DPL:Manual - DPL parameters: Controlling output volume

From FollowTheScore
Revision as of 09:46, 17 January 2008 by Gero (talk | contribs) (oneresultheader)
Jump to: navigation, search

Template:Type:Manual


Per default DPL2 will show the name of each article, including its namespace. The name will serve as a link to the article. By setting the output control parameters of DPL you can

  • show the number of articles found
  • suppress the namespace part of the articles´ names
  • cut off the articles´ names at a certain length
  • add meta data (size, author, last edited by, date of last change, ..)
  • add access date (frequency, date of last access)
  • add the articles´ contents or parts of it

Furthermore you can add a heading if there are some articles to display.

And you can define an alternate text which will be shown if there are no articles to display.


Defining Header and Footer for the Result

resultsheader

resultsheader output a headline if there is at least one article to display.

Syntax: resultsheader=some wiki text

Note: The symbol %PAGES% will be replaced by the number of pages found. %VERSION% will be replaced by the current DPL version. The symbol '\n' will be converted to a newline symbol. This is useful if you want to use wiki markup which needs to start at the begin of a line. You may need to add \n if the wiki parser requires a linefeed to understand your wikitext. See Test resultsheader. If 'oneresultheader' is specified as well, the latter will be used in case there is eactly one entry in the DPL result set. Then 'resultsheader' will only be used if there are TWO or more entries in the result set.

Example:

<DPL>
   category=Africa
   mode=userformat
   resultsheader=There are %PAGES% pages on Africa.
</DPL>

With "mode=userformat" we get complete control over the output. We do not specify "listseparators=" - so nothing will be displayed except the results header. The only thing we will get is a message about the existing number of articles.

resultsfooter

resultsfooter output a summary 'footline' if there is at least one article to display.

Syntax: resultsfooter=some wiki text

Note: The symbol %PAGES% will be replaced by the number of pages found. %VERSION% will be replaced by the current DPL version. The symbol '\n' will be converted to a newline symbol. This is useful if you want to use wiki markup which needs to start at the begin of a line.

Example:

<DPL>
   category=Africa
   mode=userformat
   resultsfooter=\n===There are %PAGES% pages on Africa.===\n
</DPL>

With "mode=userformat" we get complete control over the output. We do not specify "listseparators=" - so nothing will be displayed except the results footer. The only thing we will get is a message about the existing number of articles. We used '\n' to make sure that our headline syntax will really be recognized by the wikitext parser. See also Test resultsheader.

oneresultheader

oneresultheader output a headline if there is exactly one article to display.

Syntax: oneresultheader=some wiki text

Note: The symbol '\n' will be converted to a newline symbol. %VERSION% will be replaced by the current DPL version. This is useful if you want to use wiki markup which needs to start at the begin of a line. You may need to add \n if the wiki parser requires a linefeed to understand your wikitext. See Test resultsheader.

oneresultfooter

oneresultfooter output a footline if there is exactly one article to display.

Syntax: oneresultfooter=some wiki text

Note: The symbol '\n' will be converted to a newline symbol. This is useful if you want to use wiki markup which needs to start at the begin of a line.

noresultsheader

noresultsheader output a headline if there is no article to display (empty result).

Syntax: noresultsheader=some wiki text

Note:

Setting noresultsheader to the empty string ("noresultsheader=") will suppress the warning message from DPL" which is normally issued if no articles were found.

noresultsfooter

noresultsfooter output a footline if there is no article to display (empty result).

Syntax: noresultsfooter=some wiki text

Note:

noresultsfooter is essentially the same as noresultsheader. If there is no output to display the difference between header and footer becomes incredibly marginal ..


Adding meta data to the output

addcategories

addcategories Shows all categories to which an article belongs as a small text line after the article name.

Example:

<DPL>
  category       = Africa
  addcategories  = true
</DPL>

This will show all Africa pages together with a list of catgeories per article. Of course "Africa" will be present in each entry of the list.

Note: in 'mode=userformat' the category list can also be referenced as a built-in variable.

addpagecounter

addpagecounter Shows number of times the page has been viewed according to the definition of the 'page_counter' field on Page_table.

Example:

<DPL>
  category       = Africa
  ordermethod    = counter
  order          = descending
  addpagecounter = true
  counter        = 5
</DPL>

This will show the 5 most popular articles about Africa.

Note: in 'mode=userformat' the access counter can also be referenced as a built-in variable.

addpagesize

addpagesize Shows the size of the page.

Example:

<DPL>
  category       = Africa
  ordermethod    = size
  order          = descending
  addpagesize    = true
  counter        = 5
</DPL>

This will show the 5 biggest articles about Africa.

Note: in 'mode=userformat' the size can also be referenced as a built-in variable.

addcontribution

addcontribution shows how much a user contributed to an article.

Syntax:

addcontribution=true

If omitted, the default is false.

Using this parameter will restrict your output to articles which were edited recently (This is usually one week or one month, according to the setup of your wiki).

You will get a link to the contributor and an "asterisk bar" which indicates how many bytes in the article were changed by that user.

