add Saudi Arabia national shields #851
Conversation
src/js/screen_gfx.js
Outdated
| @@ -2,7 +2,9 @@ | |||
|
|
|||
| import rgba from "color-rgba"; | |||
|
|
|||
| const fontFamily = '"sans-serif-condensed", "Arial Narrow", sans-serif'; | |||
| const naskhFonts = | |||
| '"Noto Naskh Arabic", KacstNaskh, "Sakkal Majalla", "DecoType Naskh Regular"'; | |||
There was a problem hiding this comment.
I picked a few Naskh fonts that are available by default on most operating systems. I'm not 100% sure these are all the best choices, but they look close enough to the style as shown on Saudi shields.
There was a problem hiding this comment.
Legibility doesn't look good on my system:
Why not use the packaged Americana fontstacks that have Arabic?
There was a problem hiding this comment.
Shields are rendered client-side using fonts available to the browser. We would have to re-engineer shields to render text from a PBF fontstack instead of standard web formats.
There was a problem hiding this comment.
Oh duh good point. Don't you need some kind of <link> tag to use the Google Noto fonts?
There was a problem hiding this comment.
Something like
<link href='https://fonts.googleapis.com/css?family=Noto Naskh Arabic' rel='stylesheet'>
There was a problem hiding this comment.
Not if it's already installed on the system, as with Android. But downloading and displaying Noto on all systems is an option.
There was a problem hiding this comment.
Is there a reason we wouldn't? I assume it's just a one time download into browser cache.
There was a problem hiding this comment.
I suppose not. Though we might as well include Noto Sans Armenian as well, if consistency is what we're going for.
There was a problem hiding this comment.
I guess we'll need whatever scripts would get rendered into shields.
src/js/shield.js
Outdated
| case "arab": | ||
| routeDef.ref = latinToArabicDigits(routeDef.ref); |
There was a problem hiding this comment.
Roads in Saudi Arabia are signed solely in Eastern Arabic numerals. This PR contains code to convert digits to Eastern Arabic.
In #676, I included a Roman numeral conversion function because the Roman numerals were purely decorative. In plain text (and even some signs), the route numbers are in Western Arabic numerals. By contrast, Eastern Arabic is the local language’s numeral system. Would it be more appropriate to render the number however it’s tagged in ref, allowing mappers the flexibility to shunt the transliterated number to a subkey such as ref:Latn?
/ref #467 (comment)
There was a problem hiding this comment.
This is a philosophical question indeed. Off the cuff I would say the ref should have the Eastern Arabic numbers in them, which we would render with the right font. This PR takes a more resilient approach and papers over the difference at the expense of not exposing what we might agree is an error.
There was a problem hiding this comment.
It would be worth asking the folks that implemented the scheme though I'm not sure who to ask.
There was a problem hiding this comment.
Note that most of Saudi Arabia’s highways are also part of the Arab Mashreq International Road Network, which is signposted and tagged in Western Arabic numerals: #466.
| @@ -3687,6 +3687,20 @@ export function loadShields() { | |||
| }, | |||
| }; | |||
|
|
|||
| // Saudi Arabia | |||
| shields["SA:national"] = { | |||
There was a problem hiding this comment.
Currently, only one national route is mapped with
network=SA:national, and all other routes are missing either thenetwork=*tag or a relation altogether. ButSA:nationalclearly represents a coherent network, and this could encourage local mappers to complete the tags for other routes.
The one network=SA:national route is what Saudi Arabia calls a “highway” route, as opposed to a “city road” or “branch road”. All three classifications are part of a nationwide system. Part of Highway 40 has been mapped redundantly as network=SA:highway. If we could get a local mapper to confirm their preference, that might also have the effect of encouraging more build-out of these route relations.
