Hitachi Vantara Pentaho Community Forums
Results 1 to 9 of 9

Thread: repository explorer

  1. #1
    Brassrat Guest

    Default repository explorer

    folks,
    I have augmented the repository explorer,
    (1) you can now double-click on most entries to have some default
    action take place (such as opening a transformation or job, or
    creating a new directory (if you have the rights)). I have only tested
    this under windows xp,
    (2) the import dialog now supports multiple selections - so you can
    import, for example, all of the *.ktr files in a single file system
    directory to a selected kettle repository directory, You can select
    files containing transformations, or jobs, or repository exports (the
    only type supported in the past), each type is handled appropriately
    as the root xml tag identifies the file contents.
    (3) I added a check for transformation names that contain '/'s (such
    as transformations created by the wizard). While these names work fine
    in the repository they cause problems as file names, so i map '/'s to
    '_'s so the names are at least valid.

    hope these changes are useful,
    jay


    --~--~---------~--~----~------------~-------~--~----~
    You received this message because you are subscribed to the Google Groups "kettle-developers" group.
    To post to this group, send email to kettle-developers (AT) googlegroups (DOT) com
    To unsubscribe from this group, send email to kettle-developers-unsubscribe (AT) g...oups (DOT) com
    For more options, visit this group at http://groups.google.com/group/kettle-developers?hl=en
    -~----------~----~----~----~------~----~------~--~---

  2. #2
    Sven Boden Guest

    Default Re: repository explorer

    For 1) I wouldn't do the create directory... I wouldn't anything
    "destructive" as default option, I may kill that later

    2) I like

    3) I wouldn't do it... why do '/' while other operating systems may
    have problems with other separators.

    Regards,
    Sven


    On Jul 27, 3:04 pm, Brassrat <jgold... (AT) vestmark (DOT) com> wrote:
    > folks,
    > I have augmented the repository explorer,
    > (1) you can now double-click on most entries to have some default
    > action take place (such as opening a transformation or job, or
    > creating a new directory (if you have the rights)). I have only tested
    > this under windows xp,
    > (2) the import dialog now supports multiple selections - so you can
    > import, for example, all of the *.ktr files in a single file system
    > directory to a selected kettle repository directory, You can select
    > files containing transformations, or jobs, or repository exports (the
    > only type supported in the past), each type is handled appropriately
    > as the root xml tag identifies the file contents.
    > (3) I added a check for transformation names that contain '/'s (such
    > as transformations created by the wizard). While these names work fine
    > in the repository they cause problems as file names, so i map '/'s to
    > '_'s so the names are at least valid.
    >
    > hope these changes are useful,
    > jay



    --~--~---------~--~----~------------~-------~--~----~
    You received this message because you are subscribed to the Google Groups "kettle-developers" group.
    To post to this group, send email to kettle-developers (AT) googlegroups (DOT) com
    To unsubscribe from this group, send email to kettle-developers-unsubscribe (AT) g...oups (DOT) com
    For more options, visit this group at http://groups.google.com/group/kettle-developers?hl=en
    -~----------~----~----~----~------~----~------~--~---

  3. #3
    Sven Boden Guest

    Default Re: repository explorer

    But 2 misbehaves and corrupts the repository.... try importing twice
    the same list of files in different directories, you will get
    different copies of the same jobs/transformations... while their can
    only be one copy (a directory is just a tag to a transformation/job,
    each name should be unique).

    Regards,
    Sven


    On Jul 27, 7:59 pm, Sven Boden <list... (AT) pandora (DOT) be> wrote:
    > For 1) I wouldn't do the create directory... I wouldn't anything
    > "destructive" as default option, I may kill that later
    >
    > 2) I like
    >
    > 3) I wouldn't do it... why do '/' while other operating systems may
    > have problems with other separators.
    >
    > Regards,
    > Sven
    >
    > On Jul 27, 3:04 pm, Brassrat <jgold... (AT) vestmark (DOT) com> wrote:
    >
    > > folks,
    > > I have augmented the repository explorer,
    > > (1) you can now double-click on most entries to have some default
    > > action take place (such as opening a transformation or job, or
    > > creating a new directory (if you have the rights)). I have only tested
    > > this under windows xp,
    > > (2) the import dialog now supports multiple selections - so you can
    > > import, for example, all of the *.ktr files in a single file system
    > > directory to a selected kettle repository directory, You can select
    > > files containing transformations, or jobs, or repository exports (the
    > > only type supported in the past), each type is handled appropriately
    > > as the root xml tag identifies the file contents.
    > > (3) I added a check for transformation names that contain '/'s (such
    > > as transformations created by the wizard). While these names work fine
    > > in the repository they cause problems as file names, so i map '/'s to
    > > '_'s so the names are at least valid.

