Knowledge (XXG)

:Bots/Requests for approval/MusikBot 7 - Knowledge (XXG)

Source πŸ“

653: 1173: 494: 48: 929:.The other issue you speak of happened in week 2, I think, where it was bot operator stupidness. There was a lot of refactoring happening and this issue slipped through, and was manually fixed. Again there's so many moving parts to this task, so obviously it's hard for anyone to keep track of, which is why we want it automated!Β :) 769:
A major issue with my bot in general was discovered while testing the TAFIDaily task. That is, if I fetch the page properties in a certain way using the mediawiki-gateway gem, it escapes everything, when doing it another way does not. When I save to the page, the bot would always unescape everything.
1075:
So here the bot did everything it was supposed to (correct me if I'm wrong!), with exception of the bug in #4 which I've only now become aware of. I can simulate this on testwiki and ensure it is fixed (it just needs to go off the first embedded mb-bot comment and not all of them). Finally, as noted
518:
Note the new #5 and #6 subtasks I just added. They're coded and ready to try out. I figure if we're doing a 4-week trial I should squeeze in all of the weekly tasks into a single BRFA. I'm also going to add a on-wiki config page where you could turn off any given subtask. We just missed the mark for
823:
Do development on testwiki, but all-out testing should happen against enwiki, adding a debugger where the bot would normally save the page. This way you know exactly how the bot will perform in production, but we can catch and fix any bugs in the process without causing disruption. Had I thought of
819:
exists, and can optionally have it import any templates used on the target as well. This means I don't need to do the tedious work of copying/pasting each page from enwiki over to testwiki when testing the bot. Much of the issues outlined above would not have happened if all the relevant pages and
829:
All in all, I'm not entirely pleased with the trial, but am confident things are finally right. The only worry is as of yet, the TAFIWeekly task has not been ran by a cronjob, rather manually invoked. Next week (21 December) I hope to do this final test. The week following (28 December) I will be
794:
I did not update the mediawiki-gateway gem to point to the newest version on GitHub, which had the code for mass messaging. Prior to this it was using the copy of the gem on my local machine. Problem fixed, and I re-ran that mass message task without any issues, and also this time a timestamp was
1076:
above, this did not run via a cronjob on WMF Labs. That's the last remaining piece as far as I can tell. So clearly an extension to the trial is needed, but a mere two weeks at most. We can skip the last week of 2015 and allow it to be manually done, if you want, since I won't be around.
1153:
Hoping to allow a one/two week extension to prove this bot does the job (see week 4 above, and the minor fixes that went with it). If an extension is not approved by Sunday 00:00 GMT, we will need to do all of the above manually. Please let me know so I can notify the TAFI team. Thanks!
460:
This is all done in under 300 lines of code (a quarter of which is just handling API errors =P). I've tested as much as I can on testwiki, but obviously the real testing will have to be done in production. For the trial I will run each subtask individually, fully-monitored.
450: 671:
First and foremost, both MediaWiki and the labs server consider Monday to be the first day of the week. I did some tests and could have sworn it was Sunday. The task was quickly aborted once I realized this, and attempted again a day
925:, and this was per request. Apparently the rotation was done in the past by Theo's Little Bot, and it stopped for whatever reason. If the TAFI team doesn't want that to happen that's no problem, it can be turned off in the 486:
This is very much a housekeeping task, so there's not much room for problems as long as you follow what's being requested. I don't like giving long trials, but I figure one is necessary here since this is a weekly task.
378: 679: 443: 830:
off-wiki and unable to monitor the bot for errors. If next week goes good, I think we'll be alright. I will be sure to inform the TAFI team that I will be away during the 28 December run.
365:
part. I've found this to be the most reliable way to identify which WikiProjects we're working with. Categories and the names of the templates vary too much and are unreliable, it appears
1218: 686:. I thought maybe on testwiki these magic words just didn't exist. Lesson learned, and also all magic words are wrapped in two curly braces, not three, so there should be no confusion. 348:
currently does this, but apparently not all the time. The operator is inactive, and since MusikBot will be doing the whole process we figure it should take on this little task as well.
