5

Is \trianglelefteqslant from stmaryrd and the negated version of the symbol present in Unicode? I cannot find them, neither can shapecatcher.

Shape in different sizes: Shape in different sizes:

Related: How to get \trianglelefteq similar to \leqslant?, Another version of the command \triangleleft in amssymb

EDITED AGAIN: I took a look into the basic math block and the supplemental maths block, I searched for the occurrences of the words "triangle" and "slanted" in the Unicode Chart Databases 11.0.0 and 12.0.0, and, as of today, I found no matching characters. (It doesn't mean that such a character doesn't exist in Unicode, though.) How to use the command \trianglelefteqslant in (any)TeX is not my question and is considered off-topic. My best efforts in combining existing Unicode chars resulted in ⩽⃒ and ⩽⃓.

  • The closest things I could find are ⪨ (U+2AA8) and ⪩ (U+2AA9), which makes me suspicious about the existence of the Unicode for \trianglelefteqslant. – Ruixi Zhang Oct 24 '18 at 00:18
  • 1
    ⊴ (U+22B4) and ⊵ (U+22B5) are \trianglelefteq and \trianglerighteq. – Davislor Feb 04 '19 at 04:03
  • STIX Two’s stylistic set SS10 has slanted forms of various “or equal to” relations, but not ⊴. I don’t know of a Unicode font that has that variant. – Davislor Feb 04 '19 at 04:08
  • 1
    Unicode defines characters, not glyphs. Using a slanted bar is a stylistic choice and the symbols with horizontal and slanted bar are not distinct by meaning. If no font provides that shape, perhaps as stylistic variant, you're out of luck. – egreg Feb 04 '19 at 10:15
  • Unicode is far from perfect! This is so due to the size of the standard, plus the many years it has existed, and the fact that a character can never be removed from the standard once it has been added. So don't be surprised to find oddities and inconsistensies in unicode. In any case, if a font designer goes ahead and makes U+22B4 and U+22B5 with slanted lines, that's probably okay. It could even be a stylistic choice, with both variants found in the same font. – Harald Hanche-Olsen Feb 04 '19 at 15:03
  • Unicode is a pain, but scrapping it and starting over would be a much greater pain. I understand the sentiment, though. – Harald Hanche-Olsen Feb 04 '19 at 15:09
  • 1
    Scope-wise, this is borderline for us I suspect. I see the TeX link in that the symbol you want has TeX history, but 'is it in the Unicode standard' is not something that links to TeX-based knowledge. On the other hand, I'm not sure it's on-topic on another network site. – Joseph Wright Feb 04 '19 at 15:15
  • @user49915 Like I said, borderline; I'm mainly flagging up that you might not get much beyond comments saying 'can't find it' or whatever. Of course, it might be there or get added ... I'm no expert. – Joseph Wright Feb 04 '19 at 15:21
  • Any symbol especially emojis can be submitted and 2BF9 1 EQUALS SIGN WITH INFINITY BELOW is perhaps the latest relevant at http://www.unicode.org/pending/proposals.html –  Feb 04 '19 at 16:15
  • @user49915 I'm not discussing about unique meaning; the fact is that a reader will not be able to distinguish between the symbol with the slanted or horizontal bar. You have a point though in remarking that there are both ⩽ and ≤, which I find very curious (understatement for “wrong”). – egreg Feb 04 '19 at 16:50
  • the simplese answer for why these aren't in unicode is probably that no one requested them at the time when the mass addition of math symbols was happening. if they were recognized, it would almost certainly be with "variant selector 1" (VS1). see the table on pp.28-29 of unicode technical report #25 (can be viewed on the unicode site). i will try to follow up, but am in the process of retiring, so haven't much time. – barbara beeton Feb 04 '19 at 17:06
  • 1
    I don't know the significance but it appears that last century a number of symbols were considerd as available by composition thus 5X24 * R NOT, VERTICAL, LEFT TRIANGLE, EQUALS [composed: U+22B4 + U+20D2] whilst 5X25 * R NOT, VERTICAL, RIGHT TRIANGLE, EQUALS [composed: U+22B5 + U+20D2] I thus surmise that there are two glyphs that when combined would produce the symbols you desire (please don't flame the messenger, just correct my supposition :-) –  Feb 04 '19 at 17:07
  • I think your first link to https://tex.stackexchange.com/a/237417/170109 was in effect the same principle and I am severely dysTeXic so am happy to observe others craft their wares. –  Feb 04 '19 at 17:58

2 Answers2

10

This operator would be in the Mathematical Operators block (or its supplemental block), but it's not there, so the answer is simply

no

Maybe there exists a font which has this symbol in the private use area, but that doesn't count as Unicode.

I guess the point whether these symbols are in Unicode is rather secondary, unless you want to enter the literal Unicode symbol (which unfortunately does not exist). The fact whether they are accessible in LaTeX when using a Unicode math font is more important. The symbols in question are defined by the stmaryrd package, which works in conjunction with Unicode math. It is probably a good idea to load stmaryrd before unicode-math, such that unicode-math can overwrite symbols from stmaryrd, which actually are in Unicode.

\documentclass{article}
\usepackage{stmaryrd}
\usepackage{unicode-math}
\begin{document}
$\trianglelefteqslant \ntrianglelefteqslant$
\end{document}

enter image description here

Henri Menke
  • 109,596
  • I have cleaned up comments here. These largely focussed on whether this answers the question and the resulting vote it deserves. Voting should be purely about whether the answer addresses the question. Up/down voting should be purely based on whether an answer does that. – Joseph Wright Feb 04 '19 at 07:52
3

it's true that these symbols aren't in unicode.

the simplest answer for why this is so is probably that no one requested them at the time when the mass addition of math symbols was happening. if they were recognized, it would almost certainly be with "variant selector 1" (VS1). see the table on pp.28-29 of unicode technical report #25 (can be viewed on the unicode site).

as presenter of the stix symbols collection for addition to unicode (and co-author of tech report #25), i still have contacts within the unicode technical committee, and can suggest that additional symbols be recognized. the VS1 route is almost certainly the approach to be taken, since there is no difference in meaning (that i'm aware of) between the symbol with a horizontal equality stroke and a slanted stroke.

henri menke's identification of the desired symbol in the stmaryrd font is currently the best solution, but a unicode equivalent is certainly desirable (even though it will probably cause trouble for the mathml maintainers). i believe there are some other symbols in stmaryrd that are also in the same unfortunate limbo.

i'm currently in the process of retiring from the ams, and won't be able to do anything useful for a couple of weeks, but i intend to follow up. if there are other similar symbol instances that can be presented at the same time, please identify them in comments, and i will submit all of them together.

barbara beeton
  • 88,848