    >
    > > hope these changes are useful,
    > > jay



    --~--~---------~--~----~------------~-------~--~----~
    You received this message because you are subscribed to the Google Groups "kettle-developers" group.
    To post to this group, send email to kettle-developers (AT) googlegroups (DOT) com
    To unsubscribe from this group, send email to kettle-developers-unsubscribe (AT) g...oups (DOT) com
    For more options, visit this group at http://groups.google.com/group/kettle-developers?hl=en
    -~----------~----~----~----~------~----~------~--~---

  4. #4
    Jay Goldman Guest

    Default RE: repository explorer

    wrt to (3) the '/' is the java file separator and is included in the
    file name by the wizard - so maybe I should change the wizard? In any
    case, you can't have a file name in java that contains the file
    separator.

    -----Original Message-----
    From: kettle-developers (AT) googlegroups (DOT) com
    [mailto:kettle-developers (AT) googlegroups (DOT) com] On Behalf Of Sven Boden
    Sent: Friday, July 27, 2007 2:15 PM
    To: kettle-developers
    Subject: Re: repository explorer



    But 2 misbehaves and corrupts the repository.... try importing twice the
    same list of files in different directories, you will get different
    copies of the same jobs/transformations... while their can only be one
    copy (a directory is just a tag to a transformation/job, each name
    should be unique).

    Regards,
    Sven


    On Jul 27, 7:59 pm, Sven Boden <list... (AT) pandora (DOT) be> wrote:
    > For 1) I wouldn't do the create directory... I wouldn't anything
    > "destructive" as default option, I may kill that later
    >
    > 2) I like
    >
    > 3) I wouldn't do it... why do '/' while other operating systems may
    > have problems with other separators.
    >
    > Regards,
    > Sven
    >
    > On Jul 27, 3:04 pm, Brassrat <jgold... (AT) vestmark (DOT) com> wrote:
    >
    > > folks,
    > > I have augmented the repository explorer,
    > > (1) you can now double-click on most entries to have some default
    > > action take place (such as opening a transformation or job, or
    > > creating a new directory (if you have the rights)). I have only
    > > tested this under windows xp,
    > > (2) the import dialog now supports multiple selections - so you can
    > > import, for example, all of the *.ktr files in a single file system
    > > directory to a selected kettle repository directory, You can select
    > > files containing transformations, or jobs, or repository exports
    > > (the only type supported in the past), each type is handled
    > > appropriately as the root xml tag identifies the file contents.
    > > (3) I added a check for transformation names that contain '/'s (such


    > > as transformations created by the wizard). While these names work
    > > fine in the repository they cause problems as file names, so i map
    > > '/'s to '_'s so the names are at least valid.

    >
    > > hope these changes are useful,
    > > jay






    --~--~---------~--~----~------------~-------~--~----~
    You received this message because you are subscribed to the Google Groups "kettle-developers" group.
    To post to this group, send email to kettle-developers (AT) googlegroups (DOT) com
    To unsubscribe from this group, send email to kettle-developers-unsubscribe (AT) g...oups (DOT) com
    For more options, visit this group at http://groups.google.com/group/kettle-developers?hl=en
    -~----------~----~----~----~------~----~------~--~---

  5. #5
    Jay Goldman Guest

    Default RE: repository explorer

    Sven, thanks for the feedback,

    I could say "doctor, doctor" :-), but I will look at the code, I
    basically just used the existing code for importing repository xml
    files, just broke out the code for transformations and jobs as separate
    methods, then inspected the xml to see if it was a transform, job, or
    repository.

