Sebastian > But I may have just lost my memory of how this works.... > > - Carsten > > On Apr 19, 2010, at 1:54 AM, Sebastian Rose wrote: > >> Haaarrrgh --- >> >> >> the first patch in my previous mail does not work for all cases. Hairy >> stuff.... >> >> >> Here's what did _not_ work with my previous patch: >> >> #+html:
>> *Unsorted Patterns* >> - a >> - b >> - c >> #+HTML:
Somthing inside the last item!!! >> #+HTML:
>> >> >> >> >> >> But this one finally works: >> >> >> diff --git a/lisp/org-html.el b/lisp/org-html.el >> index fcddd50..0174e43 100644 >> --- a/lisp/org-html.el >> +++ b/lisp/org-html.el >> @@ -929,6 +929,17 @@ lang=\"%s\" xml:lang=\"%s\"> >> >> ;; Protected HTML >> (when (get-text-property 0 'org-protected line) >> + (when in-local-list >> + (let ((ind (org-get-indentation line))) >> + (while (and (car local-list-indent) (< ind (car local- >> list-indent))) >> + (org-close-li (car local-list-type)) >> + (insert (format "\n" (car local-list-type))) >> + (setq local-list-indent (cdr local-list-indent)) >> + (setq local-list-type (cdr local-list-type)) >> + (setq in-local-list local-list-indent)) >> + (insert line "\n") >> + (throw 'nextline nil))) >> + >> (let (par (ind (get-text-property 0 'original-indentation line))) >> (when (re-search-backward >> "\\(

\\)\\([ \t\r\n]*\\)\\=" (- (point) 100) t) >> >> >> Again: If it makes things easier, I could apply the change to the master >> branch and send an appropriate patch. >> >> >> >> >> >> Sebastian >> >> >> >> >> >> >> Sebastian Rose writes: >>> Hello Carsten, >>> >>> >>> there was much discussion about a terminator and I ran into a problem, >>> that made me think we need one. But then I found we had one --- it's >>> just not used on HTML export. >>> >>> >>> Below is a little file I wrote. Thanks to the `- __' items, it results >>> in the XHTML closely to what I wanted it to. >>> But only as long as I use those _undocumented_ `- __' items. Once you >>> remove them, you'll see, that the `#+html: >> inside the last list item and the XHTML will not validate. >>> >>> >>> As I looked at it, I found the most natural solution would be, to >>> terminate the list by regarding the indentation of `#+WHATEVER' and >>> `#+BEGIN_WHATEVER' if inside lists [fn:1]. >>> >>> >>> >>> The patch below (diffed against `remove-compatibility-code') makes >>> XHTML-export honor the indentation of `#+SPECIALS'. >>> >>> >>> >>> Here's the Org-file I wrote (remove and add the `- __' list items to see >>> the effect): >>> >>> >>> #+OPTIONS: toc:nil >>> #+STYLE: >>> >>> * List of design patterns >>> >>> #+HTML:

