Loading ...
Sorry, an error occurred while loading the content.

OS ARMOR Token - Remove Old Syntax

Expand Messages
  • karianna03
    Hi all, Waaayy back in the day the ARMOR token allowed: ARMORx as opposed to ARMOR.x None of our OS sheets have used the old syntax in light years (and it is
    Message 1 of 11 , Sep 10, 2009
    • 0 Attachment
      Hi all,

      Waaayy back in the day the ARMOR token allowed:

      ARMORx as opposed to ARMOR.x

      None of our OS sheets have used the old syntax in light years (and it is no longer documented, even in a deprecated form). Does anyone object if I remove the support for the old syntax in the ARMOR token?

      K
    • karianna03
      Hi all, ... And same goes for the sub tokens of ALL, EQUIPPED and NOT_EQUIPPED K
      Message 2 of 11 , Sep 10, 2009
      • 0 Attachment
        Hi all,

        > Waaayy back in the day the ARMOR token allowed:
        >
        > ARMORx as opposed to ARMOR.x
        >
        > None of our OS sheets have used the old syntax in light years (and it
        > is no longer documented, even in a deprecated form). Does anyone
        > object if I remove the support for the old syntax in the ARMOR token?

        And same goes for the sub tokens of ALL, EQUIPPED and NOT_EQUIPPED

        K
      • Michael W. Fender
        Accessing file 6639.18330.23647 ... Was it ever deprecated? -- Fluxxdog The worst crime you can commit against another human being is to make them think.
        Message 3 of 11 , Sep 10, 2009
        • 0 Attachment
          Accessing file 6639.18330.23647

          Data from user karianna03 is as follows:

          > Hi all,
          >
          > > Waaayy back in the day the ARMOR token allowed:
          > >
          > > ARMORx as opposed to ARMOR.x
          > >
          > > None of our OS sheets have used the old syntax in light years (and it
          > > is no longer documented, even in a deprecated form). Does anyone
          > > object if I remove the support for the old syntax in the ARMOR token?
          >
          > And same goes for the sub tokens of ALL, EQUIPPED and NOT_EQUIPPED

          Was it ever deprecated?
          --
          Fluxxdog

          The worst crime you can commit against another human being is to make them
          think.
        • Tom Parker
          This depends on your view of deprecation. For some of the tokens, PCGen pops out a message that says: You should be using x.y rather than xy .  But I think
          Message 4 of 11 , Sep 10, 2009
          • 0 Attachment
            This depends on your view of deprecation.

            For some of the tokens, PCGen pops out a message that says: "You should be using x.y rather than xy".  But I think this is written to the output screen for the output sheets, which means it isn't terribly obvious (given that even in a good export, a bunch of goop goes flying by on that screen).

            The output tokens can also be used in the formula system, and currently we are not validating against deprecated syntax, which means it could be hidden in LST files and would only warn people at runtime (which means you need an exhaustive test of the data to detect them)

            For both reasons, I do not believe we should be removing the code that supports the old syntax at this time.  We should be defining a strategy of how to detect them in the LST files (such as parsing the formulas at load - part of why I uploaded my formula system) and also to discuss our strategy for OS in general - meaning should we automatically convert the items to the new syntax, similar to how we can convert LST files.

            TP.
            --
            Tom Parker
            thpr@... and tppublic@...

            --- On Thu, 9/10/09, Michael W. Fender <fluxxdog@...> wrote:

            From: Michael W. Fender <fluxxdog@...>
            Subject: Re: [pcgen_developers] Re: OS ARMOR Token - Remove Old Syntax
            To: pcgen_developers@yahoogroups.com
            Date: Thursday, September 10, 2009, 11:00 AM

            Accessing file 6639.18330.23647

            Data from user karianna03 is as follows:

            > Hi all,
            >
            > > Waaayy back in the day the ARMOR token allowed:
            > >
            > > ARMORx as opposed to ARMOR.x
            > >
            > > None of our OS sheets have used the old syntax in light years (and it
            > > is no longer documented, even in a deprecated form).  Does anyone
            > > object if I remove the support for the old syntax in the ARMOR token?
            >
            > And same goes for the sub tokens of ALL, EQUIPPED and NOT_EQUIPPED

            Was it ever deprecated?
            --
            Fluxxdog

            The worst crime you can commit against another human being is to make them
            think.


