All times shown according to UTC.
| Time | Nick | Message |
|---|---|---|
| 00:38 | glee joined #miro-hackers | |
| 01:06 | ajonas joined #miro-hackers | |
| 01:12 | ajonas joined #miro-hackers | |
| 01:17 | maggie_s_ joined #miro-hackers | |
| 01:33 | maggie_s joined #miro-hackers | |
| 02:21 | maggie_s_ joined #miro-hackers | |
| 02:34 | kcw left #miro-hackers | |
| 08:59 | maggie_s joined #miro-hackers | |
| 12:42 | Jarrhead joined #miro-hackers | |
| 13:14 | z3p1 | good morning |
| 13:27 | paroneayea joined #miro-hackers | |
| 14:36 | z3p joined #miro-hackers | |
| 15:01 | z3p | bendk: so...there's a hiccup in the Unicode stuff |
| 15:01 | I've got some directories which have non-ASCII characters, but aren't Unicode | |
| 15:02 | er, I guess filenames | |
| 15:02 | er, aren't UTF-8 | |
| 15:04 | they're latin-1, or maybe windows-1255 | |
| 15:05 | but either way, Miro crashes when it scans them | |
| 15:05 | I can convert them to Unicode fine with a hack in m.l.fileutil:make_filename | |
| 15:05 | but then they'll get converted back with UTF8 which is wrong | |
| 15:12 | kcw joined #miro-hackers | |
| 15:35 | arpu joined #miro-hackers | |
| 15:35 | bendk | hey z3p |
| 15:36 | this is on a device? | |
| 16:02 | DGMurdockIII joined #miro-hackers | |
| 16:22 | CarlFK joined #miro-hackers | |
| 16:29 | zanoi joined #miro-hackers | |
| 16:32 | zanoi | could someone send me a screenshot of what the sparkle upgrade dialog looks like in miro |
| 16:32 | and with the releasenotes scrolled down as much as possible | |
| 16:33 | it should be possible to get the dialog by changing line 102 in autoupdate.py to True | |
| 16:33 | err, False I mean | |
| 16:43 | Jarrhead joined #miro-hackers | |
| 16:45 | glee | zanoi: Do you still need it? I can send that to you |
| 16:48 | zanoi: I'll send it to your email | |
| 16:51 | zanoi: Sent ... | |
| 16:53 | mattjohnson joined #miro-hackers | |
| 16:53 | zanoi | glee: thanks! |
| 16:55 | glee: that's odd. Is the release notes background really white? or had it not loaded the background picture yet? | |
| 16:56 | glee | zanoi: I believe it really is white, I left it on for a minute or so while I was fiddling with the screenshot program manpage trying to get the arguments right |
| 16:58 | zanoi | glee: then that's a bug as far as I can see |
| 17:00 | oh, maybe the beta feed is different than the release feed | |
| 17:01 | glee | zanoi: Yep, just double checked, the update-Miro402.html in the release channel thingy has a background |
| 17:02 | zanoi | glee: ok, thanks. That's actually the one I needed a screenshot of, to see how it looks in the original Sparkle :/ |
| 17:02 | glee | zanoi: I can take another one if you need it |
| 17:02 | zanoi | glee: that would be great |
| 17:02 | since it's all messed up in WinSparkle and I'm not sure how it's supposed to look | |
| 17:05 | glee | zanoi: Sent |
| 17:05 | zanoi | glee: thanks! |
| 17:07 | yay that's perfect | |
| 17:08 | glee: do you know how we produce release notes? | |
| 17:08 | glee | zanoi: What does WinSparkle use to render the page? I guess either that or maybe some css (?) stuff are broken? |
| 17:09 | zanoi | it uses IE, and yes the breakage is css related |
| 17:09 | glee | zanoi: I think Janet does the appcast updates, probably need to ask her … what I meant was … I had no idea. :-) |
| 17:09 | zanoi | ok |
| 17:11 | weirdly enough the release notes are full of html errors, but the breakage comes from valid html/css | |
| 17:11 | sigh | |
| 17:12 | glee | Sigh ..Yay for IE and yay for css. If only swearing and shouting insults at it would fix the problem…. |
| 17:13 | zanoi | then it would have been fixed long ago :) |
| 17:15 | glee | zanoi: I wonder if the MC guys could help, if they had to deal with IE/css stuff before |
