User talk:Zhuyifei1999/Archive 25

From Wikimedia Commons, the free media repository
Jump to navigation Jump to search

Your Bot YiFeiBot

Your bot found several files to be Pages_using_Information_template_with_parsing_errors [1] but they don't have parsing errors at all. There must be a mistake with the bot.--ProfessorX (talk) 19:26, 6 March 2017 (UTC)

The edit must have been triggered by special:diff/234925547, and somehow the data in the database was no longer valid. The bot has a script that queries the database for false positives and undo the edits, but idk why the query no longer works. I guess they have to be reverted manually :/ --Zhuyifei1999 (talk) 09:35, 7 March 2017 (UTC)

MediaWiki:Gadget-Cat-a-lot.js/zh-hant

感謝修改,可是我發現還是不對——我現在都想哭出來了。MediaWiki:Gadget-Cat-a-lot.js/zh-hantMediaWiki:Gadget-Cat-a-lot.js/zh-hans那個字串'cat-a-lot-xxxxx'後面現在是駝峰命名,得改成小寫和短線式的。我還說我測試過,請允許我找個地方哭一會兒。--Gyakusyuu no Amanojaku (talk) 13:39, 10 March 2017 (UTC)

... 我马上改 --Zhuyifei1999 (talk) 13:41, 10 March 2017 (UTC)
就像Special:Diff/236624454,嗚嗚嗚嗚嗚嗚……--Gyakusyuu no Amanojaku (talk) 13:43, 10 March 2017 (UTC)
我之前是仅仅检查了我子前将要做的更改,每看原文,下次我也要注意一点 --Zhuyifei1999 (talk) 13:50, 10 March 2017 (UTC)

MediaWiki:Gadget-Cat-a-lot.js

Hi! Do you have idea why in MediaWiki:Gadget-Cat-a-lot.js/translating there is

//DO NOT CHANGE THE NEXT LINE 
if (mw.config.get('wgUserLanguage') !== 'en') {
//DO NOT CHANGE THE PREVIOUS LINE

while in MediaWiki:Gadget-Cat-a-lot.js/de (but also e.g. in pl version) there is

//DO NOT CHANGE THE NEXT LINE 
if (mw.config.get('wgUserLanguage') === 'de') {
//DO NOT CHANGE THE PREVIOUS LINE

Rillke is the author of de version and last revisions of the template/"translate". IMO mw.config.get('wgUserLanguage') === 'LANGCODE' is proper condition. --jdx Re: 14:15, 10 March 2017 (UTC)

In fact, I don't see a reason these conditions should exist at all. These l10n are loaded via loadLocalization = function( lang, cb ) in MediaWiki:Gadget-Cat-a-lot.js, and this function is called conditionally when:
if ( mw.config.get( 'wgUserLanguage' ) !== 'en' ) {
	loadLocalization( mw.config.get( 'wgUserLanguage' ), maybeLaunch );
}
if ( mw.config.get( 'wgContentLanguage' ) !== 'en' ) {
	loadLocalization( mw.config.get( 'wgContentLanguage' ), maybeLaunch );
}
These additional conditions added to the l10n subpages, whether if (mw.config.get('wgUserLanguage') !== 'en') { or if (mw.config.get('wgUserLanguage') === 'de') {, not only seems unnecessary, but also appear to make wgContentLanguage in MediaWiki:Gadget-Cat-a-lot.js block almost useless (or in extreme cases, invoke unexpected behaviours) --Zhuyifei1999 (talk) 14:38, 10 March 2017 (UTC)
Thanks! I've just looked at MediaWiki:Gadget-Cat-a-lot.js and I agree with you. But I'm not a JavaScript/MediaWiki guru. --jdx Re: 14:55, 10 March 2017 (UTC)

MediaWiki:Gadget-Cat-a-lot.js/zh-hans

阁下您好,可否将Special:Permalink/236709354的内容覆盖至MediaWiki:Gadget-Cat-a-lot.js/zh-hansSpecial:Permalink/236709773的内容覆盖至MediaWiki:Gadget-Cat-a-lot.js/zh-hant,打扰到阁下还望见谅。--Jerre Jiang  Talk  06:57, 11 March 2017 (UTC)

✓ Done 没事 --Zhuyifei1999 (talk) 07:07, 11 March 2017 (UTC)

Untitled

Ok. Please, let then the images uploaded,

E. — Preceding unsigned comment added by Egargo (talk • contribs) 13:51, 18 August 2016 (UTC)

15:25, 13 March 2017 (UTC)

22:03, 20 March 2017 (UTC)

Hey Molly rose (talk) 20:10, 26 March 2017 (UTC)

A barnstar for you!