            ------------------------------------

            Yahoo! Groups Links

            <*> To visit your group on the web, go to:
                http://groups.yahoo.com/group/pcgen_developers/

            <*> Your email settings:
                Individual Email | Traditional

            <*> To change settings online go to:
                http://groups.yahoo.com/group/pcgen_developers/join
                (Yahoo! ID required)

            <*> To change settings via email:
                mailto:pcgen_developers-digest@yahoogroups.com
                mailto:pcgen_developers-fullfeatured@yahoogroups.com

            <*> To unsubscribe from this group, send an email to:
                pcgen_developers-unsubscribe@yahoogroups.com

            <*> Your use of Yahoo! Groups is subject to:
                http://docs.yahoo.com/info/terms/


          • karianna03
            ... Kind of yes, it detected the old syntax and logged an Error each time. K
            Message 5 of 11 , Sep 10, 2009
            • 0 Attachment
              > > Hi all,
              > >
              > > > Waaayy back in the day the ARMOR token allowed:
              > > >
              > > > ARMORx as opposed to ARMOR.x
              > > >
              > > > None of our OS sheets have used the old syntax in light years
              > > > (and it is no longer documented, even in a deprecated form).
              > > > Does anyone object if I remove the support for the old syntax in
              > > > the ARMOR token?
              > >
              > > And same goes for the sub tokens of ALL, EQUIPPED and NOT_EQUIPPED
              >
              > Was it ever deprecated?

              Kind of yes, it detected the old syntax and logged an Error each time.

              K
            • karianna03
              Hi Tom/All, I agree, but in this case I ve done a scan of all of our OS Sheets and I can t see a single place where this old syntax is used (checked the
              Message 6 of 11 , Sep 10, 2009
              • 0 Attachment
                Hi Tom/All,

                I agree, but in this case I've done a scan of all of our OS Sheets and I can't see a single place where this old syntax is used (checked the forumlas as well).

                You are correct in that currently the code would log the incorrect syntax to the PCGen error Console.

                I can keep this if need be, I'll try to split it out into it's own methods and mark with @deprecated Javadoc so it's easy to remove later.

                K

                >
                > For some of the tokens, PCGen pops out a message that says: "You should be using x.y rather than xy".  But I think this is written to the output screen for the output sheets, which means it isn't terribly obvious (given that even in a good export, a bunch of goop goes flying by on that screen).
                >
                > The output tokens can also be used in the formula system, and currently we are not validating against deprecated syntax, which means it could be hidden in LST files and would only warn people at runtime (which means you need an exhaustive test of the data to detect them)
                >
                > For both reasons, I do not believe we should be removing the code that supports the old syntax at this time.  We should be defining a strategy of how to detect them in the LST files (such as parsing the formulas at load - part of why I uploaded my formula system) and also to discuss our strategy for OS in general - meaning should we automatically convert the items to the new syntax, similar to how we can convert LST files.
                >
                > TP.
                > --
                >
                > Tom Parker
                >
                > thpr@... and tppublic@...
                >
                > --- On Thu, 9/10/09, Michael W. Fender <fluxxdog@...> wrote:
                >
                > From: Michael W. Fender <fluxxdog@...>
                > Subject: Re: [pcgen_developers] Re: OS ARMOR Token - Remove Old Syntax
                > To: pcgen_developers@yahoogroups.com
                > Date: Thursday, September 10, 2009, 11:00 AM
                >
                > Accessing file 6639.18330.23647
                >
                > Data from user karianna03 is as follows:
                >
                > > Hi all,
                > >
                > > > Waaayy back in the day the ARMOR token allowed:
                > > >
                > > > ARMORx as opposed to ARMOR.x
                > > >
                > > > None of our OS sheets have used the old syntax in light years (and it
                > > > is no longer documented, even in a deprecated form).  Does anyone
                > > > object if I remove the support for the old syntax in the ARMOR token?
                > >
                > > And same goes for the sub tokens of ALL, EQUIPPED and NOT_EQUIPPED
                >
                > Was it ever deprecated?
                > --
                > Fluxxdog
                >
                > The worst crime you can commit against another human being is to make them
                > think.
                >
                >
                > ------------------------------------
                >
                > Yahoo! Groups Links
                >
              • Tom Parker
                Users have OS too. We also made a commitment to convert LST data automatically, and to me, that includes the formulas too.  Since one can use any of the OS
                Message 7 of 11 , Sep 10, 2009
                • 0 Attachment
                  Users have OS too.

