Re: Module variables being set late

[Mark . Burgess]

This is another reason why I have not altered the behaviour. I do not
want to catch anyone by surprise. I have considered solving the problem
in a different way, but that will require more work,

M

On  8 Jan, Chip Seraphine wrote:
> 
> Sounds useful, although if the patch is incorproated it should be controlled 
> by a command-line option or something( since it changes behavior).
> 
> Unfortunately, this doesn't really address what is breaking me--  the problem 
> I'm having is that modules appear to be executing in a sequence that seems at 
> odds with the actionsequence definition.
> 
> On Wednesday 07 January 2004 17:37, Wipf, Stefan wrote:
>> I previously posted a simple patch for nesting imports:
>> Sorry, I don't remember which version of cfengine,
>> but the patch is simple enough.
>>
>>
>>
>> diff parse.c.original parse.c.fix
>> 127c127
>> < { struct Item *ptr;
>> ---
>>
>> > { struct Item *ptr, *nextptr
>>
>> 216c216,224
>> <    ParseFile(filename,sp);
>> ---
>>
>> >    nextptr = ptr->next;
>> >    while (VIMPORT != NULL) { VIMPORT = VIMPORT->next; }
>> >    ParseFile(filename,sp);
>> >    if (VIMPORT != NULL)
>> >       {
>> >       ptr->next = VIMPORT;
>> >       while (VIMPORT->next != NULL) { VIMPORT = VIMPORT->next; }
>> >       VIMPORT->next = nextptr;
>> >       }
>>
>> This patch will change the behavior as follows:
>>
>> example:
>>
>> cfagent.conf:
>>  import:
>>   cf.classes
>>   cf.variables
>>   cf.main
>>
>> cf.classes:
>>  import:
>>   cf.classes1
>>   cf.classes2
>>   ...
>>
>> Without the patch:
>> importing cf.classes
>> importing cf.variables
>> importing cf.main
>> importing cf.classes1
>> importing cf.classes2
>>
>> With the patch:
>> importing cf.classes
>> importing cf.classes1
>> importing cf.classes2
>> importing cf.variables
>> importing cf.main
>>
>> Mark.Burgess@iu.hio.no wrote:
>> > On  7 Jan, Wheeler, John wrote:
>> > >>If you
>> > >> follow
>> > >> the suggested structure and use cfagent.conf only to import the
>> > >
>> > > content
>> > >
>> > >> of the configuration, then it never matters that imports are done "at
>> > >
>> > > the
>> > >
>> > >> end".
>> > >
>> > > [Wheeler, John]
>> > > I didn't realize this was a limitation/feature/characteristic/property
>> > > (i.e. don't read into the adjective) of cfengine. I have many nested
>> > > imports of cfengine some as deep as 5 levels.
>> > >
>> > > Is this bad? Will this fail with future releases of cfengine?
>> >
>> > No - nesting is ok. But the nesting is evaluated linearly.
>> > Nested evaluation is technically challenging and not something
>> > I want to spent time on.
>> >
>> > M
>> >
>> > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> > Work: +47 22453272            Email:  Mark.Burgess@iu.hio.no
>> > Fax : +47 22453205            WWW  :  http://www.iu.hio.no/~mark
>> > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> >
>> > _______________________________________________
>> > Help-cfengine mailing list
>> > Help-cfengine@gnu.org
>> > http://mail.gnu.org/mailman/listinfo/help-cfengine
> 
> 
> 
> _______________________________________________
> Help-cfengine mailing list
> Help-cfengine@gnu.org
> http://mail.gnu.org/mailman/listinfo/help-cfengine



~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Work: +47 22453272            Email:  Mark.Burgess@iu.hio.no
Fax : +47 22453205            WWW  :  http://www.iu.hio.no/~mark
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~