Fahrplan thread

I can test, but I’m not a developer at all. Do I have to compile it? Or will it be available on chum (testing repository)?

Ah, no worries. I’m working on a build in chum which could just be downloaded. I’ll get back to you… thanks!

3 Likes

Ok, after a bunch of coaching from @dcaliste and @pvuorela I made a pre-release on obs for > 4 SFOS.

This branch supports adding a journey to the calendar.

https://api.merproject.org/package/binary/home:poetaster:fahrplan/fahrplan?arch=aarch64&filename=harbour-fahrplan2-2.0.38-1.25.1.jolla.aarch64.rpm&repository=sailfish_latest_aarch64

For the adventure prone among you :slight_smile: It uses a temporary file and the Calendars native import (via QDesktopService) and so can’t screw up your calendar.

I’d appreciate feedback while I work on backporting this approach to SFOS 3.x

Release 2.0.38 is out:

  • sailjail / 4.x configuration file migration
  • enables saving trip to calendar via simple QDesktop temp files
  • function to clean samei temp files
  • sailjail perms to use of UserDirs (save temp files).

On openrepos and in Chum. Tested on arm7, SFOS 3.4 and aarch64, SFOS 4.4

4 Likes

Release 2.0.40 is out on open and chum:

  • adds VRR (Rhein Rhur in Germany) parser (thanks @Thaodan !)
  • removes some defunct parsers which can’t be resurrected (SF bay, Dubai, sbbch).
  • fixes an erro in the EFA parser for Syndey, Australia.
7 Likes

Release 2.0.41 is out on openrepos and coming to chum:

  • adds Swiss, timetable.search.ch backend release
    thanks cockroach and @Thaodan for pointing it out!
3 Likes

The SBB backend is in place now :wink:

5 Likes

It seems ResRobot has broken Fahrplan in Sweden now.
Honestly i don’t understand the need for any of these changes since they don’t seem to do much good.
Unfortunately i’m not somewhere i’m able to contribute from currently, so i’ll just note some findings here in case others are looking at it too:

  • API version bump to 2.1 in url needed (someone needs to learn semantic versioning)
  • There should now be only one API key typ… current key seems to still work
  • key is now called accessId (whyyyy??)
  • id in replies should not be used, but extId, query parameter is still id, (seriously, if you need to change internal id then just add a intId and don’t break stuff)

see also: Migration guide | Trafiklab

Thanks for reporting! I’ll get to it after my vacation (if I survive the heat of southern France!)

2 Likes

And you have a PR:

1 Like

And merged!

I’ll do some testing the tomorrow and release.

Thanks!

2 Likes

Thanks for the link! As so often, Sweden and Norway (and San Francisco!) seem to be spearheading NeTEx. I wonder if we’ll finally get a global standard worthy of the name. Hmmm.

Ok, just testing on the merged code, I don’t seem to get any results. You were just using the previous keys? Maybe it works ‘once’?

Ok, it doesn’t seem to be api keys. I’m going to move this to github … I believe:

* The `Product` object has been renamed to `ProductAtStop` .
* `Product` is now the name of an array of products, containing one element. Use `ProductAtStop` for consistent results.

Are the issues.

Hello,
Iam regularily travelling from home to work and back. Fahrplan renembers my last connection and i can easily look for the next train when I travel from home to work.
When going back however, I have to do a new setup and change the departure and the destination. Many websites have a button (–> <–) to just switch departure and destination. I would love to see this feature in fahrplan.

1 Like

That’s already available from long-pressing either option. Not the most user friendly, but it works.

3 Likes

Ahh. Ok thanks a lot!

I’ve just released (currently only on openrepos) 2.0.42 bringing an api update for the Swedish ResRobot service. I’ve tested everything but GPS based lookups. Thanks @attah !

2 Likes

I can confirm GPS-based lookups are working well.

Thanks! That’s good news. Now if I could shake my summer vacation cold, all would be well. Sneeze.