Template talk:Jctint/Archive 1

SK Hwy 35 or Sk Hwy 2
Can there be added another type designation or another colour coding for segment namings? There are a few highways in Saskatchewan where only a segment is designated by an additional name. In the route box, it would be nice if it could be all joined in colour like a concurrency almost. SK Hwy 35 has one additional segment naming using this template (modified) and Sk Hwy 2 has two additional segment namings using an old table format, and hasn't been changed over to this template as of yet. Kind Regards SriMesh | talk  04:01, 13 April 2008 (UTC)

Support 'lspan' when location=none?
I've run into a corner case where I want to specify a null location for a brief concurrency, so I tried setting location=none, and then lspan=2 to push the rowspan down to cover both rows. I then discovered the template doesn't support that.

An easy fix is this change:

-{{#ifeq:{{{location}}}|none| | +{{#ifeq:{{{location}}}|none| |

Thanks! Kumba42 (talk) 01:59, 24 August 2009 (UTC)
 * The template doesn't support this feature since manually made exit lists typically don't have blank cells that span multiple rows. See Interstate 70 in Colorado. –  T M F 02:15, 24 August 2009 (UTC)
 * True, and in the few articles I've put together, if the intersections are far apart, but have no definitive location, I follow and use one blank cell per row as well. The one case that brought this up is on MD 225 where the small MD 224 overlap happens, and due to that, changes names.  I want to include the name change and list the CDPs on that road, but feel that the {{convert|0.4|mi|km|adj=on}} distance of the overlap/concurrency is large enough to warrant two separate rows.  Compounding this is that while official MDSHA maps suggest the area is called "Mason Springs", GNIS states that the area is a "Locale", which is or did have human activity, but is distinctly different from a populated place.   So from an aesthetic viewpoint, having the row span two rows looked the best (as-if there was a valid CDP there), just that the current template code doesn't offer that feature (as of writing this, I'm using a wikitable, which I can manipulate, although that will be changing shortly).  Hence, thought I'd bring it up. Kumba42 (talk) 05:14, 24 August 2009 (UTC)


 * I just discovered the 'location_special' parameter. Handy.  Toss in &amp;nbsp;, and then lspan works.  The cleaner idea is above, however, it's understandable if due to the relative rarity of these cases if it's denied. Kumba42 (talk) 05:32, 24 August 2009 (UTC)

Exit numbers
Since RJL has been established and we have a grand unified junction list standard, shouldn't jctint support exit lists? —Scott5114↗ [EXACT CHANGE ONLY] 13:17, 24 August 2010 (UTC)
 * See Jctexit and Exittop for that. –Fredddie™ 16:06, 24 August 2010 (UTC)
 * Bah. Redundant. :\ jctexit should probably be merged with jctint so that we don't have all this duplicate code running about. Just make it optional, if an exit number is included then include the column, otherwise leave it out... —Scott5114↗ [EXACT CHANGE ONLY] 08:47, 25 August 2010 (UTC)
 * And what of mixed-type highways that have a freeway segment in the middle? How would you propose that the template know to insert the blank cell for those rows where there isn't an exit number specified? would we need to add  or for at-grade sections   to every line of an exit list? Remember, jctint would have to know each time it's transcluded to make a row if it needs to include the cell for exit numbers. This method simplifies it at the moment. Use jctint and no exit number column is created. Use jctexit and the column appears. The cells are blank if the exit number isn't specified.  Imzadi   1979   →   13:15, 25 August 2010 (UTC)
 * Yes, it's probably simpler to use that way, but it's more difficult to maintain, especially since we have a great number of optional parameters for supporting the various types of county/parish/home rule municipality/independent city crap that tends to gradually increase in size due to more odd cases being discovered over time. When this needs to be changed, we wouldn't need to contend with integrating this logic into two different templates (or worse, risk forgetting to update one of them, leading the less-used one to produce surprising results). I'm sure an "exit=none" parameter could work, same as it does for unincorporated locations now. —Scott5114↗ [EXACT CHANGE ONLY] 22:24, 27 August 2010 (UTC)
 * It would certainly be feasible to make Jctint call Exitint with "exit=none", effectively turning Jctint into a shorthand template. If done correctly, there would be zero learning curve to the changes. –Fredddie™ 22:52, 27 August 2010 (UTC)
 * Check out Jctint/sandbox and Jctexit/sandbox for a demonstration. –Fredddie™ 01:44, 28 August 2010 (UTC)

The edit request below will unify the two templates. If an article calls  it will function as an exit list and add the exit column using. If the article doesn't add that first unnamed parameter with a value of "exit" it functions as it always has. This is how jctbridge and jctco already work, and how I updated jcttop to work earlier.  Imzadi 1979  →   23:09, 27 July 2011 (UTC)

Edit request from Imzadi1979, 27 July 2011
Please copy the code from Template:jctint/sandbox here. This will merge jctint and jctexit into a single template. In the end if an article calls , it will activate the exit-number column needed for an exit list, as calling   tells that template to work with the extra column displayed in an exit list versus a junction list. Once this merger is done, we can redirect jctexit to this template.to complete the merger.

