visionmedia / jspec
- Source
- Commits
- Network (13)
- Issues (62)
- Downloads (83)
- Wiki (1)
- Graphs
-
Branch:
master
-
following must support wait():
- specs
- hooks
- module hooks
todo:
- docs for wait()
Comments
-
Comments
-
>0 when failures are present
Comments
-
0 comments Created 2 months ago by visionmedia3.xxInclude jspec.growl.js in templatesGrowlxand have a wrapper checking if the binary exists..
Comments
-
to documentation and in templates?
Comments
-
Nothing works in IE... because IE sucks
Comments
-
0 comments Created 2 months ago by visionmediaIE8 mock timer "Object doesn't support this action" errorBugxIE.. I hate you, MS, I hate you. Your browser sucks, your dev tools suck, and your error messages make no sense. Is this your way of saying "no your not allowed to alter setTimeout() etc" ?
... anyways we need to fix this
Comments
-
0 comments Created 4 months ago by visionmediaBugxFix throw_error in Opera / IEMatchersxComments
-
Error cause when using --server ... reporting issue with puts()?
Comments
-
ProxyAssertion can then become subclass of Assertion.. etc
Comments
-
0 comments Created 3 months ago by visionmedia3.xxFix JSpec root when using --server and --{symlink,freeze}Bugxcontinues to use the jspec gem
Comments
-
People are getting confused.. thinking the grammar is manditory
Comments
-
and the methods which go along with it
Comments
-
1 comment Created 4 months ago by visionmediaJS load YAML manifest / config when available3.xxFor setting options / manifest
something like this http://gist.github.com/178041
make default for templates
Comments
visionmedia
Sun Sep 20 20:32:17 -0700 2009
| link
allow optional grammar use as well... ensure it works for everyone not just people using the gem
-
mock_request('/path').and_return(...)
Comments
-
When non-verbose the whitespace is a bit messed up
Comments
-
0 comments Created 4 months ago by visionmediaSet options via command-line switchesExecutablex$ jspec run --rhino --failures-only $ jspec run --server --verbose --failures-only
etc..
Comments
-
to test against with --bind and just to test faster without starting the server each time
$ jspec server start $ jspec server stopetc
Comments
-
now that mock_it() is available, move throw_error and other specific failure message assertions to their appropriate locations
Comments
visionmedia
Thu Sep 24 18:21:27 -0700 2009
| link
better coverage / shared behaviors etc
-
Documenting changes
Comments
-
CIJoe :)
Comments
-
for matchers etc... all core modules should supply several
Comments
-
Currently valid:
foo.should.be_a Function foo.should_be_a FunctionRSpec style
foo.should be_a(Function)Comments
-
currently 'before_each' is required
Comments
-
Comments
-
Comments
-
Comments
-
do not return.. utilize return value for something else useful?..
Comments
-
iteration... profile again and optimize
Comments
-
segregate everything into their own files for readability, then build with rake task.
Comments
-
Annoying to maintain several places
Comments
-
Would be nice to have, alot of people want to just run all specs without maintaining a manifest
Comments
-
Just wrap with a conditional checking of jQuery is available..
Comments
-
When 3.x becomes the mainline
Comments
-
just for fun
Comments
-
Comments
-
Comments
-
that lovely green color does not show on my windows command box. Windows does not support ANSI encodings. We need to find another way :)
Comments
-
0 comments Created 2 months ago by visionmedia3.xxbe_visible / be_hidden should use jQuery.fn.is(':hidden') etcMatchersxJSpec 2.x does not use this for historical reasons, (when jQuery did not check dimensions) however we should use this for the next version
Comments
-
0 comments Created 2 months ago by visionmedia3.xxPerformancexDetect if server has started and utilize "attatched" browsersServerxthis would allow for users to leave the server running or daemonized for quicker responses
Comments
-
4 comments Created about 1 month ago by visionmediaDetect windows and open a different default browser than SafariServerxComments
PhilHarnish
Thu Nov 26 13:02:44 -0800 2009
| link
I disagree. If anything is should be detected, it should be the default browser?
visionmedia
Thu Nov 26 13:09:01 -0800 2009
| link
forgot that windows has Safari nowadays, its tough for me to even work in the windows stuff I dont have parallels going or anything right now
PhilHarnish
Thu Nov 26 13:32:03 -0800 2009
| link
I did some googling. Here's my 2 cents.
If "browsers" are provided on the command line, use them. Else:
On Windows: system("start", filename)
Everywhere else: system("open", filename)Sound good?
