[ANN] Metacello support for GitFileTree metadata-less mode

In the last month or so we've had a couple of different discussions on
this list involving adding support to Metacello (otherwise known as
"Cypress package extension") for GitFileTree's metadata-less mode[1] and
[4] and earlier this week I released a new version of Metacello[2] that
includes an updated version of the  "Cypress package extension"[5].

Depending on the version of Pharo you are using, at some point in time,
I expect this new release of Metacello to be available in the standard
download. Until then, to install the latest version of Metacello into
Pharo execute the following in a workspace:

   Metacello new
     baseline: 'Metacello';
     repository: 'github://dalehenrich/metacello-work:master/repository';
     get.
   Metacello new
     baseline: 'Metacello';
     repository: 'github://dalehenrich/metacello-work:master/repository';
     onConflict: [:ex | ex allow];
     load

If you are using GitFileTree's metadata-less mode and Metacello, then
add the following method to your BaselineOf:

    projectClass
    ^ MetacelloCypressBaselineProject

and you are good to go.

If you are curious as to why "Cypress package extensions" are needed,
you can read this comment[6] for a description of what rules Metacello
uses when fetching/loading packages from a FileTree repository using a
baseline.

On a slightly different topic, Alistair Grant ran into a bug involving
Metacello and how it semantic version numbers[7] a couple months ago and
this release includes a bug fix ... the bug was related to the fact that
the parser was too lenient and did throw an error for some forms of
invalid Semantic version numbers ... with this fix an error is thrown
... of course it is entirely possible that there are ConfigurationOfs
out in the wild that "depend upon the old behavior" so if you get an
"invalid version number" error while working with a configuration and
discover that it is not practical to redefine the version numbers to
conform to the semantic version number format then you can add the
following method to the ConfigurationOf and the old, buggy version of
the parser will be used:

   versionNumberClass
     ^ MetacelloOldSemanticVersionNumber

If you are sharing code between GemStone and Pharo, then you will want
to make sure that you install the new version of Metacello in GemStone
as well. See the GemStone release announcement for details[8].

Dale

[1]
http://forum.world.st/Saving-to-local-git-and-quot-Loading-all-file-names-from-http-pharo5-inbox-quot-tt4897962.html#a4905790
[2] https://github.com/dalehenrich/metacello-work/pull/401
[4]
http://forum.world.st/How-do-I-convert-a-FileTree-project-to-Metadataless-GitFileTree-project-tp4903920p4905054.html
[5] https://github.com/dalehenrich/metacello-work/issues/399
[6]
https://github.com/dalehenrich/metacello-work/issues/399#issuecomment-231464884
[7] https://github.com/dalehenrich/metacello-work/issues/392
[8]
http://forum.world.st/new-release-of-Metacello-metadata-less-FileTree-repository-support-part-IV-tt4906322.html

I need to proof read my posts more carefully...

On a slightly different topic, Alistair Grant ran into a bug involving Metacello and how it semantic version numbers[7] a couple

months ago and this release includes a bug fix ... the bug was related to the fact that the parser was too lenient and did throw an

On a slightly different topic, Alistair Grant ran into a bug involving Metacello and how it PARSES semantic version numbers[7] a couple months ago and this release includes a bug fix ... the bug was related to the fact that the parser was too lenient and did NOT throw an

error for some forms of invalid Semantic version numbers ... with this fix an error is thrown ... of course it is entirely possible that there are ConfigurationOfs out in the wild that "depend upon the old behavior" so if you get an "invalid version number" error while working with a configuration and discover that it is not practical to redefine the version numbers to conform to the semantic version number format then you can add the following method to the ConfigurationOf and the old, buggy version of the parser will be used:

  versionNumberClass
    ^ MetacelloOldSemanticVersionNumber

Le 14/07/2016 à 21:49, Dale Henrichs a écrit : Hi Dale,

When I try to upgrade Metacello in the latest Pharo 6 I get an error
"receiver  of withSqueakLineEnding is nil"

See: http://puu.sh/q3zet/310f6d3086.png

--
Cyril Ferlicot

http://www.synectique.eu

165 Avenue Bretagne
Lille 59000 France

On 07/17/2016 11:19 AM, Cyril Ferlicot D. wrote: Hello Cyril,

This appears to be related to the fact that the class comment file
(README.md) is unexpectedly missing from your download ... Since the
file is present on GitHub[1] ... perhaps the download from failed in an
odd way without triggering an error leading to this problem ... I would
have expected an incomplete download error along the way  --- if you
still have a copy of the Transcript a download error message should be
present ...

