modified /packages/main/_stylerexportfosi.acl on PE

[naglists]

Hiya,

I've made a minor modification to _styerlexportfosi.acl to support my
stylersheet. I deployed my modified version in .../custom/scripts on both
machines. It performs as intended in Editor, but it's not working at all on
Publishing Engine. I have restarted Tomcat. Obviously I can directly modify
the ACL inthe PE install tree, but I think putting it in scripts should
work. Am I missing something?

--
Paul Nagai

[ptc-1787870]

Hey Paul,

I use an init.acl file in the custom/init directory to run a source
command for each ACL file in the custom/scripts directory.

J




"Paul Nagai" <->
08/17/2007 10:47 AM
Please respond to
-


To
-
cc

Subject
[adepters] - modified /packages/main/_stylerexportfosi.acl on PE






Hiya,

I've made a minor modification to _styerlexportfosi.acl to support my
stylersheet. I deployed my modified version in .../custom/scripts on both
machines. It performs as intended in Editor, but it's not working at all
on Publishing Engine. I have restarted Tomcat. Obviously I can directly
modify the ACL inthe PE install tree, but I think putting it in scripts
should work. Am I missing something?

--
Paul Nagai
----------

[naglists]

On 8/17/07, Jason Aiken <jason.aiken@jeppesen.com> wrote:
>
> I use an init.acl file in the custom/init directory to run a source
> command for each ACL file in the custom/scripts directory.


Thanks for the thought, Jason.

That's way preferable to making changes in the install tree. In order to
further enhance the transparency of such a modification, I would follow the
strategy I use for all such code in either editinit or init. That is, create
an individual, eponymous acl file. This one would be:

source__stylerexportfosi.acl

I prefer the clutter of files but being able to see what's being set where.
(I don't go as far as including the set value, for example, it's
set_gentextwarnings.acl, not set_gentextwarnings_0.acl.

Really, I just wanted to use the word eponymous.

All that said, I still don't understand why having the file in scripts
doesn't capture the package/function call. I think it should. It's working
in Editor.

--
Paul Nagai

[ptc-1787870]

I like your file naming methodology. I do that for the contents of the
scripts directory, but tend to condense source and set commands in
init.acl.

Did you say that __stylerexportfosi.acl already exists?

Wild speculation, unfounded conjecture, and foamy dribble:
Maybe the default code is sourced in the Editor environment because it's a
Styler feature. On the PE side, I can see where it may not make sense to
have this code there.




"Paul Nagai" <->
08/17/2007 01:25 PM
Please respond to
-


To
-
cc

Subject
[adepters] - RE: modified /packages/main/_stylerexportfosi.acl on PE

[naglists]

On 8/17/07, Jason Aiken <jason.aiken@jeppesen.com> wrote:
>
> I like your file naming methodology. I do that for the contents of the
> scripts directory, but tend to condense source and set commands in
> init.acl.


It is one of those arguments that can get a wee bit religious .. it's
definitely easier at times to be able to open one file and see everything
you're setting. At other times, it's nice to be able to do that from just
peering into the directory. Not sure there's a best practice ...


Did you say that __stylerexportfosi.acl already exists?


Yes. It's in /packages/main.


Wild speculation, unfounded conjecture, and foamy dribble:
> Maybe the default code is sourced in the Editor environment because it's a
> Styler feature. On the PE side, I can see where it may not make sense to
> have this code there.


It has to be called by both Editor and PE since the function I'm "fixing" is
a composition feature, not an interface feature. It provides the
count/number of a numbered item.

To dive deeper than necessary, I have step as a listitem, but also
linkablestep. I want to be able to provide authors with a list containing
only the steps that have been identified as link targets. (Otherwise, you
can imagine wading through thousands of steps (and having had to provide
meaningful titles to those thousands).) My code counts both step and
steptarget to provide an accurate count of both.

--
Paul Nagai