| [00:00:35] |
* |
tonib_ has quit IRC |
| [00:05:44] |
* |
cmlenz has joined #trac |
| [00:09:05] |
* |
mki has joined #Trac |
| [00:11:41] |
* |
nuge has quit IRC |
| [00:14:56] |
* |
nuge has joined #trac |
| [00:15:49] |
* |
cmlenz has quit IRC |
| [00:20:38] |
* |
nuge_ has joined #trac |
| [00:25:09] |
* |
nuge has quit IRC |
| [00:43:38] |
* |
rapha has joined #trac |
| [00:43:40] |
<rapha> |
Hi all! |
| [00:44:09] |
* |
cmlenz has joined #trac |
| [00:51:30] |
<rapha> |
I've searched but couldn't find out where to edit the default Trac page listing all Tracs in a directory. Can somebody help me out there? |
| [00:57:01] |
* |
ValiSystM has joined #trac |
| [00:57:14] |
* |
cboos has joined #trac |
| [00:57:54] |
<cboos> |
hi |
| [00:58:01] |
<alect> |
hi there |
| [00:59:14] |
<cboos> |
I'd like to check what others think about the little wiki feature proposed in #2877 (put space between words of wiki page names) |
| [00:59:43] |
<cboos> |
It's a small addition, configurable, and IMHO really improves readability of wikis |
| [01:00:07] |
<cboos> |
(+ will quiet a bit the CamelCase bashers...) |
| [01:04:22] |
<cboos> |
e.g. see how it looks like on Wacko Wiki http://wackowiki.com/WackoDocumentation?v=18n2 |
| [01:05:24] |
* |
zond has joined #trac |
| [01:06:02] |
<zond> |
hey there #trac. i have a question: is there a neater way of automatically generating bugs in trac (from for example email) than doing a post to the newticket page? |
| [01:07:19] |
* |
hlb has quit IRC |
| [01:08:09] |
<cboos> |
the xmlrpcplugin should does that |
| [01:08:34] |
<cboos> |
http://trac-hacks.org/wiki/XmlRpcPlugin + authenticate and then check the API link |
| [01:09:13] |
<rapha> |
cboos: do you maybe know where to edit the list-of-tracs page? |
| [01:09:27] |
<cboos> |
list-of-tracs? |
| [01:09:32] |
<cboos> |
ah the index? |
| [01:09:45] |
<cboos> |
well, I think there's a template for that now |
| [01:10:02] |
<cboos> |
yep, it's templates/index.cs |
| [01:11:09] |
<cboos> |
so, back to my question about #2877, nobody is against it? cmlenz, what about you? |
| [01:11:20] |
* |
mjg2 has joined #trac |
| [01:11:39] |
<cmlenz> |
yeah |
| [01:11:59] |
<cboos> |
'yeah' against or 'yeah' ok ? :) |
| [01:12:05] |
<cmlenz> |
that only splits up the name in links, or what? |
| [01:12:19] |
<cboos> |
yes, when rendering the default label |
| [01:12:25] |
<cmlenz> |
what about the page name in the timeline, in search, etc? |
| [01:12:32] |
<cboos> |
everywhere... |
| [01:12:41] |
<cboos> |
(when rendering that is) |
| [01:12:50] |
<cmlenz> |
the patch doesn't do that, does it? |
| [01:13:08] |
<cmlenz> |
(not talking about wikitext, but the other occurrences) |
| [01:13:41] |
<cboos> |
hm, I'm afraid I don't understand... |
| [01:13:58] |
<cmlenz> |
the timeline has the page name in the event title, for example, right? |
| [01:14:03] |
<Eridius> |
d'oh, custom query just failed me |
| [01:14:15] |
<Eridius> |
I wanted to check if summary contained "a" AND "b" |
| [01:14:16] |
<cboos> |
ah, yes |
| [01:14:18] |
<Eridius> |
but it only lets me do OR |
| [01:14:49] |
<cboos> |
Eridius: there's already a ticket about that, search for "AND/OR queries" |
| [01:14:51] |
<cmlenz> |
Eridius: yeah, that's a limitation of queries |
| [01:15:18] |
<cmlenz> |
cboos: same with search results, TitleIndex, etc... should the name be split up there too? |
