/srv/irclogs.ubuntu.com/2010/11/22/#ubuntu-bugs.txt

lifelesscharlie-tca: UI :)00:07
abuDawudCan someone confirm if this bug needs to go to 'Invalid' status? https://bugs.launchpad.net/ubuntu/+source/empathy/+bug/67833002:00
ubot2Launchpad bug 678330 in empathy (Ubuntu) "Empathy 2.32.2 Stable Release Update Request (affects: 1) (heat: 6)" [Undecided,New]02:00
micahgabuDawud: no, that's fine02:01
abuDawudSo then the upstream data should be posted and it should be wishlisted?02:02
micahgabuDawud: no, it's an Ubuntu specific request02:02
micahgabuDawud: well, wishlisted yes, but that's it02:07
abuDawudmicahg, still learning please excuse me, with an Ubuntu specific request the options would be to send for packaging, or to close it as invlaid and refer to brainstorm02:07
abuDawudcorrect?02:07
micahgabuDawud: what do you mean send for packaging?02:08
abuDawudset to wishlist and... I guess thats it02:09
abuDawudmicahg, we don't actually assign to MOTU or anything correct?02:09
micahgabuDawud: correct02:09
abuDawudokay then, since I don't have privileges to set to wishlist could someone do that for me, and in the future is IRC really the best place to note something needing to be set to this status?02:10
micahgabuDawud: yes02:11
micahgabuDawud: also, the reporter of that was a bug control member02:11
abuDawudmicahg, I saw he was a regular on launchpad, could he not wishlist it himself?02:12
micahgabuDawud: yes, he could02:12
micahgI've set it though02:12
abuDawudthanks.02:12
abuDawudmicahg, got a sec for another question? If a bug can't be reproduced using the listed steps, whats the process02:13
abuDawudeg. https://bugs.launchpad.net/ubuntu/+source/bash/+bug/67835402:13
ubot2Launchpad bug 678354 in bash (Ubuntu) "after upgrade to 10.10 alt + . is not working anymore (affects: 1) (heat: 6)" [Undecided,New]02:13
micahgabuDawud: do you mind if we continue in about 15 minutes?02:13
abuDawudmicahg, no problem, thanks for your help02:13
charlie-tcaabuDawud: on the bash bug, you might ask the reporter for specific steps to reproduce it. The way the report reads, you don't actually know what is being done.02:18
charlie-tcaAlso, sometimes the specific keyboard layout used matters. I use US Int'l , which is different than the default for English02:19
abuDawudcharlie-tca, thanks, the only thing I can think is he is trying to use right-alt. Thats why I was specific with the left alt part. Looks like an automated bug report though right?02:21
charlie-tcaThat is why we ask questions. If we guess at how to do it, we are usually wrong.02:23
charlie-tcaMost bug reports are automated now, but that doesn't mean you can't ask how to reproduce it with specific steps02:23
abuDawudcharlie-tca, I will remember that from now on, I didn't ask any questions due to the automated bug report but I guess its not a big deal to them if they are not following the bug correctly.02:24
charlie-tcaThere are a whole list of responses you can use, just copy and paste to the bug report. the URL is02:25
charlie-tcahttps://wiki.ubuntu.com/Bugs/Responses02:25
charlie-tcaNote number 202:25
abuDawudawesome, thanks charlie-tca02:25
charlie-tcaIt doesn't matter if the reports are automated or not, we still need information to reproduce them and to get them ready for the developers to work02:26
charlie-tcaabuDawud: thank you very much for helping with bugs.02:30
* yofel wonders if there's a way to dump the runtime bash configuration02:30
yofelalt+. depends at least on emacs mode being used (which should be the default)02:31
charlie-tcait depends on the keyboard setup, too. I don't have right-alt+ with this setup, but the default uses both left and right02:33
yofelright, left_alt + . works as described in emacs mode here, right_alt + . prints '…'02:34
micahgBTW, both my ALT keys + . produce the last argument02:58
abuDawudMy rightalt+. does not work, but the workaround is noted here03:03
abuDawudhttp://jorgenmodin.net/index_html/archive/2010/10/25/alt-dot-not-working-in-ubuntu-1010---workaround03:03
abuDawudOpinion? https://bugs.launchpad.net/ubuntu/+source/software-center/+bug/67831903:36
