org-mode/contrib/babel/library-of-babel.org

371 lines
13 KiB
Org Mode
Raw Normal View History

2009-08-05 14:02:25 +00:00
#+title: The Library of Babel
#+author: Org-mode People
2009-08-05 14:02:25 +00:00
#+STARTUP: odd hideblocks
2009-08-05 14:02:25 +00:00
* Introduction
The Library of Babel is an extensible collection of ready-made and
easily-shortcut-callable source-code blocks for handling common
tasks. Org-babel comes pre-populated with the source-code blocks
located in this file. It is possible to add source-code blocks from
any org-mode file to the library by calling =(org-babel-lob-ingest
"path/to/file.org")=.
This file is included in worg mainly less for viewing through the
web interface, and more for contribution through the worg git
repository. If you have code snippets that you think others may
find useful please add them to this file and [[file:~/src/worg/worg-git.org::contribute-to-worg][contribute them]] to
worg.
The raw Org-mode text of this file can be downloaded at
[[repofile:contrib/babel/library-of-babel.org][library-of-babel.org]]
2009-08-05 14:02:25 +00:00
* Simple
A collection of simple utility functions
#+srcname: echo
#+begin_src emacs-lisp :var input="echo'd"
input
#+end_src
* File I/O
** reading and writing files
Read the contents of the file at =file=. The =:results vector= and
=:results scalar= header arguments can be used to read the contents of
file as either a table or a string.
#+srcname: read
#+begin_src emacs-lisp :var file="" :var format=""
(if (string= format "csv")
(with-temp-buffer
(org-table-import (expand-file-name file) nil)
(org-table-to-lisp))
(with-temp-buffer
(insert-file-contents (expand-file-name file))
(buffer-string)))
#+end_src
Write =data= to a file at =file=. If =data= is a list, then write it
as a table in traditional Org-mode table syntax.
#+srcname: write
#+begin_src emacs-lisp :var data="" :var file="" :var ext='()
(flet ((echo (r) (if (stringp r) r (format "%S" r))))
(with-temp-file file
(case (and (listp data)
(or ext (intern (file-name-extension file))))
('tsv (insert (orgtbl-to-tsv data '(:fmt echo))))
('csv (insert (orgtbl-to-csv data '(:fmt echo))))
(t (org-babel-insert-result data)))))
nil
#+end_src
** remote files
**** json
Read local or remote file in [[http://www.json.org/][json]] format into emacs-lisp objects.
#+srcname: json
#+begin_src emacs-lisp :var file='() :var url='()
(require 'json)
(cond
(file
(with-temp-filebuffer file
(goto-char (point-min))
(json-read)))
(url
(require 'w3m)
(with-temp-buffer
(w3m-retrieve url)
(goto-char (point-min))
(json-read))))
#+end_src
**** Google docs
The following code blocks make use of the [[http://code.google.com/p/googlecl/][googlecl]] Google command line
tool. This tool provides functionality for accessing Google services
from the command line, and the following code blocks use /googlecl/
for reading from and writing to Google docs with Org-mode code blocks.
****** read a document from Google docs
The =google= command seems to be throwing "Moved Temporarily" errors
when trying to download textual documents, but this is working fine
for spreadsheets.
#+source: gdoc-read
#+begin_src emacs-lisp :var title="example" :var format="csv"
(let* ((file (concat title "." format))
(cmd (format "google docs get --format %S --title %S" format title)))
(message cmd) (message (shell-command-to-string cmd))
(prog1 (if (string= format "csv")
(with-temp-buffer
(org-table-import (shell-quote-argument file) '(4))
(org-table-to-lisp))
(with-temp-buffer
(insert-file-contents (shell-quote-argument file))
(buffer-string)))
(delete-file file)))
#+end_src
For example, a line like the following can be used to read the
contents of a spreadsheet named =num-cells= into a table.
: #+call: gdoc-read(title="num-cells"")
A line like the following can be used to read the contents of a
document as a string.
: #+call: gdoc-read(title="loremi", :format "txt")
****** write a document to a Google docs
Write =data= to a google document named =title=. If =data= is tabular
it will be saved to a spreadsheet, otherwise it will be saved as a
normal document.
