This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the
current talk page.
Can AWB create mass categories? E.g. i need to create categories from ”Football clubs established in 1900” to ”Football clubs established in 2020”. Each category must have 2 categories- parents, one stable ”Football clubs establishements by year” and second incrementable ”Establishements in {{year}}", where {{year}} has value from 1900 to 2020, for each new category incrementation +1. Its possible to do that with AWB?
XXN (
talk)
19:50, 24 October 2013 (UTC)
Thanks, but i dont understand how to use your plugin)) I know about bot policy, i will work on ro.wiki, i have already submitted an request for approval.
XXN (
talk)
23:35, 25 October 2013 (UTC)
section parameter in cite web template and AWB alert
Resolved
Hi,
I just reformatted a bunch of citations on
amphetamine for consistency. I noticed that my use of "| section=XYZ" in
template:cite web generates an error alert in AWB even though the template itself accepts the parameter and doesn't display an error notice (in refs or hidden categories). Just thought I'd put this notice here. Just to provide an example[1][2] on this page:
Extract common code from checkisbn() and issn() into new function is_valid_isxn(): Checkdigit calculation code for ISSN and for ISBN-10 is esentially the same so created a single function to do that
Date validation:
discussion – by far the largest, this change checks dates for format compliance with
MOS:DATE, checks date validity (no June 31, etc), allows year disambiguation in CITEREF identifiers when referenced authors have multiple works published in the same year without the need to use both |date= and |year=, and does not corrupt the
COinS metadata.
Is the edit summary, "clean up using
AWB" something that is automatically generated? If so, I found it to be somewhat disingenuous when I saw it recently. Rather than clean up, an editor had added
Persondata. While that was definitely the right thing to do, and a good use of AWB, it's my opinion that the edit summary should have been clearer about what was going on.
Ardric47 (
talk)
01:21, 17 November 2013 (UTC)
Ardric47, that was not an automatic edit summary. Edit summaries are entered by the editor. The only thing AWB tacks on is "using AWB" to show the edit was done using a tool. (AWB being the tool, the editor may or may not be a tool) Everything before "using AWB" is entered by editor. If a user is using a beta version of AWB, the "version" number is also given... "using AWB (9643)". You will need to contact the editor about the edit summary.
Bgwhite (
talk)
02:57, 17 November 2013 (UTC)
Checking template parameter against article title?
Resolved
Is there a way to check if a specific template parameter is identical to the article's title in a RegEx (or any other way for that matter)? I'm going to try to replace instances of {{{diversity_link}}} in
Taxobox which just link back to the article to prevent it from breaking (see
Template talk:Taxobox#Diversity link) after putting in a fix. I can see it's possible to do the opposite (i.e. add the article title to the replacement text) by
matching the title, and apparently there
used to be a bug which must have been doing essentially the same thing... but are there any magic words or equivalents for use within search strings, or does anyone know another way to do this? ‑‑xensyriaT12:00, 1 December 2013 (UTC)
Fix Script Error bug that occured when |doi_brokendate= did not contain a year value;
Fix doi() so that dois with invalid doi_brokendate categorize to "Pages with inactive DOIs" and not to "Pages with DOIs inactive since";
Change deprecated_parameter() to emit a single error message; (
discussion}
Fix bug in checkisbn() that stripped-out non-isbn characters before validation so that ISBNs were declared good as long as the stripped (not displayed) version of the isbn passed the remaining tests; (
discussion)
Year and PublicationDate promotion to Date consolidation; (
discussion)
Change validate() and the whitelist to recognize deprecated parameters; (
discussion)
How can i insert a list of words into Normal settings
I am having a spreadsheet list which consists more than 800words in column-A. All the listed words are to be replaced with another 800 words in column-B. How can i insert the spreadsheet list into the option of Find & replace's Normal settings.--
தகவலுழவன் (
talk)
11:05, 10 December 2013 (UTC)
I advise you to use an external text editor, like
Notepad++, and use the F&R of this text editor. Gives an example of formatting your list otherwise, it may help to understand the problem.
Automatik (
talk)
23:07, 12 December 2013 (UTC)
This is the sample list with few words. I have to replace nearly 1000 words in the uploading data. Using find & replace method, one by one is tedious work. So, i want to insert the list in AWB (options-->find & replace-->Normal settings) How can i ? --
தகவலுழவன் (
talk)
01:25, 14 December 2013 (UTC)
You can use an external text editor to prepare from each pair of words
Then save it in proper section (between <Replacements>...</Replacements>) of AWB setting file (.xml) and load it as a settings in AWB.
But maybe easier would be to write simple module (Tools --> Make module) with Regex.Replace and use it instead of F&R function like that:
public string ProcessArticle(string ArticleText, string ArticleTitle, int wikiNamespace, out string Summary, out bool Skip)
{
Skip = false;
Summary = "";
ArticleText=Regex.Replace(ArticleText, @"ABC", "{{ABC}}");
ArticleText=Regex.Replace(ArticleText, @"(ABC|KLM|XYZ)", "{{$1}}"); /* you can combine words if a result is similar */
/* etc. */
return ArticleText;
}
"AutoWikiBrowser requires Microsoft Windows 2000/XP or newer. It also requires Version 2 of the .NET Framework (users of Windows 2000 and Windows XP must download .NET Framework; it is included in Windows Vista and newer)."
When you go to that link, it says...
"Microsoft .NET Framework Version 2.0 Redistributable Package (x86)
The download you have chosen is no longer available. Please download Microsoft .NET Framework 3.5 Service Pack 1 instead."
If, like me, you prefer to keep a local copy of Microsoft downloads in case it says *cough* "the download you have chosen is no longer available", the full package is at...
Can anyone explain
this: that is, why does AWB insist on inverting the placement of refs #1 and #2? Most of the time the re-ordering works, but this time no. I found the problem when I did it the
first time, and now it has happened for someone else. -- Ohc ¡digame!01:55, 18 December 2013 (UTC)
AWB detects the "in use" state of the page, if the page transcludes a template {{inuse}}.
There are some other AWB features (such as Stub detection) dependent on a specific template.
The name of such a template, however, is different from language to language.
When I use AWB in huwiki, how can I change the name of the English template to the name of the corresponding Hungarian template?
Need help with regexp for detecting/adding parameter to template
Ok, so I'm looking to make a run that adds a standard parameter to all the calls of a template, but don't have even close to the knowledge of regular expressions, nor AWB to figure it out on my own. Obviously, I can easily make my list to run through all the articles that transclude the template, but I don't know how to figure out how to restrict a search for the parameter name to just the template, nor how to add that parameter to the end if it isn't there. So for an example, I want to find out if parameter whee is defined in calls of template {{ugh}}, and if it isn't add it with a default value;
so it would make {{ugh | doh = Homer Simpson | carumba = Bart }} → {{ugh | doh = Homer Simpson | carumba = Bart | whee = Maggie }}
As seen in
this link should do it for you to do exactly what you asked. You will see in this example it won't catch if parameters aren't in the exact correct order or if |doh= isn't the first parameter.
Technical 13 (
talk)
14:33, 21 December 2013 (UTC)
@
Technical 13: Ah, we have different definitions of "exactly what you asked." It appears you're looking at Vanisaac's examples, where I was looking at the requirement to "add that parameter to the end if it isn't there." I hope someone else can provide the answer to the latter, as now I'm curious too!
GoingBatty (
talk)
18:32, 21 December 2013 (UTC)
@
Technical 13: That's much closer, as it ignores the examples that already contain |whee=. But what would the proper replace statement be? {{Ugh$1|whee=Maggie}} doesn't bring back all of the original parameters.
GoingBatty (
talk)
19:09, 21 December 2013 (UTC)
Yeah, the doh doesn't matter. It may or may not be there, I only want whee to be added where it's not ever defined. So {{ugh | doh = Homer Simpson | carumba = Bart | whee = Maggie }} should be left unchanged as well.
VanIsaacWSVexcontribs20:52, 21 December 2013 (UTC)
So I'm not sure about this, but would
Find: \{\{ugh?[^(whee)]?\}\}
Replace: {{Ugh $1 |whee=Maggie}}
work right, or would it skip over an instance of whee and match some other parameter as [^(whee)], or does [^(whee)] match anything that's not an e, h, w, or parenthesis?
VanIsaacWSVexcontribs21:10, 21 December 2013 (UTC)
Thanks - I have some more questions on this, but I'll take it to your talk page, as I've probably taken a tangent from the original question here. Thanks!
GoingBatty (
talk)
22:33, 21 December 2013 (UTC)
@
Xaris333: Not sure which template you want to add, but please be aware that
Template:Talk_header#Usage states "In accordance with
Wikipedia:Talk page guidelines, this template should not be added to otherwise empty talk pages." 07:43, 27 December 2013 (UTC)
Help with an expression related to the {{convert}} template
After a recent B-class article review where a reviewer pointed out that the standard use of the {{convert}} template was inserting non-US spelling into a US-spelling article, I learned that the {{convert}} template has a |sp=us parm that may be added to enforce US spelling of, say, metres→meters and kilometres→kilometers. I'd like to do some searching for articles in the Spaceflight WikiProject, many of which I have added the {{convert}} template to in recent years, and attempt to do some clean up in this area.
Can anyone suggest how I might set up an AWB search to find articles that use the convert template AND have a unit in it that defaults to non-US spelling (like metre, kilometre, etc. FULL LIST is given
here). The convert template usually has extra characters in it so the "{{convert|" text may be separated by some arbitrary number of characters from the "m" or the "km" designation of units inside the convert template. I'm not clear on how to write an expression that will capture these?
(Later on, it might be cool to do a broader search as well for designated US-spelling articles and see where the {{convert}} template is not set for US spelling of the units where it makes a difference, but that is way beyond what I'd like to do first.)
@
N2e: Off the top of my head, here's my suggestion:
Under the List menu, be sure you uncheck "Remove non main-space".
Choose your source as "What transcludes page (all NS)".
In the What embeds field, type "Template:WikiProject Spaceflight" and click Make list. You should now have a list of 7400+ talk pages.
Under the List menu, choose "Convert from talk pages". You should now have a list of 7400+ articles.
On the Options tab, click the Find and Replace box and click Normal settings.
Set up a rule to find ({{convert\|[\d\.]+\|\w?[lLm]\d?\|(.*?))}} and replace it with $1|sp=us}}. This will take care of the units of measure with meters and liters.
Set up another rule to find ({{convert\|[\d\.]+\|(m\/s|km\/h)\|(.*?))}} and replace it with $1|sp=us}}. This will take care of the m/s and km/h.
Set up another rule to find ({{convert\|(.*?)\|sp=us)\|sp=us}} and replace it with $1}}. This will remove any duplicate additions of |sp=us.
Set up another rule to find ({{convert\|(.*?)\|abbr=on)\|sp=us}} and replace it with $1}}. This will remove any additions of |sp=us when abbreviations are on.
For each rule, click the Regex and Enabled boxes, and click OK.
Thank you,
GoingBatty. I'm glad you shared something to get me started on my journey of fixing these. When I have a bit of time in the next few days, I will try that out. I'll let you know how it goes for me when I try it out. In the meantime, as you alluded to, if someone else would care to suggest any other ways to combine the rules, that would be cool. Cheers,
N2e (
talk)
15:41, 28 December 2013 (UTC)
I have tested your set of rules and AWB process; have done a couple of sets of test edits, reviewing them each, on a few tens of articles each time. I think your AWB rules work quite well for this purpose. Thanks again!
N2e (
talk)
04:09, 30 December 2013 (UTC)
replace
I would like to replace:
| status =
| status_system =
| status_ref =
by
| status = ##status##
| status_system = IUCN##cri##
| status_ref = <ref>{{IUCN|ID = ##ID##|taxon =##spe##| assessors =|year =|title = ##spe## | assessment_year = ##Year## | version =##cri##| accessdate = 2-1-2014}}</ref>
with data list in *.txt?
Or can I add more information from a list by using AWB. Thank in advace!--
Cheers! (
talk) 14:34, January 2, 2014 (UTC)
I guess it would run using a C# or VB.NET module. But what I would do personnaly is to create a module in Lua where I put all the datas from the *.txt file, in a Lua table. Then, I create a function to access the data and I would use this function using {{subst:#invoke: in the replacement rules of AWB.
Automatik (
talk)
17:43, 5 January 2014 (UTC)
I am a veteran editor of Wikipedia, but a newbie to AutoWikiBrowser. I've poked around with AWB for a couple of hours time now, and can't seem to figure out how to cleanly and easily do article assessments for a WikiProject on article Talk pages.
I did find out how to right click the list box in the lower left corner of the AWB screen and switch a list of articles to a list Talk pages.
What I want to do is to use AWB to help get a larger percentage of the approx. 7300 articles associated with
spaceflight within the purview of
WikiProject Spaceflight. To start with something simple, I'd like to find the Categories associated with all the myriad spaceflight articles, and get them properly assessed. {{WikiProject Spaceflight |class=Category ... }}
After a couple hours of mucking around in AWB, I've been unable to set up a search that would look at some subset of spaceflight-related "Categories", bring up their Talk pages, and cleanly go through and see which are already set as {{WikiProject Spaceflight |class=Category ... }}, and which are not, while changing those that are not, where I have no idea how spaces or parm order—or |class=Category or |class=Cat or some other allowable variation—might have been used in the existing WikeProject templates on Talk pages.
Thanks, Racklever, that will be helpful as I do more work on assessing these articles. However, that is is not quite what I'm looking to do.
I'm looking to, for example, not just find the existing 509 articles that have already been assessed as |class=Category , but would also like to somehow create a list of ALL of the categories that are subcategories (or included categories if that is the better word for it) underneath Category:Spaceflight, so that I might find out which of those subcats are assessed as being under the purview of WikiProject Spaceflight. I still have not been successful in creating that list of Categories. Thanks.
