portuges")
     ("pt-br" "brazilian")
-    ("rm" "romansh")
+    ("rm" "romansh" "*")
     ("ro" "romanian")
     ("ru" "russian")
-    ("sa" "sanskrit")
-    ("se" "samin")
+    ("sa" "sanskrit" "*")
     ("sk" "slovak")
-    ("sl" "slovenian")
+    ("sl" "slovene")
     ("sq" "albanian")
     ("sr" "serbian")
     ("sv" "swedish")
-    ("syr" "syriac")
-    ("ta" "tamil")
-    ("te" "telugu")
+    ("syr" "syriac" "*")
+    ("ta" "tamil" "*")
+    ("te" "telugu" "*")
     ("th" "thai")
     ("tk" "turkmen")
     ("tr" "turkish")
     ("uk" "ukrainian")
-    ("ur" "urdu")
+    ("ur" "urdu" "*")
     ("vi" "vietnamese"))
-  "Alist between language code and corresponding Polyglossia option.")
+  "Alist between language code and corresponding Babel/Polyglossia option.
+
+For the names of the languages, the Babel nomenclature is
+preferred to that of Polyglossia, in those cases where both
+coincide.
+
+The alist supports three types of members:
+
+- Members with two elements: CODE BABEL/POLYGLOSSIA OPTION.
+
+- Members with three elements: CODE BABEL/POLYGLOSSIA OPTION
+ASTERISK (the presence of the asterisk indicates that this
+language is not loaded in Babel using the old method of ldf
+files but using ini files. If Babel is loaded in an Org
+document with these languages, the \"AUTO \" argument is just
+removed, to avoid compilation errors).
+
+- Members with four elements (for variants of languages): CODE
+BABEL-OPTION POLYGLOSSIA-OPTION POLYGLOSSIA-VARIANT")
 
 (defconst org-latex-table-matrix-macros '(("bordermatrix" . "\\cr")
 					  ("qbordermatrix" . "\\cr")
@@ -1409,14 +1374,16 @@ Return the new header."
 	header
       (let ((options (save-match-data
 		       (org-split-string (match-string 1 header) ",[ \t]*")))
-	    (language (cdr (assoc-string language-code
-					 org-latex-babel-language-alist t))))
+	    (language (nth 1 (assoc language-code
+				    org-latex-language-alist))))
 	;; If LANGUAGE is already loaded, return header without AUTO.
 	;; Otherwise, replace AUTO with language or append language if
 	;; AUTO is not present.
 	(replace-match
 	 (mapconcat (lambda (option) (if (equal "AUTO" option) language option))
 		    (cond ((member language options) (delete "AUTO" options))
+			  ((let ((l (assoc language-code org-latex-language-alist)))
+			     (and (consp l) (= (length l) 3))) (delete "AUTO" options))
 			  ((member "AUTO" options) options)
 			  (t (append options (list language))))
 		    ", ")
@@ -1462,15 +1429,17 @@ Return the new header."
 	 (concat "\\usepackage{polyglossia}\n"
 		 (mapconcat
 		  (lambda (l)
-		    (let ((l (or (assoc l org-latex-polyglossia-language-alist)
+		    (let ((l (or (assoc l org-latex-language-alist)
 				 l)))
 		      (format (if main-language-set "\\setotherlanguage%s{%s}\n"
 				(setq main-language-set t)
 				"\\setmainlanguage%s{%s}\n")
-			      (if (and (consp l) (= (length l) 3))
-				  (format "[variant=%s]" (nth 2 l))
+			      (if (and (consp l) (= (length l) 4))
+				  (format "[variant=%s]" (nth 3 l))
 				"")
-			      (nth 1 l))))
+			      (if (and (consp l) (= (length l) 4))
+				  (nth 2 l)
+				(nth 1 l)))))
 		  languages
 		  ""))
 	 t t header 0)))))
-- 
2.32.0


[-- Attachment #3: test.org --]
[-- Type: application/vnd.lotus-organizer, Size: 1525 bytes --]

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Ihor Radchenko]

Juan Manuel Macías <maciaschain@posteo.net> writes:

> I'm attaching a patch with a proposal to unify in a single constant
> (named `org-latex-language-alist')
> `org-latex-polyglossia-language-alist' and
> `org-latex-babel-language-alist', along with some necessary (minor)
> modifications in `org-latex-guess-polyglossia-language' and
> `org-latex-guess-babel-language'
>
> The new list, which is not exhaustive, is built taking as a reference
> the documentation of Babel and Polyglossia in their latest versions
> within TeX live 2021. It also assumes the latest improvements in the
> babel package (on the current state of the art regarding the babel and
> polyglossia packages, see this previous thread:
> https://list.orgmode.org/87wnmv4s87.fsf@posteo.net/). I have also
> corrected some minor inconsistencies in the previous two lists.

Thanks!
This looks like an improvement.
However, we may need to preserve the old defconsts for the time being
and declare them obsolete.

Otherwise, the patch looks fine from a cursory look.

Best,
Ihor

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Max Nikulin]

On 03/10/2021 22:28, Juan Manuel Macías wrote:
> Hi all,
> 
> I'm attaching a patch with a proposal to unify in a single constant
> (named `org-latex-language-alist')
> `org-latex-polyglossia-language-alist' and
> `org-latex-babel-language-alist', along with some necessary (minor)
> modifications in `org-latex-guess-polyglossia-language' and
> `org-latex-guess-babel-language'

I would consider structures with named fields (alists or plists) for a 
case of adding some additional settings (Font name? But it is rather 
defcustom than defconst)

("es" . (:babel "spanishmx" :poliglossia "spanish" :poliglossia-variant 
"mexican")

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

[-- Attachment #1: Type: text/plain, Size: 368 bytes --]

Ihor Radchenko writes:

> Thanks!
> This looks like an improvement.
> However, we may need to preserve the old defconsts for the time being
> and declare them obsolete.

Hi, Ihor,

I attach the new version of the patch with both variables declared
obsolete.

If everything is ok, I can add what is necessary to NEWS and to the Org Manual.

Best regards,

Juan Manuel


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-lisp-ox-latex.el-New-variable-org-latex-language-ali.patch --]
[-- Type: text/x-patch, Size: 8306 bytes --]

From 9ff77e71a8cd89b883d5ead37909c887178e4402 Mon Sep 17 00:00:00 2001
From: Juan Manuel Macias <maciaschain@posteo.net>
Date: Thu, 14 Jul 2022 13:42:50 +0200
Subject: [PATCH] * lisp/ox-latex.el: New variable `org-latex-language-alist'

(org-latex-language-alist): Unify in a single list
`org-latex-polyglossia-language-alist' and
`org-latex-babel-language-alist', and make the two variables obsolete.
---
 lisp/ox-latex.el | 173 ++++++++++++++++++++---------------------------
 1 file changed, 74 insertions(+), 99 deletions(-)

diff --git a/lisp/ox-latex.el b/lisp/ox-latex.el
index 1aab8ffd5..9e97f38db 100644
--- a/lisp/ox-latex.el
+++ b/lisp/ox-latex.el
@@ -172,144 +172,115 @@
 \f
 ;;; Internal Variables
 
-(defconst org-latex-babel-language-alist
-  '(("af" . "afrikaans")
-    ("bg" . "bulgarian")
-    ("ca" . "catalan")
-    ("cs" . "czech")
-    ("cy" . "welsh")
-    ("da" . "danish")
-    ("de" . "germanb")
-    ("de-at" . "naustrian")
-    ("de-de" . "ngerman")
-    ("el" . "greek")
-    ("en" . "english")
-    ("en-au" . "australian")
-    ("en-ca" . "canadian")
-    ("en-gb" . "british")
-    ("en-ie" . "irish")
-    ("en-nz" . "newzealand")
-    ("en-us" . "american")
-    ("es" . "spanish")
-    ("et" . "estonian")
-    ("eu" . "basque")
-    ("fi" . "finnish")
-    ("fr" . "french")
-    ("fr-ca" . "canadien")
-    ("gl" . "galician")
-    ("hr" . "croatian")
-    ("hu" . "hungarian")
-    ("id" . "indonesian")
-    ("is" . "icelandic")
-    ("it" . "italian")
-    ("la" . "latin")
-    ("ms" . "malay")
-    ("nl" . "dutch")
-    ("nb" . "norsk")
-    ("nn" . "nynorsk")
-    ("no" . "norsk")
-    ("pl" . "polish")
-    ("pt" . "portuguese")
-    ("pt-br" . "brazilian")
-    ("ro" . "romanian")
-    ("ru" . "russian")
-    ("sa" . "sanskrit")
-    ("sb" . "uppersorbian")
-    ("sk" . "slovak")
-    ("sl" . "slovene")
-    ("sq" . "albanian")
-    ("sr" . "serbian")
-    ("sv" . "swedish")
-    ("ta" . "tamil")
-    ("tr" . "turkish")
-    ("uk" . "ukrainian"))
-  "Alist between language code and corresponding Babel option.")
-
-(defconst org-latex-polyglossia-language-alist
-  '(("am" "amharic")
+(make-obsolete-variable 'org-latex-babel-language-alist
+                        "set `org-latex-language-alist' instead." "9.6")
+
+(make-obsolete-variable 'org-latex-polyglossia-language-alist
+                        "set `org-latex-language-alist' instead." "9.6")
+
+(defconst org-latex-language-alist
+  '(("am" "amharic" "*")
     ("ar" "arabic")
-    ("ast" "asturian")
+    ("ast" "asturian" "*")
     ("bg" "bulgarian")
-    ("bn" "bengali")
-    ("bo" "tibetan")
+    ("bn" "bengali" "*")
+    ("bo" "tibetan" "*")
     ("br" "breton")
     ("ca" "catalan")
-    ("cop" "coptic")
+    ("cop" "coptic" "*")
     ("cs" "czech")
     ("cy" "welsh")
     ("da" "danish")
-    ("de" "german" "german")
-    ("de-at" "german" "austrian")
-    ("de-de" "german" "german")
-    ("dsb" "lsorbian")
-    ("dv" "divehi")
+    ("de" "ngerman" "german" "german")
+    ("de-at" "naustrian" "german" "austrian")
+    ("dsb" "lsorbian" "*")
+    ("dv" "divehi" "*")
     ("el" "greek")
-    ("en" "english" "usmax")
-    ("en-au" "english" "australian")
-    ("en-gb" "english" "uk")
-    ("en-nz" "english" "newzealand")
-    ("en-us" "english" "usmax")
+    ("el-polyton" "polutonikogreek" "greek" "polytonic")
+    ("en" "american" "english" "usmax")
+    ("en-au" "australian" "english" "australian")
+    ("en-gb" "british" "english" "uk")
+    ("en-nz" "newzealand" "english" "newzealand")
+    ("en-us" "american" "english" "usmax")
     ("eo" "esperanto")
     ("es" "spanish")
+    ("es-mx" "spanishmx" "spanish" "mexican")
     ("et" "estonian")
     ("eu" "basque")
     ("fa" "farsi")
     ("fi" "finnish")
     ("fr" "french")
-    ("fu" "friulan")
+    ("fr-ca" "canadien" "french" "canadian")
+    ("fur" "friulan")
     ("ga" "irish")
     ("gd" "scottish")
     ("gl" "galician")
     ("he" "hebrew")
     ("hi" "hindi")
     ("hr" "croatian")
-    ("hsb" "usorbian")
+    ("hsb" "uppersorbian" "sorbian" "upper")
     ("hu" "magyar")
-    ("hy" "armenian")
+    ("hy" "armenian" "*")
     ("ia" "interlingua")
-    ("id" "bahasai")
+    ("id" "bahasai" "*")
     ("is" "icelandic")
     ("it" "italian")
-    ("kn" "kannada")
-    ("la" "latin" "modern")
-    ("la-classic" "latin" "classic")
-    ("la-medieval" "latin" "medieval")
-    ("la-modern" "latin" "modern")
-    ("lo" "lao")
+    ("kn" "kannada" "*")
+    ("la" "latin")
+    ("la-classic" "classiclatin" "latin" "classic")
+    ("la-medieval" "medievallatin" "latin" "medieval")
+    ("la-ecclesiastic" "ecclesiasticlatin" "latin" "ecclesiastic")
+    ("lo" "lao" "*")
     ("lt" "lithuanian")
     ("lv" "latvian")
-    ("ml" "malayalam")
-    ("mr" "maranthi")
-    ("nb" "norsk")
-    ("nko" "nko")
+    ("ml" "malayalam" "*")
+    ("mr" "maranthi" "*")
+    ("nb" "norsk" "norwegian" "bokmal")
     ("nl" "dutch")
-    ("nn" "nynorsk")
+    ("nn" "nynorsk" "norwegian" "nynorsk")
     ("no" "norsk")
     ("oc" "occitan")
     ("pl" "polish")
     ("pms" "piedmontese")
     ("pt" "portuges")
     ("pt-br" "brazilian")
-    ("rm" "romansh")
+    ("rm" "romansh" "*")
     ("ro" "romanian")
     ("ru" "russian")
-    ("sa" "sanskrit")
-    ("se" "samin")
+    ("sa" "sanskrit" "*")
     ("sk" "slovak")
-    ("sl" "slovenian")
+    ("sl" "slovene")
     ("sq" "albanian")
     ("sr" "serbian")
     ("sv" "swedish")
-    ("syr" "syriac")
-    ("ta" "tamil")
-    ("te" "telugu")
+    ("syr" "syriac" "*")
+    ("ta" "tamil" "*")
+    ("te" "telugu" "*")
     ("th" "thai")
     ("tk" "turkmen")
     ("tr" "turkish")
     ("uk" "ukrainian")
-    ("ur" "urdu")
+    ("ur" "urdu" "*")
     ("vi" "vietnamese"))
-  "Alist between language code and corresponding Polyglossia option.")
+  "Alist between language code and corresponding Babel/Polyglossia option.
+
+For the names of the languages, the Babel nomenclature is
+preferred to that of Polyglossia, in those cases where both
+coincide.
+
+The alist supports three types of members:
+
+- Members with two elements: CODE BABEL/POLYGLOSSIA OPTION.
+
+- Members with three elements: CODE BABEL/POLYGLOSSIA OPTION
+ASTERISK (the presence of the asterisk indicates that this
+language is not loaded in Babel using the old method of ldf
+files but using ini files. If Babel is loaded in an Org
+document with these languages, the \"AUTO \" argument is just
+removed, to avoid compilation errors).
+
+- Members with four elements (for variants of languages): CODE
+BABEL-OPTION POLYGLOSSIA-OPTION POLYGLOSSIA-VARIANT")
 
