And yep, I think we can only really look at stuff that is in a project. remember you can add linked folders to a project so that is one way to create "mappings"
Regards
Mark Drew
> --
> You received this message because you are subscribed to the Google Groups "CFEclipse Developers" group.
> To post to this group, send email to cfecli...@googlegroups.com.
> To unsubscribe from this group, send email to cfeclipse-de...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/cfeclipse-dev?hl=en.
>
Look forward to your feedback. I'm still finding my feet so I am sure
there's many things in there that I should have been done better...
Is there anything special I need to do to link my pull request to that
issue? Or do I just put the URL in the comment?
I'll tidy it up and do this tomorrow morning my time.
Cheers,
Andrew.
> Is there anything special I need to do to link my pull request to that
> issue? Or do I just put the URL in the comment?
Think I just figured it put - I just push pull request from the issue.
Pretty obvious huh?!
MD
WOOT! (the contributions are freaking swell too! ;])
I must say it's been *awesome* to see folks picking up the slack. At
least if I get hit by a bus, CFE will be fine, just fine. Warms the
cockles.
But that's what it'll take (a bus, or meteor, etc.) for me to not wank
around with the most awesomest open source CFML editor!
So, long story short:
I know kung- er- ANTLR!
Not really. Saying that is like saying I know regular expressions, or
God or something. But still. We're lexing and parsing cfscript, and I
mean tits style!
Some of my knarly-est script is passing. Errors are marked. Functions
defined in other cfcs are populating our contextual assists (just
piggy-backed on the stuff Andrew/Mark did).
I haven't gone /too/ nuts on the grammar, as I'm hopeful that OpenBD can
use what I've done to get <cfscript>less cfc parsing, ternary and other
things Railo has rocking. It *is* geared for Railo (no required
semi-colons, the aforementioned ternary stuff, etc.) but it's easy
enough (ha!) at this point to put it back into more of an ACF/OpenBD
style deal. We'll need at least two grammars, perhaps 3 to have all
engines covered. Better that way than to try to have one uber-grammar
with ifs and whatnot. We'll see what the OBD folks think...
And then we should go NUTS with a REAL DAMN MODEL. I don't mind using
XML as the backing store still (XPath is pretty cool for traversal), or
we could use pure objects (the ANTLR treewalker will build whatever we
want), but regardless, we need a "real" model, vs. the current XML
/style/ model (you can describe whatever you want with XML, it doesn't
have to /be/ XML itself).
Anyways, this is all in the cfml.parsing project, (which we were already
using for the parsing) and as such, we've got a stand-alone parser that
can be used for lots of different things besides an IDE.
Anyhoo, I pushed a dev build out, but I have not tested it (I'm running
it locally, obviously, but I like to try to download/install a fresh
copy myself prior to telling other people to) so it might be awful.
If you see an outline while editing script based cfcs though, you'll know.
I'll just note that this allows us to do ALL KINDS of freakishly cool
things now. Quick fixes, yadda yadda. The missing info is there in
spades.
:Den
--
After twelve years of therapy my psychiatrist said something that
brought tears to my eyes. He said, "No hablo ingles."
Or, for lots of IDE's ;-)
http://wiki.netbeans.org/New_Language_Support_Tutorial_Antlr