N2e (
talk)
21:59, 13 December 2013 (UTC)
Deprecated cauthors field does not have the same meaning as author2
Hi, everyone,
I see an editor using AWB has kindly updated templates on an article I watch. One set of tags he updated changed "coauthor" fields to "author2" fields, but those fields have never had the same meaning in the CS1 templates. On the CS1 talk page I have pointed out that there is a use case for the "coauthor" field and that perhaps this field should not be deprecated, as it currently is. I have a lot of bibliographic references for articles I work on for which the "coauthor" field is useful. Whom should I talk to about the structure of the approved fields for bibliographic templates, so it's clear which fields are useful for what? --
WeijiBaikeBianji (
talk,
how I edit)
16:40, 6 December 2013 (UTC)
The renaming of |coauthors= to |author2= is not a function built in to AWB, must have been done by the editor using custom rules. Therefore it's not a matter for AWB, it's one to agree on the CS1 templates discussion pages. Related to this,
Josh Parris posted about
this edit, suggesting that in that instance |coauthors= could simply have been removed as the parameter wasn't in use in the citation. Again though, the renaming is not done by default AWB functionality so is a matter for the CS1 talk pages.
Rjwilmsi19:20, 6 December 2013 (UTC)
Only a good idea if there isn't an already existing |author2-last=, |author-last2=, |last2=, |surname2=, |Author2=, |author2=, or |authors2=; all aliases. Replacing |coauthor= with |author2= when there is an already existing |author2= will hide one of the |author2= parameters from
Module:Citation/CS1 – there won't be an error message, but neither will CS1 render both the original |author2= and the new |author2= as was intended. Remember also that |coauthor= and |coauthors= are aliases of each other so there is no guarantee that the singular |coauthor= contains only one coauthor and, similarly, that |coauthors= always contains more than one coauthor.
It may be that this task is more than AWB is capable of handling properly, though, perhaps some clever use of the advanced search and replace functionality (which is so inadequately documented that I haven't got a clue how to use it – examples of it in use would be appreciated).
@
Trappist the monk: - Good point - I won't make changes to
WP:AWB/RTP for |coauthor=. I do have some find and replace rules that work well for up to three coauthors, but require manual review of the results before saving:
Be sure you check the Regex, After fixes, and Enabled boxes are checked.
Since I have the normal F&R settings set to Add replacements to the edit summary, I use the advanced settings for those changes I don't want to appear in the edit summary. I presume you've already read
Wikipedia:AutoWikiBrowser/Find and replace. If you could please provide a specific example of something you're trying to do with the advanced settings, hopefully someone here can help you out. Thanks!
GoingBatty (
talk)
18:54, 9 December 2013 (UTC)
At the moment I'm not trying to do anything "advanced" and since I've just started using AWB, the things I have been doing have been adequately met by the normal find and replace. Not too long ago I read somewhere that the best way to convey new concepts to people is to illustrate by example first, then explain what was done in the example: look, I can make this text green and I did it by enclosing the text in the {{
green}} template like this: {{green|make this text green}}.
I'm trying use AWB with a proxy.pac. I can access the wiki via IE and Chrome with no issue. My understanding is that AWB pulls proxy settings from IE, so if IE works then so should AWB.
This however is not the case; I, like others receive the following 407 error:
Network access error
---------------------------
The remote server returned an error: (407) Proxy Authentication Required.
My main question is what is the true status of this issue. Has it been fixed/implemented or not?
If not, someone seemed to
referencespeak to a possible workaround by hard-coding the authentication into the ApiEdit.cs class, but it was said one shouldn't do this.
All these conversations are a year old so does anyone know of any additional information or a "safe" workaround.
Or simply if currently it can not be done. If that's the case I would request it be added to the FAQs (until feature is added) due it being quite foundational to whether or not a person is able to utilize the software.
Thanks much,
--
Almack64 (
talk)
12:40, 5 November 2013 (UTC)
Same problem here, plus I get a bug report.
Ironically, it was working fine on XP, but does not work on Windows 7 Pro.
Object reference not set to an instance of an object.
Call stack:
at AutoWikiBrowser.MainForm.SetProject(String code, ProjectEnum project, String customProject, String protocol)
at AutoWikiBrowser.MainForm.LoadPrefs(UserPrefs p)
at AutoWikiBrowser.MainForm.LoadPrefs()
at AutoWikiBrowser.MainForm.MainForm_Load(Object sender, EventArgs e)
System.NullReferenceException: Object reference not set to an instance of an object.
at AutoWikiBrowser.MainForm.SetProject(String code, ProjectEnum project, String customProject, String protocol)
at AutoWikiBrowser.MainForm.LoadPrefs(UserPrefs p)
at AutoWikiBrowser.MainForm.LoadPrefs()
at AutoWikiBrowser.MainForm.MainForm_Load(Object sender, EventArgs e)
Why AWB still not have the Proxy Settings, as
pac, or I don't know the way? I find AWB has to use HTTPS, but HTTPS is blocked in China. So, if AWB's proxy is not available, it will be too difficult to use AWB for Chinese users, or even impossible to use. How we can do?
乌拉跨氪 (
talk)
06:04, 16 December 2013 (UTC)
Please see answer to "Is there any way to set AWB to not use https? (GFW blocks 443 port)" in the FAQ section above. —
Ganeshk (
talk)
12:35, 16 December 2013 (UTC)
In terms of now program status, I can't open the preferences becouse of an network access error about "WikiFunctions.Parse.SiteMatrix".
乌拉跨氪 (
talk)
14:08, 16 December 2013 (UTC)
What am I doing wrong?
In
Citation Style 1, the parameters |day= and |month= are deprecated. I have a relatively
simple script that replaces these parameters (when they are adjacent to each other and also adjacent to |year=) with |date=. In many cases editors have used |date= instead of |day= so the script also checks for combinations of |date= with |month= and |year=. When these groupings of parameters are found, the script is supposed to take the values from the three individual parameters and arrange them in dmy format as the value of a new |date= parameter at the end of the CS1 template.
{{cite web|last=White|first=Rea|title=Road course races challenge the specialsts|url=http://msn.foxsports.com/nascar/story/road-course-races-measure-competitive-ability-061710|publisher=Fox Sports|accessdate=June 18, 2012|date=18|month=June|year=2010}}
{{cite web|title=Passing is tough in Chute|url=http://chronicle.augusta.com/stories/1998/06/28/oth_232051.shtml|work=[[The Augusta Chronicle]]|accessdate=June 18, 2012|date=28|month=June|year=1998}}
should produce this:
{{cite web|last=White|first=Rea|title=Road course races challenge the specialsts|url=http://msn.foxsports.com/nascar/story/road-course-races-measure-competitive-ability-061710|publisher=Fox Sports|accessdate=June 18, 2012|date=18 June 2010}}
{{cite web|title=Passing is tough in Chute|url=http://chronicle.augusta.com/stories/1998/06/28/oth_232051.shtml|work=[[The Augusta Chronicle]]|accessdate=June 18, 2012|date=28 June 1998}}
which the Regex tester does using this normal find and replace rule set:
But, when AWB edited
2012 Toyota/Save Mart 350 it produced these conflicting results (the rest of the citations left out for clarity):
first: |date=18 |date=June 2010
second: |date=28 June 1998
So, what's going on here? Regex tester produces correct results but AWB in supervised trial doesn't. It isn't just this page, I've seen other pages do the same thing. It "feels" like AWB isn't starting at the top of the rule list as it should do – yeah, I know, feelings aren't reliable ...
Anyone out there who can see what it is that I'm doing wrong (there is a greater likelihood that I am the source of the problem here than it being something fundamentally wrong with AWB).
I tried this with ONE$1x TWO$2x THREE$3x FOUR$4x FIVE$5x as the "replace" string. This reveals that the $1 text is matching far more than you intended. You need to add a question mark near the beginning so that the ".*" is less greedy:
Thank you both. The not so greedy .*? seems to do the trick. The ONE$1x TWO$2x THREE$3x FOUR$4x FIVE$5x is really handy; glad you suggested it. It caused me think that there might be a really useful enhancement to AWB. Imagine a test or debug mode (there just happens to be room for one more tab right next to Start) where you could test your regexes against real articles. Imagine further, that the left side of the main window, instead of showing the text as it is, showed the regex matches. Imagine that on the right side of the main window where it now shows prospective changes, you could see either all of the captures, neatly ordered, or the replacement, or maybe both. Such a tool might eliminate the need for regex tester which as I have learned here doesn't always tell the truth. In this mode, saving of files would be disabled but otherwise you could skip from page to page looking for problems in your code. I suppose that this has already been thought of and for reasonable and proper reason discarded but perhaps not. Developers?
The literal string is causing AWB to die when loading settings.
Specifically the error message is:
Unexpected node type EntityReference. ReadElementString method can only be called on elements with simple or empty content. Line 150, position 97.
That location is the n following the ampersand in this portion of the regex find: (?:[-/–]|–)
If I change the regex to (?:[-/–]|.ndash;) then the settings file will load and the code works as it should (except for the likely rare cases when the position normally occupied by the ampersand is something else.
If I change the regex to (?:[-/–]|&ndash;) then the settings file also loads. This does not work in Regex Tester so I expect that it won't work in the wild.
So, my question is: How do I convince AWB that I want to match an ampersand followed by the letters n b s p and finally a semicolon? — Preceding
unsigned comment added by
Trappist the monk (
talk •
contribs) 14:02, December 22, 2013 (UTC)
@
Trappist the monk: I don't receive any errors when I load this into AWB 5.5.2.3 SVN 9803. Are you receiving an error on every page, or just certain pages? Could you please confirm if your error with "nbsp" or "ndash"? Thanks!
GoingBatty (
talk)
16:34, 22 December 2013 (UTC)
@
Trappist the monk: I deleted each of your rules that contained and the settings file loaded fine. I then manually added the rule above in the Find and Replace window, and saved the settings file. I closed AWB and opened it, and then could load the new settings just fine. I then noticed that although the Find and Replace window shows the settings file contains &nbsp;. So if you replace with &nbsp; in your settings file I think it will load.
GoingBatty (
talk)
17:24, 22 December 2013 (UTC)
I have just come to the same conclusion by a slightly different path. I changed the four s to .nbsp; in Notepad++; saved the file, and then loaded the new settings file into AWB. In Find & Replace I edited the .nbsp; back to and saved the settings. Yep, is translated to &nbsp;.
Is this a bug in AWB? If not, is there documentation that explains this? Where? Is the only html entity that has this problem?
@
Trappist the monk: Doing further testing, it seems that each instance of & is stored in the settings file as &. It doesn't appear to be a bug, but the way AWB was designed. I don't see anything in the documentation that discusses viewing/editing the settings file outside AWB. Feel free to be bold and update the documentation as you see fit.
GoingBatty (
talk)
18:11, 22 December 2013 (UTC)
Thanks for that. I don't mind doing the bold thing when I think I know what I'm doing. In this case, though I trust you and believe that you are correct, I don't know enough about the inner workings of AWB to write with confidence. Perhaps over time I will.
As further proof that I don't know enough to contemplate tweaking the documentation, I haven't been able to describe this problem correctly. I working on a script that concatenates |day= or |date=, and |month= and |year= into |date= to eliminate |month= which is deprecated. The issue that started all of this was this pair of parameters: |year=2006|month=July–August which contain: &ndash. Argh!
Even when Editor GoingBatty inasmuch tells me that I've got it wrong, I get it wrong. I'm not always an idiot, just sometimes.
I am rather new to AWB and use it currently on a Wikia page. Is there a way to use AWB to rename pages or files (e.g., rename all files from Afoobar to Bfoobar)? I cannot find anything on how to do this. My apologies if this has been addressed already or is a simple oversight. Thank you.
EvergreenFir (
talk)
04:37, 27 December 2013 (UTC)
@
GoingBatty: - Thank you for your reply! I am on the Wikia page I am using it on. I saw the "move" button on the start menu, but had to individually enter the name for each file. Was wondering if there was a way to do a mass move of sorts. I was changing all images starting with Short7 to Short8.
EvergreenFir (
talk)
04:50, 27 December 2013 (UTC)
Similar question for @
GoingBatty: or anyone: anyway to get AWB to delete stuff in, say, a certain category automatically? I see the delete button on the Start panel, but not sure if I can make it do it without prompting.
EvergreenFir (
talk)
05:30, 29 December 2013 (UTC)
For what its worth I am also an admin on a couple of Wikia sites and the delete and move buttons don't even work for me there at all. I hadn't thought to ask for a change yet but I assumed it was Wikipedia specific functionality.
Kumioko (
talk)
15:39, 29 December 2013 (UTC)
What's your status on the Wiki in question? Are you just an "admin" or are you a "founder"? That might be the difference.
Kumioko (
talk)
21:54, 29 December 2013 (UTC)
I'm an admin/crat. Again, I can move things, but I can't set it in "bot" mode to do it. I have to do each individually. Wondering if there's a way to make it automated.
EvergreenFir (
talk)
00:05, 30 December 2013 (UTC)
Oh ok, I'll play with it and see if I can get it to work on one of the wiki's I have admin rights on. Maybe in the meantime Magio or one of the other devs might have some advice or be able to help.
Kumioko (
talk)
00:12, 30 December 2013 (UTC)
@
Magioladitis:. Ok I see what I did wrong and I was able to test the functionality. I see what you mean about it being a pain. There are also some other things I notice that I think is suboptimal for a non Wikipedia site:
Deletion messages assume the summary is for Wikipedia. Recommend just giving a couple generic ones for Wikia/Non Wikipedia or allow an easier way to modify the provided listing.
It might be useful to also add a drop down for Reason group: (Vandalism, Housekeeping, Marked for deletion, etc.) before the detailed deletion message.
