User talk:Pathoschild

From Wikisource

[edit] Categories

Hey Pathoschild, thank you for informing me about the correct practice when it comes to subpage categorisation. I've proceeded to remove the categories from the Illiad and Odyssey books. Best —Anonymous Dissident^Talk 14:30, 13 February 2009 (UTC)

You're welcome, and thanks for removing them. —Pathoschild 22:21:23, 13 February 2009 (UTC)

[edit] Cross-namespace redirects

Hello Pathoschild. User:Eclecticology has questioned your 2006 implementation of the policy or guideline discouraging cross-namespace redirects at WS:S#Allowing cross-namespace redirects. Perhaps you could shed some light? Best, —Anonymous Dissident^Talk 13:21, 19 March 2009 (UTC)

[edit] Hard-coded font faces

Hello, on the topic of hard-coded font faces, I've started a discussion at Scriptorium. --Dan Polansky (talk) 09:52, 24 March 2009 (UTC)

[edit] {{DGRA}} header template

Gday. We talked recently about the standardisation of built header templates with prime part of discussion being {{DNB00}}. I have also stumbled over {{DGRA}} and wondered on whether this should just be replaced by a standard header, or whether it is salvageable. Thx. -- billinghurst (talk) 05:37, 6 April 2009 (UTC)

So long as the parameters are the same (which they appear to be, despite a comment that it requires multiple previous/next), I think it should be deprecated and replaced with {{header}}, then deleted. Jude (talk) 11:19, 18 April 2009 (UTC)

Replaced the several relevant pages with {{header}} and template deleted. -- billinghurst (talk) 12:16, 18 April 2009 (UTC)

I converted {{DNB00}} into a wrapper for {{header}}, and plopped it into Category:Specific navigation templates. —Pathoschild 18:42:24, 18 April 2009 (UTC)

Bonzer. Thx. -- 22:27, 18 April 2009 (UTC)

[edit] Regex

Hi! I'm just learning how to use regex (thanks for the Regex menu framework =D), but I've noticed you haven't added the <source> tag among the exception patterns at User:Pathoschild/standardise.js:

	/* exception pattern */
	var pattern = '<(nowiki|poem|pre)[^>]*>[\\s\\S]*?<\\/\\1>'; // double-escaping needed for RegExp()

Is it ok? Or should it be added? Helder (talk) 23:01, 2 June 2009 (UTC)

Sure, I added it to the exception list. Thanks for the suggestion. :) —Pathoschild 12:46:05, 30 June 2009 (UTC)

[edit] Welcome back

You have been missed! Hope that your break was worthwhile and fulfilling. :-) billinghurst (talk) 12:24, 30 June 2009 (UTC)

Thanks, but I won't be editing much for a while; I just check my watchlist regularly. :) —Pathoschild 12:40:28, 30 June 2009 (UTC)

Spoilsport. :-/ billinghurst (talk)

[edit] See new header template

Gday P/c. When you have the chance (not urgent) would you mind running your eye over the template {{DEL-begin}} and see whether it meets the criteria to equate with {{header}}. Once you give it the all clear, I will add it to our tag filters so it doesn't show up there. Thx. -- billinghurst (talk) 12:09, 11 August 2009 (UTC)

[edit] Please see User talk:John Vandenberg#I have lightly fubar.27d.2C need advice

John says that you have experience in cleaning up history fubars. If you have the time to have a look at this discussion, then that would be great. Thx. -- billinghurst (talk) 15:04, 17 August 2009 (UTC)

Oh, I had already left a message. I had forgotten. For reference, the files that were imported at at User_talk:Billinghurst#Import log Fubar. Thanks for helping me out here. billinghurst (talk) 04:30, 30 August 2009 (UTC)

Fixed.

• {{documentation/docname}}, {{para}}, {{Template:para/doc}}, {{pp-meta}}, {{pp-template}}, and {{purge}} had no history conflict, but I deleted them because they were unused. Feel free to restore any you need.
• {{documentation}}, {{documentation subpage}}, and {{tlx}} fixed.