 (defconst org-latex-table-matrix-macros '(("bordermatrix" . "\\cr")
 					  ("qbordermatrix" . "\\cr")
@@ -1657,14 +1628,16 @@ Return the new header."
 	header
       (let ((options (save-match-data
 		       (org-split-string (match-string 1 header) ",[ \t]*")))
-	    (language (cdr (assoc-string language-code
-					 org-latex-babel-language-alist t))))
+	    (language (nth 1 (assoc language-code
+				    org-latex-language-alist))))
 	;; If LANGUAGE is already loaded, return header without AUTO.
 	;; Otherwise, replace AUTO with language or append language if
 	;; AUTO is not present.
 	(replace-match
 	 (mapconcat (lambda (option) (if (equal "AUTO" option) language option))
 		    (cond ((member language options) (delete "AUTO" options))
+			  ((let ((l (assoc language-code org-latex-language-alist)))
+			     (and (consp l) (= (length l) 3))) (delete "AUTO" options))
 			  ((member "AUTO" options) options)
 			  (t (append options (list language))))
 		    ", ")
@@ -1710,15 +1683,17 @@ Return the new header."
 	 (concat "\\usepackage{polyglossia}\n"
 		 (mapconcat
 		  (lambda (l)
-		    (let ((l (or (assoc l org-latex-polyglossia-language-alist)
+		    (let ((l (or (assoc l org-latex-language-alist)
 				 l)))
 		      (format (if main-language-set "\\setotherlanguage%s{%s}\n"
 				(setq main-language-set t)
 				"\\setmainlanguage%s{%s}\n")
-			      (if (and (consp l) (= (length l) 3))
-				  (format "[variant=%s]" (nth 2 l))
+			      (if (and (consp l) (= (length l) 4))
+				  (format "[variant=%s]" (nth 3 l))
 				"")
-			      (nth 1 l))))
+			      (if (and (consp l) (= (length l) 4))
+				  (nth 2 l)
+				(nth 1 l)))))
 		  languages
 		  ""))
 	 t t header 0)))))
-- 
2.36.1

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Max Nikulin]

On 14/07/2022 19:34, Juan Manuel Macías wrote:
> Subject: [PATCH] * lisp/ox-latex.el: New variable `org-latex-language-alist'
Writing the previous message I forgot that currently there is no default 
option for the fontenc package (PdfLaTeX), e.g. T2A for Cyrillic. As a 
result it is not enough to specify just language to generate PDF file. 
 From my point of view it is better to have single map from language 
code to various localization options (babel, polyglossia, fontenc). 
Unfortunately plain list `org-latex-language-alist' does not allow 
further extensions. Property list would make initialization not so 
concise, but it would make the map more flexible.

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

Max Nikulin writes:

>> Subject: [PATCH] * lisp/ox-latex.el: New variable `org-latex-language-alist'
> Writing the previous message I forgot that currently there is no
> default option for the fontenc package (PdfLaTeX), e.g. T2A for
> Cyrillic. As a result it is not enough to specify just language to
> generate PDF file.  From my point of view it is better to have single
> map from language code to various localization options (babel,
> polyglossia, fontenc). Unfortunately plain list
> `org-latex-language-alist' does not allow further extensions. Property
> list would make initialization not so concise, but it would make the
> map more flexible.

Sorry, I hadn't seen your other message and I ignored it...

What you say makes sense. If I'm not mistaken, there is now nothing like
an hypothetical 'org-latex-guess-fontenc', and org defaults to the T1
option. If I remember correctly (because I haven't used pdfLaTeX in
ages), the fontenc option for Greek is LGR. And I imagine there will be
many more cases. If you or anyone wants to implement that on top of my
patch, that's fine with me. I don't have much time to do it right now.
However, my opinion is the following: I think we should focus our
efforts on finding a satisfactory solution for luatex, according to
everything that has been commented in the different sub-threads on the
subject of fonts and fallback fonts. And leave the related to pdfLaTeX
in second place, if in the end LuaTeX is going to be set as the default
engine.

In any case, I personally think that org-latex-language-alist, as it is
now in this patch, is sufficient.

Best regards,

Juan Manuel 

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

Juan Manuel Macías writes:

> In any case, I personally think that org-latex-language-alist, as it is
> now in this patch, is sufficient.

By the way, Maxim. I have been doing some tests with pdfLaTeX. I've
known for a while now that it's no longer necessary to load the inputenc
package. But it seems that it is not even necessary to load fontenc with
the encoding of each language. Try compiling this:

#+begin_src latex
  \documentclass{article}
  \usepackage[russian,polutonikogreek]{babel}

  \begin{document}

  \today

  Δαρείου καὶ Παρυσάτιδος γίγνονται παῖδες δύο, πρεσβύτερος μὲν Ἀρταξέρξης, νεώτερος δὲ
  Κῦρος· ἐπεὶ δὲ ἠσθένει Δαρεῖος καὶ ὑπώπτευε τελευτὴν τοῦ βίου, ἐβούλετο τὼ παῖδε ἀμφοτέρω
  παρεῖναι.

  \selectlanguage{russian}

  \today

  Emacs написан на двух языках: C и Emacs Lisp (Elisp, диалект Лиспа). При этом сам редактор
  является интерпретатором Elisp. По сути дела, большая часть Emacs написана на языке Elisp,
  и её можно рассматривать как расширение к основной программе.

  \end{document}
#+end_src

In TeX live 2022 the compilation is correct (I think). It seems that
Babel (or russian/greek.ldf) loads the font encodings according to the
declared languages. From the log:

#+begin_src sh
(/usr/share/texmf-dist/tex/latex/cyrillic/t2aenc.def
(/usr/share/texmf-dist/tex/latex/base/t2aenc.dfu)))
(/usr/share/texmf-dist/tex/generic/babel-greek/greek.ldf
(/usr/share/texmf-dist/tex/latex/greek-fontenc/lgrenc.def
(/usr/share/texmf-dist/tex/latex/greek-inputenc/lgrenc.dfu)
(/usr/share/texmf-dist/tex/latex/greek-fontenc/greek-fontenc.def))))
(/usr/share/texmf-dist/tex/latex/l3backend/l3backend-pdftex.def) (./fontenc.aux
 (/usr/share/texmf-dist/tex/generic/babel/locale/es/babel-spanish.tex)
(/usr/share/texmf-dist/tex/latex/cbfonts-fd/lgrcmr.fd))
(/usr/share/texmf-dist/tex/latex/cyrillic/t2acmr.fd) [1{/var/lib/texmf/fonts/ma
p/pdftex/updmap/pdftex.map}] (./fontenc.aux) ){/usr/share/texmf-dist/fonts/enc/
dvips/cm-super/cm-super-t2a.enc}</usr/share/texmf-dist/fonts/type1/public/cbfon
ts/grmn1000.pfb></usr/share/texmf-dist/fonts/type1/public/cm-super/sfrm1000.pfb
#+end_src

Is this normal or is it a new Babel feature? If it is a new feature, I
can't find it anywhere in the documentation.

Best regards,

Juan Manuel 

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Max Nikulin]

On 15/07/2022 01:17, Juan Manuel Macías wrote:
> Juan Manuel Macías writes:
> 
>> In any case, I personally think that org-latex-language-alist, as it is
>> now in this patch, is sufficient.

I agree that the lists should be merged. My point is that with unnamed 
fields and variable number of them it would not be possible to extend 
this list with additional fields. So additional step with 
`make-obsolete-variable' would be required.

> By the way, Maxim. I have been doing some tests with pdfLaTeX. I've
> known for a while now that it's no longer necessary to load the inputenc
> package. But it seems that it is not even necessary to load fontenc with
> the encoding of each language.

It looks like a promising feature.

> In TeX live 2022 the compilation is correct (I think). It seems that
> Babel (or russian/greek.ldf) loads the font encodings according to the
> declared languages. From the log:
> 
>   (/usr/share/texmf-dist/tex/generic/babel/locale/es/babel-spanish.tex)

Interesting, Spanish is not mentioned in your document.

> Is this normal or is it a new Babel feature? If it is a new feature, I
> can't find it anywhere in the documentation.

I have tried on Ubuntu-20.04 LTS focal (Latest LTS is 22.04 jammy). 
Without explicit fontenc it may work, but emits a warning

Package babel Warning: No Cyrillic font encoding has been loaded so far.
(babel)                A font encoding should be declared before babel.
(babel)                Default `T2A' encoding will be loaded  on input 
line 74.

Unfortunately in the case of

     \usepackage[russian,english]{babel}

\selectlanguage{russian} is required, without it compilation fails with

    ! LaTeX Error: Command \cyrn unavailable in encoding OT1.

With \usepackage[T2A]{fontenc} it behaves accordingly to acceptable for 
non-important documents graceful degradation. Text is readable, but no 
hyphenation applied.

So I consider explicit loading of fontenc as more reliable.

> If I'm not mistaken, there is now nothing like
> an hypothetical 'org-latex-guess-fontenc', and org defaults to the T1
> option. If I remember correctly (because I haven't used pdfLaTeX in
> ages), the fontenc option for Greek is LGR. And I imagine there will be
> many more cases.

That is why I am suggesting a mapping from language to font encoding.

> If you or anyone wants to implement that on top of my
> patch, that's fine with me.

I do not see a solution "on the top of your patch". Either an additional 
mapping should be added or your changes should be overwritten by some 
extensible structure. The former is a step backward in respect to the 
idea of merging alists, the latter might make unhappy developers of 
third party packages.

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

Max Nikulin writes:

> I have tried on Ubuntu-20.04 LTS focal (Latest LTS is 22.04 jammy).
> Without explicit fontenc it may work, but emits a warning
>
> Package babel Warning: No Cyrillic font encoding has been loaded so far.
> (babel)                A font encoding should be declared before babel.
> (babel)                Default `T2A' encoding will be loaded  on input
> line 74.

Yeah, that's something I forgot to mention. The warning is from
russianb.ldf:

-------------
\ifx\cyrillicencoding\undefined
  \if@uni@ode
    %\ifdefined\XeTeXrevision
    %  \edef\cyrillicencoding{EU1}
    %\else\ifdefined\luatexversion
    %  \edef\cyrillicencoding{EU2}
    %\fi\fi
    \edef\cyrillicencoding{TU}
  \else
    \edef\cyrillicencoding{T2A}
  \fi
  \PackageWarning{babel}%
    {No Cyrillic font encoding has been loaded so far.\MessageBreak
     A font encoding should be declared before babel.\MessageBreak
     Default `\cyrillicencoding' encoding will be loaded
    }%
  \lowercase\expandafter{\expandafter\input\cyrillicencoding enc.def\relax}%
  \AtBeginDocument{\@setcyrillicencoding}
\fi
-------------

But there is no warning in the case of Greek, where the LGR fontencoding
is not explicitly indicated either. In any case, I have commented on
this issue on the Hispanic TeX mailing list, where Javier Bezos (current
Babel maintainer) participates. It would be interesting if he could
clarify this for us...