| [01:15:24] |
<cmlenz> |
plus the html title on wiki pages |
| [01:15:28] |
<Eridius> |
whee, a contains query works with %. I would have expected either do a string match or drop the SQL-style match (i.e. turn * into %) |
| [01:16:01] |
<cmlenz> |
it does? |
| [01:16:03] |
<cboos> |
cmlenz: right, whenever we don't go through the Wiki IWikiSyntaxProvider, this won't apply (currently) |
| [01:16:19] |
<Eridius> |
cmlenz: yeah. I just searched for Summary contains "a%b" and it worked |
| [01:17:17] |
<cboos> |
cmlenz: but I could add this for consistency, of course |
| [01:19:31] |
<cmlenz> |
well, if we do it for links, we should do it everywhere... otherwise it's a mess IMO |
| [01:19:41] |
<cboos> |
cmlenz: on the other hand, at some point we should still show the real "identity" of a page somewhere, i.e. to be able to distinguish between "WikiPage" expanded and some real "Wiki Page" (i.e. a page containing space) |
| [01:21:13] |
<cboos> |
that's a rare case and I suppose that if the #2877 feature becomes available, such page with embedded space characters will become even less used |
| [01:21:44] |
* |
Eridius has left #trac |
| [01:21:46] |
* |
Eridius has joined #trac |
| [01:21:48] |
<Eridius> |
ack |
| [01:22:28] |
<cboos> |
so maybe we should keep the exact pagename at least in the <TITLE>, or do you see another place? |
| [01:22:37] |
<cmlenz> |
cboos: that's an edge case... you can always see the real page name in the URL |
| [01:23:30] |
<rapha> |
cboos: thanks! |
| [01:23:59] |
<cboos> |
so ok, I'll extend that patch on #2877 |
| [01:26:56] |
<rapha> |
Hmm |
| [01:27:21] |
<rapha> |
cboos: what version are you using? I just upgraded from 0.9.3 to 0.9.4, but /usr/share/trac/templates doesn't hold an index.cs... |
| [01:27:45] |
<cboos> |
hm, yes that must be in 0.10 only... sorry ! |
| [01:27:58] |
<cboos> |
time to upgrade :) |
| [01:28:12] |
<alect> |
heh |
| [01:28:41] |
<rapha> |
0.*10*?? |
| [01:29:01] |
<alect> |
cmlenz: i put all those ticket patches into #3029 |
| [01:29:01] |
* |
rapha can't find any such thing on http://projects.edgewall.com/trac/wiki/TracDownload |
| [01:29:29] |
<cmlenz> |
alect: yeah, thanks a lot |
| [01:29:52] |
<cmlenz> |
although I think we'll go without the adjunct listener... that's for components, milestones, etc, right? |
| [01:30:05] |
<alect> |
yeah, i just threw it in for completeness |
| [01:30:14] |
<alect> |
i didn't think you'd want to apply it |
| [01:30:18] |
* |
spike_ has joined #trac |
| [01:30:21] |
* |
spike_ is now known as spike |
| [01:30:37] |
<cmlenz> |
otherwise, looks cool |
| [01:31:17] |
* |
hlb has joined #trac |
| [01:32:18] |
<cmlenz> |
weird docstring style, though ;-) where'd you get that from? |
| [01:32:46] |
<alect> |
C++ days |
| [01:33:14] |
<alect> |
/** ... |
| [01:33:23] |
<alect> |
and align under the ellipses |
| [01:33:40] |
* |
alect is glad his c++ days are behind him |
| [01:34:41] |
<cmlenz> |
hehe |
| [01:35:03] |
<cmlenz> |
me too |
| [01:35:42] |
<Semhirage> |
awww come on, c++ isn't that bad |
| [01:36:57] |
<cmlenz> |
for the two people in the world who actually really understand it, it probably rocks ;-) |
| [01:40:48] |
* |
contingencyplan has quit IRC |
| [01:40:49] |
<rapha> |
So what about this ominous 0.10? |
| [01:41:03] |
<cmlenz> |
alect: I think it's weird that both new/ticketmodule components provide the manipulator extension point |