Co-authored-by: Minh Nguyễn <mxn@1ec5.org>
| <style id="style12736" type="text/css"> | ||
| .st2{display:none}.st0{fill:#1a1113;stroke:#000}.st1{fill:#21150d}.st2{fill:#1a1512}.st3{fill:#26201a}.st4{fill:#332822}.st5{fill:#00582b;stroke:#003000}.st6{fill:#006c35}.st7{fill:#00803f}.st8{fill:#009449}.st9{fill:#00a853}.st10{fill:none;stroke:#003000;stroke-linecap:round}.st11{fill:#adc8db;stroke:#505050}.st12{fill:#c9dbe7}.st13{fill:#e4edf3}.st14{fill:#f2f7fa}.st15{fill:#d8ad00;stroke:#9c6000;stroke-miterlimit:10}.st16{fill:#fc0}.st17{fill:#ffd429}.st18{fill:#ffdc51}.st19{fill:none;stroke:#9c6000;stroke-linecap:round}.st20{fill:#9c6000} | ||
| </style> | ||
| <path style="fill:#fff;fill-opacity:1;stroke:#006747;stroke-width:2;stroke-linecap:round;stroke-linejoin:round;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;paint-order:stroke fill markers" d="M 4.3137644,1 C 3.8492079,1.2791464 3.3643899,1.5481425 2.8135025,1.5720743 2.5301496,1.7988797 3.1220327,2.0672964 3.2487494,2.2868126 3.74522,2.6760005 3.3397435,2.4751195 3.2089685,2.7792411 3.3153332,3.164358 2.8966442,3.0307546 2.6965448,3.1329313 2.4692966,3.2866578 2.3434561,3.8461563 1.9353155,3.7120804 1.7033785,3.6741775 1.1181472,3.4075228 1.2343171,3.8659058 1.1836197,4.0687148 1.2109496,4.3795171 1.1111662,4.5751992 1.0034179,4.7348102 0.89635621,4.9578449 1.1855069,4.9539668 1.2478523,4.7118357 1.6581687,4.9648988 1.6794047,5.1926358 1.887515,5.3702657 1.9376454,5.6640074 2.1222448,5.84807 2.3751187,6.0430095 2.4307843,6.3829578 2.5992933,6.6068199 2.799276,6.8095433 2.8919397,7.1145056 3.1230109,7.3478722 c 0.3163833,0.2194855 -0.056853,0.3718614 0.1290528,0.5567673 0.1009276,0.3760652 0.4020278,0.5002508 0.7765992,0.6627499 0.4209357,0.3041748 0.3639255,0.8859852 0.6936545,1.2418362 0.1342552,0.1826742 0.010113,0.4894664 -0.00873,0.7287884 0.1863205,0.191439 0.056059,0.510349 0.1931699,0.73593 0.2490773,0.167053 0.09684,0.365887 0.3786881,0.532097 0.1157623,0.295902 0.5107813,0.137251 0.620445,0.46034 0.2852824,0.07054 0.3971888,0.374119 0.4842509,0.5973 0.1262275,0.276486 0.1943561,0.610407 0.3814566,0.81315 0.05503,0.208702 0.2075853,0.400966 0.3881644,0.584597 0.2749861,0.164692 0.3400216,0.442233 0.4706091,0.706938 0.2609456,0.21521 0.3333422,0.501996 0.2839625,0.830249 -0.090444,0.263371 0.071409,0.565306 -0.045736,0.658609 -0.052229,0.20081 0.3122414,0.403181 0.2521199,0.717143 0.1094353,0.272464 0.027472,0.524945 0.2443919,0.760762 0.028435,0.292829 0.064572,0.557238 0.1631924,0.800975 0.044448,0.269543 0.3479729,0.265548 0.6127877,0.263494 0.2374486,-0.101605 0.4304353,-0.109255 0.6999526,-0.04011 0.037881,-0.299693 0.2758696,-0.369578 0.4992786,-0.519641 0.09741,-0.173474 0.26669,-0.103092 0.474031,-0.157764 0.271578,0.0052 0.388464,-0.03311 0.613321,-0.11498 0.171011,0.02804 0.45041,-0.312783 0.647871,-0.386393 0.08291,-0.162003 0.31356,0.01466 0.529536,-0.102586 0.21142,-0.04306 0.406205,-0.257519 0.488138,-0.414741 0.122381,-0.225348 0.416162,-0.227733 0.683429,-0.238288 0.255018,-0.05095 0.385422,-0.274681 0.691106,-0.297328 0.281053,-0.07512 0.557881,-0.184409 0.765337,-0.319293 0.381056,-0.02642 0.143443,-0.527409 0.214511,-0.653441 0.08156,-0.276873 0.57915,-0.300448 0.829043,-0.474832 