itass: Italy Strade Statali

[panda80]

• The possibility to draft the tier 5 systems (sure, this will likely only happen in the remote future)
• The possibility to track travels region by region
• The possibility to compare region stats
• The possibility to see that you've clinched a region (or that you are close to....)

You have the stats for a region, but than it's more difficult to see the stats for a country. Maybe for hardcore users, that like to invest a lot of time in TM and don't care about having to write much more lines after a trip (like us), it's ok, but for normal Joe travellers it's much more difficult. Noticed it as I introduced the A9 in Germany in my daughter list, after we drove from Munich to Berlin. Before the splitting of Germany you would have written 1 line, now you have to write about 20 lines, having to invest about 20-30 minutes in introducing this one motorway to your list file. I think Chriszwolle mentioned this aspect as he send his feedback about TM. I think he didn't updated his list afterwards.

[Jim]

With the 6-field .list entries, A9 for emmaf would only need a single line:

DEU-BB A9 1 DEU-BY A9 76

[si404]

You can use the .list tool to make entries from this (got by simply clicking 'View Connected Route' on any chopped section). Doesn't take very long at all to add entries across multiple regions.

A lack of country stats are a issue though.

[Jim]

Country stats are certainly doable, just need to get it done.

[michih]

You can use the .list tool to make entries from this (got by simply clicking 'View Connected Route' on any chopped section). Doesn't take very long at all to add entries across multiple regions.

I understand how you mean it (see Jim's post) but I still have to say: NO, you can't! You don't get the entries (by regions) but one entry only. Not good for users who sort their list file by regions. However, I think it would be doable. Likely no big deal. E.g. just another check box in the list tool to switch from 6- to 4-field-entries in showroute.php for connected routes.

Country stats are certainly doable, just need to get it done.

yep.

[panda80]

With the 6-field .list entries, A9 for emmaf would only need a single line:

DEU-BB A9 1 DEU-BY A9 76

Very good improvement, removes much of my thoughts about splitting. I am ok with whatever you will decide.

[michih]

With the 6-field .list entries, A9 for emmaf would only need a single line:

DEU-BB A9 1 DEU-BY A9 76

Very good improvement, removes much of my thoughts about splitting. I am ok with whatever you will decide.

We should use https://tmstage.teresco.org/ for testing the split version

@si404 Is everything ready for a test on tmstage?

Maybe @yakra can help again with converting the user .list files?

@yakra: Would you volunteer and make some magic if we really decide for splitting?

@yakra?

[si404]

We should use https://tmstage.teresco.org/ for testing the split version

@si404 Is everything ready for a test on tmstage?

Yep, I believe it is.

[yakra]

Sorry I'm late to the party, guys.
So the Italy split needs to happen, and we're moving forward with it then? Looks like we're all prepared for it in any case...

I've uncommented the split systems and am loading them up on lab2 in order to take a look at eure2 in mapview.
I see that they are formatted systemName+'2' -- splitregion support was modeled after how FRA was done, and expects systemName+'r'. One of three things will need to happen:
1. Change these systems to systemName+'r' codes. <--It doesn't make sense to ask you to do this.
2. Change the splitregion routine to use '2' instead.
3. Change the splitregion routine to append a custom character string specified on the command line. <--This is the best option, though a wee bit more involved.

Edit: Changes made. I think. Haven't attempted compiling yet.
Edit 2: siteupdate reports a couple errors in the highway data, then crashes. I'm going to stash my changes and try option 2, just to be sure the problem wasn't introduced just now. I expect the results to be the same. This shouldn't be a showstopper though -- the crash occurs during .list rewrites, after we get the log files we need to make sure all the concurrencies check out.
Edit 3: No, wait! I'm pretty sure I didn't uncomment the split systems here on BiggaTomato. Lemme try that out.

[yakra]

HD @ 78b2123c537430631e48d034b8cc01f519bf4ab4
UD @ 5c1816a7e606ffe83b2bf4812d5a654168cf56b1

The log files are pretty big and unmanageable.

ITA-TOS E80 still contains data from:
• ITA-LAZ E80
• ITA-ABR E80Car
• ITA-LAZ E80Bor

ITA-LOM A7, ITA-PIE A7, & ITA-LIG A7 are all identical. Same as the original ITA file except for the visible border points.

On the plus side, eursf & eursf2 are clean.