> Unfortunately in the case of
>
>     \usepackage[russian,english]{babel}
>
> \selectlanguage{russian} is required, without it compilation fails with
>
>    ! LaTeX Error: Command \cyrn unavailable in encoding OT1.

Yes, this is the case of the example that I put in my previous message
(russian as second language). If a explicit babel command is not added,
the T2A fontencoding is not loaded. Therefore, you would have to add a
selectlanguage{russian} or, at a low level,
\fontencoding{T2A}\selectfont

> With \usepackage[T2A]{fontenc} it behaves accordingly to acceptable
> for non-important documents graceful degradation. Text is readable,
> but no hyphenation applied.
>
> So I consider explicit loading of fontenc as more reliable.

Wouldn't it be easier in those cases to just load the fontenc package a
second time:

#+LaTeX_Header: \usepackage[T1,T2A]{fontenc}

According to what Egreg says in this thread
(https://tex.stackexchange.com/a/79112), "The only package that's
allowed to be loaded multiple times is fontenc".

So my logic is as follows: since Org has always loaded fontenc with the T1
option by default, I don't see much point in changing that behavior now
that pdfLaTeX is, so to speak, in retirement. And in any case, the user
can change (now) the fontencoding via the babel commands or by loading
the package a second time. For the record, I'm not opposed to redoing
the patch by adding the features you're proposing. Simply, from my point
of view, I think it's not worth the effort. But it's just my opinion,
and besides, I don't use pdfLaTeX, so I may have a bias here.

Best regards,

Juan Manuel 

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

Max Nikulin writes:

> I would consider structures with named fields (alists or plists) for a
> case of adding some additional settings (Font name? But it is rather
> defcustom than defconst)
>
> ("es" . (:babel "spanishmx" :poliglossia "spanish"
> :poliglossia-variant "mexican")

I was paying more attention to the fontenc issue and I had forgotten to
comment this.

I think your proposal to use a property list makes sense. But I don't
quite see what new settings could be added without overcomplicating
things. Babel in its latest versions has several ways to load languages,
and many new commands to select fonts or associate font families to a
specific language or script. But they don't work with pdfLaTeX, so the
only thing I could think of is to keep the old babel method, valid for
all TeX engines, according to which, if a user puts in an org document:

#+language: es
#+latex_header: \usepackage[foo,AUTO]{babel}

it returns:

\usepackage[foo,spanish]{babel}

which is valid for all flavors of TeX. And I think that this way, as Org
was doing so far, it will satisfy most users.

But given the syntactical variety that babel now has (polyglossia is
simpler), I don't see how all of that could be 'translated' to Org via
Org settings. I think this is one of those cases where it's easier for
the user to just build the LaTeX preamble writing LaTeX code or create a
new 'class' for org-latex-classes. The problem with Org writing the
preamble for the user is that it will never satisfy all users. That is
why I think it is necessary for this 'automatic' preamble to be as basic
and elementary as possible (I, for example, always write the preamble
from scratch or write my own sty files). Otherwise we run the risk of
converting, or wanting to convert, Org into a clone of LaTeX, but less
flexible than LaTeX.

I agree that most Org users (like most Pandoc users) just want to
produce a clean pdf without messing with LaTeX. But if users want to do
more things in LaTeX, they should know some LaTeX, even if they prefer
to use a lightweight markup language as a latex 'interface'. That's why
I think it's great that in Org you can enter arbitrary LaTeX (or html)
code anywhere and at many levels. I think this is the real killer
feature of Org. I don't know if I'm explaining myself... But this
reflection of mine (which, of course, is debatable) would take us
further, and this is not the case here. 

Other than that, your idea of using a property list sounds good to me.
What happens is that I do not see a clear advantage (at least in the
short term).

Best regards,

Juan Manuel 

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Ihor Radchenko]

Juan Manuel Macías <maciaschain@posteo.net> writes:

> I attach the new version of the patch with both variables declared
> obsolete.

Thanks!

We usually declare obsolete variables in org-compat.el.

> If everything is ok, I can add what is necessary to NEWS and to the Org Manual.

I have some minor comments. You can address them and then go ahead with
NEWS and manual.

For Max's comment, using plist/alist would make things more clear
code-wise for future developers. I always find it annoying when I need
to go back and forth checking which element should be second or third or
forth in the list. Especially if the variable is used all around the
codebase. Though this particular case is not such -
`org-latex-language-alist' is used just in few places.

> +(make-obsolete-variable 'org-latex-babel-language-alist
> +                        "set `org-latex-language-alist' instead." "9.6")
> +
> +(make-obsolete-variable 'org-latex-polyglossia-language-alist
> +                        "set `org-latex-language-alist' instead." "9.6")
> +

As I mentioned earlier, please move the obsoletion statements to org-compat.

> -  "Alist between language code and corresponding Polyglossia option.")
> +  "Alist between language code and corresponding Babel/Polyglossia option.
> +
> +For the names of the languages, the Babel nomenclature is
> +preferred to that of Polyglossia, in those cases where both
> +coincide.
> +
> +The alist supports three types of members:
> +
> +- Members with two elements: CODE BABEL/POLYGLOSSIA OPTION.
> +
> +- Members with three elements: CODE BABEL/POLYGLOSSIA OPTION
> +ASTERISK (the presence of the asterisk indicates that this
> +language is not loaded in Babel using the old method of ldf
> +files but using ini files. If Babel is loaded in an Org
> +document with these languages, the \"AUTO \" argument is just
> +removed, to avoid compilation errors).

Two spaces between sentences please, as in
https://orgmode.org/worg/org-contribute.html

>  	;; If LANGUAGE is already loaded, return header without AUTO.
>  	;; Otherwise, replace AUTO with language or append language if
>  	;; AUTO is not present.
>  	(replace-match
>  	 (mapconcat (lambda (option) (if (equal "AUTO" option) language option))
>  		    (cond ((member language options) (delete "AUTO" options))
> +			  ((let ((l (assoc language-code org-latex-language-alist)))
> +			     (and (consp l) (= (length l) 3))) (delete "AUTO" options))

A comment explaining why "3" would be useful.

> -			      (if (and (consp l) (= (length l) 3))
> -				  (format "[variant=%s]" (nth 2 l))
> +			      (if (and (consp l) (= (length l) 4))
> +				  (format "[variant=%s]" (nth 3 l))
>  				"")
> -			      (nth 1 l))))
> +			      (if (and (consp l) (= (length l) 4))
> +				  (nth 2 l)
> +				(nth 1 l)))))

Again, comment explaining all these 2,3,4 would help.

Best,
Ihor

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

Ihor Radchenko writes:

> For Max's comment, using plist/alist would make things more clear
> code-wise for future developers. I always find it annoying when I need
> to go back and forth checking which element should be second or third or
> forth in the list. Especially if the variable is used all around the
> codebase. Though this particular case is not such -
> `org-latex-language-alist' is used just in few places.

I agree with Maxim and with you that an anonymous list is hardly usable
for future features. For this new list I followed the style of the
previous ones and I admit that I was quite conservative. The problem,
imho, is that with the current org implementation I find it difficult to
add new features. For example, Babel now has two ways of loading
languages: the old one, through ldf files, which is the one that Org
implements and that produces the traditional babel syntax; and the new
system through ini files, which also incorporates a new syntax with many
options and variants. New languages have been defined by ini files, but
cannot be loaded by the old syntax. That is the cause of the asterisks
in my list: when a language has an asterisk it means that it is only
served in babel through an ini file and, therefore, the traditional
syntax cannot be used, so it is not implemented for use in babel. And
furthermore, the new babel syntax and ini files can only be used with
the Unicode TeX engines: XeTeX and LuaTeX. This all looks like a puzzle
(I'm getting dizzy just rereading it :-)), and I don't see a clean and
sensible way to translate it to Org settings.

I also think that the current Org settings for loading languages with
Polyglossia or Babel is unhelpful and unclear. Also, it depends on
putting explicit LaTeX code in the document. A code that, in the case of
Polyglossia, is a fake LaTeX code, because something like this:

\usepackage[french,AUTO]{polyglossia}

is not really the Polyglossia syntax. And it can lead to confusion.

I think this should be reimplemented in the future using a more
org-centric syntax, using keywords (somehow keeping the
above for backwards compatibility). I don't know, maybe something like
this, with a specific keyword for language LaTeX settings:

#+latex_language: es variant:mx babel-ini:t other:en,de,el etc.

In this case, I think it would make more sense to define a more robust
list, an alist or a plist (as Maxim suggested), leaving the door open to
a fresh reimplementation of langs in latex export. I can get to work as
soon as I have some time and finish with other commitments, that will
keep me busy practically all summer.

WDYT?

Best regards,

Juan Manuel 

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Ihor Radchenko]

Juan Manuel Macías <maciaschain@posteo.net> writes:

>> For Max's comment, using plist/alist would make things more clear
>> code-wise for future developers. I always find it annoying when I need
>> to go back and forth checking which element should be second or third or
>> forth in the list. Especially if the variable is used all around the
>> codebase. Though this particular case is not such -
>> `org-latex-language-alist' is used just in few places.
>
> I agree with Maxim and with you that an anonymous list is hardly usable
> for future features. For this new list I followed the style of the
> previous ones and I admit that I was quite conservative. The problem,
> imho, is that with the current org implementation I find it difficult to
> add new features. For example, Babel now has two ways of loading
> languages: the old one, through ldf files, which is the one that Org
> implements and that produces the traditional babel syntax; and the new
> system through ini files, which also incorporates a new syntax with many
> options and variants. New languages have been defined by ini files, but
> cannot be loaded by the old syntax. That is the cause of the asterisks
> in my list: when a language has an asterisk it means that it is only
> served in babel through an ini file and, therefore, the traditional
> syntax cannot be used, so it is not implemented for use in babel. And
> furthermore, the new babel syntax and ini files can only be used with
> the Unicode TeX engines: XeTeX and LuaTeX. This all looks like a puzzle
> (I'm getting dizzy just rereading it :-)), and I don't see a clean and
> sensible way to translate it to Org settings.

Do you refer to the paragraph below when saying that Org implementation
makes it hard to add new features? The rest of the above paragraph
implies that the difficulty is on LaTeX side, not on Org side.

In any case, your existing patch is an already an improvement. I do not
deem it as requirement to apply the patch.

> I also think that the current Org settings for loading languages with
> Polyglossia or Babel is unhelpful and unclear. Also, it depends on
> putting explicit LaTeX code in the document. A code that, in the case of
> Polyglossia, is a fake LaTeX code, because something like this:
>
> \usepackage[french,AUTO]{polyglossia}
>
> is not really the Polyglossia syntax. And it can lead to confusion.

I hope that Timothy's work on more flexible and configurable template
generation can improve things.

> I think this should be reimplemented in the future using a more
> org-centric syntax, using keywords (somehow keeping the
> above for backwards compatibility). I don't know, maybe something like
> this, with a specific keyword for language LaTeX settings:
>
> #+latex_language: es variant:mx babel-ini:t other:en,de,el etc.
>
> In this case, I think it would make more sense to define a more robust
> list, an alist or a plist (as Maxim suggested), leaving the door open to
> a fresh reimplementation of langs in latex export. I can get to work as
> soon as I have some time and finish with other commitments, that will
> keep me busy practically all summer.
>
> WDYT?

This sounds reasonable. Implementing via :options-alist export property
will conform more to the overall design (see the docstring of
org-export-options-alist - it enables a lot of flexibility in terms of
customization, including the option syntax similar to what you proposed).

Best,
Ihor

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

Ihor Radchenko writes:

> Do you refer to the paragraph below when saying that Org implementation
> makes it hard to add new features? The rest of the above paragraph
> implies that the difficulty is on LaTeX side, not on Org side.

Sorry for not explaining clearly. Actually I think the problem is not
with Org or Babel (LaTeX), but rather with the "translation" from Babel
(latex) to Org. Currently the babel interface is more complex, although
it is more consistent and robust within LaTeX. The challenge is how to
style all of that for an Org user who wants to just get a correct PDF
out-of-the-box using a simple and basic syntax (of course, you don't
need to translate the whole babel: to use babel with all its power it's
better to write the LaTeX code directly). 