312: 302: 291: 922: 231: 770:
This is not a problem unless there are HTML entities somewhere on the page. The cross-task bug fix left some parts of the code accidentally unfixed, leading to
1179:
Those edits all look fine to me, and there are no objections or apparent unresolved issues. If further problems come up, they can be easily dealt with. β€”
1015: 749: 21: 990:. Notice the picture and caption links are not red-linked as they should be (since those subpages aren't created by the bot). This was because 382: 885:
Well, there was a problem when MusikBot put in Week 1 in 2015 after Week 53 at one point. Well, the potential problems with reorganising the
286:
There are a lot of elements to this task, but the implementation is fairly straightforward. I will run through the process, start to finish:
306: 153: 1112: 905: 824:
this sooner I think there would have been near to no bad edits during this trial, and the same is true with all of my other bot tasks.
606: 571: 535:
Noting I added another task, #9 above. This leaves one more to tackle, which I hope to have done by round two of the trial next week.
994: 756: 716: 295: 141: 88: 735:
auto-supplies a heading, when the API wants it's own heading. This was not clear to me, and we ended up with duplicate headings
240: 83: 659:
Not going to lie, getting this thing off the ground was rough, for an array of reasons. Here's a rundown of what happened:
1122: 1026: 498:
Approved for trialΒ (4 weeks). Please provide a link to the relevant contributions and/or diffs when the trial is complete.
491:
has options to disable individual tasks, so you can use those to turn off the particular one(s) that are being replaced.
1147: 729: 369: 165: 926: 437:
with the appropriate values. If the class assessment has gone unchanged these parameters are not passed to the template
298:}} to the end of the 2nd section. A Lua module in the template generates what will be the new TAFI in the coming weeks. 327:
that contains the name is also created. The bot does not attempt to choose an image and caption for the article (yet!)
118: 103: 464:
Note also there's more TAFI weekly tasks to come... but I figured this is a sizable "start", to say the least.
833:
So at the very least, I'm asking for one more run next week to test via a cronjob. Many thanks for your time!
488: 345: 1185: 1165: 1135: 1116: 1090: 960: 940: 909: 880: 844: 634: 610: 589: 575: 546: 530: 506: 475: 159: 1097: 1108: 901: 602: 567: 147: 1200: 985: 802:
had links to 2015 instead of 2016. For once this was not my fault, but a bug in the template, which I've
799: 40: 853: 135: 980:
I thought maybe if list out the most recent run, step by step, it might help you all evaluate the bot:
1054:
to last week's AFI talk page, including the class assessment, if it changed at all (here it did not)
1048: 693: 431: 409: 222: 208: 98: 890: 93: 781:
The mass messages for week 2 and 3 did not contain a timestamp, meaning bots will not archive them.
418:
Check the revision of the talk page a week prior and get the article assessment (stub, start, etc.)
850:
Whoa. Well, I'm gonna admit difficulty reviewing this trial as I'm not involved with the project.
620: 556: 78: 946: 351:
Notifying relevant WikiProjects that the new TAFI is within their scope. The bot does this by:
1101: 916: 894: 869: 595: 560: 519:
this week but the bot will be ready to go (supervised) for next Sunday at 00:00 GMT. Thanks!
17: 1158: 1128: 1083: 953: 949:
that is still alphabetized. You can use this to quickly find the section you're looking for
933: 861: 837: 627: 539: 523: 468: 130: 1121:
I see what you mean, it should use the abbreviated month name and not the full month name.