    I will add a check for duplicate name in a different directory ... and
    add this as an 'overwrite' case?

    wrt to (1) - the double-click only opens the dialog (if you have the
    rights to do the operation, such as creating a directory), you can just
    cancel if that is not what you want to do. Also, I think I did this to
    act like the pre-existing double-click support for database connections,
    etc where a double-click opens the create dialog. That said, I have no
    strong feelings either way, it was mostly to make opening transforms and
    jobs easier.

    -jay

    -----Original Message-----
    From: kettle-developers (AT) googlegroups (DOT) com
    [mailto:kettle-developers (AT) googlegroups (DOT) com] On Behalf Of Sven Boden
    Sent: Friday, July 27, 2007 2:15 PM
    To: kettle-developers
    Subject: Re: repository explorer



    But 2 misbehaves and corrupts the repository.... try importing twice the
    same list of files in different directories, you will get different
    copies of the same jobs/transformations... while their can only be one
    copy (a directory is just a tag to a transformation/job, each name
    should be unique).

    Regards,
    Sven


    On Jul 27, 7:59 pm, Sven Boden <list... (AT) pandora (DOT) be> wrote:
    > For 1) I wouldn't do the create directory... I wouldn't anything
    > "destructive" as default option, I may kill that later
    >
    > 2) I like
    >
    > 3) I wouldn't do it... why do '/' while other operating systems may
    > have problems with other separators.
    >
    > Regards,
    > Sven
    >
    > On Jul 27, 3:04 pm, Brassrat <jgold... (AT) vestmark (DOT) com> wrote:
    >
    > > folks,
    > > I have augmented the repository explorer,
    > > (1) you can now double-click on most entries to have some default
    > > action take place (such as opening a transformation or job, or
    > > creating a new directory (if you have the rights)). I have only
    > > tested this under windows xp,
    > > (2) the import dialog now supports multiple selections - so you can
    > > import, for example, all of the *.ktr files in a single file system
    > > directory to a selected kettle repository directory, You can select
    > > files containing transformations, or jobs, or repository exports
    > > (the only type supported in the past), each type is handled
    > > appropriately as the root xml tag identifies the file contents.
    > > (3) I added a check for transformation names that contain '/'s (such


    > > as transformations created by the wizard). While these names work
    > > fine in the repository they cause problems as file names, so i map
    > > '/'s to '_'s so the names are at least valid.

    >
    > > hope these changes are useful,
    > > jay






    --~--~---------~--~----~------------~-------~--~----~
    You received this message because you are subscribed to the Google Groups "kettle-developers" group.
    To post to this group, send email to kettle-developers (AT) googlegroups (DOT) com
    To unsubscribe from this group, send email to kettle-developers-unsubscribe (AT) g...oups (DOT) com
    For more options, visit this group at http://groups.google.com/group/kettle-developers?hl=en
    -~----------~----~----~----~------~----~------~--~---

  6. #6
    Sven Boden Guest

    Default Re: repository explorer

    If you start from partially broken code you get more broken
    code ;-) .. there was a bug out as
    http://kettle.javaforge.com/proj/tra...e&task_id=5912

    Essentially the way it worked is that transformation names (job should
    as well probably) should be unique over a repository. The directory
    they are in is just to classify them (DataStage works in the same
    way). If you save a job and change the directory it should just move
    the job e.g. (in fact in the database just overwrite the directory
    label of the job).

    Other parts of the repository depend on it, and start acting crazy
    with duplicates.

    The code to save transformations worked as it should before your
    change.