                  We also made a commitment to convert LST data automatically, and to me, that includes the formulas too.  Since one can use any of the OS tokens in a Formula (with LST files, not just the OS), that means we need to detect and convert those to maintain that commitment.  (Which is why I uploaded my formula generator)

                  I take that conversion commitment seriously, and strongly feel we need a much larger community bought into the decision if we choose to back out of that commitment (which is what would happen if we removed OS function)

                  I would also like to reserve some time in the code meeting tomorrow for you to discuss the goals of your OS project.  I am still concerned that you and I see two different projects... so I'm wondering if what you are doing will meet my needs or whether I need to launch my own OS analysis project in order to meet the needs of the core/facet changes.

                  TP.
                  --
                  Tom Parker
                  thpr@... and tppublic@...

                  --- On Thu, 9/10/09, karianna03 <martijnverburg@...> wrote:

                  From: karianna03 <martijnverburg@...>
                  Subject: [pcgen_developers] Re: OS ARMOR Token - Remove Old Syntax
                  To: pcgen_developers@yahoogroups.com
                  Date: Thursday, September 10, 2009, 11:24 AM

                  Hi Tom/All,

                  I agree, but in this case I've done a scan of all of our OS Sheets and I can't see a single place where this old syntax is used (checked the forumlas as well).

                  You are correct in that currently the code would log the incorrect syntax to the PCGen error Console.

                  I can keep this if need be, I'll try to split it out into it's own methods and mark with @deprecated Javadoc so it's easy to remove later.

                  K

                  >
                  > For some of the tokens, PCGen pops out a message that says: "You should be using x.y rather than xy".  But I think this is written to the output screen for the output sheets, which means it isn't terribly obvious (given that even in a good export, a bunch of goop goes flying by on that screen).
                  >
                  > The output tokens can also be used in the formula system, and currently we are not validating against deprecated syntax, which means it could be hidden in LST files and would only warn people at runtime (which means you need an exhaustive test of the data to detect them)
                  >
                  > For both reasons, I do not believe we should be removing the code that supports the old syntax at this time.  We should be defining a strategy of how to detect them in the LST files (such as parsing the formulas at load - part of why I uploaded my formula system) and also to discuss our strategy for OS in general - meaning should we automatically convert the items to the new syntax, similar to how we can convert LST files.
                  >
                  > TP.
                  > --
                  >
                  > Tom Parker
                  >
                  > thpr@... and tppublic@...
                  >
                  > --- On Thu, 9/10/09, Michael W. Fender <fluxxdog@...> wrote:
                  >
                  > From: Michael W. Fender <fluxxdog@...>
                  > Subject: Re: [pcgen_developers] Re: OS ARMOR Token - Remove Old Syntax
                  > To: pcgen_developers@yahoogroups.com
                  > Date: Thursday, September 10, 2009, 11:00 AM
                  >
                  > Accessing file 6639.18330.23647
                  >
                  > Data from user karianna03 is as follows:
                  >
                  > > Hi all,
                  > >
                  > > > Waaayy back in the day the ARMOR token allowed:
                  > > >
                  > > > ARMORx as opposed to ARMOR.x
                  > > >
                  > > > None of our OS sheets have used the old syntax in light years (and it
                  > > > is no longer documented, even in a deprecated form).  Does anyone
                  > > > object if I remove the support for the old syntax in the ARMOR token?
                  > >
                  > > And same goes for the sub tokens of ALL, EQUIPPED and NOT_EQUIPPED
                  >
                  > Was it ever deprecated?
                  > --
                  > Fluxxdog
                  >
                  > The worst crime you can commit against another human being is to make them
                  > think.
                  >
                  >
                  > ------------------------------------
                  >
                  > Yahoo! Groups Links
                  >




                  ------------------------------------

                  Yahoo! Groups Links

                  <*> To visit your group on the web, go to:
                      http://groups.yahoo.com/group/pcgen_developers/