 Imzadi 1979  →   22:29, 27 July 2011 (UTC)


 * Done. Mitch 32(Can someone turn on the damn air conditioning?) 23:16, 27 July 2011 (UTC)

Edit request from Imzadi1979, 12 August 2011
Please synchronize jctint/sandbox to this template. This will activate the coding needed for exit lists that have old and new numbers displayed.

 Imzadi 1979  →   21:42, 12 August 2011 (UTC)


 * Done. Mitch 32(God Bless America, Let Freedom Ring) 23:26, 12 August 2011 (UTC)

Edit request from Imzadi1979, 22 September 2011
Please sync over Template:Jctint/sandbox. This will restore the previous color handling as well as implement a new way to handle mile-/km-posts that are ranges instead of single points.

 Imzadi 1979  →   22:24, 22 September 2011 (UTC)
 * ✅. Also removed the unnecessary includeonlys and showed some output of the template on its page. Hope this is okay. &mdash; Martin (MSGJ · talk) 17:30, 23 September 2011 (UTC)

Ctdab for non-county states
I noticed that in Louisiana, if you have a disambiguated city, Foo doesn't work correctly. It inserts Foo County instead of Foo Parish like it should. Look at the link below for an example. –Fredddie™ 16:30, 1 October 2011 (UTC)


 * }

Edit request from Floydian, October 19, 2011
This will create a new parameter called jspan (junction span), which allows a custom rowspan that affects the Old, Exit, km/Mile, Destinations and Notes simultaneously. It also adds espan (exit span), dspan (distance span), namespan (interchange name span), rspan (road span) and nspan (note span) for changing those spans on an individual basis. These individual parameters will supercede the jspan parameter if entered. This code is the entire template, so delete whats there now and replace it with:
 * Code moved to Template:Jctint/sandbox to reduce clutter on the talk page. &mdash; Martin (MSGJ · talk) 

Note that I'll also have to code a filler cells template for the following row in cases where these new parameters are used. This would be fairly simple to make, but I'd like to see a few of the ways that this would be used before coding that template. -  ʄɭoʏɗiaɲ  τ ¢  05:59, 19 October 2011 (UTC)
 * What is the purpose for all of this?  Imzadi 1979  →   23:24, 19 October 2011 (UTC)
 * Making it possible to have multiline entries, such as a junction on a county/location boundary line, or interchanges with multiple exit numbers but only one km value, etc. I can't make many junction lists without those two things myself. I figured that it made sense to add the feature to each column now rather than when someone encounters a snag and can't do something. -  ʄɭoʏɗiaɲ  τ ¢  00:22, 20 October 2011 (UTC)
 * Multiline entires are kinda forbidden in MOS:RJL, which is why the templates haven't supported them. If it's all the same exit number, it's supposed to be the same line in the table, and if it's the same milepost, same thing. As for junctions that fall on or along a boundary, we've used the state_special, county_special and location_special parameters to allow a custom "X–Y" entry for the location, county or even state (in the case of Route 54, which explicitly makes the the boundary the location. See also things like M-120 (Michigan highway) which has a county tri-point, township quad-point location and runs along a series of county lines. Also M-102 (Michigan highway) (8 Mile Road) runs only on county lines except one small piece and also uses the county lines and city/township lines as locations. These changes aren't needed, so for now I have to oppose the changes.  Imzadi 1979  →   00:54, 20 October 2011 (UTC)
 * Kinda not really... but Ontario Highway 9 is going to need exemptions in order to be rebuilt with templates, and if you believe only your way is the right way and that the templates have to be built to do things that way, then I'll just stick to not using cumbersome templates that can't accomplish what they are designed for (and that is to make things easier for the average editor, not to stonewall a standard into articles that didn't previously exist nor to create some whacky setup where you have to hand code all sorts of exceptions). Split table cells are more appropriate for when crossing a border (since the line dividing the cells for location A and location B rests at the same height as the middle of the cell for the junction), dashes are more appropriate when a road acts AS the border. As for the bug affecting Safari with not displaying split level tables properly, I'm not changing a better display method because of an improperly programmed browser that affects 4% of the internet (may as well worry about internet explorer 5.5 while we're at that) by ignoring the proper w3c standards. -  ʄɭoʏɗiaɲ  τ ¢  16:29, 20 October 2011 (UTC)

I have disabled this request for now. Please continue to discuss. &mdash; Martin (MSGJ · talk) 15:56, 20 October 2011 (UTC)