ubot2Launchpad bug 678319 in software-center (Ubuntu) "Integrate update-manager in Ubuntu Software Center (affects: 1) (heat: 6)" [Undecided,New]03:36
micahgabuDawud: that might be a duplicate, I believe that's on teh roadmap for software center, just don't know which release03:37
abuDawudmicahg, looking through all the bugs for update-manager I did not see anything there, and there are no blueprints. Somewhere else to look?03:39
micahgabuDawud: look in software center03:40
abuDawudmicahg, nothing @ https://bugs.launchpad.net/update-manager and the blueprints for that 'branch' are empty03:42
micahgabuDawud: no, search for the bugs in Ubuntu03:42
micahghttps://bugs.launchpad.net/ubuntu/+source/software-center03:43
abuDawudah, cerp03:43
abuDawudderp*03:43
abuDawudmicahg, closest thing I found was an error when opening Software Center when update-manager was running, but no requests for a merge03:45
micahgabuDawud: I don't see anything either03:48
abuDawudmicahg, since it appears to be an 'off-the-hip' opinionated request, would it go to opinion?03:48
micahgabuDawud: no03:49
micahgit's actually been discussed03:49
micahgabuDawud: also, opinion is only when the maintainer believes that a bug should not be worked on03:49
abuDawudso to get to opinion it needs to show an alternative thought process in the same bug?03:50
micahgabuDawud: in Ubuntu, we don't really use opinion that often, in fact, it might not be around for much longer (it was just an experiment)03:52
abuDawudThanks for your help once again micahg.03:57
micahgabuDawud: np03:58
abuDawudhow long should I wait before submitting an app to bugcontrol. I have a history of QA stuff for a medical software company IRL :)04:27
micahgabuDawud: the suggestion is 5 bugs where you can show evidence of triaging work, unless you're an upstream developer with a current bug control member to advocate for you04:28
micahgabuDawud: https://wiki.ubuntu.com/UbuntuBugControl04:29
abuDawudmicahg, so since I can't show evidence of all the duplication stuff, it may be a couple of days...04:30
abuDawudthanks04:30
micahgabuDawud: take a while, the only difference for bug control is being able to set importance, and the triaged/won't fix status, so there's no rush04:31
njinPedro_: hello fellow, Linux key + p is a hidden shortcut ?13:51
njinpedro_:^^13:51
seb128njin, hi, what is your issue exactly?14:02
seb128njin, it's a special keybinding win-7 is using for showing display control14:03
seb128njin, recent laptops got modified bios to handle that in a special way14:03
njinseb128: bug67842314:04
njinbug 67842314:05
ubot2Launchpad bug 678423 in ubuntu "Keyboard shortcut Super+p causes entire screen to redraw (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/67842314:05
seb128bug #67842314:05
seb128bug #53947714:05
ubot2Launchpad bug 539477 in linux (Ubuntu Natty) (and 9 other projects) "Video out hot key sends super + p + return on many upcoming Dell & HP systems (affects: 20) (dups: 2) (heat: 142)" [Undecided,Invalid] https://launchpad.net/bugs/53947714:05
seb128njin, ^14:05
seb128njin, see that bug, you can maybe comment there14:05
njinseb128:thanks14:06
njinseb128: can you reproduce it typing super + p ?14:34
seb128njin, no but I'm on natty and not uptodate14:37
njinseb128: not reproducible in natty by me too, but in maverick, hitted two times it change the resolution14:42
seb128njin, the g-s-d upload was not in natty so seems right14:43
njinseb128: then g-s-d is the same of maverick, so i'm thinking on a lower level, xserver-xorg-input-keyboard as it can be easily reproduced and in Natty we have an updated version of xorg14:52
seb128no it's not14:53
seb128did you read the bug I pointed before?14:53
seb128there was a stable update to fix the handle of that keybinding14:53
seb128which is not uploaded to natty14:53
abuDawudcan someone please set bug# 678460 to wishlist after a quick review?15:59
charlie-tcabug 67846016:11
ubot2Launchpad bug 678460 in nux (Ubuntu) (and 2 other projects) "don't ship .bmp files but rather .png (affects: 1) (heat: 6)" [Critical,Triaged] https://launchpad.net/bugs/67846016:11