    Regards,
    Sven


    On Jul 27, 8:52 pm, "Jay Goldman" <jgold... (AT) vestmark (DOT) com> wrote:
    > Sven, thanks for the feedback,
    >
    > I could say "doctor, doctor" :-), but I will look at the code, I
    > basically just used the existing code for importing repository xml
    > files, just broke out the code for transformations and jobs as separate
    > methods, then inspected the xml to see if it was a transform, job, or
    > repository.
    >
    > I will add a check for duplicate name in a different directory ... and
    > add this as an 'overwrite' case?
    >
    > wrt to (1) - the double-click only opens the dialog (if you have the
    > rights to do the operation, such as creating a directory), you can just
    > cancel if that is not what you want to do. Also, I think I did this to
    > act like the pre-existing double-click support for database connections,
    > etc where a double-click opens the create dialog. That said, I have no
    > strong feelings either way, it was mostly to make opening transforms and
    > jobs easier.
    >
    > -jay
    >
    > -----Original Message-----
    > From: kettle-developers (AT) googlegroups (DOT) com
    >
    > [mailto:kettle-developers (AT) googlegroups (DOT) com] On Behalf Of Sven Boden
    > Sent: Friday, July 27, 2007 2:15 PM
    > To: kettle-developers
    > Subject: Re: repository explorer
    >
    > But 2 misbehaves and corrupts the repository.... try importing twice the
    > same list of files in different directories, you will get different
    > copies of the same jobs/transformations... while their can only be one
    > copy (a directory is just a tag to a transformation/job, each name
    > should be unique).
    >
    > Regards,
    > Sven
    >
    > On Jul 27, 7:59 pm, Sven Boden <list... (AT) pandora (DOT) be> wrote:
    > > For 1) I wouldn't do the create directory... I wouldn't anything
    > > "destructive" as default option, I may kill that later

    >
    > > 2) I like

    >
    > > 3) I wouldn't do it... why do '/' while other operating systems may
    > > have problems with other separators.

    >
    > > Regards,
    > > Sven

    >
    > > On Jul 27, 3:04 pm, Brassrat <jgold... (AT) vestmark (DOT) com> wrote:

    >
    > > > folks,
    > > > I have augmented the repository explorer,
    > > > (1) you can now double-click on most entries to have some default
    > > > action take place (such as opening a transformation or job, or
    > > > creating a new directory (if you have the rights)). I have only
    > > > tested this under windows xp,
    > > > (2) the import dialog now supports multiple selections - so you can
    > > > import, for example, all of the *.ktr files in a single file system
    > > > directory to a selected kettle repository directory, You can select
    > > > files containing transformations, or jobs, or repository exports
    > > > (the only type supported in the past), each type is handled
    > > > appropriately as the root xml tag identifies the file contents.
    > > > (3) I added a check for transformation names that contain '/'s (such

    >
    > > > as transformations created by the wizard). While these names work
    > > > fine in the repository they cause problems as file names, so i map
    > > > '/'s to '_'s so the names are at least valid.

    >
    > > > hope these changes are useful,
    > > > jay



    --~--~---------~--~----~------------~-------~--~----~
    You received this message because you are subscribed to the Google Groups "kettle-developers" group.
    To post to this group, send email to kettle-developers (AT) googlegroups (DOT) com
    To unsubscribe from this group, send email to kettle-developers-unsubscribe (AT) g...oups (DOT) com
    For more options, visit this group at http://groups.google.com/group/kettle-developers?hl=en
    -~----------~----~----~----~------~----~------~--~---

  7. #7
    Brassrat Guest

    Default Re: repository explorer

    Sven,
    I have to say i'm a little confused at the moment, after trying a few
    things in 2.5.1 and 3.0
    I'm not sure what you mean by:
    "The code to save transformations worked as it should before your
    change."

    It appears that in 2.5.1 and 3.0 you can save a transformation with
    the same name but different directory
    (not using save as ... but just create a new xform and save it. There
    is some special handing of the directory field, so when you change
    it the code knows to 'move' the transformaion. But if you just create
    a new xform and save it the name is only checked for uniqueness
    within
    the selected directory).

    Similarly, the code in RepositoryImportProgressDialog to check if the
    transformation already exists was and still is:
    long id = rep.getTransformationID(ti.getName(),
    targetDirectory.getID());

    This would not be correct if names have to be unique in the entire
    repository (processing for jobs uses getJobID which has same issue).

    The getJobID and getTransformationID methods take both a name and a
    directory id.
    Code using these methods to check for duplicate entries would allow
    entries with the same name but
    different directories (such as TransMeta#showReplaceWarning)

    Not exactly sure what to do here ... one approach would be to just
    eliminate the directory id from the getXxxxID calls
    which would really impose the 'names are unique' constraint. But from
    the shape of these calls someone thought that
    names only needed to be unique within a directory and the places that
    depend upon names being unique in the repository are wrong.