945:
Did want to mention however there is still the navigation at the top and the bottom of the
1180: 875: 619:(calendar week). I can never remember those strftime directives. Also how cool is it that 584: 513: 501: 889:
titles is that I cannot find the correct sections well enough. I usually scroll down the
816: 580: 1212: 748:
For the TAFIDaily task, I changed the location of the list of nominations to be on a
396: 338: 258: 59: 313:
Knowledge (XXG):Today's articles for improvement/Schedule#Preparing a scheduled week
865: 305:
to see what was chosen as the new TAFI, and remove it from the list of articles at
555:
Wait a sec, there's no such thing as Week 53. Surely the bot should've scheduled
1079:
Thanks to everyone for both their help and patience in working with me on this!
857: 615:
The Epoch converter is a pretty neat document, though with Ruby you can just do
354:
Checking the talk page of the article, using the API to get it's raw parsed HTML
217: 1007:
Create subpages for the new scheduled selection (except picture and caption)
874:
could you folks chime in? Is it working as expected? Any potential issues? β€”
594:
Oh, either 52 or 53. Sorry I didn't realise this. I was just too systematic.
1100:, the dates consistently put either, for example, 9 Aug 2014 or 9 Aug 2015. 1000:
was hardcoded to 2015 and not the current year (not the bot's fault). I've
893:
page to add another nomination to one of the alphabetised Level 2 titles.
762:
Lua module was still looking in the wrong place, however, producing
451:
Knowledge (XXG):Today's articles for improvement/Accomplishments/row
712:
After doing some refactoring of the code, the start date passed to
921:
The rotating of the level 2 headings you speak of are part of the
680:
Knowledge (XXG):Today's articles for improvement/Schedule/Preload
424:
Get the current revision ID of the article and current assessment
1193:
The above discussion is preserved as an archive of the debate.
678:
and others are not magic words, they are parameters passed to {{
1025:(apparently this made an extraneous entry in 2015, which is an 1199:
To request review of this BRFA, please start a new section at
702:
Some issues with adding the accomplishments, which were fixed.
195: 39:
To request review of this BRFA, please start a new section at
311:
Set up the schedule pages for the new TAFI, as described at
667:
Did not go well. A little embarrassing, but here we go:
415:
to the article's talk page. This involves a few things:
1070: 1065: 1060: 1055: 1040: 1035: 1030: 1023: 1018: 1010: 1008: 1001: 988: 803: 775: 771: 763: 736: 683: 230:
Carries out most of the weekly tasks involved with the
183: 177: 171: 113: 108: 73: 725:
First time notifying WikiProjects was attempted. The
303:
Knowledge (XXG) talk:Today's articles for improvement
292:
Knowledge (XXG) talk:Today's articles for improvement
699:
to article pages, after a few tries this was fixed.
