Home » railML newsgroups » railml.timetable » 1.1R2 - sectionTT element
1.1R2 - sectionTT element [message #685] Mon, 29 December 2008 20:23 Go to next message
Teemu Sirkiä is currently offline  Teemu Sirkiä
Messages: 5
Registered: December 2008
Junior Member
In version 1.1 R2 there are sectionTT elements like this:

<sectionTT sectionRefId="85ZVIA-85ZWIP" lineRefId="70005" trackInfo="302"
minRunTime="PT0M36S" distance="0.907"/>

In my opinion sectionRefId attribute should be changed so that there are
different attributes for the ID that starts and the ID that ends the section
like this:

<sectionTT sectionStartID="85ZVIA" sectionEndID="85ZWIP" lineRefId="70005"
trackInfo="302" minRunTime="PT0M36S" distance="0.907"/>

Here are some reasons:

- There is no documentation how those two Id's are combined in one attribute
- There is no reason to combine two things in one attribute
- What about if the ID contains "-" character?

-Teemu
Re: 1.1R2 - sectionTT element [message #686 is a reply to message #685] Mon, 05 January 2009 11:01 Go to previous messageGo to next message
Joachim Rubröder railML is currently offline  Joachim Rubröder railML
Messages: 0
Registered: November 2019
Dear Teemu,

the sectionTT element is meant for information concerning the section
between two ocpTTs of a train. The two attributes sectionRefId and
lineRefId are referencing section and line elements which are part of the
infrastructure. Unfortunately there is no such section element in the
infrastructure yet but I think there should be one. So I invented the
usage sectionRefId="85ZVIA-85ZWIP" with a combination of two ocp elements.
But it is still a RefId so it depends on the definition of a section.

With the current structure of the infrastructure I agree that a
combination sectionStartID/sectionEndID would fit better. But I hope that
there will be soon a section concept in the infrastructure.

What do you think as best connection between timetable and infrastructure
by references? How would you describe a whole network with high-level
connections from station to station in the terms of railML-infrastructure?

Kind regards,
Joachim
Re: 1.1R2 - sectionTT element [message #687 is a reply to message #686] Mon, 05 January 2009 14:33 Go to previous messageGo to next message
Teemu Sirkiä is currently offline  Teemu Sirkiä
Messages: 5
Registered: December 2008
Junior Member
"Joachim Rubröder" <coord(at)timetablerailmlorg> kirjoitti viestissä
news:gjslp9$oo9$1(at)sifaivifhgde...
> Dear Teemu,
>
> the sectionTT element is meant for information concerning the section
> between two ocpTTs of a train. The two attributes sectionRefId and
> lineRefId are referencing section and line elements which are part of the
> infrastructure. Unfortunately there is no such section element in the
> infrastructure yet but I think there should be one. So I invented the
> usage sectionRefId="85ZVIA-85ZWIP" with a combination of two ocp elements.
> But it is still a RefId so it depends on the definition of a section.
>
> With the current structure of the infrastructure I agree that a
> combination sectionStartID/sectionEndID would fit better. But I hope that
> there will be soon a section concept in the infrastructure.
>
> What do you think as best connection between timetable and infrastructure
> by references? How would you describe a whole network with high-level
> connections from station to station in the terms of railML-infrastructure?

I understand that sectionRefId is an ID but I was just thinking how to parse
a RaiML file containing timetable with current structure.

As I said, there is no documentation how that ID is generated. Can I assume
that it is always two ocpRefIds combined with "-" character? If it is so
then it is easy to get for example trackInfo between two ocpTTs and there is
no problem.

But then the other part. What if I want to find train's departure and
arrival times in particular section defined by sectionRefId. I can split
sectionRefId attribute and find those two ocpTT having splitted parts of
sectionRefId as their ocpRefId if and only if sectionRefId contains only one
"-" character.

Let's assume that I have two stations: "A-1" and "A-2". Now I have two
ocpTTs with attributes ocpRefId "A-1" and "A-2". Again, this is not a
problem but then my sectionTT would have sectionRefId attribute "A-1-A-2".
Now there are three options how I can split that back: "A and 1-A-2", "A-1
and A-2" and "A-1-A and 2". How can I know which is the correct form? This
is the reason why I don't like this structure.

Maybe there should be sectionRefId that references to that section and also
startID and endID that references to ocpTT's ocpRefId attribute?

I hope you understood my point and and the reason why I'm worried about it.

-Teemu
Re: 1.1R2 - sectionTT element [message #688 is a reply to message #687] Mon, 05 January 2009 16:57 Go to previous messageGo to next message
Joachim Rubröder railML is currently offline  Joachim Rubröder railML
Messages: 0
Registered: November 2019
Hi,
sectionRefID is an ID and you should never parse an ID for included
information!

For your two stations "A-1" and "A-2", you have to take your two ocpTTs
with ocpRefId "A-1" and "A-2" for localisation. Now you have to search
your infrastructure for any connections between A-1 and A-2.