| [01:41:06] |
<alect> |
it's in svn |
| [01:41:13] |
<cmlenz> |
... maybe move it to TicketSystem for now? |
| [01:41:21] |
<cmlenz> |
or have it only on TicketModule? |
| [01:41:38] |
<cmlenz> |
OTOH, it probably doesn't cause any harm |
| [01:42:38] |
<alect> |
yeah |
| [01:43:10] |
<alect> |
it's all cleand up in workflow of course, with the unificaiton |
| [01:43:25] |
<cmlenz> |
right |
| [01:45:20] |
<alect> |
so you coalesced all of the .py plugins under "plugins" in webadmin? |
| [01:46:39] |
<cmlenz> |
erm, I did? |
| [01:46:45] |
<cmlenz> |
I didn't think so :-P |
| [01:46:49] |
<alect> |
seem to :) |
| [01:46:53] |
<cmlenz> |
they should show up with there module name |
| [01:46:59] |
<alect> |
mmaybe i need to update.. |
| [01:47:00] |
<cmlenz> |
but I didn't test with more than one |
| [01:47:46] |
<alect> |
aha |
| [01:47:48] |
<alect> |
update |
| [01:47:51] |
<alect> |
i'm surprised it worked at all |
| [01:47:57] |
<alect> |
magical |
| [01:48:38] |
<cmlenz> |
heh |
| [01:49:14] |
<cmlenz> |
I've added a base class shared by NewticketModule and TicketModule to hold the ticket_manipulators extension point, and the validate_ticket method |
| [01:49:18] |
<cmlenz> |
seems a bit nicer |
| [01:49:39] |
<alect> |
good idea |
| [01:55:34] |
<rapha> |
Help! No matter where I look, there doesn't seem to be a 0.10 release ANYWHERE? So how do I edit that index page? |
| [01:55:44] |
<cmlenz> |
alect: I was wondering about IWikiPageManipulator... could that be used for example to add the "Tag under" field of TracTags? |
| [01:55:57] |
<cmlenz> |
rapha: there is no such release, it's in the works |
| [01:56:23] |
<cmlenz> |
rapha: the current svn trunk will one (sunny) day become 0.10 |
| [01:57:20] |
<alect> |
cmlenz: probably yeah, using some kind of markup in the page you mean? |
| [01:57:44] |
<alect> |
having it done in a form field is nice though |
| [01:57:59] |
<cmlenz> |
similar to ticket custom fields, I was thinking |
| [01:58:08] |
<alect> |
ah |
| [01:58:21] |
<cmlenz> |
i.e. in prepare_wiki_page(), the manipulator would actually add a field for the tags |
| [01:58:24] |
<rapha> |
cmlenz: Okay, I see. So if the templates/index.cs file isn't yet present in 0.94, how do I edit that page? |
| [01:58:44] |
<cmlenz> |
you just create it yourself |
| [01:59:03] |
<alect> |
definitely, if such a beast existed it would be useful |
| [01:59:40] |
<alect> |
although isn't kid going to let plugins have more control over rendering, for such situations? |
| [01:59:47] |
<alect> |
"kid" |
| [01:59:56] |
<cmlenz> |
hehe |
| [01:59:57] |
<cmlenz> |
yeah |
| [02:00:06] |
<cmlenz> |
but that's not without overhead |
| [02:00:41] |
<alect> |
hmm true, it was a tree filter or something like that |
| [02:01:32] |
<cmlenz> |
also, if multiple plugins add controls to wiki pages, we might want to exercise some control on how the are layed out... again similar to ticket custom fields |
| [02:01:48] |
<alect> |
yes, agreed |
| [02:02:16] |
<alect> |
perhaps i could refactor ticket.field to be more generic |
| [02:03:02] |
<rapha> |
cmlenz: Looking at wiki.cs, there seem to be a whole lot of <?cs ... ?> tags that can be used in *.cs files... and when I put an index.cs into my templates directory, how would Trac know that it should use that? Is this stuff documented anywhere? |
| [02:05:16] |
<cmlenz> |
