| Unit193 | micahg: Thanks for the upload. | 02:51 |
|---|---|---|
| micahg | Unit193: sure | 02:56 |
| micahg | anyone want to test this? https://bugs.launchpad.net/trusty-backports/+bug/1391487 | 02:56 |
| ubottu | Ubuntu bug 1391487 in trusty-backports "Please backport xfce4-settings 4.11.3-0ubuntu2 (universe) from utopic" [Undecided,New] | 02:56 |
| ali1234 | can someone test something for me please? | 15:31 |
| ali1234 | open mousepad, open preferences, editor tab, set tab width to 33, press okay | 15:32 |
| ali1234 | this will crash mousepad | 15:32 |
| flocculant | ali1234: ack that with the version in dev ppa | 15:37 |
| ali1234 | i'm on the dev ppa too... was hoping someone would test without it | 15:42 |
| ali1234 | i'll test in virtualbox when it's updated | 15:42 |
| ali1234 | thanks anyway tho | 15:42 |
| flocculant | I'll purge the ppa then | 15:43 |
| ali1234 | it's okay | 15:44 |
| flocculant | did it - crashed same with mousepad 0.40 | 15:45 |
| ochosi | ali1234: oh wow, interesting crash. and yeah, i can reproduce | 15:48 |
| ali1234 | found the bug | 15:49 |
| ali1234 | i'll send a patch upstream | 15:49 |
| ochosi | weird that a value that is out of schema range can cause a segfault | 15:49 |
| ali1234 | yeah | 15:49 |
| ochosi | is it in gsettings? | 15:49 |
| ali1234 | not sure | 15:49 |
| ali1234 | i think i see it | 15:56 |
| ali1234 | doing an apport... | 15:57 |
| ali1234 | bug 1472690 | 16:19 |
| ubottu | Error: Launchpad bug 1472690 could not be found | 16:19 |
| ali1234 | bug 1472690 | 16:21 |
| ubottu | bug 1472690 in mousepad (Ubuntu) "Mousepad crashes when setting tab width > 32" [Undecided,New] https://launchpad.net/bugs/1472690 | 16:21 |
| ali1234 | https://bugzilla.xfce.org/show_bug.cgi?id=12055 | 16:25 |
| ubottu | bugzilla.xfce.org bug 12055 in General "Setting tab width to 33 crashes mousepad" [Normal,New] | 16:25 |
| ali1234 | so what happens is the gsettings object becomes invalid when you try to store a wrong value in it, then later an assertion crashes mousepad because the object isn't valid | 16:26 |
| ochosi | codebrainz: ^ fun bug :) | 16:32 |
| pleia2 | if anyone wants to proof read the latest Xubuntu at... article: http://xubuntu.org/?p=3361&preview=true (need to be a Xubuntu team member w/ login to the site to view) | 20:09 |
| pleia2 | knome: ^^ | 20:09 |
| knome | will do that in an hour or so | 20:24 |
| knome | *within | 20:24 |
| pleia2 | ty | 20:24 |
| codebrainz | ali1234, ochosi, nice :) The more curious thing to me is 1) why would one want 33-wide tabs, and 2) how did they set them to 33 since the UI limits to 32 | 23:22 |
| codebrainz | I'll have a look. If it didn't hard-crash I'd just say "unsuported", but well... | 23:23 |
| ali1234 | the UI doesn't limit to 32... that's the problem | 23:23 |
| ali1234 | it limits to 128 | 23:23 |
| codebrainz | it does here | 23:23 |
| ali1234 | i put a patch on xfce bugzilla | 23:24 |
| ali1234 | no really i checked the code :) | 23:24 |
| codebrainz | http://i.imgur.com/ZMVUa8g.png | 23:24 |
| codebrainz | i mean you could force it through dconf-editor though | 23:25 |
| ali1234 | what is that... | 23:25 |
| codebrainz | screenshot of mousepad from Git master with a nasty gtk theme | 23:26 |
| ali1234 | is that a slider?? | 23:26 |
| codebrainz | yeah | 23:26 |
| ali1234 | mine doesn't have a slider | 23:26 |
| ali1234 | it doesn't look anything like that at all | 23:26 |