Retrying the git get/load might get a good result ...

Dale

[1]
https://github.com/dalehenrich/metacello-work/tree/master/repository/BaselineOfMetacello.package/BaselineOfMetacello.class

https://github.com/dalehenrich/filetree/blob/pharo6.0_dev/repository/MonticelloFileTree-Core.package/MCFileTreeStCypressReader.class/instance/addClassAndMethodDefinitionsFromEntry..st

Le 18/07/2016 à 20:33, Dale Henrichs a écrit : Hi Dale,

I can reproduce it on this computer in Pharo 5 and 6.

I also try to rewrite everything to be sure there was no weird character
during the copy paste. I wrote:

Metacello new
        githubUser: 'dalehenrich' project: 'metacello-work' commitish: 'master'
path: 'repository';
        baseline: 'Metacello';
        get.
Metacello new
        githubUser: 'dalehenrich' project: 'metacello-work' commitish: 'master'
path: 'repository';
        baseline: 'Metacello';
        onConflict: [ :e | e allow ];
        load

But I got the same.

Apparently there is a problem during the fetching of the baseline.
In the Transcript I got:

...RETRY->BaselineOfMetacello
...RETRY
...RETRY

I also deleted the github-cache before launching the command.

Here is a screen to help you a little: http://puu.sh/q67ok/74e233309c.png

I did not try on another OS (Windows 7 here), but I can try tomorrow.

BTW, this is not related but in pharo 6, the github-cache/ should not be
located in the image directory but in `SystemResolver
userLocalDirectory` instead. This is a directory created by Esteban to
reduce the noise created by the multiple files. Is it Metacello that
create this folder? Do you want me to create an issue on github?


> Dale
>
> [1]
> https://github.com/dalehenrich/metacello-work/tree/master/repository/BaselineOfMetacello.package/BaselineOfMetacello.class
>
>
> https://github.com/dalehenrich/filetree/blob/pharo6.0_dev/repository/MonticelloFileTree-Core.package/MCFileTreeStCypressReader.class/instance/addClassAndMethodDefinitionsFromEntry..st
>
>


--
Cyril Ferlicot

http://www.synectique.eu

165 Avenue Bretagne
Lille 59000 France

Cyril,

This is a FileTree issue[1] and the Pharo ports are maintained by Thierry Goubier so include @ThierryGoubier in the body and he will see the issue ..

I believe that the Pharo6.0 port is in progress as there is only a dev branch for Pharo6 at the moment[2].

Dale

[1] https://github.com/dalehenrich/filetree/issues/new
[2] https://github.com/dalehenrich/filetree/tree/pharo6.0_dev

Le 18/07/2016 à 20:33, Dale Henrichs a écrit :

Hello Cyril,

This appears to be related to the fact that the class comment file
(README.md) is unexpectedly missing from your download ... Since the
file is present on GitHub[1] ... perhaps the download from failed in an
odd way without triggering an error leading to this problem ... I would
have expected an incomplete download error along the way  --- if you
still have a copy of the Transcript a download error message should be
present ...

Retrying the git get/load might get a good result ...

Hi Dale,

I can reproduce it on this computer in Pharo 5 and 6.

I also try to rewrite everything to be sure there was no weird character
during the copy paste. I wrote:

Metacello new
	githubUser: 'dalehenrich' project: 'metacello-work' commitish: 'master'
path: 'repository';
	baseline: 'Metacello';
	get.
Metacello new
	githubUser: 'dalehenrich' project: 'metacello-work' commitish: 'master'
path: 'repository';
	baseline: 'Metacello';
	onConflict: [ :e | e allow ];
	load

But I got the same.

Apparently there is a problem during the fetching of the baseline.
In the Transcript I got:

...RETRY->BaselineOfMetacello
...RETRY
...RETRY

I also deleted the github-cache before launching the command.

Here is a screen to help you a little: http://puu.sh/q67ok/74e233309c.png

I did not try on another OS (Windows 7 here), but I can try tomorrow.