Consolidated City-County
With the changeover to this template, and the automated formatting, the exit list at Interstate 70 in Colorado is technically incorrect. It lists Denver as an independent city; actually Denver is a consolidated city-county. While there is no difference between these terms to the lay person, they are not interchangeable terms from a legal standpoint and legally there is a difference. I see two options to fix this, first change the independent_city parameter to not link to the term "independent city", so that it can be used in both instances, or create a separate but similar parameter for consolidated city-counties. This issue would also come up with any highway in San Francisco and probably a few others. Dave (talk) 06:55, 5 January 2012 (UTC)
 * Interesting I was just talking about this on IRC. If the article uses a Jctint/core wrapper, you can use indep_city_special to get around it.  Not sure if that's available with Jctint. –Fredddie™ 07:01, 5 January 2012 (UTC)
 * Didn't appear to work; thanks for the idea though. Dave (talk) 08:18, 5 January 2012 (UTC)
 * You tried it with Jctint/core, right? It doesn't work on the old Jctint. --Rschen7754 08:24, 5 January 2012 (UTC)
 * However, it sounds like we need to be more careful with this mass conversion of articles to this template. The exit list was correct when I last messed with it. =-)If I have time tonight I'll revert back to the manual template, as the Eisenhower Tunnel is another abnormality that appeared when the exit list was switched to this template. Dave (talk) 21:47, 5 January 2012 (UTC)
 * Part of the reason we're switching to templates though is so that we can make changes like adding a metric column. I'm in the process of switching CA to templates (though it's going to take a long time)... If you get a chance to come on IRC we could try and debug what's going on; it's hard to tell what the problem is from what you've written above. --Rschen7754 21:58, 5 January 2012 (UTC)
 * OK, I created COint, and coded indep_city such that Denver won't say City of Denver. Seems to be fixed. –Fredddie™ 01:09, 6 January 2012 (UTC)
 * Thanks for taking care of that Fredddie. Much better. However, I do wonder if we really need 2 templates, I don't think independent city is notable enough of a term to make it worth linking on the exit list anyways. However, this works, thanks again. One of these days I'm going to get back to actively participating on the wiki instead of just swatting flies, I promise =-) Dave (talk) 21:15, 6 January 2012 (UTC)
 * Think of them all as being analogous to the Citation family of templates. –Fredddie™ 23:57, 6 January 2012 (UTC)
 * Just curious, does the CO stand for consolidated or Colorado? If the later, we might want to think of a more elegant name, as this is going to come up again with San Francisco once California articles are transitioned over to this template. Dave (talk) 00:20, 7 January 2012 (UTC)
 * Stands for Colorado. CA is now set up the same way - see CAint. --Rschen7754 00:24, 7 January 2012 (UTC)

Edit request from Floydian, May 12, 2012
Can the /sandbox be copied over. This will transfer the remaining 2000+ transclusions onto the core templates for standardization. -  ʄɭoʏɗiaɲ  τ ¢  06:01, 12 May 2012 (UTC)

County notes not displaying
Anybody know why county notes aren't showing up on some Nevada articles? Nevada, like California, has mileposts resetting at county lines. Where I had worked on jct lists, I inserted county notes for these cases. These displayed fine prior to bringing the km conversion column online. --  LJ  ↗  22:56, 27 May 2012 (UTC)
 * U.S. Route 6 in Nevada - only the first county's note appears.
 * U.S. Route 95 in Nevada - none of the county notes appear.
 * ✅ Didn't realize NV did the same. --Rschen7754 23:03, 27 May 2012 (UTC)
 * Oops! Looks like we forgot to pass through county_note. Rschen was able to add it and it seems to be working now. –Fredddie™ 23:06, 27 May 2012 (UTC)
 * Thanks! --  LJ  ↗  23:09, 27 May 2012 (UTC)