http://projects.edgewall.com/trac/wiki/TracInterfaceCustomization#ProjectList |
| [02:05:45] |
<rapha> |
Thanks a lot! |
| [02:06:11] |
<cmlenz> |
alect: long-term, of course... but I'm inclined to add a "future-proof" prepare_wiki_page(req, page, fields) to the IWikiPageManipulator |
| [02:06:26] |
<alect> |
for 0.10? |
| [02:06:27] |
<alect> |
awesome |
| [02:06:37] |
<alect> |
and fields would just be html fragments? |
| [02:07:16] |
<cmlenz> |
well, the prepare_ticket() method isn't actually called anywhere, so I was thinking to do the same cheap trick for prepare_wiki_page() :-P |
| [02:07:31] |
<cmlenz> |
but maybe adding calls to prepare_ticket() wouldn't be so hard? |
| [02:07:33] |
<alect> |
hehe |
| [02:07:56] |
<cmlenz> |
no, that'd require trac.ticket.field stuff, of course |
| [02:07:58] |
<cmlenz> |
hmm |
| [02:08:00] |
<alect> |
not hard no, but not future compatibile |
| [02:08:02] |
<alect> |
yeah |
| [02:08:28] |
<alect> |
having it a nop should be enough for now no? |
| [02:10:23] |
<alect> |
i'm redoing my home page using trac: http://projects.swapoff.org/swapoff/wiki |
| [02:10:32] |
<alect> |
away from moinmoin |
| [02:11:08] |
<cmlenz> |
yeah, I'll make it a noop |
| [02:11:16] |
<alect> |
having the kid-style tree filters would let me do that entirely as a plugin, rather than having to hack the templates |
| [02:11:19] |
* |
alect is keen! |
| [02:11:22] |
<cmlenz> |
neat :-) |
| [02:11:25] |
<alect> |
kid! kid! kid! |
| [02:12:08] |
<cmlenz> |
I should start a branch for the template stuff sometime soon |
| [02:12:16] |
<s0undt3ch> |
alect: Error: Failed to load processor Include |
| [02:12:18] |
<s0undt3ch> |
:) |
| [02:12:30] |
<s0undt3ch> |
in http://projects.swapoff.org/swapoff/wiki/DevTodo |
| [02:12:44] |
<alect> |
yeah :) i have yet to migrate some of the moinmoin plugins |
| [02:12:46] |
<rapha> |
Are you guys all Edgewall coworkers? |
| [02:13:01] |
<alect> |
nope |
| [02:13:14] |
<rapha> |
Kay |
| [02:13:15] |
<alect> |
jborg, who sometimes comes in here, is an edgewall guy |
| [02:13:53] |
<rapha> |
I see... do you think they can be talked to in regard to implementing features first for pay? |
| [02:14:15] |
<alect> |
i'm not sure, they don't really do much trac development anymore |
| [02:14:29] |
<alect> |
what sort of features? |
| [02:15:30] |
<rapha> |
Not sure... Trac as it stands is a _pretty_ cool thing (for a single project, it plain ROCKS), but using it in a company with virtually hundreds of projects is a bit challenging. |
| [02:16:20] |
<rapha> |
I've been reading the discussion on the Wiki regarding handling multiple projects, but a way of replicating an existing installation without its data would already help a lot. |
| [02:17:22] |
<alect> |
you could check out drproject |
| [02:17:28] |
<rapha> |
(Like, give trac-admin a list of projects that a given installation will have to handle and then have some sort of template for reports and make one of these reports for each project and also add a custom field with a list of these projects) |
| [02:17:34] |
<rapha> |
drproject... never heard of it. |
| [02:17:38] |
* |
rapha looks at fm |
| [02:17:44] |
<alect> |
g00gl3 it |
| [02:18:04] |
<rapha> |
kay |
| [02:19:02] |
<rapha> |
That looks cool. |
| [02:23:02] |
<alect> |
the more i think about being able to provide extra wiki fields, the more i like that idea :) |
| [02:24:58] |
* |
misc has quit IRC |