    So, for the time being, while i am more confused i don't think that i
    have made things any more confusing than they were, :-)
    -jay

    On Jul 27, 3:10 pm, Sven Boden <list... (AT) pandora (DOT) be> wrote:
    > If you start from partially broken code you get more broken
    > code ;-) .. there was a bug out ashttp://kettle.javaforge.com/proj/tracker/itemDetails.do?navigation=tr...
    >
    > Essentially the way it worked is that transformation names (job should
    > as well probably) should be unique over a repository. The directory
    > they are in is just to classify them (DataStage works in the same
    > way). If you save a job and change the directory it should just move
    > the job e.g. (in fact in the database just overwrite the directory
    > label of the job).
    >
    > Other parts of the repository depend on it, and start acting crazy
    > with duplicates.
    >
    > The code to save transformations worked as it should before your
    > change.
    >
    > Regards,
    > Sven
    >
    > On Jul 27, 8:52 pm, "Jay Goldman" <jgold... (AT) vestmark (DOT) com> wrote:
    >
    > > Sven, thanks for the feedback,

    >
    > > I could say "doctor, doctor" :-), but I will look at the code, I
    > > basically just used the existing code for importing repository xml
    > > files, just broke out the code for transformations and jobs as separate
    > > methods, then inspected the xml to see if it was a transform, job, or
    > > repository.

    >
    > > I will add a check for duplicate name in a different directory ... and
    > > add this as an 'overwrite' case?

    >
    > > wrt to (1) - the double-click only opens the dialog (if you have the
    > > rights to do the operation, such as creating a directory), you can just
    > > cancel if that is not what you want to do. Also, I think I did this to
    > > act like the pre-existing double-click support for database connections,
    > > etc where a double-click opens the create dialog. That said, I have no
    > > strong feelings either way, it was mostly to make opening transforms and
    > > jobs easier.

    >
    > > -jay

    >
    > > -----Original Message-----
    > > From: kettle-developers (AT) googlegroups (DOT) com

    >
    > > [mailto:kettle-developers (AT) googlegroups (DOT) com] On Behalf Of Sven Boden
    > > Sent: Friday, July 27, 2007 2:15 PM
    > > To: kettle-developers
    > > Subject: Re: repository explorer

    >
    > > But 2 misbehaves and corrupts the repository.... try importing twice the
    > > same list of files in different directories, you will get different
    > > copies of the same jobs/transformations... while their can only be one
    > > copy (a directory is just a tag to a transformation/job, each name
    > > should be unique).

    >
    > > Regards,
    > > Sven

    >
    > > On Jul 27, 7:59 pm, Sven Boden <list... (AT) pandora (DOT) be> wrote:
    > > > For 1) I wouldn't do the create directory... I wouldn't anything
    > > > "destructive" as default option, I may kill that later

    >
    > > > 2) I like

    >
    > > > 3) I wouldn't do it... why do '/' while other operating systems may
    > > > have problems with other separators.

    >
    > > > Regards,
    > > > Sven

    >
    > > > On Jul 27, 3:04 pm, Brassrat <jgold... (AT) vestmark (DOT) com> wrote:

    >
    > > > > folks,
    > > > > I have augmented the repository explorer,
    > > > > (1) you can now double-click on most entries to have some default
    > > > > action take place (such as opening a transformation or job, or
    > > > > creating a new directory (if you have the rights)). I have only
    > > > > tested this under windows xp,
    > > > > (2) the import dialog now supports multiple selections - so you can
    > > > > import, for example, all of the *.ktr files in a single file system
    > > > > directory to a selected kettle repository directory, You can select
    > > > > files containing transformations, or jobs, or repository exports
    > > > > (the only type supported in the past), each type is handled
    > > > > appropriately as the root xml tag identifies the file contents.
    > > > > (3) I added a check for transformation names that contain '/'s (such

    >
    > > > > as transformations created by the wizard). While these names work
    > > > > fine in the repository they cause problems as file names, so i map
    > > > > '/'s to '_'s so the names are at least valid.

