CIF file or JSON Schedule?

[Phil Wieland]

I've been experimenting with creating a timetable database using the json schedule files available as part of the Network Rail Data Feeds, and so far it seems to be working well.

However, I notice some information is apparently missing, for example the activity codes at each location.  Or have I just looked in the wrong place?

Would I be better off using "proper" CIF files?  I'm not sure how to get these, or even if they're made available to hobby users?

I must say I was pleasantly surprised to see the files use good old 80 column punched card format!

[Peter Hicks]

Hi Phil

On 13/06/13 17:00, Phil Wieland wrote:
> I've been experimenting with creating a timetable database using the
> json schedule files available as part of the Network Rail Data Feeds,
> and so far it seems to be working well.
>
> However, I notice some information is apparently missing, for example
> the activity codes at each location. Or have I just looked in the
> wrong place?

Passing locations won't have an activity code - but you should see (at
least) TB on the first location in a schedule, T on normal station calls
and TF for the last location in a schedule.

Do you see those?

> Would I be better off using "proper" CIF files? I'm not sure how to
> get these, or even if they're made available to hobby users?

I'm working on getting a case put together for distributing 'raw' CIF
files alongside the JSON data. You *might* be able to get CIF files
from NR directly, but that's likely to ruffle some feathers.

Have you looked at the weekly 'full' extracts from ATOC on
http://data.atoc.org/ ?


Peter

[Phil Wieland]

On Thursday, June 13, 2013 5:28:19 PM UTC+1, Peter Hicks wrote:

Hi Phil

On 13/06/13 17:00, Phil Wieland wrote:
> I've been experimenting with creating a timetable database using the
> json schedule files available as part of the Network Rail Data Feeds,
> and so far it seems to be working well.
>
> However, I notice some information is apparently missing, for example
> the activity codes at each location.  Or have I just looked in the
> wrong place?
Passing locations won't have an activity code - but you should see (at
least) TB on the first location in a schedule, T on normal station calls
and TF for the last location in a schedule.

Do you see those?

There doesn't appear to be a JSON field called activity or anything like that in the location records.
 

> Would I be better off using "proper" CIF files?  I'm not sure how to
> get these, or even if they're made available to hobby users?
I'm working on getting a case put together for distributing 'raw' CIF
files alongside the JSON data.  You *might* be able to get CIF files
from NR directly, but that's likely to ruffle some feathers.

Have you looked at the weekly 'full' extracts from ATOC on
http://data.atoc.org/ ?

Aha.  That might be a better answer, I'll look into it if I find anything more important missing from the JSON files..  It's rather confusing having the same data available under different contracts in different formats.

Thanks for that.

[Jonathon Hurley]

Just to confirm, there isn't an activity field in the SCHEDULE JSON data feed.  It should appear within the locations data contained within the schedule_location element (according to the CIF user guide) but doesn't in the JSON version.

Regards,

Jonathon