| [02:26:47] |
<cmlenz> |
yeah |
| [02:27:28] |
<alect> |
the tags listing could be inserted as a "field" too |
| [02:29:32] |
<alect> |
is req.authname != 'anonymous' the best way to check the user is authenticated? |
| [02:38:25] |
<cmlenz> |
I prefer "req.authname and req.authname != 'anonymous'" |
| [02:38:40] |
<alect> |
ok |
| [02:45:52] |
<rapha> |
I can't get it to include an external stylesheet that is under trac/htdocs/css/index.css... |
| [02:46:23] |
<rapha> |
Trac itself includes /chrome/common/... but from the index template that doesn't seem to work. |
| [02:53:08] |
<cboos> |
rapha: the index.cs is a bit special: there's no Trac environment associated to it, hence the limited capabilities... |
| [02:55:43] |
<rapha> |
cboos: I'd be fine with being able to relatively link to any stylesheet on the server. Right now, only absolute links (http://.../...) work. |
| [02:58:21] |
<cmlenz> |
just use /path/on/my/server |
| [02:59:10] |
<rapha> |
Hmmm... maybe I need to fiddle around with the VHost setup some more for that to work. |
| [02:59:13] |
<rapha> |
Thanks anyway :-) |
| [03:06:00] |
<alect> |
is Ticket.select missing due to performance reasons? |
| [03:06:14] |
<alect> |
and WikiPage.select |
| [03:08:46] |
* |
tenshiKur0 has quit IRC |
| [03:25:47] |
* |
tuxipuxi has joined #trac |
| [03:36:48] |
<cmlenz> |
not really, probably just missing because we didn't need it as of yet |
| [03:38:19] |
* |
kousuke has quit IRC |
| [03:38:38] |
* |
idnar has quit IRC |
| [03:38:42] |
* |
idnar_ has joined #trac |
| [03:38:58] |
* |
idnar_ is now known as idnar |
| [03:51:14] |
<alect> |
ooh, it would be nice to be able to add controls to the wiki page "view" as well as when editing |
| [03:54:49] |
* |
Neo` has quit IRC |
| [03:54:58] |
* |
MasterC has joined #trac |
| [03:57:50] |
<cmlenz> |
how would those be submitted? |
| [03:58:31] |
<alect> |
that is a reasonable question |
| [03:58:36] |
<alect> |
magic! |
| [03:58:58] |
<cmlenz> |
heh |
| [03:59:01] |
<cmlenz> |
ajax! |
| [03:59:06] |
* |
Neo` has joined #trac |
| [03:59:11] |
<alect> |
yes! web 2.0! |
| [03:59:13] |
* |
kousuke has joined #trac |
| [03:59:25] |
<cmlenz> |
in validate_ticket, how do you suggest detecting whether the ticket is new? |
| [03:59:35] |
<cmlenz> |
ah, exists, I gues |
| [03:59:54] |
<alect> |
that's how i've been doing it yeah |
| [03:59:59] |
<alect> |
you writing a spam plugin? |
| [03:59:59] |
<cmlenz> |
yeah, that works |
| [04:00:09] |
<cmlenz> |
I'm migrating matt's stuff |
| [04:00:36] |
* |
sam` has quit IRC |
| [04:01:12] |
* |
mjg2 has quit IRC |
| [04:05:44] |
* |
sam` has joined #trac |
| [04:06:18] |
* |
mjg2 has joined #trac |
| [04:12:01] |
<alect> |
[athomas@blink:~]python2.5 -V |
| [04:12:01] |
<alect> |
Python 2.5a1 |
| [04:12:04] |
<alect> |
whoooo |
| [05:02:39] |
* |
cprov has joined #trac |
| [05:17:37] |
* |
tuxipuxi has quit IRC |
| [05:19:24] |
* |
misc has joined #trac |
| [05:19:55] |
* |
misc has quit IRC |
| [05:21:20] |
* |
misc has joined #trac |
| [05:52:09] |
* |
muness has joined #trac |
| [06:15:44] |
<alect> |
hey muness |
| [06:19:23] |
* |
cprov is now known as cprov-away |
| [06:21:48] |
<cboos> |
Q: is there a way to know, from inside an unit test, what kind of TestRunner is used? |
| [06:26:29] |
<cmlenz> |
Q back: why? |
| [06:27:05] |
<cboos> |
