qwaletee wrote:I would like to propose that any template that is intended for use beyond a USA audience should go in Template space.

qwaletee wrote:Hmm, I guess we have different definitions of clever.

Even getting the first one right will be fairly obtuse, and not a priority for me. If you would like to take a crack at it, go ahead and expand on what I did. Use a test page please (user, template, or other space). Good luck.

qwaletee wrote:For documentation, I provided the rough information. It could definitely use some work. I'm not too worried about the core template (Locking Standard) because I think its doc is OK AND I think in most cases this template would not be used by end users. The Lock Standard State "shell" template would be used instead, because less-technical editors can simply hand off the hard work to someone who knows how to deal with templates. The Lock Standard State documentation needs a lot of work. All the parameters to the core template are fully and clearly documented, though they are pain to use if you have many deviations from the default lock levels. Using the State shell template is comparatively simple: If your area is already defined, just pass the area name as a parameter, e.g., {{Lock Standard State|NORTHEAST}}. If your area is not already defined, then someone who understands the core template has to develop the correct call to the core template, and stuff it in to the shell template with the area name. Ask your wife - stuffing shels takes effort, consuming them is easy and pleasant.

Colspans - complex topic, see below.

As to Ferry and RR, I thought we set a national standard on that. If not, it isn't that hard to swap the fixed values there now for a variable option, with the same three columns as the other types.

Back to colspan, in all its hoary:

This is tougher than you might think, Pesach. It doesn't require any cleverness, just a lot of extra code. Wiki template logic is quite awkward. You don't have any variables (other than the passed parameters), so you have to repeat your full logic statement every time you use it. In this case, you have to play two sets of logic - does the column exist at all, and if so, does it repeat (and how)

For example, this is the existing freeway line definition:

Code: Select all

| {{Freeway}}
| {{{Fwy|4}}}{{#if:{{{title2|}}}|{{!}}{{!}}{{{Fwy2|{{{Fwy|4}}}}}}}}{{#if:{{{title3|}}}|{{!}}{{!}}{{{Fwy3|{{{Fwy|3}}}}}}}}

Let's say to simplify it, we would reduce functionality to 2 columns before implementing automatic column spanning:

Code: Select all

| {{Freeway}}
| {{{Fwy|4}}}{{#if:{{{title2|}}}|{{!}}{{!}}{{{Fwy2|{{{Fwy|4}}}}}}}}

This is how it MIGHT work then (untested, probably not complete and buggy, and does not support 3 columns without lots more code):

Code: Select all

| {{Freeway}}
| {{#if:{{{title2|}}}|{{#ifeq:{{{Fwy|4}}}|{{{Fwy2|{{{Fwy|4}}}}}}|colspan=2}}{{{Fwy|4}}}{{#if:{{{title2|}}}|{{#ifeq:{{{Fwy|4}}}|{{{Fwy2|{{{Fwy|4}}}}}}||{{!}}{{!}}{{{Fwy2|{{{Fwy|4}}}}}}}}}}

That unreadable glob has to be tested, fixed, repeated across all 6 driveable road types, modified for the defaults of each type (which is repeated about five times each), and for the specific variables for each of those types (repeated about 10 times each). That's about 300% of the prior code to do so.

Now, if we wanted to add the third column back, we would have to be repeated that type of code extension, except that logic is even more tortured due to the fact that all three could be the same, or the first and third could be the same, or the second and third. I'd estimate about 25x the current code to complete that.

Right now, I'm not ambitious enough to fix that for the sake of an appearance improvement that is a nice-to-have, not a "need," and could be added in later without breaking anything.

You might be thinking, OK, so go for two columns with a better appearance first, third column later. I don't want to do that, because I think the common use case will be urban/suburban/rural, which requires three columns. In NY, you already have NYC and elsewhere for two columns, and i suspect at some point, we're going to be gracious enough to allow northern and western NY to have New England's even lower level locks, because of the paucity of senior editors and the immaturity of the map.

If we had Lua added to the Wiki, more normal programming could be done, and I could spin this out in a jiffy. Readable code, reuse of code for each row.

|{{#if:{{{title2|}}}|{{#if:{{{title3|}}}|{{#ifeq:{{{Fwy|4}}}|{{{Fwy2|4}}}|{{#ifeq:{{{Fwy2|4}}}|{{{Fwy3|4}}}|colspan=3{{!}}{{{Fwy|4}}}|colspan=2{{!}}{{{Fwy|4}}}{{!}}{{!}}{{{Fwy3|4}}}}}|{{#ifeq:{{{Fwy2|4}}}|{{{Fwy3|4}}}|{{{Fwy|4}}}{{!}}{{!}}colspan=2{{!}}{{{Fwy2|4}}}|{{{Fwy|4}}}{{!}}{{!}}{{{Fwy2|4}}}{{!}}{{!}}{{{Fwy3|4}}}}}}}|{{#ifeq:{{{Fwy|4}}}|{{{Fwy2|4}}}|colspan=2{{!}}{{{Fwy|4}}}|{{{Fwy|4}}}{{!}}{{!}}{{{Fwy2|4}}}}}}}|{{{Fwy|4}}}}}