    >
    > > > > hope these changes are useful,
    > > > > jay



    --~--~---------~--~----~------------~-------~--~----~
    You received this message because you are subscribed to the Google Groups "kettle-developers" group.
    To post to this group, send email to kettle-developers (AT) googlegroups (DOT) com
    To unsubscribe from this group, send email to kettle-developers-unsubscribe (AT) g...oups (DOT) com
    For more options, visit this group at http://groups.google.com/group/kettle-developers?hl=en
    -~----------~----~----~----~------~----~------~--~---

  8. #8
    Sven Boden Guest

    Default Re: repository explorer

    To be continued tomorrow
    Sven

    On Jul 27, 11:18 pm, Brassrat <jgold... (AT) vestmark (DOT) com> wrote:
    > Sven,
    > I have to say i'm a little confused at the moment, after trying a few
    > things in 2.5.1 and 3.0
    > I'm not sure what you mean by:
    > "The code to save transformations worked as it should before your
    > change."
    >
    > It appears that in 2.5.1 and 3.0 you can save a transformation with
    > the same name but different directory
    > (not using save as ... but just create a new xform and save it. There
    > is some special handing of the directory field, so when you change
    > it the code knows to 'move' the transformaion. But if you just create
    > a new xform and save it the name is only checked for uniqueness
    > within
    > the selected directory).
    >
    > Similarly, the code in RepositoryImportProgressDialog to check if the
    > transformation already exists was and still is:
    > long id = rep.getTransformationID(ti.getName(),
    > targetDirectory.getID());
    >
    > This would not be correct if names have to be unique in the entire
    > repository (processing for jobs uses getJobID which has same issue).
    >
    > The getJobID and getTransformationID methods take both a name and a
    > directory id.
    > Code using these methods to check for duplicate entries would allow
    > entries with the same name but
    > different directories (such as TransMeta#showReplaceWarning)
    >
    > Not exactly sure what to do here ... one approach would be to just
    > eliminate the directory id from the getXxxxID calls
    > which would really impose the 'names are unique' constraint. But from
    > the shape of these calls someone thought that
    > names only needed to be unique within a directory and the places that
    > depend upon names being unique in the repository are wrong.
    >
    > So, for the time being, while i am more confused i don't think that i
    > have made things any more confusing than they were, :-)
    > -jay
    >
    > On Jul 27, 3:10 pm, Sven Boden <list... (AT) pandora (DOT) be> wrote:
    >
    > > If you start from partially broken code you get more broken
    > > code ;-) .. there was a bug out ashttp://kettle.javaforge.com/proj/tracker/itemDetails.do?navigation=tr...

    >
    > > Essentially the way it worked is that transformation names (job should
    > > as well probably) should be unique over a repository. The directory
    > > they are in is just to classify them (DataStage works in the same
    > > way). If you save a job and change the directory it should just move
    > > the job e.g. (in fact in the database just overwrite the directory
    > > label of the job).

    >
    > > Other parts of the repository depend on it, and start acting crazy
    > > with duplicates.

    >
    > > The code to save transformations worked as it should before your
    > > change.

    >
    > > Regards,
    > > Sven

    >
    > > On Jul 27, 8:52 pm, "Jay Goldman" <jgold... (AT) vestmark (DOT) com> wrote:

    >
    > > > Sven, thanks for the feedback,

    >
    > > > I could say "doctor, doctor" :-), but I will look at the code, I
    > > > basically just used the existing code for importing repository xml
    > > > files, just broke out the code for transformations and jobs as separate
    > > > methods, then inspected the xml to see if it was a transform, job, or
    > > > repository.

    >
    > > > I will add a check for duplicate name in a different directory ... and
    > > > add this as an 'overwrite' case?

    >
    > > > wrt to (1) - the double-click only opens the dialog (if you have the
    > > > rights to do the operation, such as creating a directory), you can just
    > > > cancel if that is not what you want to do. Also, I think I did this to
    > > > act like the pre-existing double-click support for database connections,
    > > > etc where a double-click opens the create dialog. That said, I have no
    > > > strong feelings either way, it was mostly to make opening transforms and
    > > > jobs easier.

    >
    > > > -jay

    >
    > > > -----Original Message-----
    > > > From: kettle-developers (AT) googlegroups (DOT) com

    >
    > > > [mailto:kettle-developers (AT) googlegroups (DOT) com] On Behalf Of Sven Boden
    > > > Sent: Friday, July 27, 2007 2:15 PM
    > > > To: kettle-developers
    > > > Subject: Re: repository explorer

    >
    > > > But 2 misbehaves and corrupts the repository.... try importing twice the
    > > > same list of files in different directories, you will get different
    > > > copies of the same jobs/transformations... while their can only be one
    > > > copy (a directory is just a tag to a transformation/job, each name
    > > > should be unique).