BTW, this is not related but in pharo 6, the github-cache/ should not be
located in the image directory but in `SystemResolver
userLocalDirectory` instead. This is a directory created by Esteban to
reduce the noise created by the multiple files. Is it Metacello that
create this folder? Do you want me to create an issue on github?

Dale

[1]
https://github.com/dalehenrich/metacello-work/tree/master/repository/BaselineOfMetacello.package/BaselineOfMetacello.class


https://github.com/dalehenrich/filetree/blob/pharo6.0_dev/repository/MonticelloFileTree-Core.package/MCFileTreeStCypressReader.class/instance/addClassAndMethodDefinitionsFromEntry..st

Le 18/07/2016 à 20:33, Dale Henrichs a écrit : Hi Dale,

I got some time to retry a little.

I tried to load in Pharo 6 a project of mine that is on GitHub and it
worked.

I tried in the same image to load Metacello but it failed.

When I go in the github-cache/ folder, I see that the Metacello folder
contains all the folder of each packages BUT they do not contains any
files[1]. The only file in the folder is the .filetree file in the
repository/ folder.

I get the problem into Pharo 5 and 6 on Windows 7. I do not have this
problem on OSX.

Also, I tried to download the zip of the repository and when I try to
extract it on Windows I get an error message saying that the zip file is
not valid[2].

When I try to clone the repository I get an error because some names are
too long[3].

--
[1] http://puu.sh/qcVoG/60e481ec1b.png
[2] http://puu.sh/qcVqE/f8a4dbd36d.png
[3] http://puu.sh/qcVkh/fbd16ba541.png

--
Cyril Ferlicot

http://www.synectique.eu

165 Avenue Bretagne
Lille 59000 France

On 7/24/16 11:29 AM, Cyril Ferlicot D. wrote: Cyril,

Thanks for providing a stack trace ... I recognize the issue --- windows
has a 256 char (I think) path length limit and a handful of the
Metacello methods exceed this limit ... I've created a dev_win-hack
branch on Metacello where I renamed the files that were exceeding the
limit[1] --- in my particular use case, git was not able to access the
files which is why I didn't recognize the issue initially --- there are
a garden basket of problems when trying to use windows, but the stack
provided the necessary detail to figure out which piece of fruit was
going bad :)

Anyway, the following should do the trick:

   Metacello new
     baseline: 'Metacello';
     repository:
'github://dalehenrich/metacello-work:dev_win-hack/repository';
     get.
   Metacello new
     baseline: 'Metacello';
     repository:
'github://dalehenrich/metacello-work:dev_win-hack/repository';
     onConflict: [:ex | ex allow];
     load

I just merged the latest master branch into the dev_win-hack branch, so
it will have the cypress package support --- but we may not be out of
the woods quite yet ... you probably need to locate your github-cache
directory as close to the root of your disk as possible. Use the
following to change the default github-cache location:

   MCGitHubRepository cacheDirectory: 'directory-path'

Good luck,

Dale

[1]
https://github.com/GsDevKit/GsDevKit_home/blob/master/docs/installation/configureOS.md#windows

Le 24/07/2016 à 22:23, Dale Henrichs a écrit : This seems to works perfectly :)

Thank you for the quick hack and answer! You rock!

I like to work a little on window for personal projects because I can
see all the bugs that people don't usually see and report them. :)

> Dale
>
> [1]
> https://github.com/GsDevKit/GsDevKit_home/blob/master/docs/installation/configureOS.md#windows
>
>


--
Cyril Ferlicot

http://www.synectique.eu

165 Avenue Bretagne
Lille 59000 France

On Mon, Jul 25, 2016 at 4:23 AM, Dale Henrichs
<[hidden email]> wrote:
Or could paths be specified in this format...
\\?\D:\very long path".

https://msdn.microsoft.com/en-us/library/aa365247(v=vs.85).aspx#maxpath

cheers -ben

On 7/24/16 4:47 PM, Ben Coman wrote: If I'm not mistaken, this kind of monkey business would have to be done
within the FileSystem code or maybe the vm ... I'm pretty sure that
FileTree is using FileReferences to access the files with long file
paths in a platform independent manner ...

Dale