>>> *Behavioural Patterns* >>> - [[file:BatchCommand][BatchCommand]] >>> - [[file:ChainOfResponsibility.org][Chain Of Responsibility]] >>> - [[file:Command.org][Command]], UndoableCommand and BatchCommand >>> - [[file:Interpreter.org][Interpreter]] >>> - [[file:Iterator.org][Iterator]] >>> - [[file:Mediator.org][Mediator]] >>> - [[file:Memento.org][Memento]] >>> - [[file:NullObject][NullObject]] >>> - [[file:Observer.org][Observer]] >>> - [[file:State.org][State]] >>> - [[file:Strategy.org][Strategy]] >>> - [[file:TemplateMethod.org][Template Method]] >>> - [[file:Visitor.org][Visitor]] >>> *Creational Patterns* >>> - [[file:AbstractFactory.org][Abstract Factory]] >>> - [[file:Builder.org][Builder]] >>> - [[file:Factory.org][Factory]] >>> - [[file:FactoryMethod.org][Factory Method]] >>> - [[file:Prototype.org][Prototype]] >>> - [[file:Singleton.org][Singleton]] >>> - __ >>> #+html:
>>> #+html:
>>> *Structural Patterns* >>> - [[file:Adapter.org][Adapter]] >>> - [[file:Composite.org][Composite]] >>> - [[file::Bridge.org][Bridge]] >>> - [[file:Decorator.org][Decorator]] >>> - [[file:Facade.org][Facade]] >>> - [[file:Flyweight.org][Flyweight]] >>> - [[file:Proxy.org][Proxy]] >>> *Unsorted* >>> - [[file:BusinessDelegate.org][Business Delegate]] >>> - [[file:DataAccessObject.org][Data Access Object]] >>> - [[file:DataTransferObject.org][Data Transfer Object]] >>> - [[file:DependencyInjection.org][Dependency Injection]] >>> - [[file:FluentInterface.org][Fluent Interface]] >>> - [[file:InversionOfControl.org][Inversion Of Control]] >>> - [[file:ModelViewControler.org][Model View Controler]] >>> - [[file:ModelViewPresenter.org][Model View Presenter]] >>> - [[file:Plugin.org][Plugin]] >>> - __ >>> #+HTML:
>>> >>> >>> >>> And, finally, the patch. I would have used the function >>> `org-export-html-close-lists-maybe' but that didn't work, so I wrote >>> similar code just in place. >>> >>> Carsten: If it makes things easier for you, I could apply the change to >>> the master branch and send an appropriate patch. >>> >>> >>> diff --git a/lisp/org-html.el b/lisp/org-html.el >>> index fcddd50..812e63c 100644 >>> --- a/lisp/org-html.el >>> +++ b/lisp/org-html.el >>> @@ -929,6 +929,15 @@ lang=\"%s\" xml:lang=\"%s\"> >>> >>> ;; Protected HTML >>> (when (get-text-property 0 'org-protected line) >>> + (when in-local-list >>> + (let ((ind (or (get-text-property 0 'original- >>> indentation line) 0))) >>> + (while (and (car local-list-indent) (< ind (car local- >>> list-indent))) >>> + (org-close-li (car local-list-type)) >>> + (insert (format "\n" (car local-list-type))) >>> + (setq local-list-indent (cdr local-list-indent)) >>> + (setq local-list-type (cdr local-list-type)) >>> + (setq in-local-list local-list-indent)))) >>> + >>> (let (par (ind (get-text-property 0 'original-indentation line))) >>> (when (re-search-backward >>> "\\(

\\)\\([ \t\r\n]*\\)\\=" (- (point) 100) t) >>> >>> >>> >>> >>> >>> Best wishes >>> >>> Sebastian >>> >>> >>> >>> >>> Footnotes: >>> >>> [fn:1] `org-end-of-item' and `org-end-of-item-list' already consider the >>> decreased indentation of the `#+html:' line the end of the list. >>> >>> You can proof that by deleting the last dot and all empty lines, >>> so that the `#+html:' line is directly below the last list >>> item. Then move point somewhere on the item and do `M-x >>> org-end-of-item RET'. >>> >>> _______________________________________________ >>> Emacs-orgmode mailing list >>> Please use `Reply All' to send replies to the list. >>> Emacs-orgmode@gnu.org >>> http://lists.gnu.org/mailman/listinfo/emacs-orgmode >> _______________________________________________ >> Emacs-orgmode mailing list >> Please use `Reply All' to send replies to the list. >> Emacs-orgmode@gnu.org >> http://lists.gnu.org/mailman/listinfo/emacs-orgmode > > - Carsten > > > -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Sebastian Rose Fachinformatiker / Anwendungsentwicklung Viktoriastr. 22 Entwicklung von Anwendungen mit freien Werkzeugen 30451 Hannover und Bibliotheken. 0173 83 93 417 sebastian_rose@gmx.de s.rose@emma-stil.de ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~