[Glass] Metacello support

[Dale Henrichs via Glass]

On Sat, Nov 8, 2014 at 2:31 AM, Dario Trussardi via Glass <
glass at lists.gemtalksystems.com> wrote:

>
>
>
> On Fri, Nov 7, 2014 at 5:25 AM, Dario Trussardi via Glass <
> glass at lists.gemtalksystems.com> wrote:
>
>> Ciao Dale,
>>
>> Dario, I am glad that you found a solution to your problem, but frankly I
> don't think the utf8 problem was caused by the code that you "fixed".
>
>
> Of course, this is the problem but not the origin of the error.
>
>
> I asked for the utf8 bytes and I would like to see the utf8 bytes as I
> suspect that the utf8 problem is still lurking there for someone else to
> trip across. If you give me the utf8 bytes I will have a chance to
> partially characterize the problem. Please?
>
> As to the (presumably) new problem. Removing the first three characters
> from a DirectoryEntry will not do anything but ignore the file,
>
>
> Removing the first two wrong characters from a DirectoryEntry name,
>  permit the system to found the relative file.
>
> With not removing the system can't manage the file and after generate the
> error.
>
> so the real question is where did this .
>
>
> Yes this is the real question.  But i hoped my informations was sufficient
> for you.
>
> file come from? If you send me the contents of the . file and the path to
> the . file, we should be able to figure that out ...
>
>
> The size of the file is 4096  ( the size of a directory entry on linux )
>
> The copy from the stack of the file is :  ' Mac OS X         2 ° â
> TEXTR*chATTR RÊB âx This resource fork intentionally left blank   ÿÿ'
>

Okay, I'm still not sure whether I'm getting the UTF8 or not ... the string
that you supplied is not valid UTF8... I suppose the best thing would be
for you to attach the file to an email message and send it to me ... do not
copy and paste it ... just attach the file directly ...



> The methodEntry,  one   aDirectoryEntry  instance data is:
>
> sourceDirectory  pathName
> '/DTRRepository/DtrFinanza.package/MATimedIvaModel.class/instance'
>
> name  '._aliquotaIva.st'
>

you say that when you remove the first two characters that things load
correctly ... that means that presumably there is a file call
'aliquotaIva.st' in the directory?

I am very curious how you came to have a file called '._aliquotaIva.st' in
an instance directory where only methods should be written with a .st
suffix especially given the fact that $. is not a valid selector character
... besides given the text above there is no way that could have come from
the image as a method fileout ... Sooooo that file must have come from
someplace else ... Personally I haven't seen files that look like this on
aMac, but I will assume that it is an artifact of a text editor or other
program that you have running in the repository directory ... the string `This
resource fork intentionally left blank` is awfully suspicious ...

I think it is safe to say that this file wasn't produced during a package
save. But given the fact that $. is not a legal selector character I
suppose that FileTree should ignore files that begine with $. to avoid
problems like this in the future...

After a little googling I found that the files are created by the apple
os[1] ... according to one answer the files are created by the finder
and/or time machine ... It's odd that we've had a bunch of people
(including myself for several years now) using FileTree on a mac and yet
this is the first time I've heard of the problem:), but I think we have
finally gotten to the root of the problem ...

The following should be a patch the problem:

          ((self fileUtils directoryFromEntry: entry) entries
            select: [ :each | each name endsWith: '.st' and: [(each name
beginsWtih '.') not ])
            do: [ :methodEntry |

ignoring files starting with $. should be safe, because it is an invalid
selector character ...

Thanks for sticking with this so that we could get to the bottom of it ...
I've submitted a FileTree bug[2].

Dale

[1]
http://apple.stackexchange.com/questions/14980/why-are-dot-underscore-files-created-and-how-can-i-avoid-them
[2] https://github.com/dalehenrich/filetree/issues/134
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gemtalksystems.com/mailman/private/glass/attachments/20141110/82656596/attachment.html>