Ok, all tests green, here are the two I added. Let me know if you
think additional scenarios need to be covered.
[Test]
public void engine_reuses_first_compile()
{
string path =
Path.GetFullPath(Path.Combine(AppDomain.CurrentDomain.BaseDirectory,
@"DslFactoryFixture\MyFirstCustom.boo"));
File.WriteAllText(path, "print 0");
try
{
engine.Compile(path);
Assert.IsFalse(engine.IsLastCompileCached);
engine.Compile(path);
Assert.IsTrue(engine.IsLastCompileCached);
}
finally
{
//clean up cached output so next test is clean
string filePath =
engine.CompilerContextCache.LastCachedFilePath;
if (File.Exists(filePath))
File.Delete(filePath);
if (File.Exists(path))
File.Delete(path);
}
}
[Test]
public void engine_recompiles_on_script_change()
{
string path =
Path.GetFullPath(Path.Combine(AppDomain.CurrentDomain.BaseDirectory,
@"DslFactoryFixture\MyCustom.boo"));
string filePath1 = null, filePath2 = null;
try
{
File.WriteAllText(path, "print 1");
engine.Compile(path);
Assert.IsFalse(engine.IsLastCompileCached);
filePath1 =
engine.CompilerContextCache.LastCachedFilePath;
File.WriteAllText(path, "print 2");
engine.Compile(path);
Assert.IsFalse(engine.IsLastCompileCached);
filePath2 =
engine.CompilerContextCache.LastCachedFilePath;
}
finally
{
//clean up cached output so next test is clean
if (File.Exists(filePath1))
File.Delete(filePath1);
if (File.Exists(filePath2))
File.Delete(filePath2);
//cleanup dynamic script
if (File.Exists(path))
File.Delete(path);
}
}
On Sep 11, 9:58 pm, webpaul <[EMAIL PROTECTED]> wrote:
> Ok, nevermind file locking issue resolved onto the caching.
>
> On Sep 11, 9:37 pm, webpaul <[EMAIL PROTECTED]> wrote:
>
>
>
> > It also appears that the boo compiler itself is hanging onto the file
> > once it is compiled. If I compile to memory I haven't been able to
> > find a way to save the in memory assembly, no save method on the
> > object and serializing doesn't work though I've found some posts on
> > people doing that. I'll keep looking.
>
> > On Sep 11, 9:28 pm, "Ayende Rahien" <[EMAIL PROTECTED]> wrote:
>
> > > Yes, persistent in this case means surviving application restar.
>
> > > On Fri, Sep 12, 2008 at 5:06 AM, webpaul <[EMAIL PROTECTED]> wrote:
>
> > > > On the storage interface there is a notify on change event. Any reason
> > > > you can think of to not just use that instead of the checksum?
>
> > > > On Sep 11, 8:37 pm, "Ayende Rahien" <[EMAIL PROTECTED]> wrote:
> > > > > Yeah :-)Load it using Assembly.Load(File.ReadAllBytes(file))
>
> > > > > Nasty, but works.
>
> > > > > On Fri, Sep 12, 2008 at 4:12 AM, webpaul <[EMAIL PROTECTED]> wrote:
>
> > > > > > Ok, it was compiling to memory, was able to get it to compile to a
> > > > > > file and cache it after that. The file is getting locked though so I
> > > > > > can't clean it up in the tests without closing mbunit. Looks like
> > > > > > this
> > > > > > is going to take longer than I thought. :)
>
> > > > > > On Sep 11, 3:22 pm, webpaul <[EMAIL PROTECTED]> wrote:
> > > > > > > That's what I was trying to do but that wasn't working out for me,
> > > > the
> > > > > > > file wasn't there for some reason. I'll try again later, maybe I'm
> > > > > > > missing something.
>
> > > > > > > On Sep 11, 2:05 pm, "Ayende Rahien" <[EMAIL PROTECTED]> wrote:
>
> > > > > > > > Sorry, the cache is persistent, it has to survive app restarts.
> > > > > > > > We already have transient cache. I am not sure that I follow why
> > > > > > serializing
> > > > > > > > the context is good. You can just start a new one with the
> > > > > > > > cached
> > > > > > generated
> > > > > > > > assembly
>
> > > > > > > > On Thu, Sep 11, 2008 at 9:21 PM, webpaul <[EMAIL PROTECTED]>
> > > > wrote:
>
> > > > > > > > > From my first look at the code that seemed to be the easiest
> > > > > > > > > way
> > > > to
> > > > > > do
> > > > > > > > > the caching. The last thing I tried was accessing the DLL
> > > > location
> > > > > > > > > created by the compiler context and when retrieving the cached
> > > > copy
> > > > > > to
> > > > > > > > > populate the object with the info but that didn't work either,
> > > > that
> > > > > > is
> > > > > > > > > in fact how the code is as-posted right now. So I tried
> > > > serializing
> > > > > > > > > the whole object and that, but neither seemed to work. I'm
> > > > > > > > > open
> > > > to
> > > > > > > > > suggestions if you think there is an easier way to do it.
>
> > > > > > > > > On Sep 11, 9:12 am, "Ayende Rahien" <[EMAIL PROTECTED]> wrote:
> > > > > > > > > > Why do you need to save the compiler context?
>
> > > > > > > > > > On Thu, Sep 11, 2008 at 4:44 PM, webpaul <[EMAIL PROTECTED]>
> > > > > > wrote:
>
> > > > > > > > > > > Working on doing this as per:
>
> > > > > >http://www.ayende.com/Blog/archive/2008/09/10/Persistent-DSL-caching-.
> > > > > > > > > ..
>
> > > > > > > > > > > Can someone suggest how to serialize or otherwise save a
> > > > > > > > > > > CompilerContext instance without modifying the Boo
> > > > > > > > > > > project as
> > > > > > well?-
> > > > > > > > > Hide quoted text -
>
> > > > > > > > > > - Show quoted text -- Hide quoted text -
>
> > > > > > > > - Show quoted text -- Hide quoted text -
>
> > > > > > > - Show quoted text -- Hide quoted text -
>
> > > > > - Show quoted text -- Hide quoted text -
>
> > > - Show quoted text -- Hide quoted text -
>
> > - Show quoted text -- Hide quoted text -
>
> - Show quoted text -
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Rhino Tools Dev" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/rhino-tools-dev?hl=en
-~----------~----~----~----~------~----~------~--~---