The delete article list doesn't contain the newly created G13. Abandoned Articles for creation submissions so you might want to add that one for Wikipedia use.
It might be useful to add a block/unblock capability too. Its fairly rare but there is sometimes a need to block a group of usernames or IP's for various reasons.
I could probably craft an AWB module to do the deletion but it will take me a few days and I wouldn't want to post the code here in the clear. I hope this helps.
Kumioko (
talk)
01:10, 30 December 2013 (UTC)
Can someone confirm to me that plugin CSV Loader for AWB, currently works fine, even without bot flag? For me - not working. --
XXN (
talk)
13:24, 28 December 2013 (UTC)
Today's scan identified
Wilhelm Rath with a PUA located at: {PUA} Publisher: Mellinger (1979) Langua". Batty, I'm willing to walk you through on a very high cliff. :)
Bgwhite (
talk)
05:41, 23 January 2014 (UTC)
This is probably something I'll be ashamed for asking, but I'm trying to remove links to a deleted standalone list article. I created the list based on "what links here" but don't see an easy way to go through them and remove all of the links. The links likely are all "See Also"s, but perhaps it's prudent to have a separate search/replace (presuming that's the best way) for isolated links like in "See Also" and another search/replace that just delinks instances of the link in the article text? I'm a regex noob, so this might be really easy... --— Rhododendritestalk |
16:21, 31 December 2013 (UTC)
@
Rhododendrites: Wich article is it? I have a script installed (I don't know what cod is for that script, sorry...) but I can unlink them for you...(if it still works). -(
t)
Josve05a (
c) 16:27, 31 December 2013 (UTC)
Comment: Red links in 'see also' sections need to be removed outright with the entire line, including the opening asterisk and the terminal line feed. -- Ohc ¡digame!04:51, 3 January 2014 (UTC)
I last used AWB on January 3rd on my own wiki (
http://absitomen.com/lexicon), I upgraded to 1.21.4 and added semantic mediawiki but other than that no changes. When I try to use AWB now, it will not connect. It gives an Error Connecting to Wiki and Error Description: The given key was not present in the dictionary.
I'm using AWB on Wikia (I'm an admin/crat on the page and the bot is registered, so no issues there) and I was wondering if there's a way to move all files or pages in a given list to a new title. For example, we have images that start with "Short8" and I want to move them all to "Short13" (e.g., Short8 foobar.jpg would become Short13 foobar.jpg). I know this can be done individually using the "Move" button on the "Start" panel. But can I get it to do it automatically? Is this possible?
How do I pull in articles from the Special lists to AWB
Does anyone know how to pull in the articles from the special lists to AWB? The only way I ave found to do it is to copy the list of articles to notepad and then remove the non article content. To clarify I am referring to the Special pages reports like Dead End pages, Wanted pages and the like.
Reguyla (
talk)
23:21, 25 January 2014 (UTC)
Thank you I will do that. Your also right that categories exist for some of these but not every wiki created categories for them and not all of them are cateforizable (such as wanted categories and templates).
Reguyla (
talk)
17:09, 27 January 2014 (UTC)
How does AWB make a list of redlinks?
One of the choices in the list maker is "Links on page (only redlinks)".
I'm writing a script in which I need to determine redlink status.
How does AWB determine what the redlinks are on a page?
I can't remember but try looking up missing or filter. I think its something like missing pages. I'll try and look it up later if you can't find it just let me know.
138.162.8.59 (
talk)
14:42, 3 February 2014 (UTC)
I've seen this answered before, people suggesting using "\r\n" to remove the new line. However, on Wikia with the AutoWikiBrowser, using "\r\n" doesn't work as far as I can tell. Plus, removing "\n" causes "\r" to be printed after every single letter, turning things like "|name = %%pagename%%" into "\r|\rn\ra\rm\re\r \r=\r \r%\r%\rp\ra\rg\re\rn\ra\rm\re\r%\r%\r".
I'm trying to wrap my head around how to use the "Advanced Settings" find and replace (making new "rules") but I can't figure it out. Basically, I'm trying to remove "|name = %%pagename%%" from articles that use a template that pulls the page name automatically (I only include the name parameter for the template documentation, but editors keep tossing in the name parameter into new articles) without having it leave a blank line. Any help would be greatly appreciated.
173.238.166.40 (
talk)
01:22, 4 February 2014 (UTC)
Try finding \|\s*name\s*=\s*%%pagename%%\s* with the replace value blank, and the Regex button checked. This works for me on
Xolnar without SingleLine checked. 02:10, 5 February 2014 (UTC) — Preceding
unsigned comment added by
GoingBatty (
talk •
contribs)
Citations wrongly reordered
This automated system improperly reorders citations; in an article, citation A might appear, and then citation B; later in the article both A and B contributed to the content of a paragraph or so. But if B contributed most of it and A only some minor detail, the citation on that paragraph needs to be B, and then A.
This system re-orders them A and then B. There is no reason for this. A discussion above about re-ordering references, which seems to be the same issue, is marked "fixed". It isn't. It still does it. And it's very annoying.
Afterbrunel (
talk)
18:34, 6 February 2014 (UTC)
It's also a violation of
WP:CITEVAR - if an editor has established a style in which adjacent references are not always sorted in increasing order (for example, by intentionally placing them out of order), that style must be respected. This is not something that can be decided by the authors of AWB; it is an issue for WP:CITE.
Users who run AWB manually should make sure that it is appropriate to rearrange references before saving AWB edits that rearrange them. Because bots cannot make that determination, they need to disable the reordering functionality. This bug has persisted in AWB for too long at this point, so I am likely to simply block bots that continue to violate WP:CITEVAR in this way, if their maintainers have been notified and have not fixed the problem by disabling the re-ordering code. — Carl (
CBM ·
talk)
20:29, 6 February 2014 (UTC)
As I mentioned on the bots talk page, why not just add {{bots|deny=AWB}} and call it a day. This isn't even a bug as you mention its a feature. It follows the Citevar so if you want it changed, change that, then AWB will have to change. Aside from that, this isn't an issue and to me if you were to use your admin tools to block it as you suggested on the bots talk page that would constitute abuse of the tools. No matter how good the bot or botop is it will never be 100%, there are always going to be exceptions and that's what this is, an exception. Rather than forumshop in the future may I suggest picking one location to have the discussion?
138.162.8.58 (
talk) 20:30, 6 February 2014
I'd used the {{bots|deny=AWB}} sparingly. It should be used when AWB does a serious problem or suggests a serious problem. From
Template:bots, it says "Avoid using the template as a blunt instrument". I've seen it used because an editor was too lazy or didn't know what the root problem was. I've seen it used to keep an editor's view of how many blank spaces there should be. I'm not sure what to do in this situation. I have my personal feelings, but I'm not sure it is the right way.
CBM, please keep the threats to block or threats to retaliate out of the discussion. Saying you are going to block anyone who doesn't abide your view is not to doing the discussion any good.
Personally, I don't see it violating
WP:CITEVAR as it is not change the citation styles. It is also used as best practices at FA. I don't know this history behind this, Carl or anybody, do you know of discussions where this issue was decided?
Bgwhite (
talk)
22:16, 6 February 2014 (UTC)
@Bgwhite: the complaint is exactly that it does change the citation style - the OP above intentionally put the citations in one style, and AWB changed them. I believe that the re-ordering code was randomly added by AWB maintainers without consideration of whether the manual of style actually supported it. And I want to be clear that I am not speaking about blocking editors here; it is bot accounts (which are not editors, nor even human) that I am speaking about. I want to emphasize that the AWB talk page is not the place where the manual of style is decided, and that bot operators who use AWB as the starting point for their bot are still responsible for configuring and editing AWB as necessary to meet the bot policies. If AWB does something which bots are not permitted to do, then bot operators must disable that feature of AWB. Stonewalling complaints by saying "that's just what AWB does" is not acceptable behavior from a bot operator. — Carl (
CBM ·
talk)
22:26, 6 February 2014 (UTC)
I do however note, that editors running AWB on their main account do the same type of edits - they also should turn it off after being notified. --
Dirk BeetstraTC11:41, 11 February 2014 (UTC)
@
Beetstra: The AWB user interface provides only one checkbox for the entire set of
general fixes; they cannot be turned on and off separately. I assume that bot owners know how to download the source code and build their own version of the software. --
John of Reading (
talk)
12:02, 11 February 2014 (UTC)
That does not necessarily make it right and it is by no means an excuse. The more reason to disable it at all in AWB itself, ánd (for the time being, on a user account) be careful and not save it when that 'fix' is performed - you may be overriding an editorial choice. One should not be using automated tools, either on a bot account or on a personal account to perform actions which do not have consensus. --
Dirk BeetstraTC12:38, 11 February 2014 (UTC)
@
John of Reading: I would bet many AWB bot owners, and surely many AWB manually users, would not know how to pick and choose which general fixes to use. I think the way to do it would be to uncheck the Apply general fixes box and load a custom module instead. @
Magioladitis: Is
this custom module up to date with all of the general fixes procedures? Thanks!
GoingBatty (
talk)
04:00, 16 February 2014 (UTC)
I was involved in a discussion about this several years ago. I thought the arguments against ordering references were unconvincing, and I never found a style guide that suggested placing important references first. I was pointed at what was purported to be an example of important-first referencing, but it was in a (to me) obscure and hard to get journal and I couldn't really get to the bottom of what was going on. I don't think threatening accounts with blocking is constructive.
Mr Stephen (
talk)
22:45, 6 February 2014 (UTC)
@
CBM: What would be the best way for editors who intentionally want references out of numerical order to note it in an article's comments so that future human editors (using AWB or not) would know that the order was intentional? I don't think I've ever seen anyone do this. Thanks!
GoingBatty (
talk)
04:11, 16 February 2014 (UTC)
I do agree with Mr Stephen .. however: Can this please be turned off - a bot should not be doing this automatically. What order the references should have is an editorial choice, and edits like [Can this feature please be standard disabled in AWB - I have just undone
this edit as, to me, this is practically a null-edit overriding what may very well have been an informed choice, as the problem that this] automatically override what may very well have been an informed choice of using the ordering (I agree, it most often is not). The problem is, that if it is a deliberate choice to order references in a certain order (as what
Afterbrunel is alluding to, and what I feel might be the case in
this edit, where the first source was way older, and probably 'closer' to the statement than the second which is of a much later date), then that is not to be automatically overridden.
Moreover, doing only this is practically a null-edit, edits like the one I just referenced should not be saved as they clutter watchlists. --
Dirk BeetstraTC11:38, 11 February 2014 (UTC)
Kerning for title and chapter leading and trailing single or double quote marks for quoted titles; (see
Quote within title parameter)
Enhance month / season year date range to check for proper left to right time sequencing of month or seasons in the range; (see
Month / season range order validation)
I have downloaded AWB to use on another wiki (
This one) after entering the line www.uesp.net/wiki/ into Site → Project → Custom box, I try to input my log in details, using the correct account details (I double checked them before posting here) it gives the error message: "The username provided doesn't exist." The username I entered was KizBOT and this account
exists according to the wiki and I can manually log in through the wiki to such an account. I can not find a relevant help doc as well relevant documentation is on Wikipedia and does not help those trying to make use of the program elsewhere. --
KizC (
talk)
15:56, 13 February 2014 (UTC)
Additional: Trying to login using the correct settings file (Default.xml) results in an additional error message: "Invalid URI: The hostname could not be parsed." Trying with just Kiz, an account that exists on both
uesp (mine) and
wikipedia (not mine) however gets the message: "The password provided is incorrect." - which means its reading Wikipedia's data rather than UESP's? Additionally, using KizC (which
exists on wikipedia alone (mine)) gives me neither a password or username error, just the Invalid URI one. --
KizC (
talk)
16:09, 13 February 2014 (UTC)
Is there a way to prevent a regex match when a specific text string exists anywhere within a template?
I have a script that replaces |coauthor= or |coauthors= with several |authorn= parameters within various
Citation Style 1 templates. The script is
here.
The script needs the ability to ignore |ref=harv because replacing |coauthor= with |authorn= will break the CITEREF link when {{
sfn}} or {{
harv}} (and family) templates are used with the CS1 template.
Examples:
The script should ignore this citation because it contains |ref=harv:
{{cite book |title=Big Book of Facts |last=Knowitall |first=Mister |coauthor=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing |ref=harv}}
The script should change this citation:
{{cite book |title=Big Book of Facts |last=Knowitall |first=Mister |coauthor=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing}}
to:
{{cite book |title=Big Book of Facts |last=Knowitall |first=Mister |author2=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing}}
because it does not contain |ref=harv.
Can this be done? Regex seems to do well when you want to match something that exists; it doesn't do so well if you want to match something that doesn't exist. I've tried look-ahead and look-behind without success which could mean that they won't work for this application or that I just don't know how to properly use them. It doesn't seem that the advanced search is appropriate because the if tab apparently applies to the whole page which is insufficiently specific for this application. I don't know about doing this task as a module.
As you said, regular expressions are much better at finding the existence of things rather than non-existence. I generally use a solution to this type of problem which is generic to situations where your primary method of change is regex replacement. That method is to first match the more specific case and change it to something that will not match the general case then change the general case. After that replacement then change the specific case back to the way it was prior to the initial change.