For that reason I think that, for now, it is more practical to keep the
old babel syntax on the Org side (as I do in my patch):

\usepackage[langs]{babel}

which is perfectly valid for pdflatex, lualatex and xelatex (except in
the languages that are loaded in babel through ini files).

> In any case, your existing patch is an already an improvement. I do not
> deem it as requirement to apply the patch.

Well, between today and tomorrow I can prepare a new version of the
patch with all your suggestions from the previous email incorporated.
And for the future I can work on a fresh re-implementation of all this,
using the :options-alist export property, as you suggest.

For example, something like this:

latex-lang: babel(sorbian) variant(upper) provide(hebrew:import,hyphenrules=+) options(bidi=default), others(french,catalan) 

returns:

\usepackage[french,catalan,uppersorbian,bidi=default]{babel}
\babelprovide[import,hyphenrules=+]{hebrew}

or this (in this case Hebrew is the main language):

latex-lang: babel provide(hebrew:main,import,hyphenrules=+) options(bidi=default), others(french,catalan) 

returns:

\usepackage[french,catalan,bidi=default]{babel}
\babelprovide[main,import,hyphenrules=+]{hebrew}

And the above is equivalent to:

latex-lang: babel(hebrew) options(bidi=default,provide=*), others(french,catalan) 

returns:

\usepackage[french,catalan,hebrew,bidi=default,provide=*]{babel}

Note that for monolingual documents, now in babel it would be enough to
indicate the language in the document class options (only for LuaTeX and
XeTeX), so it would be enough to do something like this:

#+latex_class: article
#+latex_class_options: [french,10pt,draft]
#+LaTeX_Header: \usepackage{babel}


Best regards,

Juan Manuel 

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

Juan Manuel Macías writes:

> latex-lang: babel(sorbian) variant(upper) provide(hebrew:import,hyphenrules=+) options(bidi=default), others(french,catalan) 
>
> returns:
>
> \usepackage[french,catalan,uppersorbian,bidi=default]{babel}
> \babelprovide[import,hyphenrules=+]{hebrew}

PS: In fact, I think that this new implementation would no longer depend
on a list of languages neither for babel nor for polyglossia, since it
is not connected to the #+language keyword. The old list would be kept
for backwards compatibility.

Best regards,

Juan Manuel 

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Max Nikulin]

On 18/07/2022 17:32, Juan Manuel Macías wrote:
> 
> For example, something like this:
> 
> latex-lang: babel(sorbian) variant(upper) provide(hebrew:import,hyphenrules=+) options(bidi=default), others(french,catalan)
> 
> returns:
> 
> \usepackage[french,catalan,uppersorbian,bidi=default]{babel}
> \babelprovide[import,hyphenrules=+]{hebrew}

I have never tried so complex babel configuration. Should "latex-lang" 
options affect ox-latex in any way besides adding 
\usepackage[...]{babel} and \babelprovide[...]{...}? If not, maybe it is 
better to use

#+latex_header: \usepackage[french,catalan,uppersorbian,bidi=default]{babel}
#+latex_header: \babelprovide[import,hyphenrules=+]{hebrew}

directly and add default \usepackage[french]{babel} based on 
"#+language:" only in the case when babel is not loaded through 
"#+latex_header:". Currently such strategy is implemented for inputenc 
or some other package.

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

Max Nikulin writes:

> On 18/07/2022 17:32, Juan Manuel Macías wrote:
>> For example, something like this:
>> latex-lang: babel(sorbian) variant(upper)
>> provide(hebrew:import,hyphenrules=+) options(bidi=default),
>> others(french,catalan)
>> returns:
>> \usepackage[french,catalan,uppersorbian,bidi=default]{babel}
>> \babelprovide[import,hyphenrules=+]{hebrew}
>
> I have never tried so complex babel configuration. Should "latex-lang"
> options affect ox-latex in any way besides adding
> \usepackage[...]{babel} and \babelprovide[...]{...}? If not, maybe it
> is better to use

I really think not. It's more of a way to have a more or less complete
configuration of babel (including font options) in a single line and
with a more org-centric syntax. The drawback is that the user would have
to know in this case what he is putting, and know the babel options.

> #+latex_header: \usepackage[french,catalan,uppersorbian,bidi=default]{babel}
> #+latex_header: \babelprovide[import,hyphenrules=+]{hebrew}

That's what I would do as a user. I write all my preambles in pure LaTeX
and don't let Org write anything before the \begin{document} :-)

> directly and add default \usepackage[french]{babel} based on
> "#+language:" only in the case when babel is not loaded through
> "#+latex_header:". Currently such strategy is implemented for inputenc
> or some other package.

humm, what do you think about this idea?:

If I'm not mistaken, the current behavior is as follows. If in the org
document there is this:

#+latex_header: \usepackage[english,french,AUTO]{babel}
#+language: es

it returns:

\usepackage[english,french,spanish]{babel}

But it could be extended to the new babel syntax, so that:

#+latex_header: \usepackage[french,catalan,AUTO,bidi=default]{babel}
#+latex_header: \babelprovide[import,hyphenrules=+]{hebrew}
#+language: es

returns:

\usepackage[french,catalan,spanish,bidi=default]{babel}
\babelprovide[import,hyphenrules=+]{hebrew}

Or, if we want to load the main language via babelprovide (using an ini
file instead of an ldf file):

#+latex_header: \usepackage[french,catalan,spanish]{babel}
#+latex_header: \babelprovide[main,import]{AUTO}
#+language: ru

returns:

\usepackage[french,catalan,spanish]{babel}
\babelprovide[main,import]{russian}

(in this case french, catalan and spanish are the secondary languages)

And this could be extended (in LuaTeX and XeTeX only) to the class
options (which is a new babel feature):

#+latex_class: article
#+latex_class_options: [11pt,draft,AUTO]
#+LaTeX_Header: \usepackage{babel}
#+language: ru

returns:

\documentclass[11pt,draft,russian]{article}
\usepackage{babel}

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

[-- Attachment #1: Type: text/plain, Size: 1209 bytes --]

Here is a new version of the patch, with the fixes added.

Important: I have modified in this patch org-latex-guess-babel-language
so that it recognizes the new Babel syntax alongside the old syntax.
That is, it is now possible to put:

#+LaTeX_Header: \usepackage[arguments,AUTO]{babel}
#+LaTeX_Header: \babelprovide[arguments]{AUTO}

Languages that are served in Babel *exclusively* via ini files (ie those
with an asterisk in the new list) are not added to the Babel argument
(they must be loaded via babelprovide).

However, the following situation may also occur. A user wants to load
the secondary languages via ldf files and the main language via ini
file (babelprovide):

#+LaTeX_Header: \usepackage[french,english]{babel}
#+LaTeX_Header: \babelprovide[main, import]{AUTO}
#+language: ru

This would produce in LaTeX:

\usepackage[french, english, russian]{babel}
\babelprovide[main, import]{russian}

I have not prevented this behavior as it is correct in Babel: you can
load the main language using the 'old style' and then redefine it using
babelprovide, which is a complement. Besides, maintaining this behavior
is also necessary to preserve backwards compatibility.

Best regards,

Juan Manuel 


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-lisp-ox-latex.el-New-variable-org-latex-language-ali.patch --]
[-- Type: text/x-patch, Size: 12780 bytes --]

From 2f78d6a75849819f1d3aececff70b7ffa4f36c7c Mon Sep 17 00:00:00 2001
From: Juan Manuel Macias <maciaschain@posteo.net>
Date: Tue, 19 Jul 2022 16:51:55 +0200
Subject: [PATCH] * lisp/ox-latex.el: New variable `org-latex-language-alist'

(org-latex-language-alist): Unify in a single list
`org-latex-polyglossia-language-alist' and
`org-latex-babel-language-alist', and make the two variables
obsolete. However, it may be convenient in the future to replace this
list with a more robust one. (see:
`https://list.orgmode.org/taeb0a$r62$1@ciao.gmane.io')

(org-latex-guess-babel-language): This function has been modified so
that the new Babel command `babelprovide' is also recognized. This
command is necessary to load the languages served by Babel exclusively
through an ini file. Therefore, the new Babel syntax is supported
alongside the old one.  Note that languages that are served
exclusively via an ini file are not added to the Babel argument.
---
 lisp/org-compat.el |   8 ++
 lisp/ox-latex.el   | 236 ++++++++++++++++++++++-----------------------
 2 files changed, 125 insertions(+), 119 deletions(-)

diff --git a/lisp/org-compat.el b/lisp/org-compat.el
index 6f663cc24..835ec2828 100644
--- a/lisp/org-compat.el
+++ b/lisp/org-compat.el
@@ -880,6 +880,12 @@ context.  See the individual commands for more information."
   'org-truly-invisible-p "9.6"
   "Compatibility alias for legacy misspelling of `org-truly-invisible-p'.")
 
+(make-obsolete-variable 'org-latex-babel-language-alist
+                        "set `org-latex-language-alist' instead." "9.6")
+
+(make-obsolete-variable 'org-latex-polyglossia-language-alist
+                        "set `org-latex-language-alist' instead." "9.6")
+
 ;;;; Obsolete link types
 
 (eval-after-load 'ol
@@ -888,6 +894,8 @@ context.  See the individual commands for more information."
      (org-link-set-parameters "file+sys"))) ;since Org 9.0
 
 
+
+
 \f
 ;;; Miscellaneous functions
 
diff --git a/lisp/ox-latex.el b/lisp/ox-latex.el
index 1aab8ffd5..2eed44b62 100644
--- a/lisp/ox-latex.el
+++ b/lisp/ox-latex.el
@@ -172,144 +172,111 @@
 \f
 ;;; Internal Variables
 