The Technical Barnstar
It was really nice of you to take quite a bit of your time to help me with the pop-up issue. Now I can do the work I want to without worrying about it. Many thanks! :-) Daphne Lantier 20:19, 22 March 2017 (UTC)
  • If/when it becomes safe to use that favorites gadget, can you leave me a quick note at my talk or ping me to wherever it's been resolved? Thanks. No hurry - I only use it occasionally. Daphne Lantier 20:22, 22 March 2017 (UTC)
    No problem. Though, I'm not sure how long it'll take before it is fixed; it really depends on whom I pinged --Zhuyifei1999 (talk) 00:43, 23 March 2017 (UTC)
    They're all currently active, so hopefully they'll fix it up pretty soon. I like being able to show appreciation to Commons photographers by faving their pics, and I'm sure plenty of other people here do too. Daphne Lantier 01:52, 23 March 2017 (UTC)

Hey Molly rose (talk) 20:10, 26 March 2017 (UTC)

14:46, 27 March 2017 (UTC)

Could you look at this file? The embedded data bot tagged it, but it seems to be valid. Pinging @Jcb who also deleted and undeleted the file. --jdx Re: 15:04, 29 March 2017 (UTC)

Yes, I noticed just a second too late that there was a deletion and an undeletion in the log instead of two deletions. Jcb (talk) 15:19, 29 March 2017 (UTC)
PNG and JPEG use the parser from Pillow and has some amount of likely false positives. The image data has ended (which halts parsing by Pillow) but the file continues for unknown reasons. :( --Zhuyifei1999 (talk) 16:11, 29 March 2017 (UTC)

17:53, 3 April 2017 (UTC)

Could you help, pleaseʔ

Helloǃ As I guess, this is [22] your site. I want to download full lenghth map from gallica.bnf.fr, but there in the site it says "Job did not seem to succeed. ....". Coulod you, please, help me with this? Regards, - Abkhazian1 (talk) 17:37, 4 April 2017 (UTC)

The script is broken for a while and I'm rewriting it. It'll be available during May hopefully --Zhuyifei1999 (talk) 18:07, 4 April 2017 (UTC)
@Abkhazian1: --Zhuyifei1999 (talk) 18:08, 4 April 2017 (UTC)
Thank youǃ Will be waiting for it. - Abkhazian1 (talk) 21:25, 4 April 2017 (UTC)

Missing file

Hello. I just wanted to let you know about this edit by YifeiBot to a redirect page. Not a major problem though. Green Giant (talk) 04:07, 5 April 2017 (UTC)

Well, it's technically no longer a redirect after Special:Diff/238851102. --Zhuyifei1999 (talk) 04:52, 5 April 2017 (UTC)
True, but I was thinking more along the lines of whether the bot could recognise there isn't a file attached to the page. I'm not saying this has happened before but given the number of pages in Category:Media missing infobox template, is there any way of checking if this has happened before? Green Giant (talk) 08:34, 5 April 2017 (UTC)
Not sure. I might be able to brute force this by checking all of the bots edits, but that would take a long long time... and probably not worth it. Maybe some SQL experts like Dispenser can help, though I really doubt the effort will worth it. --Zhuyifei1999 (talk) 09:19, 5 April 2017 (UTC)
Fair enough. Like I said it's not a major problem. :) Green Giant (talk) 17:00, 5 April 2017 (UTC)

Patrolling after tagging the file for speedy deletion

Is it possible for Embedded Data Bot to patrol the first revision of a file containing embedded data after they tagged it for speedy deletion? It is unnecessary for us patrollers to mark those files as patrolled since it is already tagged. Thanks, Poké95 09:22, 5 April 2017 (UTC)