In this instance the contents of the page would look like the following (all done within one AWB change-set, but the regex replacements in AWB are done in sequential order):
Initial page contents:
{{cite book |title=Big Book of Facts |last=Knowitall |first=Mister |coauthor=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing |ref=harv}}
{{cite book |title=Big Book of Facts |last=Knowitall |first=Mister |coauthor=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing}}
Contents after first regex replacement:
{{cite book |title=Big Book of Facts |last=Knowitall |first=Mister |coau,BoGuSNonMaTcHiNgStRiNg,thor=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing |ref=harv}}
{{cite book |title=Big Book of Facts |last=Knowitall |first=Mister |coauthor=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing}}
Contents after primary desired change by regex replacement:
{{cite book |title=Big Book of Facts |last=Knowitall |first=Mister |coau,BoGuSNonMaTcHiNgStRiNg,thor=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing |ref=harv}}
{{cite book |title=Big Book of Facts |last=Knowitall |first=Mister |author2=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing}}
Contents after restoring original content of cases we do not want to change via the desired general regex replacement:
{{cite book |title=Big Book of Facts |last=Knowitall |first=Mister |coauthor=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing |ref=harv}}
{{cite book |title=Big Book of Facts |last=Knowitall |first=Mister |author2=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing}}
The string ",BoGuSNonMaTcHiNgStRiNg," should be selected as one you know does not exist in your input data. This may require you to create a very long string. Better yet is to first search for the string you plan to use as a unique marker. If it is found either skip the page, use a different string, or let a human deal with it.
As mentioned, this is a generic methodology which can be used in situations where your primary method of change is regex replacement. It can be used to solve issues where you have specific cases which are not to be changed – or changed in a different manner – where the specific case is difficult to exclude from a match with a more general case you do desire to change. It is not, necessarily, the easiest method to use if there are methods of differentiation available other than regular expressions.
In this specific instance, it will probably be easier to use more than one regex pre-replacement action as it will likely be easier to match the cases where "\|\s*ref\s*=\s*harv\s*[^}\|]" occurs prior to "coauthors" separately from those where the "\|\s*ref\s*=\s*harv\s*[^}\|]" occurs after "coauthors" —
Makyen (
talk)
22:09, 1 March 2014 (UTC)
Wow, I did not know that that is possible. How is it actually implemented? Can you show me a script where your BoGuSNonMaTcHiNgStRiNg technique is used so I can more clearly understand how it works within AWB?
I did get a version of look-behind to work when looking for |ref=harv at the end of the template where, it seems, most editors put it: (\s*\|[^}]*(?<!\s*\|\s*ref\s*=\s*harv\s*)}})
An example I actually used in AWB (note: the regexes used below are non-optimum. If I had known I was going to use it as an example I would have put more effort into making it elegant instead of just functional 8-):
All entries were only in the "Find & Replace" dialog. One Regex replacement per line; all within the same dialog. From the AWB user's perspective it is just one change to the page. From the program's perspective each Find & Replace entry is a separate action with the output of the prior action feeding the input to the next action. The goal of this change set was to harmonize the use of "×" and "x", as used for "by", in describing an array. The issue was that multiple formats were used throughout the article and within section headings. I had already dropped in anchors for the current names of the section headings and I did not want those to be changed because I knew there were links to various sections. Thus, to protect the anchors the first action was to change all matching occurrences within an {{anchor}} to something that would not match in the general replacement:
(\{\{[Cc]ite[^}]*\|\s*)coauthor(\s*=[^}]*\}\}) → $1author2$2 ;change {{cite...|coauthor=...}} to {{cite...|author2=...}}
coBoGuSNonMaTcHiNgStRiNg1author → coauthor ;Put the coauthors not being changed back.
I did not test the whole thing above against a page, but I tested each individual replacement. It should result in changing:
extra coauthor text.{{cite book |title=Big Book of Facts |last=Knowitall |first=Mister |coauthor=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing |ref=harv}} extra coauthor text
extra coauthor text.{{cite book |title=Big Book of Facts |last=Knowitall |first=Mister |coauthor=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing}} extra coauthor text
extra coauthor text.{{cite book |title=Big Book of Facts |ref=harv |last=Knowitall |first=Mister |coauthor=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing}} extra coauthor text
to:
extra coauthor text.{{cite book |title=Big Book of Facts |last=Knowitall |first=Mister |coauthor=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing |ref=harv}} extra coauthor text
extra coauthor text.{{cite book |title=Big Book of Facts |last=Knowitall |first=Mister |author2=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing}} extra coauthor text
extra coauthor text.{{cite book |title=Big Book of Facts |ref=harv |last=Knowitall |first=Mister |coauthor=Squirrel, Rocket J. |publisher=Bullwinkle Industrial Publishing}} extra coauthor text
Excellent! This works. I did tweak them a bit so that a citation with |ref=harv and an empty |coauthor= or |coauthors= would not be modified so that the find rule could remove these empty parameters.
It is pretty difficult to write a set of regexes that can handle all of the scenarios for this sort of task: what's required (and simplest to write when you know how) is a procedure in a custom module. There are a number of functions I've written within WikiFunctions in AWB for template manipulation.
Rjwilmsi09:42, 4 March 2014 (UTC)
I suspect that you are correct. But, documentation or even examples of working custom modules that illustrate the basic how-to-use-custom-modules don't seem to exist. Do they? If so, where?
Category:Pages containing cite templates with deprecated parameters has some 100,000 pages that need some sort of repair. Tools are needed to shrink the category's size. This script is my second tool and it is only getting the low hanging fruit.
@
Rjwilmsi:As is usually the case, there are multiple ways to solve the problem of changing the parameter name as was desired. The specific question was about regexes. The methodology which I described is just one more tool to have available from which you can choose. It uses a few ideas which are good to know. For example: A)that when manipulating data you usually have the opportunity to use multiple steps. This allows breaking a problem down into smaller ones which are easier to solve. This is something we all forget to do, or don't think about doing sometimes. B)that every change you make to the data does not have get the data closer to the desired end product. C)that you don't have to keep the data looking perfect in every step of the process as long as you have a plan that lets you end up with data that is perfect. D)that under most situations if you are processing data, that data is yours to manipulate in the manner that makes it easiest for you to perform the task. If that means the data is not recognizable while you are processing it, that is not necessarily bad. What matters is where the data starts and where the data ends.
Custom modules, and the ability to pass data back and forth to another program, provide the ability to use whatever programming environment you desire to manipulate the page data. If there was already code available that parsed the wikitext and only called my function/method once for each template providing my function with an appropriate data structure containing parsed template parameters and then reassembled the template from the changed data structure I pass back to it, I might consider using it for the specific change which was requested. The choice would depend on multiple other factors. For example: Am I thinking that my changes might become more complex? How many pages am I expecting to manipulate? How much time is available which could be spent on coding vs. making changes? Are there other similar future problems which I expect to need to solve over which I can recoup the investment of coding a more general case solution? Are these changes attended, or unattended? Those are just a few of the things to consider when making a choice as to the methodology for solving the problem.
Even given that I already have a good portion of basic template parameter parsing written and functional (but in JavaScript at the moment), there are a considerable number of problem spaces which I would solve in the manner described above by using the regex capabilities available in AWB. [NOTE: there is a full?
wikitext parser available in Python, but I have not looked at it].
For the question asked, and assuming a relative small number of pages, I probably would have solved it using regexes. For the problem now mentioned of 100k pages with a variety of different issues, I would certainly use one programming language or another. Given the integration of C# / .NET into AWB via Custom Modules, that is potentially the best method.
Does anyone have a template parser already written as an AWB Custom Module? I know that when I was looking for such a parser a while ago, I did not find one. It is possible that I missed it.
@
GoingBatty:I have looked at the page
Wikipedia:AutoWikiBrowser/Custom Modules multiple times in the past. The primary problem with it is that it does not actually say how to program an AWB Custom Module. How one does so has to be inferred from the examples provided. The examples imply that a function named ProcessArticle with specific arguments is called by AWB, but it does not actually say "this is how you do it", or "this is the support AWB is providing". In other words, there is no description of the API of how your code is called, or any hooks that are actually available which you might call. If I had actually had cause to program an AWB Custom Module, I would have added text to that page describing how it was done. However, I have not, yet, had a need to program a AWB Custom Module. Thus, I have not felt qualified to write a description of how to do it. —
Makyen (
talk)
03:47, 5 March 2014 (UTC)
Customisation for the Chinese Wikipedia
The "autotagging" and "minor fixes" features are very useful in general, but because one or two of the rules will mess things up for the Chinese Wikipedia, I have to give it up entirely. Those include:
Done Adding a space between </ref> and the next non-space character (there are no spaces in Chinese, even after the reference tag)
Done Using the "date=" parameter rather than "time=" on maintenance tags
Done Replacing [[分類:XXX]] with [[Category:XXX]]
Is there anyway I can customise these rules, either natively, or through a plugin? Is it possible to add a more customisable menu in any future updates?
Yinweichen (
talk)
17:24, 2 March 2014 (UTC)
Yes to your question. (This is not really a problem, but the first two make the whole feature unusable.) I am using the latest version (5523). Is there such a menu, or a customisation file I can edit? One more problem: the bottom-right edit-box uses glyph width for the Latin alphabet, which makes Chinese characters clog and overlap together.
Yinweichen (
talk)
20:25, 2 March 2014 (UTC)
Yinweichen In Chinese Wikipedia (zh.wiki) 分類 is a redirect for "Category". Check
this for instance. The name used for category namespace in Chinise Wikipedia matches the English one. This is not the way the other wikipedias do it but still. --
Magioladitis (
talk)
Magioladitis I am not aware of the settings on the other zh variants. I presume they generally follow the zh Wikipedia. The trouble with timestamps is that time= supports formats "2014-03-03T21:54:14+00:00" and "2014-03-03", whereas date= only supports the Chinese year+month format "2014年3月". For new tags, the date= parameter is wrong. When fixing existing tags, the AWB tool does not recognise the time= parameter that's there, and adds the wrongly formatted date= parameter.
Yinweichen (
talk)
21:59, 3 March 2014 (UTC)
Magioladitis For {{unreferenced}}, AWB uses "time=", yet gives the Chinese date format, which should only be used for "date=". The documentation recommends the format: time={{subst:#time:c}} for most maintenance templates. Also, I've noticed that AWB tends to replace 2014-03-08T18:48:09+00:00 with 2014-03-08t18:48:09+00:00 (with a lower-case "t"). This seems to be a redundant feature.
Yinweichen (
talk)
18:49, 8 March 2014 (UTC)
This is the third issue: spaces are added after </ref>. The convention for Chinese Wikipedia is not to put a space before/after the reference tags.
Yinweichen (
talk)
20:53, 4 March 2014 (UTC)
Also please remove spacing around external links, as the text inside and outside of the link should be directly connected
Yinweichen (
talk)
22:14, 8 March 2014 (UTC)
The bottom-right edit-box uses glyph width for the Latin alphabet, which makes the wider Chinese characters clog and overlap together.
Yinweichen (
talk)
04:11, 6 March 2014 (UTC)
Yinweichen the date/time issue will be fixed in the next release. I already added the new piece of code. You can always compile your own version of AWB and use the new changes immediately. For the purpose of "Template redirects" please read the instructions given in the page itself. --
Magioladitis (
talk)
07:42, 5 March 2014 (UTC)
Which genfixes should be implemented for the Draft namespace? I suggest that any genfixes that work for mainspace be implemented for Draftspace, but would be interested in other people's opinions. Thanks!
GoingBatty (
talk)
02:36, 21 December 2013 (UTC)
Saving preferences
Hi,
When I save preference by clicking on the "Save settings" option of the "File" menu, I've some text which is saved with all my regex rules. This text is within <remove_></remove_> tags and contains a giant list of words. This makes it difficult to load preferences. Do you know where is from all this text whithin <remove_> tags? How avoid it? Thanks by advance for your answers,
Automatik (
talk)
02:47, 27 December 2013 (UTC)
I added some test words to my settings file and loaded it, but was unable to see what effect they had. I suggest you open the settings file using a text editor such as
Notepad++ and carefully remove the list of words, leaving only the <remove_></remove_> tags. Keep a backup! --
John of Reading (
talk)
08:16, 27 December 2013 (UTC)
Related to this, I have
proposed that the "suggestions may be available" link be changed so that it prompts the Find link tool to look for links to the article (instead of links to the talk page). I've updated
Template:Orphan/sandbox for people to review. Thanks!
GoingBatty (
talk)
05:30, 21 December 2013 (UTC)
@
Magioladitis: Would this also mean that we would stop orphan-untagging with AWB, or change AWB so that orphan-untagging could also happen on talk pages, or something else?
GoingBatty (
talk)
15:36, 21 December 2013 (UTC)
@
Magioladitis: OK, then I would request that you please don't remove the orphan-untagging from articlespace functionality until all the templates have been moved to talkspace. Thanks!
GoingBatty (
talk)
16:07, 21 December 2013 (UTC)
@
GoingBatty: to be honest, if I had participated to the discussion I would oppose the move. Tags on the talk for something happening to the article is a disaster. Everytime, we had something similar to the past it resulted to out-dated un-maintained tags. In the past we also had categories as "Place of birth unknown" in the talk page and then we switched them to the article. It is impossible to ask for an editor who finds that a page is orphan to go to the talk page and add the tag there. I think it was only a twist from those who want to see the template deleted. Moreover, the big problem is not the template but the 5% of Wikipedia articles that are not connected to any other page. We should encourage connectivity between pages instead of ignoring it. Image that this 5% is without disambiguation pages, SIA, etc. --
Magioladitis (
talk)
16:35, 21 December 2013 (UTC)
Orphan tagging is being removed from Twinkle too. Without Twinkle and AWB users tagging articles as orphans, I wonder how many will actually get tagged.
GoingBatty (
talk)
16:47, 21 December 2013 (UTC)
But does it really matter whether they get tagged or not? (OK, perhaps that puts me in or near the abovementioned group of "those who want the template deleted") There's just one crucial case: if an article at "Foo (disambiguator)" doesn't have a link from "Foo" (directly or via a chain", then we have a problem. That's a special case - and the article won't necessarily be an orphan as it may have other links - but to my mind is much more important than whether other articles have appropriate links to this one. Yesterday
Red or Dead (novel) was not linked from
Red or Dead. I think that's important, much more so than orphans!