0.260635,-0.03442 0.644025,-0.439445 0.768162,-0.247728 0.178655,-0.07158 0.465272,0.179615 0.605739,-0.15167 0.225364,-0.09146 -0.0087,-0.544765 0.311861,-0.678317 0.338195,-0.02463 0.506419,0.03713 0.705235,-0.08414 -0.04914,-0.625937 0.461364,-0.825775 0.938729,-0.942253 0.150754,-0.263545 -0.03863,-0.445858 -0.06129,-0.727516 0.124519,-0.16605 -0.02143,-0.586593 0.261629,-0.697469 0.03883,0.444876 0.429151,0.03411 0.392414,-0.12238 -0.0055,-0.273058 0.157486,-0.320926 0.344457,-0.435366 0.07941,-0.259705 0.168815,-0.429732 0.324196,-0.554808 0.04572,-0.243209 0.166439,-0.444061 0.121693,-0.7379902 C 20.83869,9.8127734 20.522218,9.6861652 20.431585,9.3845407 20.225535,9.0634961 19.909041,8.6877844 19.488669,8.7582545 19.208358,8.6333385 18.879523,8.5996632 18.632096,8.3597677 18.409775,8.2029922 18.324512,7.9469118 18.145442,7.7046262 18.228084,7.3904927 17.954016,7.1522435 18.062677,6.8758794 18.13613,6.6397776 18.038083,6.1279116 17.824484,6.4395324 17.65749,6.6516302 17.75338,6.8468004 17.516677,6.9552645 17.343123,7.0368255 17.207302,7.3997479 16.985047,7.508199 16.853309,7.7099414 16.66423,7.9366774 16.625785,8.1185758 16.497557,8.3155226 16.247834,8.1929588 16.019946,8.3246583 15.779121,8.5928635 15.48286,8.2518563 15.197778,8.3180283 15.078781,8.4626691 14.572212,8.6589364 14.655156,8.2633123 14.5005,8.1783003 14.264413,8.2734113 14.3739,7.9603644 14.463139,7.747792 14.561661,7.4824917 14.387665,7.2811401 14.305115,7.0836348 14.526737,6.7783622 14.202765,6.6051108 13.995404,6.4983198 13.783931,6.5730848 13.73848,6.9094551 13.608026,6.9542481 13.706253,7.7807895 13.58612,7.306844 13.498098,7.0060554 13.134601,6.8243725 13.246226,6.4680507 13.259149,6.121132 13.224122,6.1436112 12.98783,5.8459692 12.776669,5.7747115 12.67084,5.5302316 12.496535,5.4115344 12.366755,5.1907696 12.017655,5.0697559 12.000264,4.7897077 11.855838,4.5757747 11.739645,4.2199266 11.364906,4.3218688 10.98494,4.3812401 10.959643,3.7519039 10.549254,3.8545496 9.9411817,3.7641064 9.3036658,3.7462157 8.713742,3.6110542 7.6917476,2.825552 6.7062758,1.9857938 5.5718481,1.3645029 5.2146445,1.0783532 4.7379204,1.1134174 4.3137644,1 Z" id="polyF1S187P4"/> |
There was a problem hiding this comment.
If we simplify this shape a bit, I think it would become more feasible to tweak it to better accommodate the numerals at a slightly larger font size. I don’t have any concrete suggestions, since this is a pretty irregular shape, but maybe play around with it a bit.
|
It does seem a bit better with the web fonts but not sure how legible it would be to an Arabic speaker. Do we think this looks OK or should we try to enlarge or skew the shape a bit?
|
|
Shieldtest output:
|
|
Looks like Latin text in that screenshot has regressed: it’s a serif font instead of a sans-serif font. |
|
Needs to resolve conflicts. |
Roads in Saudi Arabia are signed solely in Eastern Arabic numerals. This PR contains code to convert digits to Eastern Arabic.
Currently, only one national route is mapped with
network=SA:national, and all other routes are missing either thenetwork=*tag or a relation altogether. ButSA:nationalclearly represents a coherent network, and this could encourage local mappers to complete the tags for other routes.