675: 623:will coincidentally be the TAFI over new years? =P 616: 389: 362: 358: 331: 316: 1219:Approved Knowledge (XXG) bot requests for approval 238:Links to relevant discussions (where appropriate): 232:Knowledge (XXG):Today's articles for improvement 33:The following discussion is an archived debate. 421:Get the revision ID of the article a week prior 334:to find out what the current TAFI is, and add 8: 1022:Add new scheduled selection to the archives 1014:Remove the new scheduled selection from the 778:. Ultimately I got everything squared away. 323:is the current date + 3 weeks. The subpage 357:Scan the markup for hyperlinks linking to 241:Special:PermaLink/690468498#Bot automation 307:Knowledge (XXG):Articles for improvement 791:The cronjob didn't start due to a typo. 377:Send a mass message to everyone on the 7: 392:for last week's TAFI. Then, remove 383:subst:TAFI weekly selections notice 252:Estimated number of pages affected: 738:. I repaired those pages manually. 372:}} to those WikiProject talk pages 194:02:08, Monday, November 16, 2015 ( 28: 976:Week 4 (14 December) step-by-step 798:The added scheduled selection at 315:. The bot creates the base page ( 290:Add a new "scheduled article" to 202:Automatic, Supervised, or Manual: 45:The result of the discussion was 1171: 651: 492: 46: 984:Add new scheduled selection to 296:subst:TAFI scheduled selection 294:. This involves only adding {{ 1: 1186:10:20, 20 December 2015 (UTC) 1166:19:20, 19 December 2015 (UTC) 1136:15:59, 16 December 2015 (UTC) 1117:06:28, 16 December 2015 (UTC) 1091:01:04, 16 December 2015 (UTC) 1069:Mass message project members 1059:Notify relevant WikiProjects 961:22:36, 15 December 2015 (UTC) 941:22:35, 15 December 2015 (UTC) 910:22:09, 15 December 2015 (UTC) 881:21:57, 15 December 2015 (UTC) 845:04:45, 14 December 2015 (UTC) 547:05:26, 23 November 2015 (UTC) 531:16:07, 16 November 2015 (UTC) 507:07:07, 16 November 2015 (UTC) 476:02:08, 16 November 2015 (UTC) 722:ended up being wrong. Fixed. 635:23:58, 8 December 2015 (UTC) 611:22:06, 8 December 2015 (UTC) 590:21:37, 8 December 2015 (UTC) 576:21:32, 8 December 2015 (UTC) 359:Knowledge (XXG):WikiProject_ 344:to the top of that article. 1235: 361:, using regex to grab the 370:subst:TAFI project notice 1196:Please do not modify it. 995:TAFI scheduled selection 820:templates were in place. 795:included in the message. 757:TAFI scheduled selection 717:TAFI scheduled selection 689:Some issues with adding 381:with the message body {{ 36:Please do not modify it. 1039:Detag last week's AFI 272:Already has a bot flag 22:Requests for approval 1064:Add accomplishments 786:Week 4 (14 December) 707:Week 2 (30 November) 663:Week 1 (23 November) 363:WikiProject_whatever 209:Programming language 18:Knowledge (XXG):Bots 1148:BAGAssistanceNeeded 743:Week 3 (7 December) 730:TAFI project notice 559:to Week 1 of 2016? 