#+source: gdoc-write
#+begin_src emacs-lisp :var title="babel-upload" :var data=fibs(n=10) :results silent
(let* ((format (if (listp data) "csv" "txt"))
(tmp-file (make-temp-file "org-babel-google-doc" nil (concat "." format)))
(cmd (format "google docs upload --title %S %S" title tmp-file)))
(with-temp-file tmp-file
(insert
(if (listp data)
(orgtbl-to-csv
data '(:fmt (lambda (el) (if (stringp el) el (format "%S" el)))))
(if (stringp data) data (format "%S" data)))))
(message cmd)
(prog1 (shell-command-to-string cmd) (delete-file tmp-file)))
#+end_src
example usage
: #+source: fibs
: #+begin_src emacs-lisp :var n=8
: (flet ((fib (m) (if (< m 2) 1 (+ (fib (- m 1)) (fib (- m 2))))))
: (mapcar (lambda (el) (list el (fib el))) (number-sequence 0 (- n 1))))
: #+end_src
:
: #+call: gdoc-write(title="fibs", data=fibs(n=10))
* Plotting code
2009-08-05 14:02:25 +00:00
** R
Plot column 2 (y axis) against column 1 (x axis). Columns 3 and beyond, if present, are ignored.
Enabling LoB to put results in buffer, and slowly moving towards more unified concept of function calls. Previously LoB calls were not able to produce results in the buffer. These changes go some way to allowing them to do that. [There are still some bugs to deal with]. That meant changing org-babel.el so that there is a notion of the `source block name' for a LoB line, in order to construct a #+resname (currently I've made the name the same as the function call). I'm also slowly moving towards unifying the notion of `function calls' a bit more: I've changed the org-babel-lob-one-liner-regexp so that instead of a monolithic match it now matches first the function name, and second the function arguments in parentheses. org-babel-lob-get-info makes that match, and although it still concatenates them and returns the string, the two elements can be accessed immediately afterwards using match-string. So that situation is very similar to org-babel-get-src-block-name, whose job (in this branch) is also to parse the function *name* and the function *arguments*. In a few places in the code (esp. function names), I think the word `info' should be replaced with `call' or `function call', which I believe more accurately indicates what the `info' is: a function definition, together with bound arguments/references. The function call syntax, i.e. function-name(arg1=ref1), originally introduced for references (and thereby in LoB), and which I'm proposing we use throughout, raises the question of default arguments, and those being over-ridden by supplied arguments, as in e.g. python, and R.
2009-07-11 02:59:10 +00:00
#+srcname: R-plot(data=R-plot-example-data)
#+begin_src R
plot(data)
#+end_src
Enabling LoB to put results in buffer, and slowly moving towards more unified concept of function calls. Previously LoB calls were not able to produce results in the buffer. These changes go some way to allowing them to do that. [There are still some bugs to deal with]. That meant changing org-babel.el so that there is a notion of the `source block name' for a LoB line, in order to construct a #+resname (currently I've made the name the same as the function call). I'm also slowly moving towards unifying the notion of `function calls' a bit more: I've changed the org-babel-lob-one-liner-regexp so that instead of a monolithic match it now matches first the function name, and second the function arguments in parentheses. org-babel-lob-get-info makes that match, and although it still concatenates them and returns the string, the two elements can be accessed immediately afterwards using match-string. So that situation is very similar to org-babel-get-src-block-name, whose job (in this branch) is also to parse the function *name* and the function *arguments*. In a few places in the code (esp. function names), I think the word `info' should be replaced with `call' or `function call', which I believe more accurately indicates what the `info' is: a function definition, together with bound arguments/references. The function call syntax, i.e. function-name(arg1=ref1), originally introduced for references (and thereby in LoB), and which I'm proposing we use throughout, raises the question of default arguments, and those being over-ridden by supplied arguments, as in e.g. python, and R.
2009-07-11 02:59:10 +00:00
#+tblname: R-plot-example-data
| 1 | 2 |
| 2 | 4 |
| 3 | 9 |
| 4 | 16 |
| 5 | 25 |
Enabling LoB to put results in buffer, and slowly moving towards more unified concept of function calls. Previously LoB calls were not able to produce results in the buffer. These changes go some way to allowing them to do that. [There are still some bugs to deal with]. That meant changing org-babel.el so that there is a notion of the `source block name' for a LoB line, in order to construct a #+resname (currently I've made the name the same as the function call). I'm also slowly moving towards unifying the notion of `function calls' a bit more: I've changed the org-babel-lob-one-liner-regexp so that instead of a monolithic match it now matches first the function name, and second the function arguments in parentheses. org-babel-lob-get-info makes that match, and although it still concatenates them and returns the string, the two elements can be accessed immediately afterwards using match-string. So that situation is very similar to org-babel-get-src-block-name, whose job (in this branch) is also to parse the function *name* and the function *arguments*. In a few places in the code (esp. function names), I think the word `info' should be replaced with `call' or `function call', which I believe more accurately indicates what the `info' is: a function definition, together with bound arguments/references. The function call syntax, i.e. function-name(arg1=ref1), originally introduced for references (and thereby in LoB), and which I'm proposing we use throughout, raises the question of default arguments, and those being over-ridden by supplied arguments, as in e.g. python, and R.