This connection from A-1 to A-2 (if there is any) might be part of a line
"70005" and a section "XY" in a line/section model used by a certain
program. So an export could add helpful references like sectionRefId="XY"
lineRefId="70005". But there is no rule how to define a sectionID. I just
invented one possible usage of a sectionID="85ZVIA-85ZWIP" for the
example.

So maybe its better not to use this attribute unless it has its
counterpart inside an infrastructure schema. Or you are transferring
timetable data between two programs with the same infrastructure model and
a sectionID, well known by both.

Kind regards,
Joachim
Re: 1.1R2 - sectionTT element [message #689 is a reply to message #688] Mon, 05 January 2009 17:20 Go to previous messageGo to next message
Teemu Sirkiä is currently offline  Teemu Sirkiä
Messages: 5
Registered: December 2008
Junior Member
"JoachimRubröder" <coord(at)timetablerailmlorg> kirjoitti viestissä
news:gjtal2$os3$1(at)sifaivifhgde...
> Hi,
> sectionRefID is an ID and you should never parse an ID for included
> information!
>
> For your two stations "A-1" and "A-2", you have to take your two ocpTTs
> with ocpRefId "A-1" and "A-2" for localisation. Now you have to search
> your infrastructure for any connections between A-1 and A-2.
>
> This connection from A-1 to A-2 (if there is any) might be part of a line
> "70005" and a section "XY" in a line/section model used by a certain
> program. So an export could add helpful references like sectionRefId="XY"
> lineRefId="70005". But there is no rule how to define a sectionID. I just
> invented one possible usage of a sectionID="85ZVIA-85ZWIP" for the
> example.
>
> So maybe its better not to use this attribute unless it has its
> counterpart inside an infrastructure schema. Or you are transferring
> timetable data between two programs with the same infrastructure model and
> a sectionID, well known by both.

OK. Let's suppose that I have a RailML file for example this
ICN_V1.1.R2.xml.

If I understood correctly there is no way to find out the distance between
ocpsTT "85ZUE" and "85ZVIA" or trackInfo because I don't know what is the
correct sectionIdRef between these ocpsTT?

In the version 1.1 this is very simple because section element is always
under entry element but now those elements are seperated and there is no
link between them because I'm not allowed to construct sectionRefId from
ocpRefIds.

-Teemu
Re: 1.1R2 - sectionTT element [message #690 is a reply to message #689] Wed, 07 January 2009 14:57 Go to previous messageGo to next message
Joachim Rubröder railML is currently offline  Joachim Rubröder railML
Messages: 0
Registered: November 2019
Oh, now I see the point.

The "sectionTT" in version 1.1.R2 ist still the same as the "section" in
1.1. The renaming with TT should just indicate, that it is not part of the
infrastructure. It is instead the usage of the infrastructure by a train
(this might be a bit sophisticated). In the timetable.xsd you can see,
that sectionTT is still a subelement of an "ocpTT" (= entry).

The problem is, that I made a wrong example by putting "sectionTT" on the
same level than "ocpTT". The file was valid because "sectionTT" was now
regarded an "any"-element and therefore still correct, so I missed this
error. I'm going to modify my example, will that solve your problem?

Kind regards,
Joachim
Re: 1.1R2 - sectionTT element [message #691 is a reply to message #690] Wed, 07 January 2009 17:44 Go to previous messageGo to next message
Teemu Sirkiä is currently offline  Teemu Sirkiä
Messages: 5
Registered: December 2008
Junior Member
"JoachimRubröder" <coord(at)timetablerailmlorg> kirjoitti viestissä
news:gk2cd7$p90$1(at)sifaivifhgde...
> Oh, now I see the point.

> The problem is, that I made a wrong example by putting "sectionTT" on the
> same level than "ocpTT". The file was valid because "sectionTT" was now
> regarded an "any"-element and therefore still correct, so I missed this
> error. I'm going to modify my example, will that solve your problem?

Yes, that will solve my problem totally. Now I can totally ignore
sectionRefIds because sectionTT is under ocpTT element.

-Teemu
Re: 1.1R2 - sectionTT element [message #692 is a reply to message #691] Fri, 09 January 2009 11:08 Go to previous messageGo to next message
Joachim Rubröder railML is currently offline  Joachim Rubröder railML
Messages: 0
Registered: November 2019
Hi,
I have changed the example in the timetable_V11-R2.zip, so please take a
fresh download.

Sorry for the misleading example and many thanks for detailed review!
Joachim
Re: 1.1R2 - sectionTT element [message #693 is a reply to message #692] Fri, 09 January 2009 13:51 Go to previous message
Teemu Sirkiä is currently offline  Teemu Sirkiä
Messages: 5
Registered: December 2008
Junior Member
"JoachimRubröder" <coord(at)timetablerailmlorg> kirjoitti viestissä
news:gk77mj$q2u$1(at)sifaivifhgde...
> Hi,
> I have changed the example in the timetable_V11-R2.zip, so please take a
> fresh download.
>
> Sorry for the misleading example and many thanks for detailed review!
> Joachim

Yes, now the structure looks good again.

Thanks!

-Teemu
Previous Topic: new draft V1.1-R2
Next Topic: Passing times in timetable
Goto Forum:
  


Current Time: Fri Mar 29 14:40:01 CET 2024