Sean O'Halpin writes:
> Hi,
>
> =:dir= specifies the directory for code execution, i.e. when you hit =C-c
> C-c= what is the environment in which the code is run?
I know currently it specify directory for code execution. I'm saying whether
consider extend =:dir= semantic for tangling too.
Because the header arguments could becoming very long if I repeat the
sub-directory information in executing directory =:dir= and for =:tangle=. Like
this:
#+begin_src sh :mkdirp yes :dir "data/code/project-name-1" :tangle
"data/code/project-name-1/src/project-name/main/App.java"
#+end_src
You see, The "=data/code/project-name-1=" path part is duplicate. And the header
arguments is very long. Even though I can put header arguments in ~#+HEADERS:~
line of source block. But still duplicate info inputting, isn't it?
>
> =:mkdirp= forces creation of the directory for tangle output, i.e. when you
> =org-babel-tangle= where do you put the output source code?
=:mkdirp= already support for auto force create of directory =:dir= now.
>
> They are quite separate concerns.
>
> Regards,
> Sean
>
> On Tue, 5 Mar 2019 at 06:03, stardiviner wrote:
>
> I realized one thing, the combination of ~:mkdir~, ~:dir~, and ~:tangle~
> does not work
> good enough. Can make ~:tangle~ respect ~:dir~ too. WDYT?
>
> Here is an example:
>
> #+begin_src clojure :mkdirp yes :dir "data/code/clj-crawler-demo" :tangle
> "src/clj_crawler_demo/core.clj"
> (ns clj-crawler-demo.core
>(:require [clj-http.client :as http])
>(:require [net.cgrand.enlive-html :as html]))
>
> (html/select
> (-> (http/get "https://www.baidu.com";)
> :body
> html/html-snippet)
> [:div])
> #+end_src
>
> Why need this? If ~:tangle~ respect ~:dir~, it will don't need the complete
> path to
> tangled file. And when the source block has other header argument need dir
> like
> ~:file~, this will make ~:tangle~ and other related header arguments shorter.
>
> If ~:tangle~ does not respect ~:dir~, user need to repeat the path:
>
> #+begin_src clojure :mkdirp yes :dir "data/code/clj-crawler-demo" :tangle
> "data/code/clj-crawler-demo/src/clj_crawler_demo/core.clj"
> (ns clj-crawler-demo.core
>(:require [clj-http.client :as http])
>(:require [net.cgrand.enlive-html :as html]))
>
> (html/select
> (-> (http/get "https://www.baidu.com";)
> :body
> html/html-snippet)
> [:div])
> #+end_src
>
> This is long, I know there is ~#+headers:~ for source block. But reduce info
> would
> be better. I also consider the disadvantage of this change. Some user's
> existing
> source block might be broken.
>
> Let's level this change for discussion.
>
> --
> [ stardiviner ]
> I try to make every word tell the meaning what I want to express.
>
> Blog: https://stardiviner.github.io/
> IRC(freenode): stardiviner, Matrix: stardiviner
> GPG: F09F650D7D674819892591401B5DF1C95AE89AC3
--
[ stardiviner ]
I try to make every word tell the meaning what I want to express.
Blog: https://stardiviner.github.io/
IRC(freenode): stardiviner, Matrix: stardiviner
GPG: F09F650D7D674819892591401B5DF1C95AE89AC3