2009-07-11 02:59:10 +00:00
#+lob: R-plot(data=R-plot-example-data)
#+resname: R-plot(data=R-plot-example-data)
: nil
Enabling LoB to put results in buffer, and slowly moving towards more unified concept of function calls. Previously LoB calls were not able to produce results in the buffer. These changes go some way to allowing them to do that. [There are still some bugs to deal with]. That meant changing org-babel.el so that there is a notion of the `source block name' for a LoB line, in order to construct a #+resname (currently I've made the name the same as the function call). I'm also slowly moving towards unifying the notion of `function calls' a bit more: I've changed the org-babel-lob-one-liner-regexp so that instead of a monolithic match it now matches first the function name, and second the function arguments in parentheses. org-babel-lob-get-info makes that match, and although it still concatenates them and returns the string, the two elements can be accessed immediately afterwards using match-string. So that situation is very similar to org-babel-get-src-block-name, whose job (in this branch) is also to parse the function *name* and the function *arguments*. In a few places in the code (esp. function names), I think the word `info' should be replaced with `call' or `function call', which I believe more accurately indicates what the `info' is: a function definition, together with bound arguments/references. The function call syntax, i.e. function-name(arg1=ref1), originally introduced for references (and thereby in LoB), and which I'm proposing we use throughout, raises the question of default arguments, and those being over-ridden by supplied arguments, as in e.g. python, and R.
2009-07-11 02:59:10 +00:00
2009-08-05 14:02:25 +00:00
** Gnuplot
* Org reference
** headline references
#+source: headline
#+begin_src emacs-lisp :var headline=top :var file='()
(save-excursion
(when file (get-file-buffer file))
(org-open-link-from-string (org-make-link-string headline))
(save-restriction
(org-narrow-to-subtree)
(buffer-string)))
#+end_src
#+call: headline(headline="headline references")
* Tables
** LaTeX Table export
*** booktabs
This block can be used to wrap a table in the latex =booktabs=
environment, it takes the following arguments -- all but the first two
are optional.
| arg | description |
|-------+--------------------------------------------|
| table | a reference to the table |
| align | optional alignment string |
| env | optional environment, default to "tabular" |
| width | optional width specification string |
#+srcname: booktabs
#+begin_src emacs-lisp :var table='((:head) hline (:body)) :var align='() :var env="tabular" :var width='() :noweb yes :results latex
(flet ((to-tab (tab)
(orgtbl-to-generic
(mapcar (lambda (lis)
(if (listp lis)
(mapcar (lambda (el)
(if (stringp el)
el
(format "%S" el))) lis)
lis)) tab)
(list :lend " \\\\" :sep " & " :hline "\\hline"))))
(org-fill-template
"
\\begin{%env}%width%align
\\toprule
%table
\\bottomrule
\\end{%env}\n"
(list
(cons "env" (or env "table"))
(cons "width" (if width (format "{%s}" width) ""))
(cons "align" (if align (format "{%s}" align) ""))
(cons "table"
;; only use \midrule if it looks like there are column headers
(if (equal 'hline (second table))
(concat (to-tab (list (first table)))
"\n\\midrule\n"
(to-tab (cddr table)))
(to-tab table))))))
#+end_src
*** longtable
This block can be used to wrap a table in the latex =longtable=
environment, it takes the following arguments -- all but the first two
are optional.
| arg | description |
|-----------+-------------------------------------------------------------|
| table | a reference to the table |
| align | optional alignment string |
| width | optional width specification string |
| hline | the string to use as hline separator, defaults to "\\hline" |
| head | optional "head" string |
| firsthead | optional "firsthead" string |
| foot | optional "foot" string |
| lastfoot | optional "lastfoot" string |
#+srcname: longtable
#+begin_src emacs-lisp :var table='((:table)) :var align='() :var width='() :var hline="\\hline" :var firsthead='() :var head='() :var foot='() :var lastfoot='() :noweb yes :results latex
(org-fill-template
"
\\begin{longtable}%width%align
%firsthead
%head
%foot
%lastfoot
%table
\\end{longtable}\n"
(list
(cons "width" (if width (format "{%s}" width) ""))
(cons "align" (if align (format "{%s}" align) ""))
(cons "firsthead" (if firsthead (concat firsthead "\n\\endfirsthead\n") ""))
(cons "head" (if head (concat head "\n\\endhead\n") ""))
(cons "foot" (if foot (concat foot "\n\\endfoot\n") ""))
(cons "lastfoot" (if lastfoot (concat lastfoot "\n\\endlastfoot\n") ""))
(cons "table" (orgtbl-to-generic
(mapcar (lambda (lis)
(if (listp lis)
(mapcar (lambda (el)
(if (stringp el)
el
(format "%S" el))) lis)
lis)) table)
(list :lend " \\\\" :sep " & " :hline hline)))))