That said, I've have been happy to see the orphan tag left on the main article page but producing a hidden category rather than a visible banner, just as {{coord missing}} does. The reader really doesn't need to know that there aren't any incoming links, and the de-orphaning enthusiast editor could find them through the category. Job done. Unfortunately that was a minority view.
PamD17:19, 21 December 2013 (UTC)
Apparently there was a fair number of editors that did not get an opportunity to participate in the discussion. I'm still of the opinion that fragmenting the tags by putting some on one page and some on another is counter productive and a waste of everyone's time especially considering the amount of time it will take to fix all of the scripts and tools that place tags on pages and read those tags for various reasons.
Technical 13 (
talk)
01:56, 22 December 2013 (UTC)
Yes, I did not get to voice my opinion, but I came to the exact same conclusion as PamD did upon reading the RFC. I did not know it was ongoing, otherwise I'd have voiced my opinion. If the solution retains all the functionality and doesn't involve setting a bad precedent, by all means, both parties win if it is a hidden category. Honestly, the people who are complaining about the problem are not the ones who will be involved in fixing it. I consider myself to be pretty on the ball about these things, but its obvious that the conclusion reached is based on several misconceptions. You know the whole
WP:RANDY issue? This is sorta like that, not only is the end result demonstrably worse than a much more elegant and simple solution, but the solution reached by at least two of us (independently) has worked for a larger number of pages without upsetting anyone. This is a
WP:COMMONSENSE thing, that RFC should hold no weight other than "no big orphan tags that muck up the page". So what if the better solution came a little bit late to the voting process - the end result of the RFC is still the same. Let's just be bold and do it this way... its the obvious better solution.
ChrisGualtieri (
talk)
05:21, 22 December 2013 (UTC)
I know it is a bit crazy but the reason I did not participate is that I was pretty sure this RfC will fail. The main argument has already failed many times:
Wikipedia:Perennial_proposals#Move_maintenance_tags_to_talk_pages. Distinguish editors from readers is completely out of the idea I have for Wikipedia. For me every viewer is a potential editor. No articles are complete. Major or minor, issues should be fixed and articles should improve every day. --
Magioladitis (
talk)
19:48, 22 December 2013 (UTC)
To deal with the currently existing RfC result, some AWB users may want to move the tags. To facilitate this process, I'd suggest taking another look at
a feature I requested four years ago to either perform a namespace-insensitive sort (a user could take a list of articles with Orphan tags, copy, convert to talk pages, paste original list, then sort so that each talk page immediately follows the corresponding article) or to take a list of articles and directly add each talk page to the list immediately after its article.
MANdARAX•XAЯAbИAM23:00, 22 December 2013 (UTC)
Mandarax I will deal with the list as follows: I'll load pages from the tracking categories by month, I'll add the tag in the talk page and then I'll reload the list to remove it from the article. Since now is holidays I don't expect
User:ReedyReedy or
Rjwilmsi to do large scale programming for the next short period. --
Magioladitis (
talk)
08:48, 23 December 2013 (UTC)
This is still a terrible idea. I'll open an RFC for the alternate solution. With the better option there that doesn't break a perennial, it should be smooth sailing.
ChrisGualtieri (
talk)
05:56, 28 December 2013 (UTC)
I have been passing the articletext to file for external processing. I attempted passing the article text as a parameter, but I could not get it to work that way. Any suggestions.
Snowman (
talk)
09:59, 14 January 2014 (UTC)
How can I identify the changes that apply at Wikia
Through trial and error I have found that some general fixes do not apply to Wikia and some are specific to Wikia. I have read through the instructions, manual and the archives to see if such a list exists but cannot find it. Does anyone have a list of what works at Wikia and what doesn't?
Reguyla (
talk)
23:23, 25 January 2014 (UTC)
Error in Database Scanner?
Hello, I am trying to generate a long list of all articles not containing a word and it doesn't work. I have the backup file: rowiki-20140209-pages-articles.xml - 1.3 GB.
If I try to find all articles containing "zuma", I get a list of 1687 articls, so it works
If I try to find all articles containing "a", I get a list of 32.704 articles - that's too short. In total, Romanian Wikipedia has 241.360, most of them should contain the letter "a".
If I try to find all articles NOT containing "DEFAULTSORT", I get a list of 31.955 articles - and there should be at least 150.000 articles not containing "DEFAULTSORT"
@
Ark25: On the database scanner's "Searching" tab there is a numeric field "Limit results to", which defaults to 30,000. I don't know exactly how that limit works - the one time I ran into it, I ended up with a list slightly over 30,000. Did you increase that limit before beginning the scan? --
John of Reading (
talk)
13:53, 3 March 2014 (UTC)
@
Magioladitis: Perhaps we should change the database scanner to show "n results [30,000]" if the search limit is set, so that it's clearer a limit is in place?
Rjwilmsi09:37, 4 March 2014 (UTC)
Hello! I'm an AWB n00b. Day 3. Please be gentle. Is there any way to easily replace smartquotes “ ” ‘ ’ with their straight quote equivalents, per
MOS:QUOTEMARKS? A straight find and replace is a little hairy because occasionally smartquotes matter in URLs, (
curly /
straight) and potentially other places I haven't yet figured out. Thanks,
Cyphoidbomb (
talk) 19:36, 4 March 2014 (UTC)
Wait, is that what the checkboxes for "ignore external/interwikilinks, images, nowiki, math and and ignore templates, refs, link targets and headings" are for?
Cyphoidbomb (
talk)
20:12, 4 March 2014 (UTC)
I could use this feature. Say, in AWB I have made a page list to be edited. I can filter it in some ways already. Could there be features that says & does:
"turn all pagenames into their Talk page"
"turn all pagenames into their Subject page"
"Make the list to contain all subjectpages and all talkpages of the {PAGENAME}}" (this may ~double the number of pages then).
(e/c) For the first two, right click in the list and select "Convert to talk pages" or "Convert from talk pages". For the third, I suggest you experiment with copying the list, then converting it, then pasting in the original list. --
John of Reading (
talk)
08:35, 5 March 2014 (UTC)
Is there a way to flag text/phrases for evaluation?
Hi, I'm curious if there's a way to flag text/phrases for evaluation. For example, per
WP:EUPHEMISM, articles shouldn't contain "passed away" when we mean "dead". But a recurring find/replace can't be used, because sometimes "passed away" appears in quotations or in citations, and so on. Is there a way to configure AWB to highlight a phrase so that I could then evaluate if if it needs to be changed manually? Thanks,
Cyphoidbomb (
talk)
17:09, 5 March 2014 (UTC)
Isn't this how all AWB edits work? (Except for those made through bot accounts) If you configure AWB to find "passed away" and replace it with "died", then the software will show you the diff of each edit you are about to make. It's up to you to evaluate the proposed edit and either click "Skip" or "Save". --
John of Reading (
talk)
17:34, 5 March 2014 (UTC)
John of Reading Howdy! I'm new to AWB, but your point is noted. Seems to me though, that if the default is to find/replace, and if it requires me to UN-select before saving, there is a greater potential for damage if I don't happen to catch the change. In contrast, if the software highlighted key phrases, a manual change would require some active concentration. And to pre-empt any warnings, obviously I know that active concentration should be in place while using AWB--I just mean that some form of word/phrase flagging would be convenient--find without replace.
Cyphoidbomb (
talk)
18:24, 5 March 2014 (UTC)
@
Cyphoidbomb: OK, this is possible but I think you'll find this is more work overall. At the bottom right of the "Start" tab there are some "Find" controls, and on the "Options" menu an option to "Highlight Find matches". With those you could arrange for the words "passed away" to be highlighted in yellow in each article, leaving you to make the edit manually each time. You would also need to untick all the "No changes are made" checkboxes on the "Skip" tab. --
John of Reading (
talk)
22:07, 5 March 2014 (UTC)
Awesome, I'll give it a shot, thank you! I appreciate your patience. I just don't want to mess up too badly while I learn the program. :)
Cyphoidbomb (
talk)
17:15, 6 March 2014 (UTC)
2 questions
Hello!
Is there any way to resize the Multi-wikilinks box? I find it too narrow to be of any significant use.
I've set up some find/replace rules for various things, for example I (with some exceptions) typically replace "passed away" to "died". AWB includes this change in the edit summary, but I would like to append "per
WP:EUPHEMISM" so that other editors can understand my rationale. Is there any way to do this? I generally like to nod to the policy/guidelines/MOS in my summaries. Thanks y'all,
Cyphoidbomb (
talk)
19:46, 19 March 2014 (UTC)
Cyphoidbomb, for #2, at the top of AWB, goto Options -> Default Edit Summaries. Add your summary and then save your settings so the summary will appear the next time you start up AWB. Any summaries you save will now show up in the "Default Edit Summary" box in the start tab. When using Find/Replace, AWB will append to the end of your edit summary with what was replaced.
Bgwhite (
talk)
21:18, 19 March 2014 (UTC)
I believe the question was intended to ask if there was a way to add additional predefined text when a specific F&R rule was used. Such text being in addition to the text automatically added showing the replacement. It was so a link could be provided to the specific WP guideline which detailed that such change should be made. As it is,
Cyphoidbomb would have to manually the edit summary, or manually select a predefined complete edit summary. I assume there are multiple rules being used at once. To have an edit summary which described all the edits
Cyphoidbomb would have to define an edit summary for each possible combination of edits. This would rapidly get out of hand with only a few F&R rules.
@
Cyphoidbomb: I am not aware of any way to do #2. Perhaps you should file a
feature request. As to #1, I do not see a way to resize that section of the user interface. Perhaps this should be another feature request. —
Makyen (
talk)
21:54, 19 March 2014 (UTC)
Ah, I think you've all answered aspects of my question, thank you. I think I'll go the feature request route. Much obliged, thank you!
Cyphoidbomb (
talk)
21:58, 19 March 2014 (UTC)
Changing the contents of templates Inside templates
I have an AWB script that uses "Advanced settings" and the option type "Inside templates {{...}}" in there I am looking for templates that contain "cite web" I want to replace all instances of a space with a "£" (an example, not for use in article space but for this test). That I can manage and it works fine for most cases. However in some instances {{cite web}} is within another template eg an {{Infobox}} and that is proving difficult.
This is causing problems for me because I do not want to format the content of the {{
Infobox}} but I do want to format the content of {{
cite web}} within the Infobox, so a simple string in the "if box" of the advance setting of cite web is of course picked up as a string in the {{Infobox}} and the script formats all of it, so the simple solution will not work.
It seems to me that the regular expression for start of line ^ (hat) does not work as start of template string indicator so ^cite web is not an option (I may be mistaken about this -- if so then the solution is simple).
I have tried another trick: to place cite web in an "if box" of a rule and then use a sub rule with an "if not" <ref or similar (The idea being that a {{
cite web}} template will not contain a ref tag but if {{
cite web}} is placed within another template then that template will probably contain a reference tag just before the {{
cite web}} starts).
That does not work as I thought it might. Indeed I can get some very funny results when testing this. If an identical {{
cite web}} template with identical fields exist inside and outside the {{
Infobox}} then the it changes it as expected. But if as little as one byte varies between the two instances of {{
cite web}} it does not work.
The simple solution would be if "Inside templates {{...}}" had a start of string ability but as far as I can tell it does not. So can someone please suggest a simple solution.
Yobot keeps on changing "Related topics" to "See also"...sorry, Related topics isn't wrong and no policy discourages the use of that section title, no matter how many times Yobot persists to change it. Just because
MOS:SEEALSO says "The most common title for this section is "See also", doesn't mean it is the only title and Yobot should not be changing a perfectly acceptable section title referred to elsewhere in the MOS to demand an artificial uniformity. Nothing in ANY policy or guideline says "See also" is the only "approved name". FYI, if you go back in time, the MOS:SEEALSO section used to be named "See also" and "Related topics" sections, and I direct you to this page:
[1]. --
ColonelHenry (
talk)
21:41, 24 March 2014 (UTC)
"See also" is the only one mentioned on
WP:ORDER and
MOS:SEEALSO. As
WP:MOS states for a reason for MOS, "Consistency in language, style, and formatting promotes clarity and cohesion." Readers expect the "See Also" section because 99.9% of all pages only use "See also". Using only "See also" brings consistency of style and formatting. To go against MOS and common style, there must be a good reason to do so. Just because you like "Related topics" is not a good reason. Using the term readers expect is the common sense format, not something readers don't expect.
Bgwhite (
talk)
22:17, 24 March 2014 (UTC)
I prefer the "See also" too per
WP:KISS. Of course if there is a See also section and another section should be added, Related topics is a good name for that. --
Magioladitis (
talk)
23:16, 24 March 2014 (UTC)
Can I have some statistics that 99.9% of pages use "See also"? Especially since one policy I mentioned above used to include "related topics" in the relevant section name and another active MOS guideline doesn't mention "See also" when discussing "Related topics" sections -- and since "See also" is an awkward construction to non-English speakers and in some regional English variants. Or, Bgwhite, is this just another logical fallacy appealing to a false assertion in the spirit of 500% of statistics are exaggerated.--
ColonelHenry (
talk)
14:39, 25 March 2014 (UTC)
Proposed new feature request status
Since there are a bunch of feature requests that have no replies, I am proposing some way to differentiate between new feature requests that have never been replied to or looked at and feature requests that have not received a response in terms of technical feasibility but have some discussion. This would come with a modification of the feature request template to make requests with a "new" status more eye-catching, and the addition of a new status, "checked" (or whatever) to signify that this request has been discussed. If this were added, it would make it much easier to identify requests that have been hanging around for a while with no comments or discussion. Thoughts?