line parameter?
Is there a reason why jctint does not support the line parameter in the mile cell but jctbridge does? I converted the junction list on U.S. Route 91 to use tables, but where the US 91 mileposts defer to mileposts of other numbered routes in Idaho, using  does not produce the horizontal rule as it does in jctbridge above it--this gives the look of milepost ranges (which is not correct). --  LJ  ↗  01:44, 10 June 2012 (UTC)
 * I saw kind of the same thing on US 218 a few months back. There were US 218's mileposts, but during the I-90 overlap, it had I-90's mileposts.  I thought it was confusing, so I did a little math and converted it all to US 218 mileage; we should probably do the same here.  It's not OR if all the mileposts are available and verifiable!  That being said, I don't think we should add the line parameter, but if others disagree, it's super simple to add. –Fredddie™ 21:30, 10 June 2012 (UTC)
 * That begs the debate, what is the purpose of this column, to reflect true mileage, or the mileage as posted? I guess the rest of wikipedia already made that decision for us, as if the answer was "mileage as posted" the conversion to kilometers wouldn't be useful on highways with milepost equations, discontinuities, etc. Anywho, I'm OK with Fredddie's idea. However, if implemented it would have implications on roads where the mileposts after a concurrency separation resume distance from the last milepost before the concurrency, as if the concurrency never happened. One such example that I can think of off the top of my head is Utah State Route 30. Dave (talk) 01:10, 11 June 2012 (UTC)
 * Does SR 30 officially follow the other routes, or is it officially segmented like you'd see in Arkansas?. I don't recall seeing any SR 30 shields on I-84, but it's been a couple years since I was out that way. –Fredddie™ 01:31, 11 June 2012 (UTC)
 * OK, I think I answered my own question. It seems the "gaps" on SR 30 are 0.001 miles long, but only one set of mileposts was ever drawn up. –Fredddie™ 01:37, 11 June 2012 (UTC)
 * To answer your question, SR-30 is an anomaly for Utah signage. For example along the independent sections of U.S. Route 89 (Utah) the mileposts reflect true mileage, including the distance that would be driven on the concurrent sections. However, although that's the only one I'm aware of in Utah; I'm sure many other such examples exist in other states.
 * I also haven't been out that way in years. However, I have never seen the I-84 concurrency signed at any point or at any time. However, the US-89 concurrency isn't as clear cut. It was mostly unsigned, but I recall seeing some scattered signposts with both shields present. Dave (talk) 02:01, 11 June 2012 (UTC)
 * The thing is, while both are useful, the non-Americans, like Tony1, wanted conversions, so it's become a case of "true mileage". If it were purely an "as posted", the column would disappear on all non-freeways in Michigan except US 2 because MDOT does not erect milemarkers except on freeways and part of US 2. (That's assuming that US 2 still has its mileposts on the western segment and the westernmost part of the eastern section; I haven't been that way in a while.)  Imzadi 1979  →   01:36, 11 June 2012 (UTC)
 * If nobody objects I think this should be handled on a case-by-case basis. To make everything "true mileage" would be a a lot of work for little value in states where the mileposts reset at county lines, but the countywide mileposts are published in reliable sources.Dave (talk) 02:01, 11 June 2012 (UTC)
 * Michigan's control sections reset at county lines, yet I've done true mileage by state with an exception or two. If the countywide mileposts are published in RSs with the MPs at the lines, it's simple addition to convert to true mileage. California has been the oddball because there is no way to convert the postmiles to mileposts, although CalNexus has provided true mileage for the state (see Interstate 8).  Imzadi 1979  →   02:10, 11 June 2012 (UTC)
 * OK let me rephrase that. =-) It's simple math for states where detailed milepost logs are published that either reflect true mileage or have the milepost equations and other discontinuities stated. However, for states where the published logs aren't as detailed, computing true mileage would be a lot of work for little value. Dave (talk) 02:20, 11 June 2012 (UTC)
 * OK, we're down to California then. Nevada's milepost ranges are known for each county according to what you and LJ have said in the past, and other states either have route logs with cumulative mileages listed, the ranges by county known or no source for mileposts known yet.  Imzadi 1979   →   02:27, 11 June 2012 (UTC)
 * In Nevada's specific case, NDOT does publish the start and end mileages by county, making computing true mileage in theory possible, but the situation is far from ideal. This mileage log is sporadic in the junctions that are listed, and does not state if any milepost equations are used or not. With that said, I'm fairly certain Nevada does not leave milepost equations in place for long, as I've seen crews updating the mileposts along the unaffected portions of the US-395 corridor in Carson City as sections of the freeway alignment opened to traffic. But I don't know for certain that milepost are updated statewide as alignments change. I'm ok with switching to true mileage, but am not volunteering to rush and update all the articles where I've built the major junctions table with "as posted" milage, and that's most of the articles I've done. Sigh, where are the exit list obsessed roadgeeks when you need them. =-) Dave (talk) 02:47, 11 June 2012 (UTC)
 * Don't be surprised then if someone requests such a revision at GAN or ACR as part of their review comments.  Imzadi 1979  →   02:49, 11 June 2012 (UTC)
 * Given all that, what would you suggest on the most complicated junction list I've built, U.S. Route 50 in Nevada? In that case we have a temporary routing, one county where the mileposts reflect the distance using an alignment that doesn't yet exist, and a couple of convoluted concurrences. In that case, I could switch to "true mileage" but would have to have gaps in the table as I would have to use the future alignment (at least to match NDOT's milepost log) and hope that plans don't change. (which by the way is possible, construction on the final phase was postponed yet again, due to a combination of lack of money and a feud between NDOT and the Washoe tribe over the proposed relocation of a tribal cemetery to make room for an interchange.) Dave (talk) 03:04, 11 June 2012 (UTC)
 * I think a better question is why NDOT did not give mileposts for the highway as it existed at the time. –Fredddie™ 15:07, 11 June 2012 (UTC)
 * NDOT's highway logs have stated US 50/395 to follow the unconstructed freeway bypass alignment in Carson City since at least 2001; the actual signed US highways followed what was in the books as unsigned SR 529 & SR 530. NDOT does not have a posted milepost log, having only posted "official" mileposts on maps which are not comprehensive (and recent versions don't have the mileposts anymore either...).
 * Going back to the original point though. I kinda think the line is useful for these convoluted situations, or maybe some other method as yet uninvented... --  LJ  ↗  07:26, 14 June 2012 (UTC)