    >
    > > > Regards,
    > > > Sven

    >
    > > > On Jul 27, 7:59 pm, Sven Boden <list... (AT) pandora (DOT) be> wrote:
    > > > > For 1) I wouldn't do the create directory... I wouldn't anything
    > > > > "destructive" as default option, I may kill that later

    >
    > > > > 2) I like

    >
    > > > > 3) I wouldn't do it... why do '/' while other operating systems may
    > > > > have problems with other separators.

    >
    > > > > Regards,
    > > > > Sven

    >
    > > > > On Jul 27, 3:04 pm, Brassrat <jgold... (AT) vestmark (DOT) com> wrote:

    >
    > > > > > folks,
    > > > > > I have augmented the repository explorer,
    > > > > > (1) you can now double-click on most entries to have some default
    > > > > > action take place (such as opening a transformation or job, or
    > > > > > creating a new directory (if you have the rights)). I have only
    > > > > > tested this under windows xp,
    > > > > > (2) the import dialog now supports multiple selections - so you can
    > > > > > import, for example, all of the *.ktr files in a single file system
    > > > > > directory to a selected kettle repository directory, You can select
    > > > > > files containing transformations, or jobs, or repository exports
    > > > > > (the only type supported in the past), each type is handled
    > > > > > appropriately as the root xml tag identifies the file contents.
    > > > > > (3) I added a check for transformation names that contain '/'s (such

    >
    > > > > > as transformations created by the wizard). While these names work
    > > > > > fine in the repository they cause problems as file names, so i map
    > > > > > '/'s to '_'s so the names are at least valid.

    >
    > > > > > hope these changes are useful,
    > > > > > jay



    --~--~---------~--~----~------------~-------~--~----~
    You received this message because you are subscribed to the Google Groups "kettle-developers" group.
    To post to this group, send email to kettle-developers (AT) googlegroups (DOT) com
    To unsubscribe from this group, send email to kettle-developers-unsubscribe (AT) g...oups (DOT) com
    For more options, visit this group at http://groups.google.com/group/kettle-developers?hl=en
    -~----------~----~----~----~------~----~------~--~---

  9. #9
    Sven Boden Guest

    Default Re: repository explorer

    Opinions are mixed on the uniqueness or non-uniqueness of names in the
    repository. Since the 3.0GA is still a long, long way away, I would
    suggest to leave it open for now to think on it. It was half-assed
    before, it's still half-assed now so not too much damage done :
    ( .... .

    In the end it should be decided which strategy the repository should
    take qua names and make it consistent everywhere:

    1) Names of jobs and transformations have to be unique in the
    repository, the directory is just a tag to classify items.
    2) The same name can be used in different directories.

    Both have advantages/disadvantages:

    Personally I would go for 1) the way DataStage does it. Why:
    A) having the same names in different directories will confuse people,
    making names unique forces people to differentiate
    Aa) e.g. in monitoring you can mention job A, you don't have to
    mention job A from directory X. Since names are unique.
    Ab) e.g. in spoon... currently you can't load multiple jobs with the
    same name even if they're in a different directory (and thus
    considered different by some people).
    B) Version control, reporting, ...

    The advantage of 2) is of course that's it's more "flexible". If 2)
    would be implemented you could still have 1) when you whack people
    over the head everytime they use duplicate names.

    Regards,
    Sven


    --~--~---------~--~----~------------~-------~--~----~
    You received this message because you are subscribed to the Google Groups "kettle-developers" group.
    To post to this group, send email to kettle-developers (AT) googlegroups (DOT) com
    To unsubscribe from this group, send email to kettle-developers-unsubscribe (AT) g...oups (DOT) com
    For more options, visit this group at http://groups.google.com/group/kettle-developers?hl=en
    -~----------~----~----~----~------~----~------~--~---

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Privacy Policy | Legal Notices | Safe Harbor Privacy Policy

Copyright © 2005 - 2019 Hitachi Vantara Corporation. All Rights Reserved.