| 17:17 | zanoi | glee: thanks, but it's fixed now. It's just annoying that now there is a different version of the release notes for Windows and for OS X. |
| 17:19 | glee | bummer… maybe it wouldn't be so bad if they could be auto-generated or something. Either way updating doesn't happen every day, oh well. |
| 17:22 | zanoi | glee: yep, I'll ask janet. Also I could spend some time trying to unify them. Will have to see |
| 17:24 | z3p | bendk: this is on an external disk; I'm seeing hte issue in miro_allfiles |
| 17:39 | bendk | z3p: if miro_allfiles is passed a bytestring it'll return bytestrings |
| 17:39 | so where is the original source? | |
| 17:39 | it's a watched folder? | |
| 17:39 | is it from the movies directory? | |
| 17:39 | z3p | bendk: it's passed a unicode string |
| 17:39 | bendk: it's getting strs from os.listdir | |
| 17:40 | bendk | that seems very weird |
| 17:40 | z3p | bendk: at least, that seems like what's happening |
| 17:40 | bendk: let me verify | |
| 17:41 | bendk | the docs say "Changed in version 2.3: On Windows NT/2k/XP and Unix, if path is a Unicode object, the result will be a list of Unicode objects. Undecodable filenames will still be returned as string objects." |
| 17:41 | z3p | bendk: nope it's returning strs |
| 17:41 | bendk | so maybe there's an undecodable filename? |
| 17:41 | we should add a check for that I guess | |
| 17:41 | z3p | bendk: it's undecodable as UTf8 |
| 17:41 | bendk: it's totally decodable as latin-1 or windows-1255 | |
| 17:41 | bendk | ahh |
| 17:42 | so there's sys.getfilesystemencoding() | |
| 17:42 | which I'm realizing obviously can't be correct for all filesystems | |
| 17:42 | z3p | bendk: there's no way for that to work on all filesystems |
| 17:42 | yeah | |
| 17:42 | bendk | I'm guessing that's the issue |
| 17:42 | can you file a ticket for it | |
| 17:42 | z3p | bendk: my idea was to just check a couple different encodings if the filesytemencoding bailed |
| 17:42 | bendk: will do | |
| 17:52 | bendk: #17997 | |
| 18:32 | bendk | z3p: out of curiousity, if you open that file in nautilous, does it display correctly? |
| 18:53 | DGMurdock joined #miro-hackers | |
| 18:58 | DGMurdockIII joined #miro-hackers | |
| 18:58 | DGMurdockIII joined #miro-hackers | |
| 18:58 | DGMurdockIII joined #miro-hackers | |
| 18:59 | DGMurdockIII joined #miro-hackers | |
| 18:59 | DGMurdockIII joined #miro-hackers | |
| 19:02 | DGMurdockIII joined #miro-hackers | |
| 19:03 | DGMurdockIII joined #miro-hackers | |
| 19:33 | z3p | bendk: it says (invalid encoding) and displays the stray character with a diamond ? |
| 19:46 | bendk | z3p: thanks |
| 19:46 | so, I think we may have to revert all of these filename changes :( | |
| 19:46 | for this exact reason | |
| 19:47 | even if we decode it correctly, how will we know how to re-encode it properly to pass it to the video renderer | |
| 19:47 | or migrate, delete the file, etc. | |
| 19:48 | I think we should ideally be following the conversions here: http://developer.gnome.org/gli[…]t-Conversion.html | |
| 19:48 | which says keep the filenames as bytestrings all the time and only decode when you want to display them for the user | |
| 19:48 | z3p | bendk: which is basically what you're doing |
| 19:49 | bendk: except bytes instead of unicode | |
| 19:49 | bendk: alternatively, I offered an idea in the ticket | |
| 20:00 | bendk | yeah, basically |
| 20:01 | but to use bytestrings means I would have to do all the work I just did again | |
| 20:01 | plus more work to make sure that we can display them correctly | |
| 20:01 | at this point maybe it's better to just back things out and try again later | |
| 21:48 | DGMurdockIII joined #miro-hackers | |
| 23:20 | arpu joined #miro-hackers |