HelpOnParsers

  1. Parsers
    1. ParserBase
      1. cplusplus
      2. java
      3. pascal
    2. python
    3. CSV
    4. reStructuredText
      1. Supported elements
      2. Unsupported elements
      3. Example
      4. Links

Parsers

Parsers go through the contents of a page to create a sequence of formatter calls which in sequence create some readable output. MoinMoin will choose the parser for a page using different techniques: FORMAT Processing Instruction (see HelpOnProcessingInstructions) and code display regions (see HelpOnFormatting).

A #FORMAT pi can be used to tell MoinMoin which parser to use for the whole page content. By default this is the wiki parser. Example:

#FORMAT cplusplus 
... some C++ source ...

With the use of code display regions, a parser can be applied to only a part of a page (this was a processor region in earlier versions of Moin). You specify which parser to call by using a bang path-like construct in the first line. A bang path is a concept known from Unix command line scripts, where they serve the exact same purpose: the first line tells the shell what program to start to process the remaining lines of the script. Example:

{ { {#!CSV
a,b,c
d,e,f
} } }

For more information on the possible markup, see HelpOnEditing.

ParserBase

!ParserBase is a parser utility class used to produce colorized source displays. It is easily extended. The HTML Formatter will render such code displays with switchable linenumbers, if the browser supports DOM and !JavaScript.

A !ParserBase colorization parser understands the following arguments to a #FORMAT pi or a hashbang line. Just add those arguments after the name of the parser (#FORMAT python start=10 step=10 numbering=on or #!python numbering=off).

numbers

if linenumbering should be added. defaults to 'on'. possible values: 'on', 'off' (no linenumbers, but javascript to add them), 'disable' (no line numbers at all)

start

where to start with numbering. defaults to 1

step

increment to the linenumber. defaults to 1

MoinMoin comes with a few examples from which you can go on:

cplusplus

   1 int main(int argc, char **argv) {
   2   return 0;
   3 }

java

   1 import java.util.Date;
   2 import java.util.Calendar;
   3 
   4 public class IntDate
   5 {
   6   public static Date getDate(String year, String month, String day)
   7     {
   8       // Date(int, int, int) has been deprecated, so use Calendar to
   9       // set the year, month, and day.
  10       Calendar c = Calendar.getInstance();
  11       // Convert each argument to int.
  12       c.set(Integer.parseInt(year),Integer.parseInt(month),Integer.parseInt(day));
  13       return c.getTime();
  14     }
  15 }

pascal

   1 function TRegEx.Match(const s:string):boolean;
   2 var
   3     l,i : integer;
   4 begin
   5     result := MatchPos(s,l,i);
   6 end;

python

Colorizes python code. It is not derived from ParserBasem, but it allows the same arguments as the !ParserBase parsers.

   1 def hello():
   2     print "Hello World!"

CSV

The CSV parser works on so-called comma separated values, though the comma is now usually a semicolon. The first line is considered to contain column titles that are rendered in bold, so when you don't want table headers, leave the first line empty.

The bang path can contain "-index" arguments, to hide certain columns from the output; column indices are counted starting from 1.

Any other argument is treated as the seperator. This allows you to use e.g. commas (,) instead of semicolons (;). If you do not supply a separator, ; will be used. See the examples below.

!MoinMoin 1.3 - clipping of the patch history:

patch-366 make _normalize_text public method Nir Soffer
patch-367 fixed failing test wikiutil: good system page names Nir Soffer
patch-368 Fixed DeprecationWarning in RandomPage.py and an unused import in twistedmoin.py Alexander Schremmer
patch-369 remove duplicate code in formatter.base Thomas Waldmann
patch-370 fixed long int in mig3 Thomas Waldmann
patch-371 fixed unicode error on eventlog Nir Soffer
patch-372 fixed util.web.makeQueryString and Page.url Nir Soffer
patch-373 fixed again non ascii http_referer Nir Soffer
patch-374 CSV.py supports different separators now Alexander Schremmer
patch-375 improved searchform behavior on Mozilla/Firefox Nir Soffer
patch-376 More correct script for actions menu init Nir Soffer

MoinMoin Version History:

Version Date
0.11 2002-03-11
0.10 2001-10-28
0.9 2001-05-07
0.8 2001-01-23
0.7 2000-12-06
0.6 2000-12-04
0.5 2000-11-17
0.4 2000-11-01
0.3 2000-10-25
0.2 2000-08-26
0.1 2000-07-29

reStructuredText

The reStructuredText parser is experimental. It works well, but many elements are still unsupported. Before you can use it, you need to install the Python docutils package, which provides the additional reStructuredText support that MoinMoin needs.

Supported elements

Emphasis, strong emphasis, inline literal, headers, bullet lists, enumerated lists, definition lists, field lists, literal blocks, and block quotes. And perhaps others.

Unsupported elements

Line blocks, tables, directives, references of all kinds, footnotes, etc.

Example

#!rst (-)
This is a *very* simple example. If you see two asterisks around the word "very", in the previous sentence, then docutils is improperly installed (or not installed at all). When docutils is there, the word is displayed in italics, and this whole block of text is not displayed in a special source-code-like format, but as a normal part of the page.

Links

last modified 2004-12-11 15:03:12