                  <*> Your email settings:
                      Individual Email | Traditional

                  <*> To change settings online go to:
                      http://groups.yahoo.com/group/pcgen_developers/join
                      (Yahoo! ID required)

                  <*> To change settings via email:
                      mailto:pcgen_developers-digest@yahoogroups.com
                      mailto:pcgen_developers-fullfeatured@yahoogroups.com

                  <*> To unsubscribe from this group, send an email to:
                      pcgen_developers-unsubscribe@yahoogroups.com

                  <*> Your use of Yahoo! Groups is subject to:
                      http://docs.yahoo.com/info/terms/


                • karianna03
                  Hi Tom, ... Yeah I was just thinking that. ... Fair enough, I agree actually. As an aside we can automatically convert the bad syntax internally and still
                  Message 8 of 11 , Sep 10, 2009
                  • 0 Attachment
                    Hi Tom,

                    > Users have OS too.

                    Yeah I was just thinking that.

                    > We also made a commitment to convert LST data automatically, and to
                    > me, that includes the formulas too.  Since one can use any of the OS
                    > tokens in a Formula (with LST files, not just the OS), that means we
                    > need to detect and convert those to maintain that commitment. 
                    > (Which is why I uploaded my formula generator)
                    >
                    > I take that conversion commitment seriously, and strongly feel we
                    > need a much larger community bought into the decision if we choose
                    > to back out of that commitment (which is what would happen if we
                    > removed OS function)

                    Fair enough, I agree actually. As an aside we can automatically convert the bad syntax internally and still display a warning to the user that the old syntax is deprecated and that they should make a permanent change in their OS. I'll make sure that solution is followed.

                    Do we then put the deprecated syntax docs back in place in the documentation? Can't quite remember what we do for LST here.

                    > I would also like to reserve some time in the code meeting tomorrow
                    > for you to discuss the goals of your OS project.  I am still
                    > concerned that you and I see two different projects... so I'm
                    > wondering if what you are doing will meet my needs or whether I need
                    > to launch my own OS analysis project in order to meet the needs of
                    > the core/facet changes.

                    Sure, I'm simply going through and making sure that the token does what it says in the docs. Every once in awhile I also find little things that I want to clarify there and then (while I'm thinking about it). I'm _not_ going to spend my time 100% purely checking code vs docs, that would drive me insane :).

                    The comparing docs vs code takes time as I'm also using a local troubleshooting sheet to output the various permutations (you can't rely on the Unit tests 100%) of the tokens, which in turn means I have to try and create some characters which exercise a majority of these tokens. That in turn causes me to find other bugs and issues in PCGen :).

                    As an aside if you're looking to have this analysis done by date X then please let me know now.