charlie-tcahm, not going to do that. critical to wishlist seems wrong16:12
jcastropedro_: holler at me when you announce the banshee hug day pls16:41
pedro_oops16:41
pedro_jcastro, i'll do that today afternoon, just being too busy atm16:42
jcastropedro_: nod, I just wanted to catch you before you did. :)16:42
=== deryck is now known as deryck[lunch]
njinpedro_: hello, can i disturb ?18:20
=== deryck[lunch] is now known as deryck
njingdm-binary[1097]: WARNING: Unable to load file '/etc/gdm/custom.conf': Нет такого файла или каталога18:21
njinand various segfault during shutdown18:21
njincan be  a linuxissue ?18:22
=== yofel_ is now known as yofel
=== bilalakhtar is now known as papercutter
=== papercutter is now known as bilalakhtar
Tiibiidiihi18:55
Tiibiidiii have some general questions18:55
Tiibiidiifor example: now some bugs that i've reported some time ago, are going to expire18:56
Tiibiidiiand i'm realizing that for some bugs18:56
Tiibiidii(tipically these are the "most interesting" bugs: memory leaks, crashes...)18:57
Tiibiidiii've not been actually able to replicate them18:57
Tiibiidii(contrarly to user interface bugs or bugs found when testing and alpha/beta of ubuntu)18:57
yofelonly incomplete bugs can expire, so either add the missing information to the bug or change the status if the information was added18:58
Tiibiidiiand i was wondering.... do you think it's still relevant to report them?18:58
Tiibiidiiyofel, that's the problem18:58
Tiibiidiii reported all the information that was needed18:58
Tiibiidiiwhen i had this avilable18:59
Tiibiidiibut since i can't replicate the bug18:59
Tiibiidiii cannot add more detail18:59
Tiibiidiii mean: i don't like this... because the bug is definitely still there somewhere18:59
Tiibiidiibut indeed, keeping track of it seem quite useless19:00
geekosopherTiibiidii: if the problem was in alpha/beta versions, but not in final, that seems the bugs are fixed19:00
Tiibiidii(actually, one time i remember i was able to replicate the bug, but wasn't able to report a valgrind log, due to valgrind crashing)19:00
Tiibiidiino, they were reported with a stable release (at least the one i'm talking about)19:01
Tiibiidiii mean...19:01
Tiibiidiii'm not a member of the bugsquad or anything19:01
Tiibiidiii don't actively search bugs most of the time19:02
Tiibiidiii just happen to report them when it happens19:02
Tiibiidii(btw, do you think it's fine this way? i mean: joining the bug squad seems like another responsibility to have to keep track of)19:02
yofelwell, the auto-expiring is a bit annoying here, but we have so many bugs that those that we don't have stepst to reproduce them for are hardly going to get fixed19:02
geekosopherTiibiidii: that itself is an important contribution19:03
Tiibiidiiand so i was wondering how do you do in these cases19:03
Tiibiidiiok19:03
Tiibiidiiso, do you report them nonetheless, and just don't bother if they expire without you being able to contribute any additional information?19:03
Tiibiidii(or wait until you've been able to define the steps needed to reproduce the bug, and only then trying to report it?)19:04
Tiibiidii(thank you for your attention in the meanwhile)19:04
yofelI would let them expire, If a bug still needs more information and I can't provide it, the bug is probably not going to get fixed, you can always reopen the bug or file a new bug once you can reproduce the issue again.19:05
Tiibiidiiok19:05
Tiibiidiii was also thinking19:06
yofeland no, report the bug once you see it, could be a one time failure that many people hit for example19:06
Tiibiidiiok, good to know it's fine this way19:06
geekosopherTiibiidii: it is not just you, if someone else faces the same problem, he/she will confirm the bug and probably provide the required information, so it should be reported nontheless19:06
Tiibiidiii like launchpad a lot, and i find bugzilla quite clunky...19:06
Tiibiidiibut for many projects, the upstream is not in launchpad19:07
Tiibiidiiand since the resources for ubuntu developers are quite scarce...19:07
geekosophernobody is perfect ;)19:07
