Is there any way to fetch the talk page title of a given page title through MediaWiki API?
I know that I can get talkid using prop=info. But the problem is that there is no pageid for a talk page that does not exist yet. Also there are some obvious ways to get talk page title by adding a prefix to subject title, but it seems to me that they are all language/setup dependent...
leo's answer: So, given a title, that might not exist, you want to know what the talk page would be if the page was created, in a wiki with custom namespaces that you do not know? Then I would just have grabbed the list of all namespaces from the API: http://www.mediawiki.org/w/api.php?action=query&meta=siteinfo&siprop=namespaces, and looked my prefix up there. The talk page namespace will be in NS+1.
(Posted as community wiki so that this question isn't listed as unanswered.)
Related
I want to extract the pages mentioned in the infobox and templates of pages.
E.g. From this page:
https://en.wikipedia.org/wiki/DNA
I want to extract all of the links in the infobox, like: "Genetics", "Introduction to Genetics" etc.
I want to do it, by using the sql dumps, possibly avoiding to parse the xml of whole pages, and I don't want to do it with APIs.
I could not find a way.
While Pagelinks does include also the links of infoboxes, I cannot find a way to exclude them.
I thought Templatelinks may have that info, but it is not: I could not find the pageids of the corresponding links in infoboxes.
Where is this information stored?
Or which kind of tables should I look at?
I consulted previous questions:
where can I find the infobox templates used in wiki?
and Mediawiki reference:
https://www.mediawiki.org/wiki/Manual:Templatelinks_table#Schema_summary
but could not find a solution.
That is a sidebar rather than an infobox: https://en.wikipedia.org/wiki/Template:Genetics_sidebar
I don't think there's a way of doing it other than parsing the content of the template to extract the links or using the API: e.g. https://en.wikipedia.org/w/api.php?action=query&prop=links&titles=Template:Genetics%20sidebar&pllimit=100&plnamespace=0
Something like this should also work but it's not returning any results for me:
SELECT * from pagelinks
where pl_title = 'Genetics_sidebar'
and pl_namespace = 0
and pl_from_namespace = 10
https://quarry.wmcloud.org/query/71442
Is there a way to get the intro content from wikipedia page to my mediawiki page? I was thinking of using wikipedia's api but i dont know how to parse the url on my page and also with templates. I just want a query that will display the introduction part of a wikipedia page on my page?d
I used the External_Data Extension and Wikipedia's api to achieve this.
The API
http://en.wikipedia.org/w/api.php? action=query&prop=extracts&format=json&exintro=&titles=[title of wikipedia page]
How I used it
{{#get_web_data:
url=http://en.wikipedia.org/w/api.php? action=query&prop=extracts&format=json&exintro=&titles={{PAGENAME}}
|format=JSON|data=extract=extract}}
How I displayed the extract on pages
{{#external_value:extract}}
I however need to figure out how to get only a paragraph from the return text. Will probably use a parser function.
When I make api query sometimes I have list with few pages. For example
http://en.wikipedia.org/wiki/Ask gives a lot of pages, I need website "Ask.com, a web search engine, formerly Ask Jeeves"
can I make query only for some category ("websites")?
How I can check category for each page in response?
Thanks
There is no trivial way to do what you're asking. You could do something like this:
Get the list of pages the disambiguation page list. You could do this by listing the links on that page (action=query&prop=links).
Get the categories of all the pages from the previous step and use that to decide which one is the one you're looking for. This is not that simple, because Ask.com is not directly in Category:Websites, it's in one of its subcategories.
I have list with few pages, for example http://en.wikipedia.org/wiki/Ask
The problem is that you're not getting a list of pages, you just are getting an ordinary page which is in the disambiguation pages category. To get the list, you need to get the links in that page.
can I make query only for some category ("websites")?
No, mediawiki does not support that.
How I can check category for each page in response?
Use the links property as a title list generator and get the categories of each page in the response. In your case, that would be http://en.wikipedia.org/w/api.php?action=query&titles=Ask&generator=links&prop=categories (don't forget to continue the query).
If you are OK with "full-text search" for "ask",
you can do that like this:
http://en.wikipedia.org/w/api.php?format=json&action=query&generator=search&gsrsearch=ask%20incategory:%22Online%20companies%22&prop=info
As you can see, "search" text is [ask incategory:"Online companies"]
The same solution also can be seen at:
Wikipedia API: how to search for a term in a specific category
I'm talking about using the wiki syntax. Normally, I create an article by making a link and then editing it, such as
[[secure-auth|Secure 2-Factor Authentication]]
But this results in the article being titled "secure-auth" on that page. I want the URL to be foo.com/mywiki/secure-auth , but the article title to be "Secure 2-Factor Authentication".
You can't have an article with URL different from the article title. But there are some ways to work around that:
Create a redirect from “secure-auth” to “Secure 2-Factor Authentication”.
Use {{DISPLAYTITLE}} to modify how the page title looks like.
Configure your web server (most likely Apache or IIS) to rewrite the URL the way you want.
Is there any microformat/standard for implementing search form on the site?
(access keys, naming etc.)
Any good practices?
The only things I can think of are that searches should be GET requests, and that you may want to implement a RESTful API that allows developers to query JSON and XML results in addition to HTML
If you are trying to implement a plugin for browsers like IE and Firefox to allow for search/autocomplete in the search box of the browser, check out this: https://developer.mozilla.org/en/creating_opensearch_plugins_for_firefox
Here are some conventions I follow. Forgive me if these are not exactly on topic with microformats, or "technically not" in the way I describe the various parts to my answer.
I have found validation in these few standards I've copied from others:
HTML form ID = "search"
Action URL for the form is //root-of-site/search/
Search Results URL construct:
//root-of-site/search?q=queryClause1+Clause2&AnotherParamName=foo
[personally that structure bugs me a little because search-forward-slash appears to be a directory, and the search-question-mark looks like a page taking a query string, and IMO a page should have a suffix. I've been tempted to use search.cgi or search.app, but I see the big guys using /search?q= and so it is]
ID of search query is "q" (this is nearly universal in adoption)