I concur, the horizontal line should be implemented, unless and until something better is implemented. For all the huffing and puffing about using "true mileage" instead of "mileage as posted", there are certain situations where it is just not practical, and we shouldn't be so rigid as to not accommodate that reality.Dave (talk) 17:50, 14 June 2012 (UTC)
 * I don't have the bit (hint hint); all that needs to be added to the template is line. –Fredddie™ 18:26, 14 June 2012 (UTC)
 * ✅ --Rschen7754 18:43, 14 June 2012 (UTC)

GA DabLink
Using state=GA links to Georgia instead of Georgia (U.S. state). Rwalker (talk) 14:17, 18 June 2012 (UTC)
 * The problem lies with, which this template uses to get Georgia from state=GA. Unfortunately, if StateName is changed, it's going to create a whole host of redlinks as suddenly we'll be linking to Valdosta, Georgia (U.S. state) and not Valdosta, Georgia, etc. –Fredddie™ 22:56, 18 June 2012 (UTC)
 * Now that I've done a little digging, any instances of should be replaced with . –Fredddie™ 00:06, 19 June 2012 (UTC)

NVint
Parameter indep_city_special doesn't seem to pass through to NVint, one has to use jctint instead. --  LJ  ↗  11:36, 3 November 2012 (UTC)

WYint and Wyoming Highway 225
In Wyoming Highway 225, the mileage and the notes both span 2 rows at one point, but WYint doesn't have mspan and nspan, and when I tried adding them it forced the second destination into a sixth column for some reason. Can someone who knows template syntax better than I do fix this? TheCatalyst31 Reaction•Creation 01:42, 24 November 2012 (UTC)
 * Maybe try it again? I can't tell what the problem is without having the example. --Rschen7754 01:47, 24 November 2012 (UTC)
 * Nope, still having the same problem. I put the table with the problem in my sandbox if you want to take a look at it. TheCatalyst31 Reaction•Creation 01:59, 24 November 2012 (UTC)
 * I've fixed it in your sandbox. --Rschen7754 02:02, 24 November 2012 (UTC)
 * Thanks, didn't know you needed to do that. TheCatalyst31 Reaction•Creation 02:05, 24 November 2012 (UTC)

Assumption with units
The template makes an assumption that the source units for distances are the the primary unit when displayed. That is limiting and I would prefer the following behaviour: This is a particular nuisance for the UK, as miles are the primary unit, but government works in km. Without a fix it will significantly slow down roll out of this template to British roads. Above is best solution IMO, and point 3 will ensure no breakage.
 * 1) Primary unit definition is defined by sub-templates, or direct calls.
 * 2) If its defined, allow source figures to be either km or mile.
 * 3) If primary unit not defined, look for mile or km reference as at present.

However, this fix would have global applicability. For instance, on routes that cross the US-Canada or US-Mexico borders, both km and mile source figures could be used.--Nilfanion (talk) 22:58, 2 July 2013 (UTC)
 * Would it be simpler to implement the same behavior that flip does for ?
 * I ask because #1 already exists and #2 and #3 are moot because of #1. Check out  and look for the unitdef (unit definition) parameter.  Every Jctint wrapper has that parameter. –Fredddie™ 23:27, 2 July 2013 (UTC)
 * I should note that I am specifically referring to the other XXint templates, not Jctint. This template predates the core/wrapper system the XXint templates use now, and was subsequently changed so nothing would break during the switch to the core/wrapper system. –Fredddie™ 23:30, 2 July 2013 (UTC)
 * Maybe, I am not familiar enough with syntax here - its a bit of a maze - and better doc will assist. Thing is I can see benefit from being able to use km and mile sources in the same table. For instance, deriving some figures from location signs (km) and relying on Google to fill in the rest. Can the jctint, and also the other templates like jctbridge, give that flexibility? If so how exactly?--Nilfanion (talk) 23:50, 2 July 2013 (UTC)
 * Not currently. I have an idea, which isn't pretty, but it would work.  It seems miles should be the first unit in the column, we could then have a km parameter which would fill convert back to miles to fill in the first column.  Technically the template would convert it back to km, which is why I said it wasn't pretty.  I'll get a mockup working. –Fredddie™ 00:02, 3 July 2013 (UTC)


 * }

It works like I thought it would, but it's wrought with rounding errors. You'll notice the 200.00 comes out as 199.99. It's undoubtedly from the quick-and-dirty conversion scheme implemented in my example. 200, 200.0, and 200.000 convert as intended, though. –Fredddie™ 00:34, 3 July 2013 (UTC) Thanks. And yes probably, and if you are doing a double conversion that is prone to rounding error. The thing to avoid is saying 199.99 km, if the sign (or any other sort of source) says 200.00 km. But I am of opinion nearest 100m is excessive accuracy in any case, having a less precise figure than the source all but eliminates that risk.--Nilfanion (talk) 11:06, 3 July 2013 (UTC)

Location rewrite idea
I know there has been some effort to rewrite (parts of) this template in Lua, but I think we can simplify the core template before we get to that. My idea is to move the complexity of locations out of the core template as much as possible. When I originally designed the core, I added a complex switch to allow for locations that were primary topics, that is Toronto instead of Toronto, Ontario. It works, but I think it's putting an unnecessary load on the servers at a time when our articles are already being targeted for not loading quickly.