I wrote a quick and dirty script to fix edit history merges; when run on the source's history page, it generates a script that will check/uncheck revisions when restoring the target's history. I'll paste it here for future reference, since I otherwise won't keep it around.

/*######################
## Generate a unique revision signature
######################*/
function gen_signature( li ) {
	/* extract data */
	var text = li.innerHTML; // parse string instead of DOM, much faster
 
	var date    = text.match( /\d+:\d+, \d+ [a-z]+ \d+/i )[0];
	var user    = text.match( /mw-userlink"[^>]*>([^<]+)<\/a>/i )[1];
	var comment = text.match( /comment"[^>]*>\((.+)\)<\/span>/i );
 
	/* flatten comment & strip HTML */
	if( comment )
		comment = comment[1].replace( /<[^>]+>([^\n]+?)<\/[^>]+>/ig, '$1' );
	else
		comment = '';
 
	/* return signature */
	return date.replace( /[^a-z0-9]/ig, '' ) + '|' + user.replace( /[^a-z0-9-]/ig, '' ) + '|' + comment.replace( /[^a-z0-9_-]/ig, '' );
}
 
/*######################
## On a history page, generate script to run on Special:Undelete
######################*/
function parse_history() {
	/* get history */
	var form    = document.getElementById( 'mw-history-compare' );
	var history = form.getElementsByTagName( 'li' );
 
	/* generate signatures */
	var signatures = [];
	for( var i = 0, len = history.length; i < len; i++ )
		signatures.push( gen_signature(history[i]) );
 
	/* build script */
	var script = '/*######################\n## Generate a unique revision signature\n######################*/\n'
		+ gen_signature + '\n\n'
		+ '/*######################\n## Bad revision signatures \n######################*/\n'
		+ 'var signatures = \'#' + signatures.join( '#' ) + '#\'\n\n'
		+ '/*######################\n## Toggle revisions \n######################*/\n'
		+ uncheck_bad_revisions	+ '\n'
		+ 'uncheck_bad_revisions();'
 
	/* output */
	pre = document.getElementById( 'fix-fubar-history' );
	if( !pre ) {
		var div = document.createElement( 'div' );
		div.setAttribute( 'style', 'padding:1em; border:1px solid #F00;' );
 
		var header = document.createElement( 'h3' );
		header.appendChild( document.createTextNode('Fix fubar history') );
 
		var text = document.createTextNode( 'Delete the imported page, then run the script below on its undelete page to check good revisions and uncheck bad ones.' );
 
		var pre = document.createElement( 'pre' );
		pre.setAttribute( 'id', 'fix-fubar-history' );
 
		div.appendChild( header );
		div.appendChild( text );
		div.appendChild( pre );
		form.insertBefore( div, form.childNodes[0] );
	}
	pre.innerHTML = script;
}
parse_history();
 
 
/*######################
## On undelete page, check/uncheck revisions based on signatures
######################*/
function uncheck_bad_revisions() {
	/* default reason */
	document.getElementById( 'wpComment' ).value = 'Fixed merged edit history';
 
	/* get history & offset to deletion log */
	var history = document.getElementById( 'undelete' ).getElementsByTagName( 'li' );
	var offset  = 0;
	while( history[offset].getAttribute('class') == 'mw-logline-delete' )
		offset++;
 
	/* (un)check boxes */
	var count = 0;
	for( var i = offset, len = history.length; i < len; i++ ) {
		if( signatures.indexOf('#' + gen_signature(history[i]) + '#') == -1 ) {
			history[i].getElementsByTagName('input')[0].checked = true;
			history[i].setAttribute( 'style', '' );
		}
		else {
			history[i].getElementsByTagName('input')[0].checked = false;
			history[i].setAttribute( 'style', 'background:#FCC;' );
			count++;
		}
	}
 
	alert( count + ' bad revisions.' );
}

—Pathoschild 17:58:44, 30 August 2009 (UTC)