| codebrainz | are you using gtk2 or 3? | 23:26 |
| ali1234 | gtk2 | 23:26 |
| ali1234 | http://i.imgur.com/FdwA3fZ.png | 23:27 |
| ali1234 | why do you have a separate window just for the tab width? i don't even... | 23:27 |
| codebrainz | ah ... I thought you meant Document->Tab Size->Other ... | 23:28 |
| ali1234 | i didn't even know that was a thing | 23:28 |
| Unit193 | Gtk2 mousepad is nicer. :/ | 23:28 |
| ali1234 | yeah the per-document setting is fine | 23:28 |
| ali1234 | btw | 23:28 |
| ali1234 | reason i wanted this was because i had a datafile which was tab-delimited | 23:28 |
| codebrainz | ali1234: did you have any look at why the hard-crash? | 23:29 |
| ali1234 | and i wanted the data columns to line up, and was too lazy to import in a spreadsheet | 23:29 |
| ali1234 | so i just tried to make tab stop huge -> crash | 23:29 |
| ali1234 | and the crash happens because if you try to set the gsettings value to something invalid, the whole gsettings object becomes invalid, then a later assertion crashes | 23:29 |
| codebrainz | ugh, nice | 23:30 |
| ali1234 | didn't look any deeper than that | 23:30 |
| codebrainz | since it's hard-coded, I guess I won't investigate too deeply either, as it shouldn't be allowed to go out of range (modulo the bug) | 23:30 |
| ali1234 | on a completely unrelated note, text files with 5MB of text and no line breaks makes mousepad unhappy | 23:30 |
| codebrainz | s/mousepad/many editors/ :) | 23:31 |
| SwissBot | codebrainz: You did something wrong... Try s/you/me/ or tell me "help sed" | 23:31 |
| ali1234 | but basically no text editors can handle that | 23:31 |
| codebrainz | 5mb sure, but without the line break, even on Scintilla editors like Geany it'll bomb | 23:31 |
| codebrainz | IIRC Gtksourceview keeps a list of line strings, so I guess it only has one huge link | 23:32 |
| ali1234 | gedit will warn you if you try to open such a file, i think | 23:32 |
| ali1234 | mousepad opens it, thinks for like four hours, then draws the whole line wordwrapped but on a single line, so it looks like when the paper gets stuck in a typewriter and you just keep going | 23:33 |
| codebrainz | oh wow, with line-wrapping too, nice :) | 23:33 |
| codebrainz | I'm not even sure how you'd warn about that without scanning and counting how many newlines | 23:34 |
| ali1234 | well you must have scanned through it before displaying it | 23:35 |
| ali1234 | i could look at how gedit does it some time, if you're interested | 23:35 |
| codebrainz | i'd have to look to be sure, but I don't think mousepad does much besides dump the file contents straight into GSV buffer | 23:36 |
| codebrainz | ali1234: I'm not strictly opposed, but it's kind of outside of mousepad's scope to handle such files (IMO) | 23:36 |
| codebrainz | but someone made a simple patch that didn't add an additional pass of scanning the buffer just for that, I would probably be ok with it | 23:37 |
| ali1234 | would be nice if it didn't crash though | 23:38 |
| ali1234 | i tink this might be the cause of "mousepad becomes unresponsive after some time" | 23:38 |
| codebrainz | oh | 23:38 |
| ali1234 | i mean technically it doesn't crash... if you are patient enough | 23:39 |
| codebrainz | maybe it could just check if the file is greater than N kb/mb and put up a dialog "The mouspad text editor is not suitable to huge files, and if it doesn't have sufficient line breaks it will probably eat your breakfast" | 23:39 |