-(defconst org-latex-babel-language-alist
-  '(("af" . "afrikaans")
-    ("bg" . "bulgarian")
-    ("ca" . "catalan")
-    ("cs" . "czech")
-    ("cy" . "welsh")
-    ("da" . "danish")
-    ("de" . "germanb")
-    ("de-at" . "naustrian")
-    ("de-de" . "ngerman")
-    ("el" . "greek")
-    ("en" . "english")
-    ("en-au" . "australian")
-    ("en-ca" . "canadian")
-    ("en-gb" . "british")
-    ("en-ie" . "irish")
-    ("en-nz" . "newzealand")
-    ("en-us" . "american")
-    ("es" . "spanish")
-    ("et" . "estonian")
-    ("eu" . "basque")
-    ("fi" . "finnish")
-    ("fr" . "french")
-    ("fr-ca" . "canadien")
-    ("gl" . "galician")
-    ("hr" . "croatian")
-    ("hu" . "hungarian")
-    ("id" . "indonesian")
-    ("is" . "icelandic")
-    ("it" . "italian")
-    ("la" . "latin")
-    ("ms" . "malay")
-    ("nl" . "dutch")
-    ("nb" . "norsk")
-    ("nn" . "nynorsk")
-    ("no" . "norsk")
-    ("pl" . "polish")
-    ("pt" . "portuguese")
-    ("pt-br" . "brazilian")
-    ("ro" . "romanian")
-    ("ru" . "russian")
-    ("sa" . "sanskrit")
-    ("sb" . "uppersorbian")
-    ("sk" . "slovak")
-    ("sl" . "slovene")
-    ("sq" . "albanian")
-    ("sr" . "serbian")
-    ("sv" . "swedish")
-    ("ta" . "tamil")
-    ("tr" . "turkish")
-    ("uk" . "ukrainian"))
-  "Alist between language code and corresponding Babel option.")
-
-(defconst org-latex-polyglossia-language-alist
-  '(("am" "amharic")
+(defconst org-latex-language-alist
+  ;; TODO: replace this list with a property list (the actual
+  ;; implementation is not very robust).
+  '(("am" "amharic" "*")
     ("ar" "arabic")
-    ("ast" "asturian")
+    ("ast" "asturian" "*")
     ("bg" "bulgarian")
-    ("bn" "bengali")
-    ("bo" "tibetan")
+    ("bn" "bengali" "*")
+    ("bo" "tibetan" "*")
     ("br" "breton")
     ("ca" "catalan")
-    ("cop" "coptic")
+    ("cop" "coptic" "*")
     ("cs" "czech")
     ("cy" "welsh")
     ("da" "danish")
-    ("de" "german" "german")
-    ("de-at" "german" "austrian")
-    ("de-de" "german" "german")
-    ("dsb" "lsorbian")
-    ("dv" "divehi")
+    ("de" "ngerman" "german" "german")
+    ("de-at" "naustrian" "german" "austrian")
+    ("dsb" "lsorbian" "*")
+    ("dv" "divehi" "*")
     ("el" "greek")
-    ("en" "english" "usmax")
-    ("en-au" "english" "australian")
-    ("en-gb" "english" "uk")
-    ("en-nz" "english" "newzealand")
-    ("en-us" "english" "usmax")
+    ("el-polyton" "polutonikogreek" "greek" "polytonic")
+    ("en" "american" "english" "usmax")
+    ("en-au" "australian" "english" "australian")
+    ("en-gb" "british" "english" "uk")
+    ("en-nz" "newzealand" "english" "newzealand")
+    ("en-us" "american" "english" "usmax")
     ("eo" "esperanto")
     ("es" "spanish")
+    ("es-mx" "spanishmx" "spanish" "mexican")
     ("et" "estonian")
     ("eu" "basque")
     ("fa" "farsi")
     ("fi" "finnish")
     ("fr" "french")
-    ("fu" "friulan")
+    ("fr-ca" "canadien" "french" "canadian")
+    ("fur" "friulan")
     ("ga" "irish")
     ("gd" "scottish")
     ("gl" "galician")
     ("he" "hebrew")
     ("hi" "hindi")
     ("hr" "croatian")
-    ("hsb" "usorbian")
+    ("hsb" "uppersorbian" "sorbian" "upper")
     ("hu" "magyar")
-    ("hy" "armenian")
+    ("hy" "armenian" "*")
     ("ia" "interlingua")
-    ("id" "bahasai")
+    ("id" "bahasai" "*")
     ("is" "icelandic")
     ("it" "italian")
-    ("kn" "kannada")
-    ("la" "latin" "modern")
-    ("la-classic" "latin" "classic")
-    ("la-medieval" "latin" "medieval")
-    ("la-modern" "latin" "modern")
-    ("lo" "lao")
+    ("kn" "kannada" "*")
+    ("la" "latin")
+    ("la-classic" "classiclatin" "latin" "classic")
+    ("la-medieval" "medievallatin" "latin" "medieval")
+    ("la-ecclesiastic" "ecclesiasticlatin" "latin" "ecclesiastic")
+    ("lo" "lao" "*")
     ("lt" "lithuanian")
     ("lv" "latvian")
-    ("ml" "malayalam")
-    ("mr" "maranthi")
-    ("nb" "norsk")
-    ("nko" "nko")
+    ("ml" "malayalam" "*")
+    ("mr" "maranthi" "*")
+    ("nb" "norsk" "norwegian" "bokmal")
     ("nl" "dutch")
-    ("nn" "nynorsk")
+    ("nn" "nynorsk" "norwegian" "nynorsk")
     ("no" "norsk")
     ("oc" "occitan")
     ("pl" "polish")
     ("pms" "piedmontese")
     ("pt" "portuges")
     ("pt-br" "brazilian")
-    ("rm" "romansh")
+    ("rm" "romansh" "*")
     ("ro" "romanian")
     ("ru" "russian")
-    ("sa" "sanskrit")
-    ("se" "samin")
+    ("sa" "sanskrit" "*")
     ("sk" "slovak")
-    ("sl" "slovenian")
+    ("sl" "slovene")
     ("sq" "albanian")
     ("sr" "serbian")
     ("sv" "swedish")
-    ("syr" "syriac")
-    ("ta" "tamil")
-    ("te" "telugu")
+    ("syr" "syriac" "*")
+    ("ta" "tamil" "*")
+    ("te" "telugu" "*")
     ("th" "thai")
     ("tk" "turkmen")
     ("tr" "turkish")
     ("uk" "ukrainian")
-    ("ur" "urdu")
+    ("ur" "urdu" "*")
     ("vi" "vietnamese"))
-  "Alist between language code and corresponding Polyglossia option.")
+  "Alist between language code and corresponding Babel/Polyglossia option.
+
+For the names of the languages, the Babel nomenclature is
+preferred to that of Polyglossia, in those cases where both
+coincide.
+
+The alist supports three types of members:
+
+- Members with two elements: CODE BABEL/POLYGLOSSIA OPTION.
+
+- Members with three elements: CODE BABEL/POLYGLOSSIA OPTION
+ASTERISK (the presence of the asterisk indicates that this
+language is not loaded in Babel using the old method of ldf
+files but using ini files. If Babel is loaded in an Org
+document with these languages, the \"AUTO \" argument is just
+removed, to avoid compilation errors).
+
+- Members with four elements (for variants of languages): CODE
+BABEL-OPTION POLYGLOSSIA-OPTION POLYGLOSSIA-VARIANT")
 
 (defconst org-latex-table-matrix-macros '(("bordermatrix" . "\\cr")
 					  ("qbordermatrix" . "\\cr")
@@ -1644,31 +1611,54 @@ Insertion of guessed language only happens when Babel package has
 explicitly been loaded.  Then it is added to the rest of
 package's options.
 
-The argument to Babel may be \"AUTO\" which is then replaced with
-the language of the document or `org-export-default-language'
-unless language in question is already loaded.
+The optional argument to Babel or the mandatory argument to
+`\babelprovide' command may be \"AUTO\" which is then replaced
+with the language of the document or
+`org-export-default-language' unless language in question is
+already loaded.
 
 Return the new header."
-  (let ((language-code (plist-get info :language)))
-    ;; If no language is set or Babel package is not loaded, return
-    ;; HEADER as-is.
-    (if (or (not (stringp language-code))
-	    (not (string-match "\\\\usepackage\\[\\(.*\\)\\]{babel}" header)))
+  (let* ((language-code (plist-get info :language))
+	 (language (nth 1 (assoc language-code
+				 org-latex-language-alist)))
+	 ;; If no language is set or Babel package is not loaded, return
+ 	 ;; HEADER as-is.
+	 (header (if (or (not (stringp language-code))
+			 (not (string-match "\\\\usepackage\\[\\(.*\\)\\]{babel}" header)))
+		     header
+		   (let ((options (save-match-data
+				    (org-split-string (match-string 1 header) ",[ \t]*"))))
+		     ;; If LANGUAGE is already loaded, return header
+		     ;; without AUTO.  Otherwise, replace AUTO with language or
+		     ;; append language if AUTO is not present.  Languages that are
+		     ;; served in Babel exclusively through ini files are not added
+		     ;; to the babel argument, and must be loaded using
+		     ;; `\babelprovide'.
+		     (let ((l (assoc language-code org-latex-language-alist)))
+                       ;; Three elements imply that LANGUAGE is served
+                       ;; in Babel only by means of an ini file.
+                       ;; Therefore it will not be added to the Babel
+                       ;; argument.  TODO: this should be improved
+                       ;; when `org-latex-language-alist' is replaced
+                       ;; by a more robust list.
+		       (if (and (consp l) (= (length l) 3))
+                           header
+			 (replace-match
+			  (mapconcat (lambda (option) (if (equal "AUTO" option) language option))
+				     (cond ((member language options) (delete "AUTO" options))
+					   ((member "AUTO" options) options)
+					   (t (append options (list language))))
+				     ", ")
+			  t nil header 1)))))))
+    ;; if `\babelprovide[args]{AUTO}' is present, AUTO is
+    ;; replaced by LANGUAGE.
+    (if (not (string-match "\\\\babelprovide\\[.*\\]{\\(.+\\)}" header))
 	header
-      (let ((options (save-match-data
-		       (org-split-string (match-string 1 header) ",[ \t]*")))
-	    (language (cdr (assoc-string language-code
-					 org-latex-babel-language-alist t))))
-	;; If LANGUAGE is already loaded, return header without AUTO.
-	;; Otherwise, replace AUTO with language or append language if
-	;; AUTO is not present.
-	(replace-match
-	 (mapconcat (lambda (option) (if (equal "AUTO" option) language option))
-		    (cond ((member language options) (delete "AUTO" options))
-			  ((member "AUTO" options) options)
-			  (t (append options (list language))))
-		    ", ")
-	 t nil header 1)))))
+      (let ((prov (match-string 1 header)))
+	(when (equal "AUTO" prov)
+	  (replace-regexp-in-string (format
+				     "\\(\\\\babelprovide\\[.*\\]\\)\\({\\)%s}" prov)
+				    (format "\\1\\2%s}" language) header t))))))
 
 (defun org-latex-guess-polyglossia-language (header info)
   "Set the Polyglossia language according to the LANGUAGE keyword.
@@ -1710,15 +1700,23 @@ Return the new header."
 	 (concat "\\usepackage{polyglossia}\n"
 		 (mapconcat
 		  (lambda (l)
-		    (let ((l (or (assoc l org-latex-polyglossia-language-alist)
+		    (let ((l (or (assoc l org-latex-language-alist)
 				 l)))
 		      (format (if main-language-set "\\setotherlanguage%s{%s}\n"
 				(setq main-language-set t)
 				"\\setmainlanguage%s{%s}\n")
-			      (if (and (consp l) (= (length l) 3))
-				  (format "[variant=%s]" (nth 2 l))
+                              ;; Four elements implies that there is a
+                              ;; variant (4) for LANGUAGE when
+                              ;; declared by Polyglossia (3). TODO:
+                              ;; this should be improved when
+                              ;; `org-latex-language-alist' is
+                              ;; replaced by a more robust list.
+                              (if (and (consp l) (= (length l) 4))
+				  (format "[variant=%s]" (nth 3 l))
 				"")
-			      (nth 1 l))))
+			      (if (and (consp l) (= (length l) 4))
+				  (nth 2 l)
+				(nth 1 l)))))
 		  languages
 		  ""))
 	 t t header 0)))))
-- 
2.37.1

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Max Nikulin]

On 19/07/2022 22:01, Juan Manuel Macías wrote:
> +			 (replace-match
> +			  (mapconcat (lambda (option) (if (equal "AUTO" option) language option))
> +				     (cond ((member language options) (delete "AUTO" options))
> +					   ((member "AUTO" options) options)
> +					   (t (append options (list language))))
> +				     ", ")

In my opinion this code should not attempt to be excessively clever. If 
user skipped AUTO then do not append language. Test for duplicated 
options is redundant as well. Such cases may still be a reason to issue 
a warning (e.g. by `org-lint').

On the other hand I would consider adding babel by default without 
explicit header. To suppress loading users may add
#+latex_header: % \usepackage{babel}

I like that you decided to avoid inventing of a DSL just to have 
org-like options that are translated to to a couple of preamble 
commands. From my point of view it would not help novices and would make 
things more complicated for experienced LaTeX users.

> #+latex_class: article
> #+latex_class_options: [11pt,draft,AUTO]
> #+LaTeX_Header: \usepackage{babel}
> #+language: ru

While AUTO is suitable for \usepackage{babel} and \babelprovide, for 
class option the placeholder should be clearly associated with babel, 
e.g. BABEL_LANG instead of AUTO.

I am not familiar enough with setting used to generate preview of 
equations or images from LaTeX source blocks, so I am not completely 
sure that suggested changes do not affect these features in some 
negative way.

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

Max Nikulin writes:

> On 19/07/2022 22:01, Juan Manuel Macías wrote:
>> +			 (replace-match
>> + (mapconcat (lambda (option) (if (equal "AUTO" option) language
>> option))
>> + (cond ((member language options) (delete "AUTO" options))
>> + ((member "AUTO" options) options)
>> + (t (append options (list language))))
>> +				     ", ")
>
> In my opinion this code should not attempt to be excessively clever.
> If user skipped AUTO then do not append language. Test for duplicated
> options is redundant as well. Such cases may still be a reason to
> issue a warning (e.g. by `org-lint').

I completely agree. I've kept that old part of the code for backwards
compatibility and because, at the end of the day, it doesn't break
anything new. I mean, if a user declares the main language using
babelprovide and this code (too intrusive) puts the main language in the
Babel argument too (something the user doesn't want in origin), that
syntax is supported by Babel. Babel simply takes into account the main
language declared with babelprovide, if the 'main' option has been
added. But even if it doesn't return any errors, it's unnecessary
redundancy.

Anyway, yes, it's too intrusive. I am in favor of removing that part,
but I don't know if it will have any effect on backwards compatibility.

(BTW, I think I didn't explain in this thread the advantages of using
babelprovide over the traditional ldf file system. With ini files the
user has more control over the loaded language and there are many
options, including the ability to associate a font with a script.
Furthermore the user can create his own custom ini files, and define his
own languages. For example, I could write an ini file with specific
values for Spanish, or even an imaginary language. On the one hand, it
is an important advance. But on the other hand it adds more fuel to the
current great latex pandemic: multiplicity).