#+end_src
** Elegant lisp for transposing a matrix.
#+tblname: transpose-example
| 1 | 2 | 3 |
| 4 | 5 | 6 |
#+srcname: transpose
#+begin_src emacs-lisp :var table=transpose-example
(apply #'mapcar* #'list table)
#+end_src
#+resname:
| 1 | 4 |
| 2 | 5 |
| 3 | 6 |
2009-08-05 14:02:25 +00:00
* Misc
#+srcname: python-identity(a=1)
Enabling LoB to put results in buffer, and slowly moving towards more unified concept of function calls. Previously LoB calls were not able to produce results in the buffer. These changes go some way to allowing them to do that. [There are still some bugs to deal with]. That meant changing org-babel.el so that there is a notion of the `source block name' for a LoB line, in order to construct a #+resname (currently I've made the name the same as the function call). I'm also slowly moving towards unifying the notion of `function calls' a bit more: I've changed the org-babel-lob-one-liner-regexp so that instead of a monolithic match it now matches first the function name, and second the function arguments in parentheses. org-babel-lob-get-info makes that match, and although it still concatenates them and returns the string, the two elements can be accessed immediately afterwards using match-string. So that situation is very similar to org-babel-get-src-block-name, whose job (in this branch) is also to parse the function *name* and the function *arguments*. In a few places in the code (esp. function names), I think the word `info' should be replaced with `call' or `function call', which I believe more accurately indicates what the `info' is: a function definition, together with bound arguments/references. The function call syntax, i.e. function-name(arg1=ref1), originally introduced for references (and thereby in LoB), and which I'm proposing we use throughout, raises the question of default arguments, and those being over-ridden by supplied arguments, as in e.g. python, and R.
2009-07-11 02:59:10 +00:00
#+begin_src python
a
#+end_src
#+srcname: python-add(a=1, b=2)
#+begin_src python
a + b
#+end_src
* GANTT Charts
The =elispgantt= source block was sent to the mailing list by Eric
Fraga. It was modified slightly by Tom Dye.
#+source: elispgantt
#+begin_src emacs-lisp :var table=gantttest
(let ((dates "")
(entries (nthcdr 2 table))
(milestones "")
(nmilestones 0)
(ntasks 0)
(projecttime 0)
(tasks "")
(xlength 1))
(message "Initial: %s\n" table)
(message "Entries: %s\n" entries)
(while entries
(let ((entry (first entries)))
(if (listp entry)
(let ((id (first entry))
(type (nth 1 entry))
(label (nth 2 entry))
(task (nth 3 entry))
(dependencies (nth 4 entry))
(start (nth 5 entry))
(duration (nth 6 entry))
(end (nth 7 entry))
(alignment (nth 8 entry)))
(if (> start projecttime) (setq projecttime start))
(if (string= type "task")
(let ((end (+ start duration))
(textposition (+ start (/ duration 2)))
(flush ""))
(if (string= alignment "left")
(progn
(setq textposition start)
(setq flush "[left]"))
(if (string= alignment "right")
(progn
(setq textposition end)
(setq flush "[right]"))))
(setq tasks
(format "%s \\gantttask{%s}{%s}{%d}{%d}{%d}{%s}\n"
tasks label task start end textposition flush))
(setq ntasks (+ 1 ntasks))
(if (> end projecttime)
(setq projecttime end)))
(if (string= type "milestone")
(progn
(setq milestones
(format
"%s \\ganttmilestone{$\\begin{array}{c}\\mbox{%s}\\\\ \\mbox{%s}\\end{array}$}{%d}\n"
milestones label task start))
(setq nmilestones (+ 1 nmilestones)))
(if (string= type "date")
(setq dates (format "%s \\ganttdateline{%s}{%d}\n"
dates label start))
(message "Ignoring entry with type %s\n" type)))))
(message "Ignoring non-list entry %s\n" entry)) ; end if list entry
(setq entries (cdr entries)))) ; end while entries left
(format "\\pgfdeclarelayer{background}
\\pgfdeclarelayer{foreground}
\\pgfsetlayers{background,foreground}
\\renewcommand{\\ganttprojecttime}{%d}
\\renewcommand{\\ganttntasks}{%d}
\\noindent
\\begin{tikzpicture}[y=-0.75cm,x=0.75\\textwidth]
\\begin{pgfonlayer}{background}
\\draw[very thin, red!10!white] (0,1+\\ganttntasks) grid [ystep=0.75cm,xstep=1/\\ganttprojecttime] (1,0);
\\draw[\\ganttdatelinecolour] (0,0) -- (1,0);
\\draw[\\ganttdatelinecolour] (0,1+\\ganttntasks) -- (1,1+\\ganttntasks);
\\end{pgfonlayer}
%s
%s
%s
\\end{tikzpicture}" projecttime ntasks tasks milestones dates))
#+end_src