So, I'd like to propose stripping out any location coding presently in the template that is not required to create the table. Instead, location formatting would be left to the wrapping templates (like ) already in use. The wrappers were thought up as a way to allow for by-location customization, and this is no different. –Fredddie™ 01:01, 24 October 2013 (UTC)

lspan parameter not working as it should?
In this set of templates, the first  instance includes &amp;nbsp; and 2. My browser shows that the  isn't working so the whole second row is left-shifted one cell:

Here is the code that is produced and, sure enough,  is set to 1 not 2 as it should be.

If I change &amp;nbsp; to &amp;nbsp; then 2 does the right thing:

Here is the resulting html for that:

I presume that this is not how the template is supposed to work. If I use &amp;sp; then it works:

But, I should not have to do that. The template should do whatever is required to maintain columnar alignment and editors should not have to insert special characters to make it work right.

—Trappist the monk (talk) 01:31, 11 November 2013 (UTC)


 * You should be using none for each row. We don't span blank cells across the rows. Don't use a non-breaking space in the location because you'll link to " " (there's a redlink there if you move your cursor over the space between the quotes.)  Imzadi 1979  →   01:33, 11 November 2013 (UTC)


 * Perhaps the template doesn't currently span blank cells across rows but I think it should; or, at the very least, the documentation for lspan should reflect that it doesn't and explain why it doesn't.


 * Someone thought that the use of &amp;nbsp; is legitimate because, along with,  tests location (disguised as sub2) for both of those values.  Who thought that mapping a parameter with a contextually meaningful name to a contextually meaningless name was a good idea?


 * Using &amp;nbsp; does not create a wikilink to.


 * It is very common for templates to allow editors to omit or leave parameters empty. In  that is not the case.  location cannot be omitted or left empty because when that happens, all of the columns to the right are left-shifted.  To correct that, editors must set &amp;nbsp; or none. This is just wrong.  location, when omitted or empty, is semantically the same as none.  When location is omitted or empty,  shoud output   as it does for none and so maintain proper columnar alignment without user intervention.


 * —Trappist the monk (talk) 11:54, 11 November 2013 (UTC)
 * It was designed that way because for international support. Contextually meaningless names like sub2 give no preference to one region over another.  We in the US use County/Location, but that doesn't mean that works for any other region. –Fredddie™ 13:00, 11 November 2013 (UTC)


 * If someone from another language wiki is going to go to the trouble of translating this suite of templates into another language, it will be just as difficult for them as it is for me to mentally map contextually meaningful parameter names in to the contextually meaningless parameter names in .  This seems like obfuscation hoping for clarity.


 * —Trappist the monk (talk) 14:17, 11 November 2013 (UTC)


 * – This is incorrect, whenever location_special is used, location is empty/omitted. - Evad37 &#91;talk] 13:20, 11 November 2013 (UTC)


 * When I said that none is semantically the same as omitting or leaving location blank, I meant that. The end result should be the same: an empty cell in the Location column of the table. It should not be necessary for editors to explicitly state "this cell intentionally blank" (none) simply to preserve columnar alignment.  That is the job of the template.  The template should present a properly formatted table regardless of which parameters are present.


 * The documentation says that location_special overrides location, which it should do because location_special is unique. However, in this example, with Bryce Canyon and Special we get this:


 * Setting none also prevents the display of location_special. If special is really supposed to override whatever is in location, then the template isn't working as it should.


 * —Trappist the monk (talk) 14:17, 11 November 2013 (UTC)
 * This template will never be able to automatically produce a properly formatted table row in every case, because the right format depends on what parameters are used in jcttop, or other jcttop/core wrappers. If the location column is turned off, because the road is in a single location, then an empty first cell would break that table's formatting. In cases such as this, should an editor have to explicitly state "don't put in an empty cell"?
 * As for location_special, I agree that it should override location, but the logic is probably easier to write in lua, I imagine, rather than introducing more complicated parser functions. - Evad37 &#91;talk] 17:03, 11 November 2013 (UTC)


 * Yep, I've come to the realization that this is just a mess. I hope that whomever it is that undertakes a Lua conversion chooses a path that doesn't just replicate the several nested templates but creates a unified whole.


 * At the very least, the documentation of this version should reflect what it actually does. I'll tweak it to so that it complies with what I know.


 * —Trappist the monk (talk) 19:58, 11 November 2013 (UTC)

re: Template:Jctint/doc‎ edit summary: Parameters: don't use non-breaking spaces like that, creates other issues

What other issues? I think a more complete explanation of these other issues is in order. The template supports &amp;nbsp; just as it supports none. If &amp;nbsp; is inappropriate then needs to be changed to reject &amp;nbsp;. Until that change is made, if it is made, the documentation should either state that &amp;nbsp; is equivalent to none or it should explicitly state that while &amp;nbsp; is supported it is deprecated because . Leaving &amp;nbsp; out of the documentation when editors can see it in use in article space simply makes it look as though the template developers neglected to include &amp;nbsp; in the documentation, a not uncommon occurrence.