259:Exclusion compliant 774:when it should be 621:Person of the Year 617:DateTime.now.cweek 557:Person of the Year 228:Function overview: 489:Theo's Little Bot 402:from that article 379:notification list 346:Theo's Little Bot 284:Function details: 1226: 1198: 1183: 1175: 1174: 1164: 1163: 1161: 1152: 1146: 1134: 1133: 1131: 1089: 1088: 1086: 1053: 1047: 1034:Tag current AFI 999: 993: 959: 958: 956: 947:nomination board 939: 938: 936: 920: 878: 873: 854:Northamerica1000 843: 842: 840: 761: 755: 734: 728: 721: 715: 698: 692: 677: 655: 654: 633: 632: 630: 618: 587: 545: 544: 542: 529: 528: 526: 517: 504: 496: 495: 474: 473: 471: 455: 449: 436: 430: 414: 408: 401: 395: 391: 364: 360: 343: 337: 333: 326: 322: 318: 188: 187: 50: 49: 38: 1234: 1233: 1229: 1228: 1227: 1225: 1224: 1223: 1209: 1208: 1207: 1194: 1181: 1172: 1159: 1157: 1155: 1150: 1144: 1129: 1127: 1125: 1084: 1082: 1080: 1051: 1045: 997: 991: 954: 952: 950: 934: 932: 930: 914: 876: 851: 838: 836: 834: 759: 753: 732: 726: 719: 713: 696: 690: 657:Trial complete. 652: 649: 628: 626: 624: 585: 540: 538: 536: 524: 522: 520: 511: 502: 493: 483: 469: 467: 465: 453: 447: 446:page using the 444:accomplishments 434: 428: 412: 406: 399: 393: 341: 335: 324: 320: 246:Edit period(s): 133: 129: 124: 63: 47: 34: 26: 25: 24: 12: 11: 5: 1232: 1230: 1222: 1221: 1211: 1210: 1206: 1205: 1189: 1188: 1143: 1141: 1140: 1139: 1138: 1073: 1072: 1067: 1062: 1057: 1042: 1037: 1032: 1020: 1012: 1005: 978: 977: 972: 970: 969: 968: 967: 966: 965: 964: 963: 923:TAFIDaily task 828: 826: 825: 821: 817:Special:Import 813: 812: 808: 807: 796: 792: 788: 787: 783: 782: 779: 767: 745: 744: 740: 739: 723: 709: 708: 704: 703: 700: 687: 684:Stupid mistake 673: 665: 664: 648: 645: 644: 643: 642: 641: 640: 639: 638: 637: 553: 552: 551: 550: 549: 482: 479: 458: 457: 440: 439: 438: 425: 422: 419: 403: 386: 375: 374: 373: 366: 355: 349: 328: 309: 299: 123: 122: 116: 111: 106: 101: 96: 91: 86: 81: 76: 74:Approved BRFAs 71: 64: 62: 57: 56: 55: 29: 27: 15: 14: 13: 10: 9: 6: 4: 3: 2: 1231: 1220: 1217: 1216: 1214: 1204: 1202: 1197: 1191: 1190: 1187: 1184: 1178: 1170: 1169: 1168: 1167: 1162: 1149: 1137: 1132: 1124: 1120: 1119: 1118: 1114: 1110: 1106: 1104: 1099: 1098:WP:TAFIACCOMP 1095: 1094: 1093: 1092: 1087: 1077: 1071: 1068: 1066: 1063: 1061: 1058: 1056: 1050: 1043: 1041: 1038: 1036: 1033: 1031: 1028: 1024: 1021: 1019: 1017: 1013: 1011: 1009: 1006: 1003: 996: 989: 987: 983: 982: 981: 975: 974: 973: 962: 957: 948: 944: 943: 942: 937: 928: 924: 918: 913: 912: 911: 907: 903: 899: 897: 892: 888: 884: 883: 882: 879: 871: 867: 863: 859: 855: 849: 848: 847: 846: 841: 831: 822: 818: 815: 814: 810: 809: 805: 801: 797: 793: 790: 789: 785: 784: 780: 777: 773: 768: 765: 758: 751: 750:separate page 747: 746: 742: 741: 737: 731: 724: 718: 711: 710: 706: 705: 701: 695: 688: 685: 681: 674: 670: 669: 668: 662: 661: 660: 658: 647:Trial results 646: 636: 631: 622: 614: 613: 612: 608: 604: 600: 598: 593: 592: 591: 588: 582: 579: 578: 577: 573: 569: 565: 563: 558: 554: 548: 543: 534: 533: 532: 527: 515: 510: 509: 508: 505: 499: 490: 485: 484: 480: 478: 477: 472: 462: 452: 445: 441: 433: 426: 423: 420: 417: 416: 411: 404: 398: 387: 384: 380: 376: 371: 367: 356: 353: 352: 350: 347: 340: 329: 314: 310: 308: 304: 300: 297: 293: 289: 288: 287: 285: 281: 279: 276: 273: 269: 267: 264: 261: 260: 255: 253: 249: 247: 243: 242: 239: 