PTNA: news for Public Transport Network Analysis

[ToniE (Toni Erdmann)]

@ToniE

I also change the layout of the Web-pages

@Nielkrokodil @mcliquid

Did it:

If the GTFS:stop_name differs from the ‘normalized’ name, I’ll set

• background colour=lightgreen
• title=original name, so that you can see the GTFS:stop_name if you move the mouse over it.

Example:

• Proposal
• Stops

Still not perfect expansion/normalization, but that can be coded quite easily and will be added with the following GTFS-Import.

[mcliquid]

@ToniE localhost links

[ToniE (Toni Erdmann)]

Well, as the code wasn’t yet active on ptna.openstreetmap.de …

But now:

• Proposal
• Stops

[ToniE (Toni Erdmann)]

I just added analysis of public-transport for

• FR-CVL-TAO
  • Orléans Métropole, France
  • for Transport de l’Agglomération Orléanaise
  • gtfs data is also available FR-CVL-TAO

to ptna

[ToniE (Toni Erdmann)]

I just added analysis of public-transport for

• DE-MV-KGVP

  • Landkreis Vorpommern-Greifswald
  • for Kooperationsgemeinschaft „Vorpommern“

• DE-MV-VLP

  • Landkreises Ludwigslust-Parchim, Amt Neuhaus im Landkreis Lüneburg
  • for Verkehrsgesellschaft Ludwigslust-Parchim

• DE-ST-marego

  • Region Magdeburg
  • for Magdeburger Regionalverkehrsverbund

to ptna

[Nielkrokodil]

I just saw two “Abzw” which were not changed / green.
PTNA - GTFS Analysen stop Nr. 25 “Steinberg Abzw Rohrbach”. Is this only updating after some time?

[ToniE (Toni Erdmann)]

Oops. “Abzw.” is considered, but "Abzw " not. Fixed that.

[ToniE (Toni Erdmann)]

For de_*, “Abzw.” and "Abzw " are expanded to “Abzweig”
For de_AT, “Abzw.” and "Abzw " are expanded to “Abzweigung”

Hope that fits

[Nielkrokodil]

I live in eastern Austria and I only know the word “Abzweigung”. So I guess that is ok.

I am afraid something went wrong again. See here PTNA - GTFS Analysen all stops with “Abzw” are now green but empty.

[ToniE (Toni Erdmann)]

Fixed. I called the “normalization” manually and that deleted (set to “”) all entries (normalized names).
Thanks.

[ToniE (Toni Erdmann)]

Yesterday, I added analysis of public-transport for

• DE-NI-NHP
  • Landkreis Hameln-Pyrmont, Germany
  • for Nahverkehr Hameln-Pyrmont

to ptna

[Nielkrokodil]

@ToniE

@Nielkrokodil

2. Eine Filterfunktion zusätzlich zum Sortieren wäre hilfreich. Manche Fehler kann ich von Remote leicht lösen (z.B. falsch sortierte Routenmitglieder). Manche sind nur mit Ortskenntnis zu lösen (z.B. Zufahrtsbeschränkungen auf Wegen). Diese remote nicht (oder nur schwierig) behebbare Fehler würde ich gerne ausblenden können.

Gute Idee, aber recht schwierig/aufwändig zu lösen. Die Anzahl der verschiedenen Meldungen ist groß. Ich müsste sie kategorisieren und aus-/einblendbar machen (via Javascript/CSS?) wie beim OSM-Inspector / keepright.at auch.

I am taking up this topic again. Would it be possible to implement a filter without categories? As a free field, and as soon as you type it is filtered live? Like the search function for tags on community.openstreetmap.org, for example. Filtering would be helpful for all columns.

[ToniE (Toni Erdmann)]

That might be a “pure text based filter” only, as it must work for different languages. I’ll think and google about this. Let’s see.

[ToniE (Toni Erdmann)]

PTNA struggles with the up-to-date-ness of Overpass-Api data of some servers.

This happens for the UTC+01 timezone only. For UTC+01, 6 jobs are running in parallel using other servers than the “official” Overpass server. Their data seems not to be up-to-date but rather more then 6 hours old at the time of the query. PTNA refuses this data and starts a new attempt with the official server - one job after the other. This all takes some time. The UTC+01 jobs are therefore not done between 2 AM and 4 AM but still ongoing.

See the progress of catching up with the jobs at PTNA - Statistics

[ToniE (Toni Erdmann)]

I’m going to change things in ptna : remove non-ASCII characters from file names, …

This affects

• AT-OÖVV which will be renamed to AT-OOeVV
• FR-ARA-Drôme which has already been renamed to FR-ARA-Drome

File names, GTFS-feed names, … are checked to not include “strange” characters like ; & / > < …
This is done with a positive rather than negative list, allowing only a specific set of charaters [0-9a-zA-Z_.-]

I’ll take care for the necessary changes in CSV files, gtfs:feed= names, …

[ToniE (Toni Erdmann)]

I’ve found a simple JavaScript example. Two things to consider:

• performance filtering over lots of tables with lots of rows while typing

• placing the input field not on top of the HTML file but rather in the top-right corner of the screen

  • always visible even when scrolling through the HTML
  • similar to the stuff in the right-bottom corner in the DIFF files

[ToniE (Toni Erdmann)]

@ToniE

I’ll take care for the necessary changes in CSV files, gtfs:feed= names, …

All changes done.

[ToniE (Toni Erdmann)]

@ToniE

PTNA struggles with the up-to-date-ness of Overpass-Api data of some servers.

Problems are still there.

Interim solution: start analysis for UTC+01 at 00:12 AM (UTC+01), not at 02:00 AM with sequential analysis (1 job after the other, instead of 6 jobs in parallel) with standard Opverpass-API server.

[ToniE (Toni Erdmann)]

I just added analysis of public-transport for

• FR-BRE-Star
  • Rennes Métropole, France
  • for STAR

to ptna on GitHub pull request #98

[ToniE (Toni Erdmann)]

I just added analysis of public-transport for

• Estonia
  • Estonia
  • for all transport associations/agencies

to ptna. gtfs will follow later on.

Discussion:

This includes the analysis of the complete State of Estonia in one HTML.
The corresponding GTFS data includes a list of 36 agencies (‘network’ in OSM terminology).

We can discuss, whether a single analysis file is OK and we structure that internally for the ‘networks’.

Alternatively, I can configure PTNA to create 36 separate analysis files. That’s no issue, mostly c&p.

Edit: 2023-12-16 17:36 PTNA found 18 different values for ‘network’ plus 243 routes without ‘network’ being set.