> On the other hand I would consider adding babel by default without
> explicit header. To suppress loading users may add
> #+latex_header: % \usepackage{babel}

I don't understand this very well. What would happen, then, to users who
prefer to use Polyglossia, or those who prefer to explicitly add babel
or polyglossia code?

> I like that you decided to avoid inventing of a DSL just to have
> org-like options that are translated to to a couple of preamble
> commands. From my point of view it would not help novices and would
> make things more complicated for experienced LaTeX users.

Yes, in the end I realized that this was getting into a slippery slope,
especially for the reasons you mention...

>> #+latex_class: article
>> #+latex_class_options: [11pt,draft,AUTO]
>> #+LaTeX_Header: \usepackage{babel}
>> #+language: ru
>
> While AUTO is suitable for \usepackage{babel} and \babelprovide, for
> class option the placeholder should be clearly associated with babel,
> e.g. BABEL_LANG instead of AUTO.

What you say makes sense. However, this was ultimately not implemented
in the current version of the patch because the argument of
org-latex-guess-babel-language is a #+latex_header keyword.

> I am not familiar enough with setting used to generate preview of
> equations or images from LaTeX source blocks, so I am not completely
> sure that suggested changes do not affect these features in some
> negative way.

I think that shouldn't affect previsualization, because if I remember
correctly the preamble used for previews is the value of
org-format-latex-header. But I'm not sure...

Best regards,

Juan Manuel 

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Max Nikulin]

On 20/07/2022 02:31, Juan Manuel Macías wrote:
> Max Nikulin writes:
> 
>> On 19/07/2022 22:01, Juan Manuel Macías wrote:
>>> +			 (replace-match
>>> + (mapconcat (lambda (option) (if (equal "AUTO" option) language
>>> option))
>>> + (cond ((member language options) (delete "AUTO" options))
>>> + ((member "AUTO" options) options)
>>> + (t (append options (list language))))
>>> +				     ", ")
>>
>> In my opinion this code should not attempt to be excessively clever.
>> If user skipped AUTO then do not append language. Test for duplicated
>> options is redundant as well. Such cases may still be a reason to
>> issue a warning (e.g. by `org-lint').
> 
> I completely agree. I've kept that old part of the code for backwards
> compatibility and because, at the end of the day, it doesn't break
> anything new.

I am sorry, I missed the old code below the added lines.

>> On the other hand I would consider adding babel by default without
>> explicit header. To suppress loading users may add
>> #+latex_header: % \usepackage{babel}
> 
> I don't understand this very well. What would happen, then, to users who
> prefer to use Polyglossia, or those who prefer to explicitly add babel
> or polyglossia code?

Certainly if polyglossia or explicit babel related commands are detected 
then default babel configuration is not added to preamble. The idea is 
to add babel if a user have not expressed her intentions explicitly.

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

Max Nikulin writes:

>>> On the other hand I would consider adding babel by default without
>>> explicit header. To suppress loading users may add
>>> #+latex_header: % \usepackage{babel}
>> I don't understand this very well. What would happen, then, to users
>> who
>> prefer to use Polyglossia, or those who prefer to explicitly add babel
>> or polyglossia code?
>
> Certainly if polyglossia or explicit babel related commands are
> detected then default babel configuration is not added to preamble.
> The idea is to add babel if a user have not expressed her intentions
> explicitly.

Ah, I see. I think it's a nice idea. I guess a basic babel setup would
be added to preamble. Something like:

#+language: lang

--> \usepackage[lang]{babel}

But I think also users who use custom preamble templates included in
org-latex-classes or those who load the entire preamble via an external
file (a .sty or .tex file) will want to avoid this. Maybe it would be
nice to add a defcustom, with the following values:

- load babel with the value of #+language, when there is no explicit code from babel (default?)

- load polyglossia, idem but for polyglossia

- nil

- any other arbitrary string?

If the user loads babel or polyglossia explicitly, with AUTO and all
that, then the current rules in
org-latex-guess-babel/polyglossia-language would be applied.

WDYT?

Best regards,

Juan Manuel

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Max Nikulin]

On 21/07/2022 04:30, Juan Manuel Macías wrote:
> Max Nikulin writes:
> 
> #+language: lang
> 
> --> \usepackage[lang]{babel}
> 
> But I think also users who use custom preamble templates included in
> org-latex-classes or those who load the entire preamble via an external
> file (a .sty or .tex file) will want to avoid this. Maybe it would be
> nice to add a defcustom, with the following values:

A custom variable may be convenient, however originally I considered 
only per-document configuration. If a user do not like babel added by 
default then there are some alternatives as

#+latex_header: \usepackage[AUTO]{polyglossia}

or

#+latex_header: % \usepackage{babel} % suppress babel and polyglossia

Certainly with the following variant

#+latex_header: % \usepackage{polyglossia} % no babel or polyglossia

The commented out command is a kind a hack, but I consider it as 
acceptable for advanced users who have custom classes or who need to 
compile document without babel (or polyglossia) for some reason.

If you feel that defcustom should be added as well, variants may be
- nil to suppress babel and polyglossia
- 'babel or 'polyglossia symbols to get language from #+language: 
keyword or from LANGUAGE or LANG environment. Unsure which LC_... 
variable may have greater priority.
- string for exact latex code added to preamble.

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

Max Nikulin writes:

> The commented out command is a kind a hack, but I consider it as
> acceptable for advanced users who have custom classes or who need to
> compile document without babel (or polyglossia) for some reason.

Yes, I think the ability to control this per document is also necessary.

I would vote for a custom variable, at the global level (I agree with
the options you suggest) and at the document level, to economize,
perhaps this would be enough to avoid the code of both babel and
polyglossia:

#+latex_header: NOLANG

(BTW, in the current version of my patch I'm afraid the commit message
is malformed, and there's an asterisk where it shouldn't be, apologies).

Best regards,

Juan Manuel 

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Max Nikulin]

On 21/07/2022 22:39, Juan Manuel Macías wrote:
> Max Nikulin writes:
> 
> I would vote for a custom variable, at the global level (I agree with
> the options you suggest) and at the document level, to economize,
> perhaps this would be enough to avoid the code of both babel and
> polyglossia:
> 
> #+latex_header: NOLANG

Form my point of view it is unnecessary magic. Originally #+latex_header 
is intended for valid LaTeX code and "% \usepackage{babel} % disable" 
does not make code invalid (being a kind of magic though). If you 
consider such cast as too verbose then even
     #+options: latex-l10n:nil
might be better. I am unsure if babel or polyglossia are parsed as 
strings or as symbols in such context. I am still against a DSL for 
"#+options:" to generate complex babel commands in favor of explicit 
"#+latex_header:".

On 18/07/2022 23:21, Juan Manuel Macías wrote:

>> \documentclass[11pt,draft,russian]{article}
>> \usepackage{babel}

I have realized that it resembles
     \documentstyle[russian,epsfig,wrapfig,12pt]{article}
from the previous century and LaTeX-2.09. Due to lack of support in 
babel, several alternatives to add Russian language existed and one of 
them required the \documentstyle option.

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

Max Nikulin writes:

> Form my point of view it is unnecessary magic. Originally #+latex_header 
> is intended for valid LaTeX code

OK, so why not just:

#+latex_header: % NOLANG

?

I think this has in its favor: a) it is simple and obvious to remember;
b) it's a single string, and of course c) it's valid LaTeX code. And it
can be easily controlled from org-latex-guess-babel/polyglossia-language
with a conditional.

Anyway, whatever the choice, I would vote for Org not to load babel or
polyglossia by default, and for the default option of the custom
variable that handles that globally to be nil. For example, I'm in the
group of users that load babel using an external preamble (a .tex or a
.sty file or a 'latex-class'), and frankly I don't want to have to add a
new line to all my documents to prevent Org from reloading babel for me
and return an error on the compilation.

Best regards,

Juan Manuel 

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

Juan Manuel Macías writes:

> OK, so why not just:
>
> #+latex_header: % NOLANG
>
> ?

Forget this. On second thought, I think that the option you proposed
"#+LaTeX_Header: % \usepackage{babel}" is much better.

(I don't know where my head was and I didn't remember there was a
string-match, so your suggestion is the shortest way. However, the line
would have to be (with arguments):

#+LaTeX_Header: % \usepackage[something]{babel}

or

#+LaTeX_Header: % \usepackage[something]{polyglossia}

or modify the regexp in org-latex-guess-babel/polyglossia-language.

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Ihor Radchenko]

Juan Manuel Macías <maciaschain@posteo.net> writes:

> Here is a new version of the patch, with the fixes added.

Thanks! Considering that the followup discussion deviated from the
substance of the patch towards related design issues, I have decided to
marge the patch as is. We can decide on handling AUTO staff later - it
would be a major change to remove and the details of an alternative are
not yet worked out.

Applied onto main via 97cfb959d after adding some double spaces
between sentences and upcasing the beginning of some sentences.

Best,
Ihor

BUG Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Kai von Fintel]

I believe that that this patch has made it impossible to export to latex, because the function =org-latex--format-spec= in =ox-latex.el= still refers to the old variables, which are now not bound:

> Debugger entered--Lisp error: (void-variable org-latex-babel-language-alist)
>  org-latex--format-spec(

etc.

— Kai.

Re: BUG Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Ihor Radchenko]

Kai von Fintel <fintel@mit.edu> writes:

> I believe that that this patch has made it impossible to export to latex, because the function =org-latex--format-spec= in =ox-latex.el= still refers to the old variables, which are now not bound:
>
>> Debugger entered--Lisp error: (void-variable org-latex-babel-language-alist)
>>  org-latex--format-spec(

org-latex-babel-language-alist is the new variable introduced in the
patch. You seem to be loading "mixed" Org versions - built-in + latest.

Best,
Ihor

Re: BUG Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Kai von Fintel]

On 23 Jul 2022, at 16:59, Ihor Radchenko wrote:

> Kai von Fintel <fintel@mit.edu> writes:
>
>> I believe that that this patch has made it impossible to export to latex, because the function =org-latex--format-spec= in =ox-latex.el= still refers to the old variables, which are now not bound:
>>
>>> Debugger entered--Lisp error: (void-variable org-latex-babel-language-alist)
>>>  org-latex--format-spec(
>
> org-latex-babel-language-alist is the new variable introduced in the
> patch. You seem to be loading "mixed" Org versions - built-in + latest.

No, I’m looking just at the HEAD version of =ox-latex.el= at https://git.savannah.gnu.org/cgit/emacs/org-mode.git/tree/lisp/ox-latex.el

The new variable is =org-latex-language-alist=, defined on line 175f.

But, =org-latex—format-spec=, defined on line 1860ff, uses the old variable names:

(defun org-latex--format-spec (info)
  "Create a format-spec for document meta-data.
INFO is a plist used as a communication channel."
  (let ((language (let ((lang (plist-get info :language)))
		    (or (cdr (assoc-string lang org-latex-babel-language-alist t))
			(nth 1 (assoc-string lang org-latex-polyglossia-language-alist t))
			lang))))

This borks my latex exports.

— Kai.

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

Ihor Radchenko writes:

> Thanks! Considering that the followup discussion deviated from the
> substance of the patch towards related design issues, I have decided to
> marge the patch as is. We can decide on handling AUTO staff later - it
> would be a major change to remove and the details of an alternative are
> not yet worked out.
>
> Applied onto main via 97cfb959d after adding some double spaces
> between sentences and upcasing the beginning of some sentences.

Thanks, Ihor. And sorry again for my typos...

If it's okay with you, I can send another patch with the updated
information in NEWS and the Manual. And we can continue the subsequent
discussion related of babel, polyglossia et alii in another thread.

Best regards,

Juan Manuel 

Re: BUG Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Ihor Radchenko]

Kai von Fintel <fintel@mit.edu> writes:

>>>> Debugger entered--Lisp error: (void-variable org-latex-babel-language-alist)
>>>>  org-latex--format-spec(
>>
>> org-latex-babel-language-alist is the new variable introduced in the
>> patch. You seem to be loading "mixed" Org versions - built-in + latest.
>
> No, I’m looking just at the HEAD version of =ox-latex.el= at https://git.savannah.gnu.org/cgit/emacs/org-mode.git/tree/lisp/ox-latex.el
>
> The new variable is =org-latex-language-alist=, defined on line 175f.
>
> But, =org-latex—format-spec=, defined on line 1860ff, uses the old variable names:
>
> (defun org-latex--format-spec (info)
>   "Create a format-spec for document meta-data.
> INFO is a plist used as a communication channel."
>   (let ((language (let ((lang (plist-get info :language)))
> 		    (or (cdr (assoc-string lang org-latex-babel-language-alist t))
> 			(nth 1 (assoc-string lang org-latex-polyglossia-language-alist t))
> 			lang))))
>
> This borks my latex exports.