APerson (
talk!)
02:28, 26 March 2014 (UTC)
Automating sort keys "by year"
I placed this as a bot request, but, for the moment, I was wondering whether AWB can help me.
The specific bot request I made was to do things such as:
Fix longstanding bug that broke citation terminal punctuation if the value assigned to |postscript= is multicharacter (like html entities); Moved citation template's default assignments for |separator=, |postscript, and ref=harv from the invoking template into the module; Added support for |postscript=none; (
discussion)
Limit acceptable years in dates to current year+1; (
discussion)
Expand date validation; all allowable date formats should now be supported; (
discussion)
Move |postscript= and |separator= default initialization into Module:Citation/CS1/sandbox;
Add subject and subject link for cite interview migration;
Add artist, albumlink, albumtype, notestitle, publisherid for cite AV media notes migration;
Add lccn error detection;
Delete albumtype; merge deprecated parameters albumlink, artist, director, notestitle, publisherid, titleyear as aliases of other parameters; remove these parameters after 1 October 2014;
to Module:Citation/CS1/Whitelist:
Add subject and subjectlink for cite interview migration;
Add artist, albumlink, albumtype, notestitle, publisherid for cite AV media notes;
Invalidate albumtype; deprecate artist, albumlink, director, notestitle, publisherid, titleyear; these last to be invalidated after 1 October 2014;
I am not aware of any such bot. However, if you have rollback and they are the latest edits you can at least open the contributions page and have a single click per each action.
Rjwilmsi12:57, 1 April 2014 (UTC)
LOL, no special:nuke permission for me. I'll resort to botrequest or biological arms: a mouse & some fingers. thx you both. -
DePiep (
talk)
16:51, 1 April 2014 (UTC)
Replace all content?
I'm using Autowikibrowser on another wiki and want to know if there is a way to replace all the content on a page with a template - I'm replacing text on multiple categories with a template and would like to know if this task can be automated.
NemesisAT (
talk)
20:47, 1 April 2014 (UTC)
I would start with something like: find \{\{ft\|([^}]*)}} and replace with $1 You may have to throw in a few \s* to catch unexpected spaces. See
Regular expression and test it using the tool.
Mr Stephen (
talk)
21:21, 12 April 2014 (UTC)
I started this discussion at
User talk:Bgwhite#AWB: cite book, where a number of people chimed in, but it seems to be not user-related but purely AWB related, so better to have it here.
According to
Template:Cite book, there is no rule to automatically change the parameter "date" to "year" when only the year is given (there may be a reason to do so on individual article basis). However, this is a standard AWB change. Note that the "year" parameter isn't even present in the standard "cite" tool in the editing toolbox, so it seems strange that a parameter you can't choose on the one hand, should be the obligatory one for AWB on the other hand.
I would suggest changing AWB so that it no longer makes this change, but perhaps there are convincing arguments to the contrary...
Fram (
talk)
12:48, 22 April 2014 (UTC)
The
Citation Style 1 templates that are supported by the various editors ({{
cite web}}, {{
cite news}}, {{
cite book}}, {{
cite journal}}) are rendered into proper citations by
Module:Citation/CS1. Before that tool was developed, CS1 templates were rendered by {{
citation/core}}. There were certain limitations in {{citation/core}} that required separate |year= and |date= parameters when used with {{
sfn}} and the {{
harv}} family shortform citation templates. The restriction was necessary particularly when disambiguating the year when an author published multiple works in the same year. With Module:Citation/CS1 those limitations no longer apply. Except for the six, soon to be five, CS1 citations that still use {{citation/core}}, |year= is no longer required.
I think that, as a general fix, changing |date= to |year=, even when it contains only a year value, is unnecessary.
Sources? If it is a year, it should be year. If there is a date, it must include date, month, year. I don't see anything wrong with that, I really appreciated such type of AWB setting.
OccultZone (
Talk)
15:57, 22 April 2014 (UTC)
Previously, month and year were recorded separately. Now they are recorded together in a single date parameter, and the "month" and "day" parameters are deprecated. In future "date" will hold either a full or partial date. There's a table listing the acceptable formats. --
79.67.241.210 (
talk)
16:10, 22 April 2014 (UTC)
Sources for what? When is a year not a date? Would you have a separate parameter for month and year or season and year combinations? All are dates.
In the cases where the citation is handled by
Module:Citation/CS1 and where |date= is empty, CS1 concatenates the individual parameters |day=, |month=, (yes, those two are still out there) and |year= into an internal variable called Date. When |date= is not empty, its value is assigned to the variable Date so when there is only one of |date= or |year= both are treated identically. From Date CS1 makes the CITEREF anchor used by {{
sfn}} and the like. Date is also the source for the date included in the citation's
COinS metadata.
Because of the limitations I mentioned in the previous post, citations with both |date= and |year= still exist. When a CS1 citation contains both, if |ref=harv then CS1 uses |year= to make the CITEREF anchor. It does this for compatibility because in the old {{
citation/core}} days, year disambiguation could not use the value in |date=.
But, all of this is not really relevant to the point of this discussion. The question is: Should AWB general fixes change |date= to |year=? To which my answer is: no. It is ok to continue to use |year= if you'd like; it is still an active and valid CS1 parameter. I think that ultimately, |year= should go away because like |day= and |month= it will have outlived its usefulness. But that is a topic for another day.
rev 10083 Following Lua conversion of citation templates, no longer necessary to rename date parameter to year for Harv support when date is just a year value
Rjwilmsi17:32, 23 April 2014 (UTC)
category check
Is there some way to do the following: 1) Check on whether a category I am about add to an article exists or not so if it does not exist, I can go create the category first or correct my edit; or 2) Check whether a category already in an article exists or not so if it does not exist, I can go create it when warranted. Thanks.
Hmains (
talk)
23:17, 25 April 2014 (UTC)
The tag
dl represents a definition list and
dt is the term which should be defined. So ;... is not the same as <h6>...</h6>! From
Wikipedia:Accessibility:
Do not make pseudo-headings using bold or semicolon markup. Screen readers and other machines can only use correctly formatted headings...
Stephan Kulla. You are confusing many things. AWB and
WP:CHECKWIKI are different projects. Don't associate the two together. <h6> *is* html that should be avoided. It was converted to wikicode, thus the edit summary is correct. Editors should not have to know html and wikicode to edit pages. Also, <h6> is a section header. You had a level 2 heading followed by a level 6 heading. This is strictly forbidden in
WP:ACCESSIBILITY#Headings.
Bgwhite (
talk)
05:30, 28 April 2014 (UTC)
What is the proper tool to learn to use to clean up articles which repeat citations, such as
the current version of
Lester Breslow? In that article a few citations are generated repeatedly whereas the references ought to be named. Is it a function of AWB to condense these to one? If not AWB, is there any tool which does this? Thanks.
Blue Rasberry (talk)13:49, 28 April 2014 (UTC)
@
J. Johnson: As long as we're already talking about this, I think you mentioned on my bot's talk page that you had a concern about this functionality. Care to elaborate?
GoingBatty (
talk)
16:58, 28 April 2014 (UTC)
Could it be possible to use AWB to fill Wikidata items with the Wikipedia labels? Also, are there plans to enable AWB for Wikidata?
Jane (
talk)
11:04, 3 May 2014 (UTC)
Request to expand #4 of AWB rules and include a behavioural guidelines
User:OccultZoneAnother editor is making many minor/insignificant edits using AWB. I tried talking to him at
User_talk:OccultZone#.23_4. A couple of weeks ago, I wrote an essay "AWB spammer" then had it deleted because it became too much aggressive. Please consider adding AWB behavioural guidelines. Tito☸Dutta 08:18, 10 May 2014 (UTC) removed name to highlight the suggestion.Tito☸Dutta08:21, 10 May 2014 (UTC)
Hi, as a semi-new-ish user of AWB, I don't understand what the beef is here and I know that I'd personally benefit from understanding the background (i.e. the presumably discussion-based rationale) that led to the rule. (I really wish that policy/guideline/MOS pages actually linked to the discussions that led to the rules...) Though I'm not an AWB power-user, when I do use it, sometimes one of my find/replace filters will only find minimal problems on a page, say, to change smart quotes to straight quotes per the MOS. I would probably consider that both minor and insignificant, but it does improve the consistency of the project. Should I not do that? Is that a misuse of the tool? Are there specific examples that you're asking to include in this proposed expansion of the behavioral guidelines?
Cyphoidbomb (
talk)
08:31, 10 May 2014 (UTC)
Hi
Cyphoidbomb, the aim should be to help the encyclopedia. If "I" am making thousands of edits just to increase my edit count, it is not acceptable. I have checked some of
your,
AWBedits. You are doing fine. I mean edit like
this, and if someone keeps on making many such edits, that becomes annoying. --Tito☸Dutta08:42, 10 May 2014 (UTC)
I know that usually happens when too many edits are made. I understand that they should be avoided too. But I am not certain about the rules that you want. Like what it should be? A minimal change still happens to be a change. There should be something like "You cannot make edit if its less than 5 bytes" or something like they do on forums? That is what I thought right after checking your post here. Although like I said, I would like to hear what type of rule you want.
OccultZone (
Talk)
08:46, 10 May 2014 (UTC)
It really is irritating when my watchlist keeps coming up with edits like
this one. There's clearly scope for misinterpreting what counts as an inconsequential edit.
Cyphoidbomb mentioned changing curly quotes to straight quotes: that's a good edit because it has a constructive effect on the rendered page, so that what the reader sees is more MoS-compliant than it was before.
One thing we need IMO under "you are recommended not to" are examples of inconsequential edits: things like changing underscores in pagenames to spaces, moving a tag when that positional change has no effect on what the reader sees, "fixing" working redirects, removing odd bits of whitespace. With a reminder that however many inconsequential changes AWB suggests for a page, we're recommended not to do them unless at the same time we make a significant edit that does have a constructive effect on the rendered page. cheers,
Struway2 (
talk)
10:01, 10 May 2014 (UTC)
To me it is very clear that #4 includes the above example edits. It says: "An edit that has no noticeable effect on the rendered page is generally considered an insignificant edit. If in doubt, or if other editors object to edits on the basis of this rule, seek consensus at an appropriate venue before making further edits."
A. Changing the underscores to spaces has no noticeable effect on the page and thus is clearly an insignificant edit. B. An editor objected on the basis of that rule. It is then required of the editor desiring to make such edits to stop making them and "seek consensus at an appropriate venue before making further edits."
While more explanation for #4 might be desired, the current issue is that the rule is clearly not being followed both in that it should have been known prior to the edits that they fell under the description of inconsequential and that the process required when an editor objects was not being followed.
If what was desired for the project was to just have AWB general fixes applied to articles in mass then that is what bots are for. —
Makyen (
talk)
11:02, 10 May 2014 (UTC)
I agree. He replied here in this thread at 8:46, then he made
this edit, one whitespace removal, one link to a dab page. I don't know why he removed portals
here. anyway, whatever the reason is, edit summary is incomplete.
This is not the best type of edits, where one adds persondata (actually AWB automatically generates it, so, no work here), but does not fill "short description" as it takes few seconds to do it manually. --Tito☸Dutta11:29, 10 May 2014 (UTC)
"Portal" shouldn't be used, but "portalbar", and it should be in footer of see also, not before the "see also".
OccultZone (
Talk)
11:38, 10 May 2014 (UTC)
OccultZone OK the problem is the following: Many of us make a lot of edits but a) you should always go back and fix your mistakes before moving especially when you get warning from bots like Bracketbot b) you should not save when you see an obvious bug c) you should always report bug in software d) you should show evidence that you check your edits e) you should interact with other editors. --
Magioladitis (
talk)
12:08, 10 May 2014 (UTC)
I've uploaded a new snapshot to sourceforge
here, rev 10102. It is a release build. The official AWB 5.5.2.4 release will be made in the next couple of week. Thanks
Rjwilmsi10:47, 8 May 2014 (UTC)
The new handling after a double click in the diff window, where you now stay in the same place rather than returning to the top, is wonderful! Thanks for all of the hard work that the AWB team puts in – I see that you reached the milestone of ten thousand revisions about seven weeks ago.
MANdARAX•XAЯAbИAM05:36, 11 May 2014 (UTC)
Filter
I can't for the life of me figure out how to filter pages from the list without running individual "Apply" operations on single words. For example, if I want to filter out all titles containing "sandbox" and all titles containing "noticeboard", I have to do two filter operations. Is there any way to filter the list based on excluding any of several words? I've tried "sandbox OR noticeboard", etc., to no avail. A little more documentation on the "Filter" feature would be appreciated. Thanks!—
D'Ranged 1talk00:10, 18 May 2014 (UTC)
On the "List Filter" popup at the bottom center of the "Title" grouping box there is a checkbox that says "Regular expression". If you check that you can use
regular expressions in the "Remove titles containing" and "Keep titles containing" fields. The short version of regular expressions for what you want is that OR is indicated by a "|" (pipe) character. —
Makyen (
talk)
03:37, 18 May 2014 (UTC)
This section currently has five points. I suggest we remove the sentence 'The Wikipedia tenet "be bold" is not ... lacking demonstrable consensus' from the last point, since it is already mentioned at point 3, and merge 'If challenged, the onus ... make on a large scale.' with point 3.
Jayakumar RG (
talk)
12:34, 17 May 2014 (UTC)
Recently there have been 297 entries added to
Wikipedia:AutoWikiBrowser/Template redirects increasing the number of rules by more than 50%. This groups of entries appears to be a wholesale change made in the template naming scheme by a project. I only checked one of the new entries, but it appeared to have a very low transclusion count. The one new template redirect rule I checked was {{AllenOH-geo-stub}}. It has a total of
8 transclusions (now 7), only
2 of which are in mainspace.