| codebrainz | with an "I agree" button | 23:39 |
| ali1234 | yeah i might be misremembering on gedit actually... it might just do that | 23:40 |
| ali1234 | let me check | 23:40 |
| codebrainz | we discussed such a thing for Geany, as it doesn't handle massive files well if syntax highlighting is enabled (or if on single line) | 23:40 |
| ali1234 | hmm gedit has a loading progress bar with a cancel button now | 23:41 |
| codebrainz | neat | 23:41 |
| ali1234 | but attempting to scroll the file killed it | 23:41 |
| ali1234 | it actually lets you view the file before it finished loading too | 23:41 |
| codebrainz | mousepad seems to handle massive files quite well, assuming they have line breaks | 23:41 |
| codebrainz | (at least last time i checked) | 23:42 |
| codebrainz | ali1234: i guess it needs to get lower bound set too | 23:43 |
| codebrainz | changing it to zero also crashes it | 23:43 |
| ali1234 | doh | 23:44 |
| codebrainz | you mind if I amment your commit? | 23:44 |
| ali1234 | go for it | 23:44 |
| codebrainz | *amend | 23:44 |
| ali1234 | i tested setting -1 in gsettings | 23:44 |
| codebrainz | (just to avoid two micro commits for same issue) | 23:44 |
| ali1234 | but not 0 | 23:44 |
| ali1234 | -1 also seems to be the default in the schema | 23:45 |
| codebrainz | what happens with -1? | 23:45 |
| ali1234 | it seems to use 4 as a default | 23:45 |
| codebrainz | ok, fine | 23:45 |
| ali1234 | but you can't have -1 without also allowing 0 | 23:46 |
| ali1234 | i didn't understand what that was all about, so i left it | 23:46 |
| codebrainz | ali1234: did you put your patch in a separate branch? | 23:47 |
| ali1234 | i don't think so? | 23:48 |
| ali1234 | i just git pulled | 23:48 |
| codebrainz | I mean i want to git commit --amend to it, which will re-write the sha1 of it | 23:48 |
| codebrainz | so you'll have to undo it or it might give grief when you pull next time | 23:48 |
| ali1234 | no problem, i'm used to that stuff | 23:48 |
| codebrainz | okie | 23:48 |
| codebrainz | http://git.xfce.org/apps/mousepad/commit/?id=15f4e5a47bd85d2cd831f65d6a66e548ad4d5708 | 23:50 |
| ali1234 | thanks | 23:51 |
| codebrainz | and to you | 23:52 |
| codebrainz | note to self: review all prefs to make sure there are no other crashes accessible from UI :) | 23:52 |
| ali1234 | going to amend this too? http://git.xfce.org/apps/mousepad/tree/mousepad/org.xfce.mousepad.gschema.xml?id=15f4e5a47bd85d2cd831f65d6a66e548ad4d5708#n107 | 23:52 |
| codebrainz | where is it from? | 23:53 |
| codebrainz | oh the schema | 23:53 |
| codebrainz | ali1234: it _might_ be that I smarty made -1 use a safe hardcoded default value | 23:54 |
| codebrainz | *smartly | 23:54 |
| ali1234 | hmm gedit loaded that 5MB text file | 23:54 |
| ali1234 | wordwrapped it all, and it works absolutely fine | 23:54 |
| codebrainz | i don't see why I'd put -1 instead of 0 or 1 otherwise | 23:54 |
| ali1234 | must be gtk3? | 23:54 |
| codebrainz | could be. i think ./configure --enable-gtk3 will force it with mousepad, if you want to test | 23:55 |
| ali1234 | i'll give it a try | 23:55 |
| ali1234 | this works fine, as long as you wait for it to finish loading before you try to do anything | 23:55 |
| ali1234 | oh wait, no, i just didn't get to the looooong line | 23:56 |
| ali1234 | now it froze | 23:56 |
| codebrainz | hehe | 23:56 |
| codebrainz | you need better files :) | 23:56 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