—Trappist the monk (talk) 22:21, 11 November 2013 (UTC)
 * See my previous comment. Using a non-breaking space when the state is defined, which it is for all of the state-specific templates like MIint that reference the documentation page, means you'll be linking to "&amp;nbsp;, Michihan" or similar. That's why you should not use that code, and that's why I removed that portion of your addition.  Imzadi 1979  →   23:30, 11 November 2013 (UTC)


 * I think you are mistaken. Using &amp;nbsp; does not create a link to &amp;nbsp;, Michigan.  Here is a side by side comparison of two  templates. Note in the rendered HTML that there is no link to &amp;nbsp;, Michigan.  If you float your mouse cursor over the empty Location cells in the rendered tables, the pointer does not change to the pointing finger.


 * I can see no differences between the two.


 * —Trappist the monk (talk) 01:31, 12 November 2013 (UTC)
 * Up until four days ago, Imzadi would have been correct. changed the   to a    – but only for the almost always disambiguated locations, and not for the primary topic locations - Evad37 &#91;talk] 02:14, 12 November 2013 (UTC)
 * I've added the switch for the primary topic location as well now. -- WOSlinker (talk) 07:30, 12 November 2013 (UTC)

re:.

I just don't understand. On the one hand, Editor Imzadi1979 appears to support changes to the template that reflect how editors are using the template in article space. But, on the other hand, Editor Imzadi1979 is opposed to documentation changes that reflect how editors are using the template in article space.

I understand that &amp;nbsp; is not the preferred mechanism to create blank location cells. But, it's out there, it's in use, and  is a common filler in html tables.

So here's some alternative wording:
 * location: name of the location without wiki markup—the location is automatically linked by the template. When an intersection is not in any location as is often the case in states outside of the northeast, either none (preferred) or (deprecated) will create a blank cell.
 * lspan: number of rows the location spans. If value=1, omit this parameter. Ignored when none or.

This text explicitly identifies both parameter values and shows which is preferred. Similarly, it documents exactly when lspan is ignored.

—Trappist the monk (talk) 19:21, 13 November 2013 (UTC)
 * Graying it out is fine, but it has been deprecated (and in fact until recently it cause problems so it shouldn't be given "approved" status).  Imzadi 1979  →   20:25, 13 November 2013 (UTC)

Broke it!
Woo, I think I broke something

Help? --  Admr Boltz  00:53, 2 January 2014 (UTC)
 * Imz resolved this off wiki ( |mile=none ) --  Admr Boltz  00:57, 2 January 2014 (UTC)

length_ref protected edit request
Please add |length_ref= to the long list of jctint/core parameters, per Template talk:Jctint/core. Thanks. --NE2 05:26, 5 January 2014 (UTC)
 * ✅ --Rschen7754 05:29, 5 January 2014 (UTC)

mspan on NVint
The |mspan parameter doesn't seem to work on NVint. Can this be fixed? --  LJ  ↗  00:39, 3 March 2014 (UTC)
 * ✅ should be fixed. –Fredddie™ 00:42, 3 March 2014 (UTC)
 * Thanks, Freddie! --  LJ  ↗  04:51, 3 March 2014 (UTC)

namespan parameter
Does anyone know about the namespan parameter in this template? If so, it does not work. The spanning part worked, but it moves the road to the notes column. -- Bloonstdfan360  Transit center! 04:40, 3 March 2014 (UTC)
 * If the spanning works, then the parameter works. Any subsquent line in the namespan, you can simply use  where XXint is the Jctint variant you're using. So if 3, you would use  twice after namespan and then continue with . –Fredddie™ 04:57, 3 March 2014 (UTC)
 * It works now. I followed the advice above. -- Bloonstdfan360  Transit center! 05:59, 3 March 2014 (UTC)

piped links
In a case such as Burton, Texas in this table, where there is more than one city in a state with the same name, is there a way to pipe the link so that just the city name is shown, rather than having to include the county name in the visible link? Nick Number (talk) 15:06, 14 March 2014 (UTC)
 * yes, this edit fixed that for you using Washinton.  Imzadi 1979  →   15:14, 14 March 2014 (UTC)
 * Ok, thanks, I missed that switch in the docs. Nick Number (talk) 15:36, 14 March 2014 (UTC)

type=trans
Would it be acceptable to use trans for freeway to non-freeway transitions? I added it as a test to U.S. Route 75 in Nebraska the other day, but I don't think the wikilink in is appropriate for this case. Thoughts? –Fredddie™ 03:00, 22 July 2014 (UTC)
 * The link is there because it was intended to indicate a transition between different routes on a named road, or between sections with/without routes. Perhaps you would be better off with a new type for that situation, maybe "Freeway terminus" or "Freeway transition" or something. - Evad37 &#91;talk] 03:17, 22 July 2014 (UTC)
 * That's more in line with what I was thinking. The transition is what's important, so another type could be added with the blue color à la the purple toll/ETC type. –Fredddie™ 12:02, 22 July 2014 (UTC)

Template-protected edit request on 4 December 2014
Could a country parameter be added?