Also: this shouldn't have the 3.x tag IMHO
visionmedia
Thu Nov 26 13:50:36 -0800 2009
| link
that is similar to whats currently implemented (could be better tho)
for example FF:
def visit uri system "open -g -a Firefox '#{uri}'" if macos? system "firefox #{uri}" if linux? system "#{File.join(ENV['ProgramFiles'] || 'c:\Program Files', '\Mozilla Firefox\firefox.exe')} #{uri}" if windows? end -
everything :D
move all readme to the site. make the site more like io's
dont forget to change fixture stuff!Comments
-
Comments
-
2 comments Created about 1 month ago by visionmedia3.xxInstallables should work on various platformsInstallablexOr at least wrap in #macos? and issue warnings etc
Comments
PhilHarnish
Thu Nov 26 13:01:32 -0800 2009
| link
What installables? JavaScript is cross platform.
visionmedia
Thu Nov 26 13:10:48 -0800 2009
| link
src/installables.rb uses some unix specific paths, installs env.js, rhino, prototype, blah blah all in spec/support
-
Comments
-
For Safari just use open -a -g like the rest
Comments
-
When jspec run [path] is a javascript file default to rhino
0 comments Created about 1 month ago by visionmediacurrently you would need to:
jspec run --rhino path/to/your.jsComments
-
0 comments Created about 1 month ago by visionmedia3.xxPrevent Browser::Default from opening when using --serverBugxwith no --browsers specified all supported browsers will run, which may
duplicate a browser if Default runs as wellComments
-
move environment/* files back to spec/
1 comment Created about 1 month ago by visionmediato fix FF's crabby pants
Comments
PhilHarnish
Fri Nov 27 17:57:21 -0800 2009
| link
Also resolves XHR relative URL issues in Chrome/Chromium (manifests itself in failing fixture tests).
-
Report containing file in formatted output
0 comments Created about 1 month ago by PhilHarnishWhenever possible the origin of the specs should be listed.
Locating a failing spec in a dozen files can be tedious.
Comments
-
Tagged specs would allow for running logical groups of tests or, conversely, identifying groups of problems when they fail.
For example:
I'm routing out a bunch of failing JSpec specs which are failing in IE only. Now that I understand the source of the problem the longterm fix is to tag them as flaky in IE, citing something like, "These JSpec features do not function correctly in your browser."One such IE failure goes something like this:
function Foobar () {}
alert(Foobar.name)
// IE: undefined / Everyone else: "Foobar"Comments
PhilHarnish
Fri Nov 27 18:29:19 -0800 2009
| link
To add: The failing tests aren't actually wrong. At worst, IE has less helpful error output than Firefox/Chrome/etc. To put it another way, the average user's specs will fail or pass consistently but the red error text will be different in IE than other browsers.
This is only a problem when you look at JSpec's specs which require the text to look a certain way.
We could either: fix the IE bugs (like jQuery fixes IE bugs) or tag em and forget em.
visionmedia
Sun Nov 29 19:39:21 -0800 2009
| link
yeah tough call. Personally I dont want to go patching up environments, but we do need JSpec to be consistent, especially in situations like this which may cause a lot of headaches providing false-negatives (first time ive had to use that one haha..).
I suppose we could do something like this:
it "" if (env.supports.verboseExceptions) -{ foo() }.should throw_error(SomeError, /some message/) else -{ foo() }.should throw_error() end
visionmedia
Sun Nov 29 19:39:52 -0800 2009
| link
not sure how common these problems will be though, if this exception one is the main one then maybe we should consider patching IE's lame ass
-
People will have to mock and stub like mad since JSpec is not queued yet but I need it for express
Comments
-
0 comments Created 26 days ago by visionmediaUse pure ruby for installing RhinoInstallablexBetter cross-platform compat
Comments
-
They do not appear to be working correctly. Also we should have
a different construct for shared behaviors, they should not be run they
should only provide as a list collection of specsComments
-
Both providing as documentation for doing so, and ... well testing for node.js
Perhaps auto-detect and bake support in.
Comments
visionmedia
Thu Dec 03 14:37:48 -0800 2009
| link
And remove the exports.JSpec or whatever I had before
visionmedia
Thu Dec 03 16:24:43 -0800 2009
| link
BLAH! fixtures are broken in node.js
-
When showing the output in the browser, it would be great to show the line numbers of the failing specs when you hover over the spec.
Comments
-
When showing the output in the browser, it would be great to show the origin filename to see where the spec came from.
Comments
-
Not sure when or why this went wrong, but I liked the assertion graphs on
the far rightComments
-
Display exception stack in some useful manor
0 comments Created 13 days ago by visionmediaTo large to flat-out display unconditionally ... option? DOM could display this in a nice way but Terminal cant
Comments
-
It would be nice to have a way of adding global hooks so all suites can share and execute them.
Something like JSpec.addHook(hook, body).Comments
- 3.x▾
- Bug▾
- Envjs▾
- Executable▾
- Formatters▾
- Grammar▾
- Growl▾
- Installable▾
- Matchers▾
- Mock XHR▾
- Modules▾
- Node.js▾
- Performance▾
- Server▾
- Templates▾
- Utilities▾
- Apply to Selection
-
Change Color…
Preview:preview
- Rename…
- Delete