You will also have the built-in variables %CONTRIBUTOR%, %CONTRIB% and %CONTRIBUTION% available to be used in your own format statement.

adduser

adduser Requires ordermethod=[...,]firstedit or ordermethod=[...,]lastedit (where the [...,] signifies a complex ordermethod with extra parameters). If firstedit (lastedit), 'adduser=true' displays the user who made the first (last) revision of the page. In this way the parameter is equivelent to the addauthor (addlasteditor) parameter (see below).

Syntax:

adduser=true

If omitted, the default is false.

Example:

<DPL>
  category    = Africa
  ordermethod = lastedit
  adduser     = true
</DPL>

This will output a list of pages belonging to [[Category:Africa]], prepending each result with the author of the last revision.

Note: if 'mode=userformat' the user can be referenced as the built-in variable %USER%.

addauthor

addauthor show the user who created the article

Syntax:

addauthor=true

If omitted, the default is false.

Example:

<DPL>
  category    = Africa
  addauthor   = true
</DPL>

This will output a list of pages belonging to [[Category:Africa]], which shows the user who edited the initial revision of that page.

Note: addauthor and addlasteditor cannot be used within the same DPL query


addlasteditor

addlasteditor show the user who edited the most recent revision of a page

Syntax:

addlasteditor=true

If omitted, the default is false.

Note: addauthor and addlasteditor cannot be used within the same DPL query

addpagetoucheddate

addpagetoucheddate Shows date/time of last change to the page according to the definition of the 'page_touched' field on Page_table.

Requires ordermethod=[...,]pagetouched or ordermethod=[...,]title. ([...,] means complex ordermethods with extra param before are allowed.)

Example:

addpagetoucheddate=true

If omitted, the default is false.

Example:

<DPL>
  category=Africa
  ordermethod=pagetouched
  addpagetoucheddate=true
</DPL>

This list will output a list of pages belonging to [[Category:Africa]], prepending each result with the date of last change (formatted according to your local mediawiki date display preferences or the user's preferences if logged in).

Note: in 'mode=userformat' this date can also be referenced as a built-in variable. The formatting of the date can be influenced using userdateformat=

addeditdate

addeditdate Requires ordermethod=[...,]firstedit or ordermethod=[...,]lastedit. ([...,] means complex ordermethods with extra param before firstedit

Purpose:

If firstedit (resp. lastedit), 'addeditdate=true' shows the date of the first revision/creation (resp. last revision) of the page.

Example:

addeditdate=true

If omitted, the default is false.

Example:

<DPL>
  category=Africa
  ordermethod=lastedit
  addeditdate=true
</DPL>

This list will output a list of pages belonging to [[Category:Africa]], prepending each result with the date of last revision (formatted according to your local mediawiki date display preferences or the user's preferences if logged in).

Note: in 'mode=userformat' this date can also be referenced as a built-in variable. The formatting of the date can be influenced using userdateformat=

addfirstcategorydate

addfirstcategorydate Shows the date/time the article got added to one of the listed include categories. If there are more than one categories listed and an article belongs to more than one of them, the result is ambiguous.

From a logical point of view it is recommended to include one category only with 'category' parameter or to make sure that each of the article in the result belongs to only one of the categories listed. Conflicts with other "add*date" (addeditdate, etc.) parameters to avoid confusion.

Example:

addfirstcategorydate=true

If omitted, the default is false.

Example:

<DPL>
  category             = Africa
  addfirstcategorydate = true
</DPL>

This list will output a list of pages belonging to [[Category:Africa]], prepending each result with the time and date (formatted according to your local mediawiki date display preferences or the user preferences if user logged in).

Note: in 'mode=userformat' this date can also be referenced as a built-in variable. The formatting of the date can be influenced using userdateformat=


Include contents from the articles in the result set

include

include include pages (whole content) or include certain sections of articles or template parameters.

This functionality is based on the ideas and work of Steve Sanbeg and his extension Labeled Section Transclusion. DPL comes with a modified version of Sanbeg´s source, so there is no need for additional installation.

With include you can incorporate one or more of the following

  • the whole article as it is
  • a certain chapter -- identified by its headline
  • a certain chapter -- identified by its position (sequence number, regardless of level)
  • parameter(s) used in template calls
  • the output of a surrogate template (phantom templat) which is used instead of the original template
  • pieces of text which are marked by special section markers

There is a close correlation between the output of the include statement and table and tablerow. You should understand those statements if you want to create tabular output from included content.

Syntax:

include whole article

To include a whole page, use a wildcard:

include=*
you can use includemaxlength=[n] to restrict the included text to a portion of the article. The text is truncated in a way which will not spoil tag structures or bracket nestings. It will be cut at a word boundary if possible.

include chapters

To include sections (chapters) by a reference to their heading: use the heading name with a "#" as prefix. This will include text from the first occurrence of the heading 'heading1' (resp. 'heading2') until the next heading of the same or lower level. (Refer to Transcluding visual headings.). Specifying a single "#" will include the text up to the first heading. Note that the text comparison is case insensitive.