It would need a minor rewrite -- The current queue system is pagetitle-based, not rcid-based, and to patrol the change you need the rcid (unless the bot re-parse the version history). I currently don't have too much time for non-mission-critical functionalities, though you or someone else may submit a pull request at - well, you know where (can't say it too publicly as idk if some abusers are watching this page) - and I'll be happy to review and deploy it. --Zhuyifei1999 (talk) 09:35, 5 April 2017 (UTC)

You reverted the bot's error at this file. Any idea what could have gone wrong? I'm not particularly familiar with the bot's operations, so maybe there's an obvious reason that I've overlooked. My reason for asking is that I'm hesitant to delete images under the "contains embedded data" criterion if there's more than a tiny chance that the bot has mis-tagged a page. Nyttend (talk) 22:29, 5 April 2017 (UTC)

(talk page stalker) @Nyttend: FWIW, the bot had flagged it as 'unknown mime'... most of the files that end up in the suspect category specifically state they have an embedded RAR archive. - Reventtalk 23:12, 5 April 2017 (UTC)
Thank you, but how would that happen? Both the extension and the "MIME type" line (below the image itself) say that it's an ordinary PNG. Nyttend (talk) 23:39, 5 April 2017 (UTC)
@Nyttend: exiftool indicates "Warning : [minor] Trailer data after PNG IEND chunk". My guess would be that whatever program was used to create it uses PNG as a native file format, and appends info that's not really part of the standard. Adobe Fireworks, in particular, tends to do that. - Reventtalk 23:52, 5 April 2017 (UTC)
(Edit conflict)The file is padded with a ton of space characters (ascii 0x20), and the bot thought that as some form of embedded data. And no, the whole point of this embedded data detection is finding them from otherwise ordinary files. --Zhuyifei1999 (talk) 23:53, 5 April 2017 (UTC)

POTY question

Do you know the answer to this question [23] about the contest rules? Thanks Royalbroil 13:49, 8 April 2017 (UTC)

Thanks for the notification. The rules should be three only as seen in Commons:Picture_of_the_Year/2016/Rules#Voting_rules. I've notified Steinsplitter about the issue; he's more familiar with the translation and interface than I do. --Zhuyifei1999 (talk) 14:08, 8 April 2017 (UTC)
Thanks to both of you, @Zhuyifei1999: and @Steinsplitter: , for fixing it! I was just getting around to explain which page had the issue. Royalbroil 14:12, 8 April 2017 (UTC)

Embedded data bot

Hello. The Arabic-speaking Wiki Zero pirates have moved onto .tif files. I count 15 or 20 different users today. Example: File:Windows.tif. It would be great if you could search for this. Magog the Ogre (talk) (contribs) 03:19, 31 March 2017 (UTC)

I don't have much time, but I'll try. :( --Zhuyifei1999 (talk) 03:34, 31 March 2017 (UTC)
@Magog the Ogre: FYI: Special:Diff/239647361 --Zhuyifei1999 (talk) 08:22, 3 April 2017 (UTC)
Awesome! Bad news, though; they're doing XCF too now. Magog the Ogre (talk) (contribs) 00:47, 4 April 2017 (UTC)
Oh wait, you already got that. What about this file though? Xcf.xcf. Magog the Ogre (talk) (contribs) 00:48, 4 April 2017 (UTC)
Running the script locally:
$ python -c 'import detection; print detection.detect("testfiles/Xcf.xcf")'
{'mime': ('application/x-rar', 'RAR archive data, v81,'), 'pos': 95979, 'posexact': True}
$ unrar L testfiles/Xcf.xcf

UNRAR 5.30 beta 2 freeware      Copyright (c) 1993-2015 Alexander Roshal

Enter password (will not be echoed) for testfiles/Xcf.xcf: 

Checksum error in the encrypted file testfiles/Xcf.xcf. Corrupt file or wrong password.
Archive: testfiles/Xcf.xcf
Details: RAR 4, SFX, encrypted headers
  0 files
On labs the bot use the latest version of the file utility (the mime-sniffing one) from git so the file type details (RAR archive data, v81,) may look different if the bot went through this file. Btw: the bot runs through the RC so if anything was not implemented when the bot receive the RC entry then uh oh :/ --Zhuyifei1999 (talk) 06:18, 4 April 2017 (UTC)
Could you take a look at these two files? File:Kol-hamza.jpg, File:Hachtag12.jpg. Magog the Ogre (talk) (contribs) 03:06, 11 April 2017 (UTC)
$ grep 'Rar' Kol-hamza.jpg
$ unrar L Kol-hamza.jpg