To me this brings up the question of: When should we add entries to
AWB Typos or
AWB Template redirects vs. just changing a low number of pages?
Both of these features cause AWB to search for the changes listed on every page edited by anyone using General Fixes (I assume only when General Fixes are selected). These changes are attempted as long as the rule is listed in the appropriate page. In other words, basically forever, unless someone cleans the list. It appears to me that these rules should be used primarily for typos or template redirects where we expect editors to continuously and repeatedly enter the wrong word or the template that is redirected. The facility, as I understand it, is to help with ongoing issues which should be addressed when they are encountered.
For entries like the one I checked on above, it appears to me that the appropriate thing to do would be to bite the bullet once; run through the changes once on the low number of pages affected and be done with it. Perhaps even then remove the redirect from the old naming scheme so that it is not used?
I know that we are generally not to make changes which do not affect what is displayed by the article. A run just to bulk-change a list of redirects would certainly not make any visible change to the displayed pages. However, it seams to me to be the lesser of two negative impacts: 1. a set of template changes on a limited number of pages marked as a minor change with a edit summary that states that the template was renamed, or 2. everyone using AWB always looking for all of these changes on the off chance that one of these small number of easily identified articles is being edited.
Is there policy on this issue? What do others think?
While this particular set of additions happened to be what got me to check and ask the question, I do not intend to single out this set of changes. There was also a somewhat related discussion in
this discussion on the Typos page.—
Makyen (
talk)
08:59, 21 May 2014 (UTC)
@
Makyen: I don't believe there are official policies or guidelines as to when it is appropriate to add entries to the typos or template redirect lists, but I'm happy to share my opinion.
For typos, I'll agree with
Chris the speller that there should be a significant number of typos to be corrected and no false positives.
I update the template redirect list in those cases when AWB is set up to make a visible change with a template and I want AWB to do the same with each of the redirects as well. For example:
When I am fixing dates in {{cite news}}, I also want the dates fixed in {{cit news}}
Since AWB has functionality to move some maintenance templates (e.g. {{notability}}) to the top of the article and include them in {{multiple issues}}, I have updated the list to include redirects so AWB will move them too (e.g. {{Chemical-importance}}
When someone is making a wholesale change in the template naming scheme, I hope they will consider whether the redirect can be deleted after the name change, and then removed from
WP:AWB/TR. Hope this helps!
GoingBatty (
talk)
13:14, 21 May 2014 (UTC)
I disagree with the addition of all these stub templates and redirects too. For maintance templates the feature helps in general fixes. The same holds for infobox redirects that do not follow the "Infobox..." convention. Since AWB works fine with any "...-stub" template there is no reason to bypass the redirect. --
Magioladitis (
talk)
13:22, 21 May 2014 (UTC)
3rd party user scripts
Is it possible to combine such user script (shown under See also) to interpose (if there are automatically)? I thought so, since one is so signed in. That would be great or did I ignore something!? --
πϵρήλιο15:54, 12 May 2014 (UTC)
Ok, short said, I use some simple (automatic) "cleanup" js-user-scripts (as they shown under "See also" here), they don't work with AWB together. Is there a possibility, you know? Greetings --
πϵρήλιο14:08, 13 May 2014 (UTC)
Not really, but it would be nice to have, (if it could be in a sort of a support feature to made it compatible like wikEd has such feature.) But I think it must be in such a way as plugin!? --
πϵρήλιο00:28, 15 May 2014 (UTC)
Background: The powers that be have been working on a new back end for Special:Search. The current search is "lsearchd" (that's a lower-case "L"), and the new search is "CirrusSearch". They seem fairly close to rolling out the new back end, since "CirrusSearch was enabled as the primary search method on all Wikipedias with less than 100,000 pages on May 30" per Village Pump (tech). You may use CirrusSearch by checking the "New search" box under the "Beta" menu. If this is checked, AWB will also use that back end when making a list of articles using the source "Wiki search(text)". If you want to use the new search on your usual browser on occasion without affecting AWB, just append "&srbackend=CirrusSearch" to the URL.
More background: The search index for lsearchd is not being updated (since May 29), and we can not expect it to be updated, per
this bug on Bugzilla. So, if you search for "buisness" every week and fix them, you will not find any new ones, only the ones you already fixed before May 29.
Good news: If you use CirrusSearch, its index is updated in real time (lagging no more than a few seconds at most times), whereas you had to wait up to 24 hours for the lsearchd index to be updated. CirrusSearch also has some new features, such as a fuzzy search, where you can specify how close together the words in your search argument have to be. searching for "dropped ordinance"~1 will find "the plane dropped its ordinance", but "dropped ordinance"~0 will not find intervening words.
Bad news: CirrusSearch ignores hyphens. A search for "wholly-owned" works just like a search for "wholly owned", so its use for finding and fixing mishyphenation is very limited. I commented on the
talk page for the new CirrusSearch. Or maybe I'm the only person fixing hyphens, I dunno.
What next? It might be good for AWB to provide a way to specify which search to use for source "Wiki search(text)", but there might not be much time to roll that out. Their target was "early 2014". The asteroid is on its way, so get your things in order.
Christhe spelleryack17:47, 3 June 2014 (UTC)
AWB at outsider project wikiskripta.eu (wikilectures.eu)
Hi,
I would like to employ AWB at wikis for medical specialists and students
wikiskripta.eu and
wikilectures.eu, but me and my colleagus cant get it running the regular way. Our bot API works just fine and our version is 1.22. The error AWB is giving is:
Error connecting to wiki
An error occured while connecting to the server or loading project information from it. Please ensure that your internet connection works and such combination of project/language exist. Enter the URL in the format "en.wildpedia.org/wr (including path where index.php and api.php reside).
Error description: Neočekavana deklarace XML. Deklarace XML musf být prvním uzlem v dokumentu. Před ní nejsou povoleny žadné prazdné znaky. Radek 2, pozice 3
I sometimes wish AWB didn't change the order of references. I can see at
WP:AutoWikiBrowser/General fixes#ReorderReferences (ReorderReferences) why it does it but is it really a good idea? Suppose two named references have been previously used and are numbered 1 and 2. Then there is a sentence making two rather distinct claims referenced by 2 and 1 in that order. I think it is better to have (and keep) the references in that order. It is unimportant that the references are in ascending numerical order. It would be good if AWB could be changed but, if not, is there a way I can stop it doing this in significant cases? Another question – in the documentation I see this feature is fully compatible with "the 'named references at end' formatting option". Is this another way of referring to what is described at
Help:List-defined references?
Thincat (
talk)
09:01, 11 June 2014 (UTC)
(Partial answer) If you stick an HTML comment between two references, AWB will not swap them. And yes, I think "named references at end" is referring to list-defined references. --
John of Reading (
talk)
09:58, 11 June 2014 (UTC)
Twice now at
Corrour Lodge AWB has removed a section heading.
[3][4] I think this change was not beneficial. I think it may be to do with the section heading being the same as the article name because the same happened
at another article where there was more justification though I still think it was not a good edit. Now, I guess I can fool AWB but why is the heading being removed and is it a good thing to be doing?
Thincat (
talk)
09:41, 11 June 2014 (UTC)
@
Thincat: Yes I made the first one, and the other one. You cannot have section with the same title as your article. We use different words for describing the entity. For example, if it is a person, most common section would be "biography".
OccultZone (
Talk)
09:46, 11 June 2014 (UTC)
@
OccultZone: well, I think there are occasions when the same heading is justified but if you are right, why don't you replace the heading with one you approve of rather than simply remove it? What should the section heading be in this case?
Thincat (
talk)
09:53, 11 June 2014 (UTC)
I rather agree a changed heading would be editorially sensible. However, the way to achieve this is not by removing the heading completely.
Thincat (
talk)
10:28, 11 June 2014 (UTC)
@
Thincat: I *think* this code was written to catch the case where a new article starts with a redundant copy of the name of the article, but as currently implemented it does too much. It's caught me out a few times, and I've had to go back and re-insert a heading; doubtless there are other cases where I have accidentally damaged an article and not noticed. I would prefer to see this as an "Alert" rather than an automatic edit, so that the editor using AWB can decide whether to make a change. --
John of Reading (
talk)
10:10, 11 June 2014 (UTC)
Yes, I can see that generally a section heading would best not be the same as the article title (and if at or near the top could well be removed). In other situations it would be far better to change the heading than to remove it. Care is needed removing (or even changing) section headings because sometimes they are in use as anchors.
Thincat (
talk)
10:22, 11 June 2014 (UTC)
@
Magioladitis: but bigger question was that why there is even a need to name subsection after the page title? It has to be different. I had closely observed those 3 edits.
OccultZone (
Talk)
10:41, 11 June 2014 (UTC)
I have now changed (hopefully improved) the heading. I have also removed the {{bots|deny=AWB}} tag
Magioladitis so helpfully put there because (despite my interventions here) I do think AWB can sometimes make useful edits. However, I shall slide the tag up my sleeve for future use!
Thincat (
talk)
10:56, 11 June 2014 (UTC)
I was running a batch today and
genfixes was moving {{oldafdfull}} after the WikiProject banners, but I thought oldafdfull qualified as an "article history" template per
WP:TPL? I'm having trouble reproducing this now, but I wanted to at least make sure my logic was right czar
♔13:23, 12 June 2014 (UTC)
@
Czar: Afd should not come before the wikiprojects. But when someone has added them before the wikiprojects, they shouldn't be notified for that. I had questioned before and they blamed on the script. It is a non-controversial edit.
OccultZone (
Talk •
Contributions •
Log)
13:32, 12 June 2014 (UTC)
@
Czar: in which page did this happen? Can you please provide an example? AWB does not recognise all redirects of all talk page banners/tags. --
Magioladitis (
talk)
13:34, 12 June 2014 (UTC)
@
Magioladitis, as mentioned, I went back over my edits when writing this post and I couldn't reproduce it, though it happened to me twice. Edit: okay, I got it. Try
Talk:Global StarCraft II League with genfixes on—it moves oldafdfull to the bottom. czar
♔13:44, 12 June 2014 (UTC)
Hi all, I think it would be really great if the "Make list" will have an option to search pages based on a set of Regex expressions defined in a page, such as
User:AlexNewArtBot/Romania, or defined in the interface. This could be of great help when making lists of articles which should be tagged with a certain project tag, e.g. {{WikiProject Romania}} in this case. Any thoughts? --
Codrin.B (
talk)
08:42, 19 June 2014 (UTC)
I make my list with AWB. I can add+assess article at the same time too(no AWB here). If you are used to with AWB, you will know that it can be also used for things other than editing articles.
OccultZone (
Talk •
Contributions •
Log)
15:30, 19 June 2014 (UTC)
Use the built in database scanner (you will of course have to download the 10GB db file which expands to about 45GB). — {{U|
Technical 13}}(
e •
t •
c)15:50, 19 June 2014 (UTC)
Plugin: Send AWB to back?
I've been working on a plugin to use MS Word's spellchecker, and I've got it working apart from one thing: AWB blocks the spellchecker window AND becomes unresponsive until the spellcheck is complete. My work-around atm is to alt-shift-tab to bring the spellchecker window forward. So, is there some way of telling AWB to send itself to the back? Or minimize? --
Niightblade (
talk)
09:36, 23 June 2014 (UTC)
Width of the list box
Hello AWB experts! I am just starting out with AWB, and this seems obvious, but I can't seem to find a way to widen the list box. Since I want to work with pages in "Wikipedia talk:Articles for creation/", all I see is the first letter of each page title. What should I do? —
Anne Delong (
talk)
17:13, 30 June 2014 (UTC)
With the mouse, hover over a page name, and the full name will pop up. This is only good for one page at a time; you can't see 17 pages at a glance.
Christhe spelleryack19:35, 30 June 2014 (UTC)
Yes, I have done that, but since all of the pages in the category begin with the same 38 letters, I can't look through the titles other than in the very inefficient way mentioned by
Chris the speller. Since there are thousands of pages in the category, it looks as though AWB may not be the right tool for working with these pages. In the future, when the AfC pages are all in the Draft namespace instead of at "Wikipedia talk:Articles for creation", I will try again. I have been having some success, though, in using AWB to fix up some the thousands of bad links created when the Canadian Encyclopedia decided to change all of its URLs. —
Anne Delong (
talk)
10:58, 1 July 2014 (UTC)
Every so often AWB is upgraded. When this is done, the old version is flagged so that you can no longer use it until you run the updater. So why is the upgrade always labeled as an optional upgrade? Either it should be listed as a required upgrade or else the software changed so that old versions can continue to run. —
RRabbit42 (
talk)
16:34, 21 June 2014 (UTC)
This isn't an issue of whether a computer can update the software. I'm addressing a misleading statement about updates. For your issue, you may need to completely uninstall the program and clean out the registry if you're running it under Windows. —
RRabbit42 (
talk)
14:29, 28 June 2014 (UTC)
@
RRabbit42: Well I mentioned the situation of my laptop, because it was different with my computer. Computer could work with the older version, but laptop couldn't. No idea though. As for your query, I would say that it should be listed as 'required update', because AWB probably won't work without the update.
OccultZone (
Talk •
Contributions •
Log)
11:24, 1 July 2014 (UTC)
Reference ordering
Resolved
I was wondering why in a spot of multiple references are they sometimes recommended by AutoWikiBrowser to be reordered -- is this to keep the footnote numbers in numerical order, or for a different purpose? I skipped on these changes until I could find out the reason for this.
Stevie is the man!Talk •
Work15:19, 1 July 2014 (UTC)
AutoWikiBrowser is written on C#. Why you do not support Mono natively?