Many thanks. Lifesaver! Is leaving me with the code representative of confidence or the challenge to fubar again? <eyeroll> billinghurst (talk) 22:17, 30 August 2009 (UTC)

I'm curious about the script above... What do you mean by "fix edit history merges"? Helder (talk) 11:48, 2 September 2009 (UTC)

[edit] Comment at MediaWiki talk:Usejs/Comment highlighting.js

I am clueless on what the file even does, so I will just alert you, and my job is done!-- billinghurst (talk) 00:36, 2 September 2009 (UTC)

[edit] |override_translator request for {{header}}

Man crawling across something broken and spiky.

I am trying to find someone who will be able to amend {{header}} so that it is able to carry the parameter | override_translator = ... where it basically doesn't try to wiki the contents of the field, and just takes the entered content, akin to |override_author in the same template. I cannot find any takers, and I have tried numerous people. I have looked at the file, however, the levels of nesting on #if and #ifeq statements, with pipes etc. is sending my brain around the twist, and it is not one that I want to just play with. Would you be able to help? Apologies for having to ask a busy person. -- billinghurst (talk) 15:13, 3 October 2009 (UTC)

Done. It will use most text as-is, but it will still handle "?", "not mentioned", "unknown", and "wikisource" as special keywords; that can be changed easily enough if needed. —Pathoschild 18:31:54, 03 October 2009 (UTC)

Brilliant, those keywords are definitely still needed. /me does a little dance. -- billinghurst (talk) 22:59, 3 October 2009 (UTC)

[edit] User:Pathoschild/Template:PD-age

Perhaps you would like to look at Help:Copyright_tags#Automated_public_domain_licenses to upgrade your templates to combine US and automatically updated non-US copyright status over years.--Jusjih (talk) 03:39, 6 October 2009 (UTC)

[edit] Another header needing alignment {{EAm06}}

Adding to your WHENEVER list, just come across another pseudo-header {{EAm06}} that could do with your magical alignment with {{header}}. Thx. billinghurst (talk) 23:53, 24 October 2009 (UTC)

and {{Collier 1701}} <shrug> billinghurst (talk)

Done, along with {{EB11}} and {{CAB03}}. —Pathoschild 02:30:41, 25 October 2009 (UTC)

[edit] Again thanks!

Probably inelegant, however, 'tis functional Template:Active projects/var‎ and Template:Active projects/base. -- billinghurst (talk) 13:01, 1 November 2009 (UTC)

[edit] How to identify non-transcluded validated and proofread pages

Hi P/child. Trying to work out how we could look to identify pages in Page: namespace that are in Category:Proofread and Category:Validated however have not been transcluded into the main namespace, aka missed pages. Also something that we could look to identify on a regular basis so we could undertake the necessary maintenance work to transclude pages. Ideas? Thanks. billinghurst (talk) 11:39, 12 November 2009 (UTC)

If I may interject, the API supports an "embeddedin" query, which will give you a list of pages that transclude a given page. You can restrict the transclusion namespace, and you can limit it to only return one result, which suffices if all you want to know is whether there are any transclusions e.g. http://en.wikisource.org/w/api.php?action=query&list=embeddedin&einamespace=0&eititle=Page:Makers_of_British_botany.djvu/15&eilimit=1

It would be pretty easy to write a pywikipedia bot to walk a category and identify pages that aren't embedded in the main namespace. I'll even do it for you if you don't want to do it yourself. I could tweak a bot I have which does something roughly similar, so it is probably only half an hour's work.

Hesperian 13:16, 12 November 2009 (UTC)

Sweet, I am a pro, I don't mind who I have to huggle (or their friends) to get the job done. It is the end result that is my aim. Thx. billinghurst (talk) 19:32, 12 November 2009 (UTC)

Done. Running now. Shall we continue this discussion at User talk:Billinghurst#Non-transcluded validated pages. Hesperian 00:46, 13 November 2009 (UTC)