I'd like to reformat the msg given by the AssertionError, in a way readable on the console |
| [06:27:38] |
<cboos> |
but I guess this reformatting would break other runners (unless they don't parse the AssertionError, but do it differently...) |
| [06:29:00] |
* |
Erestar has joined #trac |
| [06:29:16] |
<cboos> |
(btw, I'm in the middle of moving everything around, so I hope you're not working on other parts of the wiki tests right now :) ) |
| [06:29:16] |
* |
danbeck has joined #trac |
| [06:29:48] |
<cboos> |
(besides r3152, which is OK of course) |
| [06:30:37] |
<alect> |
heh: # An administrator is allowed to spam |
| [06:31:17] |
<cmlenz> |
cboos: how would you make it more readable except for using linebreaks? |
| [06:32:09] |
<cboos> |
using line breaks but also make both of the text start on the same col, so the diff is easy to grasp |
| [06:32:29] |
<cmlenz> |
I don't see how that would break other test runners |
| [06:32:46] |
<cboos> |
well, ok then |
| [06:32:48] |
<alect> |
what would be mighty useful is if assertEqual() did some kind of introspection of the types, and displayed the differences usefully |
| [06:33:09] |
<alect> |
when you have two ticket fields with 20 attributes each, finding the differences is quite annoying |
| [06:41:36] |
* |
_idostyle has joined #trac |
| [06:44:19] |
* |
boorad has joined #trac |
| [06:45:00] |
* |
asmodai has quit IRC |
| [06:45:05] |
* |
asmodai has joined #Trac |
| [06:56:53] |
* |
idostyle has quit IRC |
| [06:57:01] |
* |
_idostyle is now known as idostyle |
| [07:06:41] |
<muness> |
alect: hi |
| [07:07:17] |
<alect> |
hey |
| [07:07:56] |
<muness> |
I noticed the checkins. Time for a release? |
| [07:08:44] |
<alect> |
there's a bug with tags with hyphens or plus symbols in them... |
| [07:08:54] |
<alect> |
but it's actually a bit tricky |
| [07:09:21] |
<alect> |
because it can be a valid link if expression_space=false |
| [07:09:25] |
* |
matt_good_ has joined #trac |
| [07:09:48] |
<muness> |
oh, yeah. |
| [07:09:54] |
<muness> |
makes sense. |
| [07:09:58] |
<alect> |
yeah :\ |
| [07:10:02] |
<alect> |
so...i dunno |
| [07:10:18] |
<muness> |
I'd say unless there is a really good reason, just say no hyphens, plus symbols or commas in tags. |
| [07:10:48] |
<alect> |
there is one reason...when you use the reporter or cc ticket fields, they can have email addresses in them :\ |
| [07:10:59] |
<muness> |
ugh. |
| [07:11:11] |
<muness> |
and those can have both hyphens and plus symbols. |
| [07:11:21] |
<alect> |
so stuff like the address here: http://trac-hacks.org/tags/alexander.klimetschek%40hpi.uni-potsdam.de |
| [07:11:31] |
<muness> |
maybe use a different symbols in the expression language? |
| [07:11:33] |
<alect> |
that's not showing anything because of the hyphen :( |
| [07:11:57] |
<muness> |
(I am not fond of that suggestions, but it'd save us grief) |
| [07:12:16] |
<alect> |
you can quote the tags and it works: http://trac-hacks.org/tags/'alexander.klimetschek%40hpi.uni-potsdam.de' |
| [07:12:22] |
* |
matt_good has quit IRC |
| [07:12:30] |
<alect> |
but of course, quoting them then doesn't work when expression_space=false |
| [07:12:32] |
<alect> |
can't win!!! |
| [07:12:49] |
<muness> |
why does quoting them not work when expression_space=false? |
| [07:13:03] |
<muness> |
(btw, why isn't that just expression=false?) |
| [07:13:35] |
<alect> |
yeah dunno why it isn't :) |
| [07:13:42] |