Why only Windows is supported natively? I thought Wikipedia is free and open project. Why users of Wikipedia must use Windows? --
Mvitaminus (
talk)
09:56, 1 July 2014 (UTC)
Per FAQ, "AWB might also run under
Mono. See
here for a status report of AWB vs Mono v2.6." Toolserver has been shutdown, and today is 1st day. So you may have to wait. Like Magioladitis said, you are free to assist.
OccultZone (
Talk •
Contributions •
Log)
11:21, 1 July 2014 (UTC)
Any person can use the AWB without permission in ta.wikimedia projects. This is because of the absence of approval mode. How can we set the permission/approval log--
தகவலுழவன் (
talk)
02:32, 3 July 2014 (UTC)
This is about a template on huwiki with a date parameter (format: yyyy-mm-dd). I would like to make a list of articles transcluding templates with expired date values. How can I do this?
GoingBatty, I remember
Magioladitis saying the stats were moved over to WMFLabs. Code was changed in May I think. WMFLabs' webserver has been real flaky the past several weeks. It keeps going down and individual projects have to manual restart their own webserver instance. So, this might be the cause.
Bgwhite (
talk)
22:13, 5 July 2014 (UTC)
Typo rule trying to change the sort key of a category
I wonder if AWB should ever mess with sort keys for categories. In
List of Georg-August University of Göttingen people, AWB wants to change the sort key from "Göttingen" to "Gottingen"; this screws up the sorting of the category page for "Lists of people by university or college in Germany". Aren't we always likely to do more harm than good trying to "fix" category sort keys?
Christhe spelleryack19:41, 10 July 2014 (UTC)
Thanks for that, John. I filed that bug report a year ago and hadn't looked at it for many months. It's spooky that "Göttingen" in particular was added to that discussion just a few days before I posted here. I'll comment over there on the bug report.
Christhe spelleryack03:03, 11 July 2014 (UTC)
Looks quite heavy duty, I was wondering if I could do it by hand with AWB.
OccultZone: sorry, please can you explain how to wikiproject tag with AWB? Thanks,
Matty.00710:44, 12 July 2014 (UTC)
First make sure what page you have to edit, save the link to a list.
Second know that which wikiproject should be added to the page.
You can just copy-paste, AWB will directly open the talk page for you or convert the main page to talk page. You can also use the 'find and replace' option of AWB, which automatically places for your or any modules(scripts) that are provided for that purpose.
OccultZone (
Talk •
Contributions •
Log)
11:02, 12 July 2014 (UTC)
Bug - typo fixing in image links
With
this edit AWB's typo fixer changed the image parameter in the infobox and so broke the image. While I realise that it is my responsibility and I should have checked better, would it be possible to fix this to prevent it happening again? Thanks,
BethNaught (
talk)
19:33, 12 July 2014 (UTC)
Ok so I've been working on filling Persondata templates (it's a nice simple mindless task to chip away at), but I've noticed that the
project page is pretty inactive, and there are
oldish comments about Wikidata eventually replacing Persondata. However there are still new Persondata templates being added and there doesn't seem to be any real movement for converting to Wikidata. So basically there's this big dead rolling ball of Persondata without anyone behind the controls.
So I was wondering if you guys had any insight, since you know about using bots, and this is probably the main platform that adds Persondata. Do you have any plans, or have you had any discussions regarding converting to Wikidata?
Do you know anyone willing and able to take the reigns of the Persondata WikiProject and phase it out?
Everyday I've been filling persondata, there are members who have filled thousands of pesondata. Thanks for informing about WikiProject. I will probably switch to persondata once I am done of these other maintenance tasks.
OccultZone (
Talk •
Contributions •
Log)
17:54, 9 July 2014 (UTC)
@
OccultZone:Msmarmalade, persondata will be phased out. It is only a matter of time. The German Wikipedia Chapter is in charge of Wikidata and they started persondata. So the timing is probably upto them.
Bgwhite (
talk)
04:36, 11 July 2014 (UTC)
After we pass to phase 2 of Wikidata, Persondata will most probably become obsolete and we 'll start removing it but I ve no idea when this is. Phase 1 is still incomplete. --
Magioladitis (
talk)
12:52, 11 July 2014 (UTC)
Wikidata Wikidata is a metadata system still in development. After Wikidata passes into the second stage, Persondata will likely become obsolete, and be removed. Until that time, users can continue adding Persondata.
I think Wikidata has passed the 2nd and 3rd stages. They had bots gather info from persondata and put it into Wikidata. I'm not sure after that.
Bgwhite (
talk)
06:04, 13 July 2014 (UTC)
@
Msmarmalade: I hope Persondata would not become obsolete until all of the information was moved to Wikidata. Until there is an agreed upon plan to obsolete Persondata, I think it would be best NOT to add speculation to the official documentation. Thanks!
GoingBatty (
talk)
15:04, 13 July 2014 (UTC)
@
GoingBatty: That's fair enough, I mainly just want some brief explanation at least on the project page for people like me who wonder what's happening with wikidata. What about this wording instead?
Wikidata is a metadata system still in development. Eventually, information from Persondata will be moved to Wikidata, and Persondata may become obsolete. However, for now, users are still helping by adding Persondata and are encouraged to do so.
@
Msmarmalade: That's much better - thanks! Any sense of whether the Persondata values be migrated to wikidata, or whether wikidata would be created from scratch? Thanks!
GoingBatty (
talk)
16:53, 14 July 2014 (UTC)
@
Msmarmalade: It's fine with me if you add it to the Persondata page. My question wasn't necessary, since you wrote "information from Persondata will be moved to Wikidata". Thanks!
GoingBatty (
talk)
02:28, 17 July 2014 (UTC)
I apologize if this question was already answered (I couldn't find anything in the archives, and if it has, feel free to trout me), but I heard somewhere that there was a plugin for AWB that could fix double redirects. A Google search proved fruitless though, so I was wondering if anyone had the plugin, or a link to where I can obtain it.
Unreferenced to Refimprove changes existing date to current date
Resolved
When General Fixes changes the Unreferenced template to Refimprove, it removes the date that's there and replaces it with the current date. I checked the bug list and apparently this was fixed in 2012, so it either wasn't or its crept back in. I know dates on tags are used for categorisation purposes, but they're also used by people fixing the lack of refs, working their way through from oldest to newest. The tags are also used as value judgements in content discussions. If an IP user changed the date of a tag from April 2011 to August 2014 it may be classed as a disruptive or vandal edit. But we're allowed to get away with it because we use AWB. -
X201 (
talk)
09:11, 8 August 2014 (UTC)
I think this what we were asked to do. Change the date with the current date. This is because refimprove is a new tag on the page and the page has to be recategoried. --
Magioladitis (
talk)
10:36, 8 August 2014 (UTC)
While classifying articles with AWB (C class, stub class). I often just click the link, go to the article. Grade it. come back to AWB, press control V (which pastes |class=). Then write the appropriate class. But sometimes, I accidentally write start right after I press control V, I accidently hold the control button which saves the page. Is there a way to disable this? Thanks, TheQ Editor (Talk)16:52, 14 August 2014 (UTC)
No, there's no option to disable the keyboard shortcuts. However, there are a couple of ways you could do it: one would be to make your own build of AWB with the keyboard shortcut disabled in the code. Another option would be to use the open source
AutoHotKey utility and configure it to override certain keyboard shortcuts (e.g. do nothing) when AWB is active.
Rjwilmsi09:18, 15 August 2014 (UTC)
"Reset to default settings" doesn't seem like the best wording
I was thinking of opening a bug for this, but I wanted to get feedback first. "Reset to default settings" in the File menu goes back to the application's original settings, rather the default settings I have set up. My default settings can be retrieved by closing the application and re-opening it, or by going to File, Recent settings, Default settings. This is confusing, at least to me. Perhaps "Reset to default settings" should be reworded to "Reset to original settings". Thoughts?
Stevie is the man!Talk •
Work11:23, 20 August 2014 (UTC)
Have .net 4, asks me for .net 2 and downloads .net 3.5? Should I?
I am using Windows XP at the moment. When I check add/remove programs it says I have .net Framework 4 Client Profile and Framework Extended. Isn't that .net 4? Well AWB said I needed .net 2 and gave me a link to a cloudnet group
[7] which automatically downloaded to me, but it said, your app is requesting .net 2, but .net 3.5 is best. So I downloaded it, but I am a bit iffy to go installing anything confusing without someone experienced saying go ahead... I searched the archives but the same terms are swamped by Wine discussion, or this just hasn't come up... I have looked at every hit for ".net" in the bugs section... it's not there... What should I do, or, will it just not work on this older computer? ~
R.
T.
G00:30, 24 August 2014 (UTC)
I just wonder, if I go ahead and install the 3.5 package, will it potentially damage the .net 4 currently installed? ~
R.
T.
G15:55, 24 August 2014 (UTC)
test if page exists in a custom module.
Hi,
I'm developing a
custom module for template maintenance on French Wikipédia.
I was wondering if there is an easy way to determine if another page exists or not, using some WikiFunctions or anything else ?
@
Stevietheman: Trappist the monk is not referring to an AWB custom module. Trappist is referring to the code that displays CS1 citations in Wikipedia, so there is no need to upgrade AWB to see these changes.
GoingBatty (
talk)
16:29, 24 August 2014 (UTC)
@
Stevietheman: Changes Trappist makes to the CS1 module may require additional changes to AWB. For example, the addition of hyphenated parameter name aliases may require a change to the AWB code that checks for valid citation parameters and/or removal of entries from
WP:AWB/RTP.
GoingBatty (
talk)
17:26, 24 August 2014 (UTC)
Anyone know if AWB is being used for Icelandic Wikipedia (is.wikipedia.org)? Haven't started using/learning it for that or the English one.. yet. I just noticed that it can delink months/dates with AWB. I see that a lot there - they didn't get the memo.. (delinking should be ok - no rule there explicitly says you should link or not, and I'll try get it to say don't).
Only legal date format in Iceland(ic Wikipedia) is DMY in this format: 1. janúar 2014. Note the first period and then lower case. [After the year, nothing except comma or period as other normal/English-like rules would dictate.] The months are (we used ISO8859-1/Latin1 if it matters):
Note: We abolised the z, but some still write marz for mars (or it could be found in quotes). We do not have c officially but ok to use sometimes but is incorrect for december (but could be a "typo".. for desember). Any other (unimplemented?) country specific stuff I could think of? 15:28, 22 August 2014 (UTC)
Comp.arch you can help customisation of AWB in Icelandic by providing me translations of the edit summary in Icelandic, giving me information of the names of the tags used, etc. --
Magioladitis (
talk)
06:06, 23 August 2014 (UTC)
Yes, I probably could.. "Edit summary", you mean what would end up in them? Is there a list I can find that I would look at? I'm not using AWB, maybe I should install and it will be clear. Didn't take a look at the SVN. Most Icelanders would be ok with using/understanding English help/manual if there is one (if there is a lot to translate), but the actual functionality (or anything else) regarding Icelandic issues isn't in AWB I assume and would need changes in code?
comp.arch (
talk)
13:37, 25 August 2014 (UTC)
Select all in Database scanner
Hi,
How select all the pages of a list of the Database scanner (to paste the selection)? Select all manually can be very tedious. Thanks.
Automatik (
talk)
21:43, 25 August 2014 (UTC)
@
Automatik: Click on the top line to select it, and then either press Ctrl-Shift-End, or use the scrollbar to go to the bottom of the list and then Shift-click on the last line. --
John of Reading (
talk)
06:20, 26 August 2014 (UTC)
Is it possible to use the Make Module option with the Database Scanner? I've got a custom module that dumps the contents of a template field to a file for later analysis, prior to a clean-up run in the future. The module needs to run on 19,000+ articles, there's no time pressure on the age of the data, so running via the database scanner would be better than chucking the load onto the WP servers. Does anyone know if this is possible? or have an alternative suggestion? Thanks.-
X201 (
talk)
15:19, 1 September 2014 (UTC)
No, though if you're able to write a medium/high complexity custom module you should be able to compile a custom version of AWB from SVN to replace the logic of one of the default scanner options.
Rjwilmsi09:46, 2 September 2014 (UTC)
In my current task I am often needing to create a new find+replace rule that is a modified copy of the one just above or below it. Currently to duplicate a rule I am inserting a new blank row, copy+pasting the "find", copy+pasting the "replace", and ticking some of the checkboxes. I can do this quite quickly - I've had lots of practice - but am I missing a neat way to do this? --
John of Reading (
talk)
20:15, 7 September 2014 (UTC)
You would not be out of line to make a feature request for an additional option on the right-click menu – in addition to "Insert row" and "Delete row", let's have "Duplicate row".
Christhe spelleryack14:16, 8 September 2014 (UTC)
Auto Wiki Browser
I've seen other users using this tool, and I really want to know how to use it myself. I might just have to thoroughly read the instructions in the article. Unless someone could give me some simple instructions. --
C.Syde (
talk |
contribs)
22:39, 24 September 2014 (UTC)
When I started AWB minutes ago, I received some update request named "Something 2.0.2.0". I clicked OK, but by now I cancan not relate it to AWB. First: the warning was early (way too) in the AWB-startup process, and second because my actual AWB (start screen) says version "5.5.5.0". What did I click OK for? Is this OK? -
DePiep (
talk)
21:18, 26 September 2014 (UTC)
I am trying to perform my first edit with AWB. However, the message, "API reports this user is blocked from editing" appears when I click "start". I was approved to use AWB on
July 9, 2014. What do I need to do to fix this issue? Thanks.
Mitchumch (
talk)
11:07, 28 September 2014 (UTC)
Add support for valid date formats Summer yyyy–yy and Summer yyyy–yyyy;
(discussion)
There is enough here that there is a deal of documentaion to be done. I think that I'll begin that and not bother to hide it prior to the live update – it's just easier that way.