This is unrelated to the error you are seeing. void-variable error is
thrown for org-latex-babel-language-alist, so I still suggest checking
the place in your config where you are loading Org.

However, you did raise an important omission in the patch - the variable
value was lost in the patch. It was not supposed to happen. I now
restored (in 127e7fee4) the defconst statements and moved them into
org-compat.el. Thus, they can be used by third-party code for the time
being.

Best,
Ihor

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Ihor Radchenko]

Juan Manuel Macías <maciaschain@posteo.net> writes:

> Ihor Radchenko writes:
>
> Thanks, Ihor. And sorry again for my typos...

No problem.

> If it's okay with you, I can send another patch with the updated
> information in NEWS and the Manual. And we can continue the subsequent
> discussion related of babel, polyglossia et alii in another thread.

Sure. Or you can keep the same thread, probably changing the subject to
something people can notice better (M-x message-change-subject).

Best,
Ihor

Re: BUG Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Kai von Fintel]

On 23 Jul 2022, at 17:22, Ihor Radchenko wrote:

> Kai von Fintel <fintel@mit.edu> writes:
>
>>>>> Debugger entered--Lisp error: (void-variable org-latex-babel-language-alist)
>>>>>  org-latex--format-spec(
>>>
>>> org-latex-babel-language-alist is the new variable introduced in the
>>> patch. You seem to be loading "mixed" Org versions - built-in + latest.
>>
>> No, I’m looking just at the HEAD version of =ox-latex.el= at https://git.savannah.gnu.org/cgit/emacs/org-mode.git/tree/lisp/ox-latex.el
>>
>> The new variable is =org-latex-language-alist=, defined on line 175f.
>>
>> But, =org-latex—format-spec=, defined on line 1860ff, uses the old variable names:
>>
>> (defun org-latex--format-spec (info)
>>   "Create a format-spec for document meta-data.
>> INFO is a plist used as a communication channel."
>>   (let ((language (let ((lang (plist-get info :language)))
>> 		    (or (cdr (assoc-string lang org-latex-babel-language-alist t))
>> 			(nth 1 (assoc-string lang org-latex-polyglossia-language-alist t))
>> 			lang))))
>>
>> This borks my latex exports.
>
> This is unrelated to the error you are seeing. void-variable error is
> thrown for org-latex-babel-language-alist, so I still suggest checking
> the place in your config where you are loading Org.
>
> However, you did raise an important omission in the patch - the variable
> value was lost in the patch. It was not supposed to happen. I now
> restored (in 127e7fee4) the defconst statements and moved them into
> org-compat.el. Thus, they can be used by third-party code for the time
> being.

Thanks, Ihor. Your change fixed my issue.

I do think that the code on lines 1864 and 1865 of =ox-latex.el= (https://git.savannah.gnu.org/cgit/emacs/org-mode.git/tree/lisp/ox-latex.el#n1864) should not use the old variable names. Since you’ve now defined the old variables in =org-compat.el=, my exports work, so I’m ok for the moment. But I don’t understand why they are still used in the definition of =org-latex--format-spec=.

But of course I’m far from familiar with the org code base.

— Kai.

Re: BUG Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Ihor Radchenko]

Kai von Fintel <fintel@mit.edu> writes:

>>> (defun org-latex--format-spec (info)
>>>   "Create a format-spec for document meta-data.
>>> INFO is a plist used as a communication channel."
>>>   (let ((language (let ((lang (plist-get info :language)))
>>> 		    (or (cdr (assoc-string lang org-latex-babel-language-alist t))
>>> 			(nth 1 (assoc-string lang org-latex-polyglossia-language-alist t))
>>> 			lang))))
>>>
>>> This borks my latex exports.
> ...
> Thanks, Ihor. Your change fixed my issue.
>
> I do think that the code on lines 1864 and 1865 of =ox-latex.el= (https://git.savannah.gnu.org/cgit/emacs/org-mode.git/tree/lisp/ox-latex.el#n1864) should not use the old variable names. Since you’ve now defined the old variables in =org-compat.el=, my exports work, so I’m ok for the moment. But I don’t understand why they are still used in the definition of =org-latex--format-spec=.
>
> But of course I’m far from familiar with the org code base.

Hmm. You are actually right.
Juan, can you please take a look. It looks like you missed
"org-latex--format-spec" in the patch. It should use the new
org-latex-language-alist variable instead.

Best,
Ihor

Re: BUG Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

Hi Kai,

Kai von Fintel writes:

> I do think that the code on lines 1864 and 1865 of =ox-latex.el=
> (https://git.savannah.gnu.org/cgit/emacs/org-mode.git/tree/lisp/ox-latex.el#n1864)
> should not use the old variable names. Since you’ve now defined the
> old variables in =org-compat.el=, my exports work, so I’m ok for the
> moment. But I don’t understand why they are still used in the
> definition of =org-latex--format-spec=.

I think you're right. My bad, sorry for that: I have not accounted for
`org-latex--format-spec' in my patch. I will send a new patch soon to fix
this.

Best regards,

Juan Manuel 

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Max Nikulin]

On 22/07/2022 21:07, Juan Manuel Macías wrote:
> 
> Forget this. On second thought, I think that the option you proposed
> "#+LaTeX_Header: % \usepackage{babel}" is much better.
> 
> (I don't know where my head was and I didn't remember there was a
> string-match, so your suggestion is the shortest way. However, the line
> would have to be (with arguments):
> 
> #+LaTeX_Header: % \usepackage[something]{babel}
> 
> or
> 
> #+LaTeX_Header: % \usepackage[something]{polyglossia}
> 
> or modify the regexp in org-latex-guess-babel/polyglossia-language.

It was you who suggested

     \documentclass[russian]{article}
     \usepackage{babel}

At least at first glance it works in texlive-2019 (Ubuntu-20.04), so it 
is not a feature from the latest release. Unless commented out,
"#+latex_header: \usepackage{babel}" may be considered as an instruction 
to add value of #+language: to \documentclass option. I do not expect 
that adjusted regexp will cause a problem.

> Anyway, whatever the choice, I would vote for Org not to load babel or
> polyglossia by default, and for the default option of the custom
> variable that handles that globally to be nil. For example, I'm in the
> group of users that load babel using an external preamble (a .tex or a
> .sty file or a 'latex-class'), and frankly I don't want to have to add a
> new line to all my documents to prevent Org from reloading babel for me
> and return an error on the compilation.

Doesn't the purpose of a custom variable is to set it to a suitable 
value in the init file making it unnecessary adding configuration to 
each org file? I did not expected that setting it to e.g. nil will be 
real burden for users like you.

My vote is to configure babel by default if it is possible to provide 
default preamble that allows reasonable quality of PDF for simple Org 
documents with no explicit configuration.

Ideally, the following should be possible out of the box. When no 
advanced features are involved, user should be able to just export 
document to e.g. having printed version of their notes during a meeting, 
to send a summary to the boss. etc. It should be possible for users 
completely ignorant in respect to LaTeX.

Likely language is not enough and e.g. paper size should be guessed 
(LC_PAPER?) as well.

If a document require careful tuning then a couple of extra lines in the 
org file or a couple of additional custom variables in the init file 
should not be a problem.

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Max Nikulin]

On 23/07/2022 12:01, Ihor Radchenko wrote:
> Juan Manuel Macías writes:
> 
>> Here is a new version of the patch, with the fixes added.
> 
> Thanks! Considering that the followup discussion deviated from the
> substance of the patch towards related design issues,

I believed that the subject of discussion is how much values should be 
added to the alist in addition to babel and polyglossia options: 
fontenc, paper size, etc.

I hope, not so much third party packages use the changed variables, so 
committing the patch can hardly be harmful even in the case of 
additional change of the list format in future.

Re: BUG Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

[-- Attachment #1: Type: text/plain, Size: 372 bytes --]

Hi Ihor and Kai,

Ihor Radchenko writes:

> Hmm. You are actually right.
> Juan, can you please take a look. It looks like you missed
> "org-latex--format-spec" in the patch. It should use the new
> org-latex-language-alist variable instead.

Attached a new patch to fix (I hope) the org-latex-language-alist issue
in org-latex--format-spec.

Best regards,

Juan Manuel 


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-lisp-ox-latex.el-Fix-obsolete-variables-for-babel-an.patch --]
[-- Type: text/x-patch, Size: 1554 bytes --]

From 95ce88336f6d2106968250379767ce2fe690fe2c Mon Sep 17 00:00:00 2001
From: Juan Manuel Macias <maciaschain@posteo.net>
Date: Sat, 23 Jul 2022 17:42:50 +0200
Subject: [PATCH] lisp/ox-latex.el: Fix obsolete variables for babel and
 polyglossia

* (org-latex--format-spec): Use the new variable `org-latex-language-alist'
---
 lisp/ox-latex.el | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/lisp/ox-latex.el b/lisp/ox-latex.el
index 6cd751409..ee059d5ce 100644
--- a/lisp/ox-latex.el
+++ b/lisp/ox-latex.el
@@ -1860,10 +1860,12 @@ INFO is a plist used as a communication channel."
 (defun org-latex--format-spec (info)
   "Create a format-spec for document meta-data.
 INFO is a plist used as a communication channel."
-  (let ((language (let ((lang (plist-get info :language)))
-		    (or (cdr (assoc-string lang org-latex-babel-language-alist t))
-			(nth 1 (assoc-string lang org-latex-polyglossia-language-alist t))
-			lang))))
+  (let ((language (let* ((lang (plist-get info :language))
+		         (l (assoc lang org-latex-language-alist)))
+                    (cond ((and (consp l) (= (length l) 4))
+	                   (nth 2 (assoc-string lang org-latex-language-alist t)))
+	                  ((and (consp l) (< (length l) 4))
+	                   (nth 1 (assoc-string lang org-latex-language-alist t)))))))
     `((?a . ,(org-export-data (plist-get info :author) info))
       (?t . ,(org-export-data (plist-get info :title) info))
       (?s . ,(org-export-data (plist-get info :subtitle) info))
-- 
2.37.1

Improvements in the default LaTeX preamble (was: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists)

[Juan Manuel Macías]

Max Nikulin writes:

> My vote is to configure babel by default if it is possible to provide
> default preamble that allows reasonable quality of PDF for simple Org
> documents with no explicit configuration.
>
> Ideally, the following should be possible out of the box. When no
> advanced features are involved, user should be able to just export
> document to e.g. having printed version of their notes during a
> meeting, to send a summary to the boss. etc. It should be possible for
> users completely ignorant in respect to LaTeX.
>
> Likely language is not enough and e.g. paper size should be guessed
> (LC_PAPER?) as well.
>
> If a document require careful tuning then a couple of extra lines in
> the org file or a couple of additional custom variables in the init
> file should not be a problem.

I think that my position, after all these discussions here and in other
threads, needs a couple of clarifications:

- I am in favor of Org producing a consistent "standard" preamble for
  LaTeX out-of-the-box, so that users get a technically optimal PDF
  without messing with LaTeX code. This would include not only issues
  related to document languages (babel and polyglossia) but also fonts
  (especially XelaTeX and LuaLaTeX support), page layout (with geometry
  package), publishing support and some other things that can be
  proposed here. In this regard, I have changed the subject of this
  thread, if that's okay with you.

- I can agree that all of the above is by default. But it seems
  essential to me that there is at least the possibility of giving all
  that a nil value at a global level, for users who need more control
  and want to write (La)TeX code or want to load the entire preamble
  from an external document (a .tex file or a .sty file) . Which is not
  incompatible with document-level control and fine-tuning (*only* if
  the above is enabled).

Best regards,

Juan Manuel 

Re: BUG Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Ihor Radchenko]

Juan Manuel Macías <maciaschain@posteo.net> writes:

>> Hmm. You are actually right.
>> Juan, can you please take a look. It looks like you missed
>> "org-latex--format-spec" in the patch. It should use the new
>> org-latex-language-alist variable instead.
>
> Attached a new patch to fix (I hope) the org-latex-language-alist issue
> in org-latex--format-spec.

Thanks!

> +  (let ((language (let* ((lang (plist-get info :language))
> +		         (l (assoc lang org-latex-language-alist)))
> +                    (cond ((and (consp l) (= (length l) 4))
> +	                   (nth 2 (assoc-string lang org-latex-language-alist t)))
> +	                  ((and (consp l) (< (length l) 4))
> +	                   (nth 1 (assoc-string lang org-latex-language-alist t)))))))

Can you please add the comments, similar to what I requested earlier.
These magic length of 4 may be hard to grasp otherwise.