TheWombatGuru (talk) 16:37, 4 December 2014 (UTC)
 * Yes, indeed it would be a good idea for roads crossing national boundaries. But how would it be displayed?  What would be the change to ?  Once these have been decided and there is consensus, the changes can be made. All the best: Rich Farmbrough, 19:34, 4 December 2014 (UTC).


 * Also, there are national-specific templates for other countries, like NLDint. Like this template, it uses jctint/core. As for highways that cross national boundaries, a country column would be somewhat superfluous to the state/province column in the case of ones that cross from the US to Canada.
 * As for other situations, like an E-road in Europe, I would advise an editor that E-roads are going to be very similar to Interstate or US Highways in the US in one respect: if you're at the level of creating detailed, tabular, road junction lists, you should be breaking into sub articles in most cases. You'll notice that Interstate 75 lacks such a table, but Interstate 75 in Michigan has one. I would tell an editor writing up an article about an E-road crossing from one country to another to either create separate "E1 in " articles, or if it only involved two countries, to skip a country column and explicitly note the boundary crossings list U.S. Route 131.  Imzadi 1979  →   20:10, 4 December 2014 (UTC)

incorrect conversion?
On Wyoming Highway 73 I noticed a junction was at mile 4.64 which was automatically converted to 2.88 km. This is the wrong conversion direction. The kilometer value should be greater than the mile value. I'm not sure where that conversion is happening. Is that on this template, and if so, would someone fix it? I'm prohibited from editing this page. Runner1928 (talk) 16:01, 25 January 2015 (UTC)
 * I added mi to and that fixed the issue. Thanks for the heads up. –Fredddie™ 16:25, 25 January 2015 (UTC)

redlinked locarion
A version of this template is used in the article Sterling Highway. The road starts at an intersection in the middle of nowhere by a small, non-notable lake, "Tern Lake Junction". This is a redlink and will probably always be one as these two roads and a small, shallow lake (more of a pond really) are the only thing there. Is there some way to get the template to not try and wikilink the location? Beeblebrox (talk) 02:58, 4 February 2015 (UTC)
 * Yes.  Imzadi 1979  →   03:21, 4 February 2015 (UTC)
 * Thanks! Beeblebrox (talk) 05:08, 4 February 2015 (UTC)

ORint milepoint
I noticed today that Oregon Route articles use a "milepoint" in the junction list, rather than the typical mile and conversion seen in most states. The milepoint often includes the underlying highway number and may also sometimes have a letter/symbol attached (so it's similar to a text string entry used with California postmiles). ORinttop modifies the table header to use one "milepoint" column in place of the typical two "mile"/"km" columns. However, ORint does not appear to be recognize the  parameter correctly, as the templates generate empty mile and km conversion cells, as opposed to just one cell with the milepoint data. Contrast to the jctbridge series, which do correctly display the milepoint info in the single cell. See Oregon Route 22 for an example of both in use. --  LJ  ↗  21:14, 6 February 2016 (UTC)
 * on it at ORint/sandbox & Template:ORint/testcases - Evad37 &#91;talk] 03:46, 7 February 2016 (UTC)
 * Done . Pages may need a WP:PURGE to update them. - Evad37 &#91;talk] 04:13, 7 February 2016 (UTC)
 * Evad37, thank you for your assistance. This will be helpful in getting several Oregon Route junction lists converted to templates. --  LJ  ↗  04:46, 7 February 2016 (UTC)

Related request: Can we get the  enabled for ORint? This will allow display of the horizontal rule between  and   parameters, useful where the underlying highway changes. A dash (indicating a range of milepoints) currently displays instead. --  LJ  ↗  23:27, 13 February 2016 (UTC)
 * Should be working. –Fredddie™ 23:33, 13 February 2016 (UTC)
 * Thanks, Fredddie! --  LJ  ↗  00:43, 14 February 2016 (UTC)

Mexico
It's been a while and I have no idea what needs to be fixed. Using a Mexican state abbreviation doesn't work; see Mexican Federal Highway 45. (I also have no idea how to generate the proper table heading row.) --NE2 19:58, 4 March 2016 (UTC)
 * Jctint passes the value of state to StateName, which only expands abbreviations from the US and Canada. But before we go any further, (1) why do you want a state column for a highway in a single state? (2) Is there a more appropriate subdivision to use, i.e. a similar spot in the hierarchy to County in the US or to Local government area in Australia? - Evad37 &#91;talk] 01:12, 5 March 2016 (UTC)
 * It's not all in a single state as it's over 1400 km long. I created  as a barebones endeavor, but there needs to be a state switch as Hidalgo needs to be Hidalgo. –Fredddie™ 01:15, 5 March 2016 (UTC)