Once these issues are fixed I can give it another go and see what's left over.

If there's any info left in the logs after the next pass, boil it down with:

Code: [Select]

for s in {eure,itaa,itass}{,2}; do echo $s; grep -h "has [^0] concurrent segments with same name+banner in different region(s) (1 expected)" $s-concurrencies.log | sed 's~ has [0-9]\+ concurrent segments.*~~' | tr ' ' ';' > $s.csv; doneand then the same thing with just 0

[yakra]

A little more info...

eure:
ITA E62 +X826 +TOS/LOM has 0 concurrent segments with same name+banner in different region(s) (1 expected)
ITA E62 SS33 +X821 has 0 concurrent segments with same name+banner in different region(s) (1 expected)

itass2:
ITA-LAZ SS1Mon SP140 LAZ/TOS has 0 concurrent segments with same name+banner in different region(s) (1 expected)
ITA-MAR SS77Dir SS361 SS78 has 0 concurrent segments with same name+banner in different region(s) (1 expected)

itass:
ITA SS77Dir SS361 SS78 has 0 concurrent segments with same name+banner in different region(s) (1 expected)

SS77Dir suggests a banner/abbrev mixup:
itass2.csv:itass2;ITA-MAR;SS77;Dir;;;itamar.ss077dir;
itass.csv:itass;ITA;SS77;;Dir;;ita.ss077dir;

ita.ss077dir;;;;ABBREV_AS_CON_BANNER;itass_con.csv#L189
ita.ss106tar;;;;ABBREV_AS_CON_BANNER;itass_con.csv#L228
ita.ss212ben;;;;ABBREV_AS_CON_BANNER;itass_con.csv#L337

There are 2 different underlying causes, so I've opened a pull request to clarify the manual.

[si404]

Looking at Mapview of the regions to see what leaks out of the regions so I can do a first round of fixes at some point

• ITA-ABR - fine
• ITA-BAS - fine
• ITA-CAL - fine
• ITA-CAM - SS749 and SS17SMC in wrong region, E842 needs splitting
• ITA-EMI - A21, E70, E31, A22, E45
• ITA-FRI - A28
• ITA-LAZ - SS5
• ITA-LIG - A6, A12, E31, SS62 not split
• ITA-LOM - A7, A21 not split, A8/A26 missing, A22 missing
• ITA-MAR - SS685 segments need swapping
• ITA-MOL - SS158
• ITA-PIE - A6, A7, A21, A8/A26, E62 all leak out
• ITA-PUG - E842, SS655 need splitting
• ITA-SAR - an island, all fine
• ITA-SIC - an island, all fine
• ITA-TOS - E80, A12, SS62, E31 need splitting. SS64 has two sections that should be in Emilia
• ITA-TRE - fine
• ITA-UMB - SS685 segments need swapping
• ITA-VAL - all fine
• ITA-VEN - SS52

[si404]

The log files are pretty big and unmanageable.
<snip?
Once these issues are fixed I can give it another go and see what's left over.

I've gone through and hopefully dealt with all the stuff leaking out of their rightful regions.

One issue I found that would give an awful lot of errors is that I didn't add the split points into itass. That won't take long as I have files to do that.

Now to get it all uploaded without Github crashing on me. 

Edit: https://github.com/TravelMapping/HighwayData/pull/5309

[michih]

New national road numbers in Lombardy Region (former provincial roads): https://www.skyscrapercity.com/threads/484944/post-176112156

[yakra]

Unrelated to the split -- looks like there's nothing wrong with this; it's just uniquely weird:
https://travelmapping.net/hb/showroute.php?r=ita.ss027var&lat=45.771953&lon=7.305458&zoom=17
This route looks like it's unbannered SS27 (Variney), as opposed to a Variante without an Abbrev.
That makes it SS27Var, which ends at... SS27Var. A Variante. This is a normal waypoint label, as it's standard to leave off Abbrevs from labels unless needed to disambiguate.
Unbannered SS27 picks up again on the other side. Is this similar to how A1 will become A1(M) for a while, or IRL M6 becomes N6?
(What's the difference between Var and not? Anything related to this SS 27 -> SR 27 signage change?)

South of there:
ITA SS27Var ends at T2, but there's no corresponding route in the HB though it's signed down south.
E27 is missing a point here or here or wherever you decide the correct coords are.