Best,
Ihor

Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Ihor Radchenko]

Max Nikulin <manikulin@gmail.com> writes:

> On 23/07/2022 12:01, Ihor Radchenko wrote:
>> Juan Manuel Macías writes:
>> 
>>> Here is a new version of the patch, with the fixes added.
>> 
>> Thanks! Considering that the followup discussion deviated from the
>> substance of the patch towards related design issues,
>
> I believed that the subject of discussion is how much values should be 
> added to the alist in addition to babel and polyglossia options: 
> fontenc, paper size, etc.

Adding things like paper size is a much more debatable topic.
Considering the number of expected developments in this area, including
the earlier discussion on XeLaTeX/LuaTeX and preamble generation by TEC,
adding the new specific defaults will need to be a subject of extensive
discussion and testing. I do not see much point delaying this patch,
which provides an immediate improvement in the codebase, until we
complete all that.

> I hope, not so much third party packages use the changed variables, so 
> committing the patch can hardly be harmful even in the case of 
> additional change of the list format in future.

This alist is rather internal variable representing current defaults. I
am not even sure how it can be useful for third-party packages. It
should not be modified by packages either (defconst).

Best,
Ihor

Re: BUG Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

[-- Attachment #1: Type: text/plain, Size: 549 bytes --]

Ihor Radchenko writes:

> Can you please add the comments, similar to what I requested earlier.
> These magic length of 4 may be hard to grasp otherwise.

Hi Ihor,

Here is the new patch. I have realized that it is not necessary to put a
cond, since in this case it is only necessary to obtain the name of the
language for the metadata, so this new patch is simpler.

Anyway, I think I'm going to prioritize working on a new
org-latex-language-alist that is a plist, to avoid all this stuff about
numbers and lengths...

Best regards,

Juan Manuel


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-lisp-ox-latex.el-Fix-obsolete-variables-for-babel-an.patch --]
[-- Type: text/x-patch, Size: 1497 bytes --]

From 7f3ddd0d2e6e06becd0d43575be88b77b8d699a4 Mon Sep 17 00:00:00 2001
From: Juan Manuel Macias <maciaschain@posteo.net>
Date: Sun, 24 Jul 2022 13:20:25 +0200
Subject: [PATCH] lisp/ox-latex.el: Fix obsolete variables for babel and
 polyglossia

* (org-latex--format-spec): Use the new variable `org-latex-language-alist'.
---
 lisp/ox-latex.el | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/lisp/ox-latex.el b/lisp/ox-latex.el
index 6cd751409..aea602982 100644
--- a/lisp/ox-latex.el
+++ b/lisp/ox-latex.el
@@ -1860,10 +1860,11 @@ INFO is a plist used as a communication channel."
 (defun org-latex--format-spec (info)
   "Create a format-spec for document meta-data.
 INFO is a plist used as a communication channel."
-  (let ((language (let ((lang (plist-get info :language)))
-		    (or (cdr (assoc-string lang org-latex-babel-language-alist t))
-			(nth 1 (assoc-string lang org-latex-polyglossia-language-alist t))
-			lang))))
+  (let ((language (let ((lang (plist-get info :language))
+                        ;; Here it would suffice to obtain the second
+                        ;; element, which always returns the name
+                        ;; language name in `org-latex-language-alist'
+			(nth 1 (assoc-string lang org-latex-language-alist t))))))
     `((?a . ,(org-export-data (plist-get info :author) info))
       (?t . ,(org-export-data (plist-get info :title) info))
       (?s . ,(org-export-data (plist-get info :subtitle) info))
-- 
2.37.1

Improvements in the default LaTeX preamble: templates? (was: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists)

[Juan Manuel Macías]

Ihor Radchenko writes:

> Adding things like paper size is a much more debatable topic.
> Considering the number of expected developments in this area, including
> the earlier discussion on XeLaTeX/LuaTeX and preamble generation by TEC,
> adding the new specific defaults will need to be a subject of extensive
> discussion and testing. I do not see much point delaying this patch,
> which provides an immediate improvement in the codebase, until we
> complete all that.

I agree that adding more elements to the standard preamble is a complex
matter. LaTeX is already horribly complex and multiple, and it is
difficult to satisfy all kinds of users with a standard code. It
occurred to me that an alternative to modifying Org's code in this
regard could be to have some kind of "LaTeX template library". I think
Pandoc has something similar too, if I remember correctly. Those
templates could be on Org or provided by third parties somewhere else,
like Worg. In Org, we also have a great system for creating LaTeX
documents templates, which is the org-latex-classes list. A large number
of elements could be defined in a 'single' class for any type of
document.

Some time ago I shared here a function I wrote (it's a bit raw and
little tested) to be able to convert a LaTeX document (the preamble)
into a lisp expression to be added to org-latex-classes:

https://list.orgmode.org/87czgly8x8.fsf@posteo.net/

On the one hand Org would offer a more or less basic preamble out of the
box and on the other hand there could be a "org-latex-classes library"
(extensible by users) that would support a multitude of LaTeX document
types or users. This would include preamble types with a bunch of
components: page layout, fonts, lua code, custom LaTeX commands, custom
lua functions, article types, book types, etc. It would suffice to add:

#+latex_class: foo

and the magic is done.

Best regards,

Juan Manuel 

Re: Improvements in the default LaTeX preamble: templates? (was: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists)

[Ihor Radchenko]

Juan Manuel Macías <maciaschain@posteo.net> writes:

> I agree that adding more elements to the standard preamble is a complex
> matter. LaTeX is already horribly complex and multiple, and it is
> difficult to satisfy all kinds of users with a standard code. It
> occurred to me that an alternative to modifying Org's code in this
> regard could be to have some kind of "LaTeX template library". I think
> Pandoc has something similar too, if I remember correctly. Those
> templates could be on Org or provided by third parties somewhere else,
> like Worg. In Org, we also have a great system for creating LaTeX
> documents templates, which is the org-latex-classes list. A large number
> of elements could be defined in a 'single' class for any type of
> document.

LaTeX is just one export backend to worry about. From broader
perspective, we can have a generic template library.

ox.el currently allows export backends to define document template as a
function, which is the most generic way. However, we can come up with
something more customizable - customizable in a consistent way, in
contrast to the current disarray with various export backends
approaching the boilerplate code differently.

TEC is working on something along these lines. See https://tecosaur.github.io/emacs-config/config.html#cleverer-preamble

Note that we also have inner templates that apply to individual exported
elements.

Best,
Ihor

Re: Improvements in the default LaTeX preamble: templates?

[Juan Manuel Macías]

Ihor Radchenko writes:

> LaTeX is just one export backend to worry about. From broader
> perspective, we can have a generic template library.

Nice idea, I agree. I was targeting LaTeX specifically because of the
questions that have been raised in this thread and other parallel threads.
And because LaTeX is an extremely complex beast. And it has become much
more complex over the years[1]. No two LaTeX documents are alike just as no
two LaTeX users are alike. Just take a look at tex.stackexchange.com to
realize that reality...

(ConTeXt can be a good alternative for those who don't want to mess with
the complexity of LaTeX. In ConTeXt you don't need to load a package for
everything ---modules at most, but that's another story---, so almost
everything is out-of-the-box there).

[1] And there is also the problem of multiplicity: three TeX engines
coexisting at the same time, LaTeX2ε coexisting with LaTeX 3, etc.

> TEC is working on something along these lines. See
> https://tecosaur.github.io/emacs-config/config.html#cleverer-preamble

Thanks for the pointer! I did not know it and it seems to me a
tremendously interesting work. I'll keep an eye on it.

In my workflow, I am used to writing the configuration of a LaTeX
document (aka, "the preamble") through .sty files that I build according
to the requirements of each project. That is, I write my own packages.
That's probably why I have a bias of opinion here (I use LaTeX for
typesetting and editorial production, so I need more control), but I
tend to think that a LaTeX preamble is something so ductile that
achieving a certain degree of automation is an arduous task. At least
one automation that covers all possible use cases. That's where I got
the idea of being able to have a library of templates to cover different
types of documents, even the most idiosyncratic ones. Or, at least, that
they can serve as an inspiration to other users.

Best regards,

Juan Manuel 

Re: BUG Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Ihor Radchenko]

Juan Manuel Macías <maciaschain@posteo.net> writes:

> Here is the new patch. I have realized that it is not necessary to put a
> cond, since in this case it is only necessary to obtain the name of the
> language for the metadata, so this new patch is simpler.

Thanks for the update!

The patch has some misplaced parenthesis.

> -  (let ((language (let ((lang (plist-get info :language)))
> -		    (or (cdr (assoc-string lang org-latex-babel-language-alist t))
> -			(nth 1 (assoc-string lang org-latex-polyglossia-language-alist t))
> -			lang))))
> +  (let ((language (let ((lang (plist-get info :language))
> +                        ;; Here it would suffice to obtain the second
> +                        ;; element, which always returns the name
> +                        ;; language name in `org-latex-language-alist'
> +			(nth 1 (assoc-string lang org-latex-language-alist t))))))

Your (nth 1 ...) sexp is inside the let definition:
(let ((lang ...)
      (nth 1 ..))
  nil)

Please pay attention to the compiler warnings.

Also, the original code contained the clause:
(or (get lang from the alist1)
    (get lang from the alist2)
    lang ; Fallback to provided language if not known.
    )

Your variant does not have the fallback part. Is it intentional?

Best,
Ihor

Re: BUG Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Juan Manuel Macías]

[-- Attachment #1: Type: text/plain, Size: 782 bytes --]

Hi Ihor,

A thousand apologies for my horrible carelessness with the parentheses.
I should have checked the code. Here goes the patch again corrected. I
hope it's alright now.

Ihor Radchenko writes:

> Also, the original code contained the clause:
> (or (get lang from the alist1)
>     (get lang from the alist2)
>     lang ; Fallback to provided language if not known.
>     )
>
> Your variant does not have the fallback part. Is it intentional?

Yes, I removed it because I thought it was not necessary, because after
all the user must put a supported language as the value of #+language.
Anyway, in case it breaks something backwards I have replaced it. Now
the or expression is:

(or (nth 1 (assoc-string lang org-latex-language-alist t))
	lang)

Best regards,

Juan Manuel 


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-lisp-ox-latex.el-Remove-Babel-and-Polyglossia-alists.patch --]
[-- Type: text/x-patch, Size: 1324 bytes --]

From 483cf69e0ca56c560c3bd53db13887a63d529ec9 Mon Sep 17 00:00:00 2001
From: Juan Manuel Macias <maciaschain@posteo.net>
Date: Tue, 26 Jul 2022 18:01:52 +0200
Subject: [PATCH] lisp/ox-latex.el: Remove Babel and Polyglossia alists

* (org-latex--format-spec): The new variable is now `org-latex-language-alist'
---
 lisp/ox-latex.el | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/lisp/ox-latex.el b/lisp/ox-latex.el
index 121a3f84c..14728f0ba 100644
--- a/lisp/ox-latex.el
+++ b/lisp/ox-latex.el
@@ -1861,8 +1861,10 @@ INFO is a plist used as a communication channel."
   "Create a format-spec for document meta-data.
 INFO is a plist used as a communication channel."
   (let ((language (let ((lang (plist-get info :language)))
-		    (or (cdr (assoc-string lang org-latex-babel-language-alist t))
-			(nth 1 (assoc-string lang org-latex-polyglossia-language-alist t))
+                    ;; Here it would suffice to obtain the second
+                    ;; element, which always returns the name
+                    ;; language name in `org-latex-language-alist'
+                    (or (nth 1 (assoc-string lang org-latex-language-alist t))
 			lang))))
     `((?a . ,(org-export-data (plist-get info :author) info))
       (?t . ,(org-export-data (plist-get info :title) info))
-- 
2.37.1

Re: BUG Re: [PATCH] ox-latex.el: Unify in one single list Babel and Polyglossia languages alists

[Ihor Radchenko]

Juan Manuel Macías <maciaschain@posteo.net> writes:

> A thousand apologies for my horrible carelessness with the parentheses.
> I should have checked the code. Here goes the patch again corrected. I
> hope it's alright now.

Thanks!
Applied onto main via d37c0ee5f after adding full stop after sentences
in the commit message and amending some wording.
https://git.savannah.gnu.org/cgit/emacs/org-mode.git/commit/?id=d37c0ee5fa7dc4be4bbe3aa9b6f4e79d4b1e638d

Best,
Ihor