Tiibiidiii mean: ideally we should report all the bugs... and then when relevant report them upstream19:08
charlie-tcaTiibiidii: it has taken 4 years to figure out how to reproduce reliably a panel crash in xubuntu19:08
Tiibiidiiuh19:08
charlie-tcaBut i can now reliably reproduce it19:09
Tiibiidiithe point is... often bugs in launchpad take a lot of time to be triaged (if ever), and often the triager is not a proper developer19:09
yofelthat's called lack of manpower, sadly :/19:09
charlie-tcaif the bug can not be reproduced, it is very difficult for the developer to reproduce it too19:09
Tiibiidiiyeah19:09
Tiibiidiiso i was thinking that from now i could directly report them upstream19:09
geekosophercharlie-tca: really!?19:10
geekosopherhehe19:10
charlie-tcayup19:10
charlie-tcareporting it directly upstream if it can be reproduced results in the same action19:10
Tiibiidiithe problem is:  as i said before, i'm not doing it proactively... i already created accounts and used them in the mozilla bugzilla, in the gnome bugzilla ( and maybe also in a novell bugzilla)19:10
Tiibiidiidoes everybody of you have so many bug trackers accounts?19:11
Tiibiidiior maybe you specialize only with bugs in some certain software/frameworks19:11
charlie-tcaabuDawud: bug 67846019:11
ubot2Launchpad bug 678460 in nux (Ubuntu) (and 2 other projects) "don't ship .bmp files but rather .png (affects: 1) (heat: 6)" [Critical,Triaged] https://launchpad.net/bugs/67846019:11
Tiibiidii(e.g. you happen to report mainly only firefox and thunderbird bugs)19:11
charlie-tcawas filed by a developer, triaged and marked importance by the developer. We do not change status once the bug is changed to triaged19:12
* yofel has accounts for: launchpad, sourceforce, gnome, kde, mozilla, quassel (and whatever I can't remember ATM) - debian doesn't need an account at least)19:12
charlie-tcaI have so many I quit counting19:12
Tiibiidiiok, so it's a common problem19:12
abuDawudcan someone please set the status of https://bugs.launchpad.net/ubuntu/+source/openoffice.org/+bug/678125 to triaged?19:12
ubot2Launchpad bug 678125 in openoffice.org (Ubuntu) (and 1 other project) "No A6 format preset in 'page' dialog (affects: 1) (heat: 8)" [Wishlist,Confirmed]19:12
charlie-tcaabuDawud: Please do not touch the bugs that are already "triaged" status19:12
abuDawudcharlie-tca, yea i have made two big mistakes today... part of the learning process I guess19:13
charlie-tcaas long as we learn, it is okay to make mistakes. It is when we repeat them a number of times that it becomes a problem.19:13
charlie-tcaabuDawud: done19:14
Tiibiidiia last question... this one is a little bit controversial... about meta-bugs19:14
Tiibiidiido you use evolution?19:14
charlie-tcaI do not use evolution.19:15
Tiibiidiibecause i often chose to use thunderbird instead19:15
Tiibiidiiok19:15
Tiibiidiithen i tried again to use it19:15
Tiibiidiifor the usual reason:19:15
Tiibiidiidogfeeding19:15
charlie-tcaI have not used it for 2 or 3 or 4 years19:15
abuDawudthanks charlie-tca, learnin as I go here. More mistakes to come most likely...19:15
Tiibiidii(i mean... if i want to measure how ubuntu is improving, and to report bug, and etc.)19:15
charlie-tcaGreat! Keep on, improvements usually come from mistakes.19:16
Tiibiidii(i have to keep using evolution...) it has the plus that's better integrated in the system19:16
Tiibiidiibut, as i think you could agree, it has quite a lot of defects19:16
Tiibiidiiand i've seen this "not using evolution" as a common trend among the ubuntu contributors19:16
Tiibiidiiso i was thinking19:17
Tiibiidiithis is a bug... but not actually...19:17
Tiibiidiii mean: it's a problem, but it doesn't have a single and easily identificable cause19:17
charlie-tcaEvolution is a great email package for those who need all the things offered by it. For many of us, it is really overkill for reading email.19:17
Tiibiidiiand the solution could be the improvement of evolution, or the switch of the default mail client... this thing would need a serious proposal and some blueprint19:18