235: 233: 229: 225: 224: 221: 219: 214: 212: 210: 205: 203: 199: 197: 193: 189: 185: 182: 179: 176: 173: 170: 167: 164: 161: 158: 155: 152: 149: 146: 143: 140: 137: 132: 128: 120: 117: 115: 112: 110: 107: 105: 102: 100: 97: 95: 92: 90: 87: 85: 82: 80: 77: 75: 72: 70: 66: 65: 61: 58: 53: 44: 42: 37: 31: 30: 23: 19: 1195: 1192: 1176: 1142: 1102: 1078: 1074: 979: 971: 917:Qwertyxp2000 895: 886: 870:Qwertyxp2000 832: 827: 666: 656: 650: 596: 561: 497: 463: 459: 283: 282: 277: 274: 271: 270: 265: 262: 257: 256: 251: 250: 245: 244: 237: 236: 227: 226: 216: 215: 207: 206: 201: 200: 191: 190: 180: 174: 168: 162: 156: 150: 144: 138: 126: 125: 68: 51: 35: 32: 1160:MusikAnimal 1130:MusikAnimal 1105:wertyxp2000 1085:MusikAnimal 1049:Former TAFI 955:MusikAnimal 935:MusikAnimal 898:wertyxp2000 862:SMcCandlish 839:MusikAnimal 694:Former TAFI 629:MusikAnimal 599:wertyxp2000 564:wertyxp2000 541:MusikAnimal 525:MusikAnimal 470:MusikAnimal 442:Update the 432:Former TAFI 410:Former TAFI 218:Source code 192:Time filed: 131:MusikAnimal 1123:No problem 1096:Umm... In 891:WP:TAFINOM 764:this error 676:{{{week}}} 514:The Earwig 481:Discussion 220:available: 204:Automatic 184:ANI search 178:rights log 166:page moves 154:edit count 114:rights log 104:page moves 60:MusikBot 7 1177:Approved. 811:Takeaways 456:template. 172:block log 127:Operator: 109:block log 1213:Category 1113:contribs 1027:easy fix 906:contribs 607:contribs 572:contribs 321:new_date 319:) where 275:(Yes/No) 263:(Yes/No) 234:project 142:contribs 84:contribs 52:Approved 20:‎ | 1201:WT:BRFA 986:WT:TAFI 887:Level 2 868:, and 866:Coin945 800:WT:TAFI 581:Well... 248:Weekly 41:WT:BRFA 1182:Earwig 927:config 877:Earwig 858:Evad37 752:. The 672:later. 586:Earwig 503:Earwig 388:Check 368:Add {{ 330:Check 301:Check 223:GitHub 1002:fixed 804:fixed 213:Ruby 89:count 16:< 1109:talk 1044:Add 1016:list 1004:this 902:talk 776:this 772:this 682:}}. 603:talk 568:talk 427:Add 405:Add 397:TAFI 339:TAFI 280:Yes 268:Yes 211:(s): 160:logs 136:talk 119:flag 99:logs 79:talk 69:BRFA 196:UTC 148:SUL 94:SUL 1215:: 1156:β€” 1151:}} 1145:{{ 1126:β€” 1115:) 1111:| 1081:β€” 1052:}} 1046:{{ 1029:) 998:}} 992:{{ 951:β€” 931:β€” 908:) 904:| 864:, 860:, 856:, 835:β€” 760:}} 754:{{ 733:}} 727:{{ 720:}} 714:{{ 697:}} 691:{{ 625:β€” 609:) 605:| 583:β€” 574:) 570:| 537:β€” 521:β€” 500:β€” 466:β€” 454:}} 448:{{ 435:}} 429:{{ 413:}} 407:{{ 400:}} 394:{{ 385:}} 342:}} 336:{{ 325:/1 254:7 198:) 1203:. 1107:( 1103:Q 919:: 915:@ 900:( 896:Q 872:: 852:@ 806:. 766:. 601:( 597:Q 566:( 562:Q 516:: 512:@ 390:] 332:] 317:] 278:: 266:: 186:) 181:Β· 175:Β· 169:Β· 163:Β· 157:Β· 151:Β· 145:Β· 139:Β· 134:( 121:) 67:( 54:. 43:.

Index

Knowledge (XXG):Bots
Requests for approval
WT:BRFA
MusikBot 7
BRFA
Approved BRFAs
talk
contribs
count
SUL
logs
page moves
block log
rights log
flag
MusikAnimal
talk
contribs
SUL
edit count
logs
page moves
block log
rights log
ANI search
UTC
Programming language
Source code
GitHub
Knowledge (XXG):Today's articles for improvement

Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.

↑