<alect> |
seemed like a good idea at the time |
| [07:14:15] |
<alect> |
it doesn't work because it doesn't have code to deal with quoting...it just grabs everything after the /tags/ prefix |
| [07:14:39] |
<muness> |
We can change the re to handle quoting. |
| [07:14:46] |
<muness> |
make it more consistent. |
| [07:15:00] |
<alect> |
going forward i'd like to exclusively use expressions |
| [07:15:15] |
<alect> |
much more powerful, and it will make the codebase claner |
| [07:15:20] |
<muness> |
true. |
| [07:15:27] |
<alect> |
yeah maybe that's the way forward |
| [07:15:30] |
<alect> |
handle quoting |
| [07:15:35] |
* |
alect hmsmmms |
| [07:16:08] |
<muness> |
what sort of things would break if we stick with only expressions right now, make commas the same as plus symbols in the expression language. |
| [07:16:10] |
<muness> |
? |
| [07:16:29] |
<muness> |
the expression language then becomes a superset of the old language, right? |
| [07:16:51] |
<alect> |
that is actually the case (re. commas) |
| [07:17:05] |
<alect> |
yeah |
| [07:17:11] |
<muness> |
so, why not drop the other parsing. expressions rule! |
| [07:17:41] |
<alect> |
only slight problem seems to be that IWikiSyntaxProvider seems to steal the quotes |
| [07:17:54] |
<alect> |
eg. tag:'foo@bar.com'+closed |
| [07:17:59] |
<alect> |
it only grabs up to the + |
| [07:18:09] |
<muness> |
that sucks. |
| [07:18:13] |
<muness> |
sounds like a bug to me. |
| [07:18:29] |
* |
cboos jumps in the conversation |
| [07:18:33] |
<alect> |
heh |
| [07:18:36] |
<alect> |
hi |
| [07:18:47] |
<cboos> |
yes this sounds like a bug... |
| [07:18:50] |
<alect> |
i assume it's for quoting links with spaces in them |
| [07:19:06] |
<cboos> |
what is expected is that there is a space after the '...' |
| [07:19:33] |
<cboos> |
ah, wait, you're talking about the short form... |
| [07:19:41] |
<cboos> |
outside of [], right? |
| [07:19:48] |
<alect> |
yep |
| [07:19:55] |
<cboos> |
ok, then it's a feature :) |
| [07:20:00] |
<muness> |
how? |
| [07:20:15] |
<alect> |
hehehe |
| [07:20:32] |
<cboos> |
I mean it's the expected behavior so far, but of course we can change that... |
| [07:21:29] |
<cboos> |
but what are you trying to achieve here, give 'foo@bar.com'+closed as a target? |
| [07:21:37] |
<muness> |
yes |
| [07:21:39] |
<cboos> |
If so, simply use tag:"'foo@bar.com'+closed" |
| [07:21:54] |
<cboos> |
should work ;) |
| [07:22:08] |
<alect> |
i tried that |
| [07:22:41] |
* |
mgalvin has joined #trac |
| [07:23:01] |
<alect> |
it ends up with: http://localhost:8080/devel/tags/%27%27asmodai%40in-nomine.org%27%2Bclosed%27 |
| [07:23:09] |
<alect> |
which is: ''asmodai@in-nomine.org'+closed' |
| [07:23:19] |
<alect> |
ie. not that helpful :) |
| [07:23:49] |
<alect> |
however, i think i can get around this by using get_wiki_syntax() no? |
| [07:23:57] |
<alect> |
provide my own parser |
| [07:24:40] |
* |
gluegadget has quit IRC |
| [07:25:08] |
<alect> |
alright, so...make expressions the default |
| [07:25:19] |
<alect> |
and roll our own wiki syntax doohicky |
| [07:25:28] |
<cboos> |
alect: I'll check what tag:"'foo@bar.com'+closed" does... |
| [07:26:32] |
<alect> |
ok |
| [07:26:58] |
<alect> |
i'm looking forward to your unicode guide christian |
| [07:27:17] |
<alect> |
i must admit to being unsure of where and how to apply unicode magic |
| [07:27:26] |
<muness> |