UNRAR 5.30 beta 2 freeware      Copyright (c) 1993-2015 Alexander Roshal

Kol-hamza.jpg is not RAR archive
$ xxd Kol-hamza.jpg | grep 'ffd9'
00003200: feb8 dfe7 1253 ffd9 0038 4249 4d04 2100  .....S...8BIM.!.
000148d0: 5ef2 59fd ffd9 f6d0 d556 6f3b b76b b893  ^.Y......Vo;.k..
$ xxd Kol-hamza.jpg | grep 'ff d9'
00001830: b8df e712 53ff d9ff ed1a 4650 686f 746f  ....S.....FPhoto
0001b5e0: 95ff d92a bc95 43bc 7373 a017 3915 1bbb  ...*..C.ss..9...
$ python -c 'import detection; print detection.detect("testfiles/Kol-hamza.jpg")'Traceback (most recent call last):
  File "detection/pillow.py", line 35, in detect
    image.tobytes()
  File "/home/zhuyifei1999/embeddeddata/env/local/lib/python2.7/site-packages/PIL/Image.py", line 676, in tobytes
    self.load()
  File "/home/zhuyifei1999/embeddeddata/env/local/lib/python2.7/site-packages/PIL/ImageFile.py", line 244, in load
    raise_ioerror(err_code)
  File "/home/zhuyifei1999/embeddeddata/env/local/lib/python2.7/site-packages/PIL/ImageFile.py", line 59, in raise_ioerror
    raise IOError(message + " when reading image file")
IOError: broken data stream when reading image file
WARNING: FIXME: Failed detection
None
Corrupted file? --Zhuyifei1999 (talk) 04:04, 11 April 2017 (UTC)

18:34, 10 April 2017 (UTC)

Double flickr review

FlickreviewR 2 seems to review every file twice at the moment. I guess this is not the intended behaviour. --Didym (talk) 21:04, 10 April 2017 (UTC)

Should be fixed. Somehow the tool labs grid started another job in tools-exec-1414 when there is already one on tools-exec-1402 --Zhuyifei1999 (talk) 23:56, 10 April 2017 (UTC)

A barnstar for you!

The Technical Barnstar
For your work on User:Embedded Data Bot, really impressive. Thibaut120094 (talk) 00:26, 11 April 2017 (UTC)
Thanks --Zhuyifei1999 (talk) 03:57, 11 April 2017 (UTC)

whats up

All i did was upload a picture — Preceding unsigned comment added by Kingkg012 (talk • contribs) 16:59, 11 April 2017 (UTC)

@Kingkg012: Have you read the tutorial on the right? --Zhuyifei1999 (talk) 17:04, 11 April 2017 (UTC)

i look over it — Preceding unsigned comment added by Kingkg012 (talk • contribs) 18:03, 11 April 2017 (UTC)

I will thx--KarinaMartinez1 (talk) 18:07, 16 April 2017 (UTC)KarinaMartinez1

Source code of FlickreviewR 2

Hi Zhuyifei, firstly many thanks for your work on bots which keeps the functioning smooth and easy. Secondly, I request you to provide me with the source code of User:FlickreviewR 2. We are in process of developing a bot to review files from Category:Unreviewed photos from indiannavy.nic.in. So the source code of FlickreviewR 2 will help us to make that. --Krishna Chaitanya Velaga (talk · mail) 01:34, 17 April 2017 (UTC)

Sure. [28] All of the three bots have their scripts in that repo. Though, I'm too lazy to organize those scattered scripts and publish to somewhere like GitHub. --Zhuyifei1999 (talk) 02:07, 17 April 2017 (UTC)
Many thanks for providing that. I confused between which one is to be taken as a reference. --Krishna Chaitanya Velaga (talk · mail) 09:13, 17 April 2017 (UTC)