                    K

                    > TP.
                    > --
                    >
                    > Tom Parker
                    >
                    > thpr@... and tppublic@...
                    >
                    > --- On Thu, 9/10/09, karianna03 <martijnverburg@...> wrote:
                    >
                    > From: karianna03 <martijnverburg@...>
                    > Subject: [pcgen_developers] Re: OS ARMOR Token - Remove Old Syntax
                    > To: pcgen_developers@yahoogroups.com
                    > Date: Thursday, September 10, 2009, 11:24 AM
                    >
                    > Hi Tom/All,
                    >
                    > I agree, but in this case I've done a scan of all of our OS Sheets and I can't see a single place where this old syntax is used (checked the forumlas as well).
                    >
                    > You are correct in that currently the code would log the incorrect syntax to the PCGen error Console.
                    >
                    > I can keep this if need be, I'll try to split it out into it's own methods and mark with @deprecated Javadoc so it's easy to remove later.
                    >
                    > K
                    >
                    > >
                    > > For some of the tokens, PCGen pops out a message that says: "You should be using x.y rather than xy".  But I think this is written to the output screen for the output sheets, which means it isn't terribly obvious (given that even in a good export, a bunch of goop goes flying by on that screen).
                    > >
                    > > The output tokens can also be used in the formula system, and currently we are not validating against deprecated syntax, which means it could be hidden in LST files and would only warn people at runtime (which means you need an exhaustive test of the data to detect them)
                    > >
                    > > For both reasons, I do not believe we should be removing the code that supports the old syntax at this time.  We should be defining a strategy of how to detect them in the LST files (such as parsing the formulas at load - part of why I uploaded my formula system) and also to discuss our strategy for OS in general - meaning should we automatically convert the items to the new syntax, similar to how we can convert LST files.
                    > >
                    > > TP.
                    > > --
                    > >
                    > > Tom Parker
                    > >
                    > > thpr@ and tppublic@
                    > >
                    > > --- On Thu, 9/10/09, Michael W. Fender <fluxxdog@> wrote:
                    > >
                    > > From: Michael W. Fender <fluxxdog@>
                    > > Subject: Re: [pcgen_developers] Re: OS ARMOR Token - Remove Old Syntax
                    > > To: pcgen_developers@yahoogroups.com
                    > > Date: Thursday, September 10, 2009, 11:00 AM
                    > >
                    > > Accessing file 6639.18330.23647
                    > >
                    > > Data from user karianna03 is as follows:
                    > >
                    > > > Hi all,
                    > > >
                    > > > > Waaayy back in the day the ARMOR token allowed:
                    > > > >
                    > > > > ARMORx as opposed to ARMOR.x
                    > > > >
                    > > > > None of our OS sheets have used the old syntax in light years (and it
                    > > > > is no longer documented, even in a deprecated form).  Does anyone
                    > > > > object if I remove the support for the old syntax in the ARMOR token?
                    > > >
                    > > > And same goes for the sub tokens of ALL, EQUIPPED and NOT_EQUIPPED
                    > >
                    > > Was it ever deprecated?
                    > > --
                    > > Fluxxdog
                    > >
                    > > The worst crime you can commit against another human being is to make them
                    > > think.
                    > >
                    > >
                    > > ------------------------------------
                    > >
                    > > Yahoo! Groups Links
                    > >
                    >
                    >
                    >
                    >
                    > ------------------------------------
                    >
                    > Yahoo! Groups Links
                    >
                  • Tom Parker
                    My idea with the docs has generally been: If the deprecated syntax isn t in the docs, don t bother. Don t encourage people to use it :) If others disagree, and
                    Message 9 of 11 , Sep 10, 2009
                    • 0 Attachment
                      My idea with the docs has generally been:

                      If the deprecated syntax isn't in the docs, don't bother. Don't encourage people to use it :)

                      If others disagree, and we can rope Eric into doing the work...

                      TP.
                      --
                      Tom Parker
                      thpr@... and tppublic@...


                    • karianna03
                      Hi Tom/All, So how do we want to go about formally deprecating OS Tokens? IIRC, with the LST Tokens we would create a DEPRECATED version of the plugin that
                      Message 10 of 11 , Sep 11, 2009
                      • 0 Attachment
                        Hi Tom/All,

                        So how do we want to go about formally deprecating OS Tokens? IIRC, with the LST Tokens we would create a DEPRECATED version of the plugin that people could reference/use. Is that something we'd like to for OS?

                        K

                        > My idea with the docs has generally been:
                        >
                        > If the deprecated syntax isn't in the docs, don't bother. Don't encourage people to use it :)
                        >
                        > If others disagree, and we can rope Eric into doing the work...
                        >
                        > TP.
                        > --
                        >
                        > Tom Parker
                        >
                        > thpr@... and tppublic@...
                        >
                      • thpr
                        ... I would like to have a method of detecting deprecated items in OS that isn t involved with spitting out messages at runtime. IMHO, this is a different case
                        Message 11 of 11 , Sep 11, 2009
                        • 0 Attachment
                          --- In pcgen_developers@yahoogroups.com, "karianna03" <martijnverburg@...> wrote:
                          >
                          > Hi Tom/All,
                          >
                          > So how do we want to go about formally deprecating OS Tokens? IIRC, with the LST Tokens we would create a DEPRECATED version of the plugin that people could reference/use. Is that something we'd like to for OS?
                          >
                          > K

                          I would like to have a method of detecting deprecated items in OS that isn't involved with spitting out messages at runtime.

                          IMHO, this is a different case than LST tokens because those we only deprecated/removed once we could warn people at load. In fact there are still BONUS tokens that have been deprecated since something like 5.6 that I haven't removed because I can't convert them automatically (yet)

                          TP.
                        Your message has been successfully submitted and would be delivered to recipients shortly.