Note that this functionality is also available as a separate parser function (called dplchapter).
  • Normally your text pattern will be compared literally and it must match the whole headline of the chapter. If you use a double hash sign (##), however, the text will be taken as a regular expression. It will automatically be enclosed within slashes and the matching will be case insensitive and will refer to the headline as a whole (^regexp$). So if you only want to match part of the text you must add ".*" around your pattern. It is your responsibility to provide correct regexp syntax. Otherwise you may see runtime errors, get no output etc.
include = ##.*omethin.*
The above will match a Headline like "This is something old" or "That Was Something Old".
  • Add a limit and an optional link text in square brackets to delimit the size of the included text portion to a certain number of characters. If text is cut off a link will appear which points directly to the chapter which was included. Using "0" as limit will only show the link (if the corresponding chapter in the article is not empty). When truncating included text portions care is taken not to spoil wiki syntax (i.e. maintain a symmetry of brackets and braces, make sure that nowiki and pre tags are balanced). Therefore the size of the included text can deviate from what you specified.
If there is no chapter with the specified heading the output will be empty.
include = #heading1,#heading2[limit linktext],....

linktext can be

plain text
then this text will be used as a link
text starting with "img="
then the link will appear as an image link (you must provide the image in a suitable size; no scaling will happen); note that image linking only works if there is another mediawiki extension installed (LinkedImages).
a text containing %SECTION%";
this symbol will be replaced by "article#heading". It is up to you to create a link from this text using normal wiki syntax if you wish.
  • To include sections (chapters) by a reference to their position in the text: use a simple number with a "%" as prefix. "%0" will include the text BEFORE the first chapter, "%1" will include the first chapter, %6 the sixth and so on. Chapers are counted regardless of their level. No nesting logic is applied. "%-1" is a special value which refers to the last chapter of an article (%-2 will not work, however). "%2[30]" will include the first 30 characters of the second chapter. %SECTION% will contain the heading of the selected chapter. See Test include sections by number.
include = %1[100]

includes the first 100 characters of the first chapter

include contents related to templates

  • You can include one or more parameters of a template call by specifying a list of names or position numbers separated by colons. Names are use for named parameters, numbers for positional parameters.
include={animal}:age:size:1
This will include the values of the two parameters named 'age' and 'size' and the value of the first unnamed parameter in all calls of the template "animal".
  • Instead of collecting parameter values you can include the output of a 'surrogate' template. DPL calls this template instead of the original template with the identical parameter list (plus the two additional parameters %PAGE% and %TITLE%). You will get the output of the surrogate template. To make this happen you specify the name of a template within single curly braces and add a suffix. As said before, the template name including the suffix will be called instead of the original template, with the same parameters, and the result will contain just the output of your alternate template (see example). Note that we pass four additional parameters to each template, called
    •  %PAGE%
    •  %TITLE%
    •  %DATE%
    •  %USER%

They allow the template to produce a link to the source file from where it was included, show the date from the DPL query (e.g. lasteditdate) and show the user from the query (e.g. the user who made the last edit).
If a page listed in your result does not use the specified template you will get the possibility to react properly as the DPL engine will call a template with the additional suffix ".default".

Example: includepage={myTemplate}.dpl
If a page does not contain a reference to "Template:myTemplate" then "Template:myTemplate.dpl.default" will be invoked with %PAGE% and %TITLE% as parameters.

include specially labeled text portions

  • To include sections which are labeled with special tags you just mention the tag name. Refer to Labeled Section Transclusion for details on how to label sections in your pages accordingly:
include=sec1,sec2,...

If the section name starts with a '*' the asterisk will be cut off and the rest of the text will be taken as a rtegular expression. If the section name is '**' it will match all section names. In both cases the tag name will precede the output, separated by '::'.

See Test:section inclusion

include nothing

  • To include nothing from the page (no inclusion), leave blank (this is default):
include=

combinations of the above possibilities

All of the above can be combined, you can even refer to the same template or heading more than once.

include = {template1}suffix1,{template1}:p1:p2:2,#headline,marker,....

Example: include = foo,#bar[200 ..more..],#bang[0 img=my.gif],{boo}.dpl

This will include

  • a text portion which is marked by special tags named "foo"
  • the contents of a chapter named "bar"; only the first 200 characters of wiki text will be shown; if the chapter is larger than that, a link with the text "..more.." will point to the source
  • an image link based on "my.gif" to the chapter "bang" in the article - if there is such a chapter and it is not empty
  • the output of template "boo.dpl" being called with the same parameters that were used to call template "boo" in the original page.

Note: See also DPL:Manual - DPL parameters: Criteria for page selection#includematch.

includepage

includepage this can be used as a longer name for "include".

includemaxlength

includemaxlength Delimit the size of an included article to a maximum of [n] characters of wiki source text or less.

Care is taken to respect pairs of braces and brackets as far as possible. Otherwise we might confuse the result by half-cut syntax elements of transcluded sections. Therefore the output might be shorter or even larger than [n] characters.

Syntax:

includemaxlength=[n]

Note: this parameter is only used in combination with includepage=*; truncation of included chapter contents is specified by adding a limit in sqauare brackets within the 'includepage=' statement.