Tiibiidiiyeah, so i was thinking about meta-bugs19:18
yofelabuDawud: bug 678371 - what's your reasoning for marking that a duplicate?19:18
ubot2Launchpad bug 678371 in apport (Ubuntu) "apport-gtk speaker test message refers to "Close" button as "OK" (dup-of: 667451)" [Undecided,New] https://launchpad.net/bugs/67837119:18
charlie-tcaThunderbird was proposed at UDS this time, and will probably be proposed for the 11.10 release19:18
ubot2Launchpad bug 667451 in apport (Ubuntu) ""ubuntu-bug audio" hangs on speaker test (affects: 3) (dups: 2) (heat: 153)" [Undecided,Fix committed] https://launchpad.net/bugs/66745119:18
Tiibiidiibut the point is... meta-bugs seems that could be quite subjective19:18
charlie-tcaum, maybe I missed something. What is the "meta-bug" in evolution that you are referring to?19:19
Tiibiidiii mean... the sabdfl reported bug #1, but what is the practical use-case and procedure for reporting a meta-bug (if there is one)19:19
ubot2Launchpad bug 1 in tilix (and 20 other projects) "Microsoft has a majority market share (affects: 612) (heat: 3026)" [High,New] https://launchpad.net/bugs/119:19
Tiibiidiiit doesn't exist19:19
Tiibiidiii was just evaluating the possibility of reporting one19:19
Tiibiidiiand i realized that maybe reporting a meta-bug like this one could be quite controversial (for all the reasons i listed: subjective, huge work, need blueprint, no single cause, etc.etc.)19:20
charlie-tcaTBH, there is no such thing as a meta-bug of that nature. That was done because the distribution was brand new. It provided a purpose before the first release of Ubuntu19:21
Tiibiidiiyeah, but it's not the only one19:21
Tiibiidiifor example19:21
Tiibiidii(it has been fixed)19:21
Tiibiidiihttps://bugs.launchpad.net/ubuntu/+source/gwibber/+bug/61474219:21
ubot2Launchpad bug 614742 in gwibber (Ubuntu Maverick) (and 3 other projects) "[meta] cannot add facebook account, facebook does not update and more (affects: 306) (dups: 76) (heat: 1422)" [Undecided,Fix released]19:21
Tiibiidiisince there were too many gwibber+facebook bugs19:22
Tiibiidiidevelopers "decided to focus" on this19:22
Tiibiidiias well, evolution is a program with A LOT of bugs reported19:22
charlie-tcaMost bugs marked as a meta bug will be there because it gives a master bug for the developers to work with. It allows them to combine all the suggestions without losing the original issues19:23
charlie-tcaIf they do not concern a single idea, they will not be combined into a single bug.19:23
greg-gTiibiidii: meta bugs come after many specific bugs and group them together per the discretion of the developers of the application19:23
Tiibiidiithat's why i was thinking about it... but maybe metabugs of this kind are better left to the ubuntu developers themselves to report for19:23
Tiibiidiiok, so...19:26
Tiibiidii(i realize that now this issue has been brought to attention during the uds)19:26
Tiibiidiibut, in a similar case... if the number of bugs grows this huge https://bugs.launchpad.net/ubuntu/+source/evolution19:27
charlie-tcaDo they all concern the same issue ( the gwibber bug was 'facebook' specific) ?19:28
Tiibiidiithere isn't a standard way for someone to warn that things are not as optimal....19:28
charlie-tcaall applications have bugs, if they don't, they usually stagnate and disappear, since they are never improved.19:28
Tiibiidiino, not exactly... but there a lot of segmentation fault for example19:28
Tiibiidiihttps://bugs.launchpad.net/ubuntu/+source/evolution?field.searchtext=sigsegv&orderby=-importance&search=Search&field.status%3Alist=NEW&field.status%3Alist=INCOMPLETE_WITH_RESPONSE&field.status%3Alist=INCOMPLETE_WITHOUT_RESPONSE&field.status%3Alist=CONFIRMED&field.status%3Alist=TRIAGED&field.status%3Alist=INPROGRESS&field.status%3Alist=FIXCOMMITTED&field.assignee=&field.bug_reporter=&field.omit_dupes=on&field.has_patch=&field.has_no_package=19:28
Tiibiidii(whoops, sorry for the ugly url)19:29
Tiibiidiii mean: i think it's a general problem of quality for evolution19:29
charlie-tcathe segmentation fault can be caused by many different things, which require many different fixes19:29
