[Gin] Add a basic unit testing framework

[Gin] Add a basic unit testing framework

Previously, we were using JavaScript bindings to gtest to unit test JavaScript
code in Gin and Mojo. The gtest bindings were very basic and not very
idiomatic.

This CL introduces a simple AMD module call "expect" to help us write more
idiomatic unit tests. The API for "expect" is based on the popular Jasmine unit
testing framework for node.js. I investigated just importing one of Node's many
unit testing frameworks, but they all try to do too much (e.g., drive the
entire test harness via Node's file system interface).

The "expect" modules doesn't try to drive the testing process. We just let
gtest handle that. Instead, "expect" just provides a simple language in which
to write test assertions. We'll likely evolve our testing strategy over time,
but hopefully this CL is an improvement over the primitive gtest bindings.

[email protected]
[email protected]
BUG=none

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=237145

[jochen (gone - plz use gerrit), 2013-11-25 08:19:24]

lgtm

can you add a BUG= line?

Can you make sure the reference to the Jasmine license doesn't trip the
checklicense.py check? Do we need to add this to chrome://credits?

https://codereview.chromium.org/85223002/diff/40001/gin/converter.cc
File gin/converter.cc (right):

https://codereview.chromium.org/85223002/diff/40001/gin/converter.cc#newcode174
gin/converter.cc:174: std::string V8ToString(v8::Handle<v8::Value> value) {
i wonder whether this should be a template

template<typename T>
T FromV8(v8::Handle<v8::Value> value) {
  if (value.IsEmpty())
    return T();
  T result;
  if (!ConvertFromV8(value, &result))
    return T();
  return result;
}

wdyt

[abarth-chromium, 2013-11-25 16:29:13]

On 2013/11/25 08:19:24, jochen wrote:
> can you add a BUG= line?

What bug should I reference?  I closed the bug about creating JavaScript
bindings for Mojo because that more or less works now.  Is there a benefit to
having a bug to reference?

> Can you make sure the reference to the Jasmine license doesn't trip the
> checklicense.py check? Do we need to add this to chrome://credits?

I'm pretty sure we don't need to add anything to chrome://credits.  It's similar
to gtest and gmock in the sense that we don't ship the code.  In license.py,
those directories have the NOT_SHIPPED annotation:

    # Special-case modules that aren't in the shipping product, so don't need
    # their license in about:credits.
    if metadata["License File"] != NOT_SHIPPED:

> https://codereview.chromium.org/85223002/diff/40001/gin/converter.cc
> File gin/converter.cc (right):
> 
>
https://codereview.chromium.org/85223002/diff/40001/gin/converter.cc#newcode174
> gin/converter.cc:174: std::string V8ToString(v8::Handle<v8::Value> value) {
> i wonder whether this should be a template
> 
> template<typename T>
> T FromV8(v8::Handle<v8::Value> value) {
>   if (value.IsEmpty())
>     return T();
>   T result;
>   if (!ConvertFromV8(value, &result))
>     return T();
>   return result;
> }

The problem is that C++ can't guess T from the return type.  You'd have to write
FromV8<std::string> everywhere.  Can add that if you like and then add a string
specialization, like we have in other direction.

[jochen (gone - plz use gerrit), 2013-11-25 16:32:01]

On 2013/11/25 16:29:13, abarth wrote:
> On 2013/11/25 08:19:24, jochen wrote:
> > can you add a BUG= line?

ok, I assumed it was part of a series of patches. Some ppl use BUG=none to
indicate that there's no bug, but that's up to you.

> 
> What bug should I reference?  I closed the bug about creating JavaScript
> bindings for Mojo because that more or less works now.  Is there a benefit to
> having a bug to reference?
> 
> > Can you make sure the reference to the Jasmine license doesn't trip the
> > checklicense.py check? Do we need to add this to chrome://credits?
> 
> I'm pretty sure we don't need to add anything to chrome://credits.  It's
similar
> to gtest and gmock in the sense that we don't ship the code.  In license.py,
> those directories have the NOT_SHIPPED annotation:
> 
>     # Special-case modules that aren't in the shipping product, so don't need
>     # their license in about:credits.
>     if metadata["License File"] != NOT_SHIPPED:

sg

> 
> > https://codereview.chromium.org/85223002/diff/40001/gin/converter.cc
> > File gin/converter.cc (right):
> > 
> >
>
https://codereview.chromium.org/85223002/diff/40001/gin/converter.cc#newcode174
> > gin/converter.cc:174: std::string V8ToString(v8::Handle<v8::Value> value) {
> > i wonder whether this should be a template
> > 
> > template<typename T>
> > T FromV8(v8::Handle<v8::Value> value) {
> >   if (value.IsEmpty())
> >     return T();
> >   T result;
> >   if (!ConvertFromV8(value, &result))
> >     return T();
> >   return result;
> > }
> 
> The problem is that C++ can't guess T from the return type.  You'd have to
write
> FromV8<std::string> everywhere.  Can add that if you like and then add a
string
> specialization, like we have in other direction.

True. I guess it's not worth it then,.

[commit-bot: I haz the power, 2013-11-25 17:00:06]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/85223002/40001

[commit-bot: I haz the power, 2013-11-25 18:31:40]

Retried try job too often on android_dbg for step(s) slave_steps
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=android_db...

[commit-bot: I haz the power, 2013-11-25 18:34:05]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/85223002/40001

[abarth-chromium, 2013-11-25 18:43:14]

@joth: Can you take a look at webview_licenses.py ?  Thanks!

[abarth-chromium, 2013-11-25 20:35:25]

On 2013/11/25 18:43:14, abarth wrote:
> @joth: Can you take a look at webview_licenses.py ?  Thanks!

@joth: I'm going to TBR the webview_licenses.py change to you.  This code isn't
related to android_webview at all.  If there's an issue, I'm happy to fix it in
a followup CL.

[commit-bot: I haz the power, 2013-11-25 20:37:04]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/85223002/70001

[abarth-chromium, 2013-11-25 21:26:38]

Committed patchset #3 manually as r237145 (presubmit successful).

[joth, 2013-11-25 22:20:27]

lgtm

[abarth-chromium, 2013-11-25 22:31:26]

On 2013/11/25 22:20:27, joth wrote:
> lgtm

Thanks!