sounds great. except make the expressions the only way to parse. Any reason for others? |
| [07:28:00] |
<muness> |
i.e. the old way of parsing isn't needed, right? |
| [07:28:10] |
<alect> |
agreed |
| [07:28:21] |
<muness> |
Let's call it 0.4 then. |
| [07:28:27] |
<muness> |
? |
| [07:28:29] |
<cboos> |
well, using the above tag: expression in a wiki link resolver callback (i.e. the one given in get_link_resovers), you would get the following string: "\'foo@bar.com\'+closed" for the target |
| [07:29:08] |
<cboos> |
... and from there you certainly can make an usable link out of it ... |
| [07:29:13] |
<alect> |
hmm, so why does this: tag:"'asmodai@in-nomine.org'+closed" |
| [07:29:27] |
<alect> |
yield this: http://localhost:8080/devel/tags/%27%27asmodai%40in-nomine.org%27%2Bclosed%27 |
| [07:29:41] |
<alect> |
ie. ''asmodai@in-nomine.org'+closed' |
| [07:29:49] |
<alect> |
there is strangeness afoot |
| [07:30:36] |
<alect> |
also muness, i thought tag:/ could refer to the index |
| [07:30:58] |
<alect> |
i considered tag:index, but people might want to use index as a tag |
| [07:31:03] |
<muness> |
yeah. |
| [07:31:24] |
<muness> |
we already use . as a special character, so adding / is fine by me. |
| [07:31:41] |
<alect> |
okay, so 0.4 you think |
| [07:31:47] |
<muness> |
tag: bugs me a little doesn't seem intuitive. |
| [07:31:49] |
<alect> |
and then do a branch for the api changes? |
| [07:31:59] |
<muness> |
Yes. |
| [07:32:13] |
<alect> |
excellent |
| [07:32:18] |
<muness> |
tagged: ? |
| [07:32:29] |
<muness> |
findtags: ? |
| [07:32:47] |
<muness> |
(we can keep tag: , just add another that's more intuitive) |
| [07:33:23] |
<alect> |
i like tagged: |
| [07:33:34] |
<alect> |
tagged:athomas+closed |
| [07:34:00] |
<muness> |
yeah, you can just read that. |
| [07:34:05] |
* |
jerome has left #trac |
| [07:34:22] |
<alect> |
tag: sounds too much like a verb |
| [07:34:30] |
<muness> |
it confused me at first. |
| [07:34:40] |
<muness> |
I thought it was like the TagIt macro. |
| [07:34:48] |
<alect> |
yeah i recall that :) |
| [07:34:52] |
* |
alect smacks himself around |
| [07:35:01] |
<muness> |
I like having a verb that can be used to tag things. |
| [07:35:23] |
<muness> |
but I don't know if we can do that now. |
| [07:35:29] |
<muness> |
or should. |
| [07:35:37] |
<muness> |
might confuse too many people. |
| [07:35:47] |
<alect> |
how would that work? |
| [07:35:55] |
<muness> |
I like it because I like the idea of a wiki syntax as a dsl. |
| [07:35:57] |
<alect> |
what with the tag field |
| [07:36:08] |
<muness> |
Exactly. |
| [07:36:14] |
<muness> |
It'd be too confusing. |
| [07:36:26] |
<muness> |
Which would be the authority (maybe just accumulate them?) |
| [07:36:30] |
<alect> |
true |
| [07:36:46] |
<alect> |
i always found the TagIt macro kind of crufty.. |
| [07:36:52] |
<alect> |
just felt too error prone i guess |
| [07:36:57] |
<muness> |
It is. |
| [07:37:00] |
<alect> |
hehe |
| [07:37:22] |
<muness> |
But I liked being able to clone a wiki via copy/paste. |
| [07:37:28] |
<cboos> |
however, using a "link" resolver for that is a bit cumbersome too... |
| [07:38:00] |
<cboos> |
e.g. in TracMercurial, tag: would like to a specific tagged changeset... like tag:tip or tag:0.8 |
| [07:38:02] |
<muness> |
no worries - I don't miss TagIt much. |
| [07:38:37] |
<alect> |
any suggestions cboos? |