yofelblame launchpad for adding redundant information to the URL :(19:29
Tiibiidiihaving this many bugs uncared is not a good thing19:29
Tiibiidii<Tiibiidii> having this many bugs uncared is not a good thing <-- again, i'm not complaining for the lack of work-force...19:30
charlie-tcaI see a few in new status, but it looks most of them are cared about. They are not sitting in new status, they are instead looked at19:30
charlie-tcaIf you care that much, instead of taking all this time and effort to complain, start triaging them. Or better yet, fixing them.19:31
charlie-tcaHow many could you have triaged in the last 45 minutes?19:32
Tiibiidiiyou're right... i worry too much about these things19:33
Tiibiidiibut i think that i wouldn't be able to make such a difference as you imply with this mere 40 minutes19:34
abuDawudtiibiidii, if everyone had that mindset nothing would ever get done, right?19:36
Tiibiidiieh, you're right too... however thank you for answering... at least now i have done this little reality check about how to work with bugs19:40
charlie-tcaprobably see a lot more difference than just talking19:41
yofelabuDawud: ping? (apport bug)19:50
abuDawudyofel, sorry whats up?19:59
yofelabuDawud: why did you mark bug 678371 as a duplicate?20:00
ubot2Launchpad bug 678371 in apport (Ubuntu) "apport-gtk speaker test message refers to "Close" button as "OK" (dup-of: 667451)" [Undecided,New] https://launchpad.net/bugs/67837120:00
ubot2Launchpad bug 667451 in apport (Ubuntu) ""ubuntu-bug audio" hangs on speaker test (affects: 3) (dups: 2) (heat: 153)" [Undecided,Fix committed] https://launchpad.net/bugs/66745120:00
abuDawudthe cancel message bug is in that other bug20:01
abuDawudthere was another exact dupe of 678371 marked as a duplicate of it20:02
yofelyes, the cancel message bug, but here he's talking about 'Close' being used instead of 'OK', not about Cancel hanging20:02
yofelok, I mixed something up (looked at bug 678360)  - he's indeed talking about both issues in 667451, which is wrong since we never track more than one issue per bug20:04
abuDawudoh dammit. I was cleaning duplicates of that cancel hanging issue and wrapped that feature request in20:04
ubot2Launchpad bug 678360 in apport (Ubuntu) "apport-gtk hangs when cancel button is clicked during audio test tone (dup-of: 667451)" [Undecided,Confirmed] https://launchpad.net/bugs/67836020:04
ubot2Launchpad bug 667451 in apport (Ubuntu) ""ubuntu-bug audio" hangs on speaker test (affects: 3) (dups: 2) (heat: 153)" [Undecided,Fix committed] https://launchpad.net/bugs/66745120:04
abuDawudwhats the proper process to break a bug in two?20:05
yofelI would undupe 678371 and 678360 and mark 667451 as a dup of 678360 - that has steps to reproduce and we get the issues properly seperated20:06
abuDawudmark an older report a dupe of a newer bug? Isn't that counter intuitive?20:07
yofelactually wait - I'm tired (the date it was reported doesn't matter, the amount of information does) the other one is already marked fix committed so it's better not to touch it20:08
yofelabuDawud: undupe 678371 and post a comment on 667451 that the 'close' issue is tracked there20:10
abuDawudyofel, thanks for the note regarding report order not mattering, i'll discard that thought process20:10
abuDawudi'll do that.20:10
abuDawudthanks for correcting me20:10
yofelthanks for helping ;)20:10
abuDawudyofel, can you mark 678371 as triaged for me?20:20
yofelabuDawud: sure, Importance? (https://wiki.ubuntu.com/Bugs/Importance)20:22
abuDawudlow, correct?20:22
yofelack20:22
yofelabuDawud: done20:23
abuDawudty sir.20:23
BUGabundoo/21:01
Awsoonnhi all, what package is the network proxy config tool in?21:29
hggdhAwsoonn: if you know the binary name, you can search by 'dpkg -S <binary>', or 'apt-file search <binary>'21:54
charlie-tcaHello, hggdh. been hiding, huh?21:57
hggdhcharlie-tca: hey21:57
hggdhjust a bit, just a bit ;-)21:57
charlie-tcaIt's okay. Hiding sometimes is good for a person21:58
hggdhbut I have been discovered now...22:00

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!