regex101: Advanced chemical formula parser

Negative Lookbehind(?<![^\s>])

Negated Character Class[^\s>]

\sany whitespace character

>the literal > (62₁₀ / 76₈ / 3E₁₆)

Comment: LEFT boundary: preceding char is whitespace, '>' (end of a tag), or start-of-string

Group 1( # capture 1: the whole formula (?![^<>]*>) # guard: don't start inside an HTML tag's attribute list (?: # ============ HEAD — one or more element/bracket "units" ============ (?: # a "unit": an element/bracket run, then any trailing counts (?: (?:H[eogsf]?|L[iavru]|B[eahkri]?|C[arofmusenld]?|N[eiahopdb]?|O[sg]?|F[rlem]?|M[godtcn]|A[lrsgutmc]|S[icerngmb]?|P[uabotmrd]?|Kr?|T[icebmsalh]|V|Z[nr]|G[ade]|R[buhenagf]|Yb?|I[nr]?|Xe|E[urs]|D[ysb]|W|U) # any element symbol (all 118) |[()\[\]] # ...or a bracket )+ (?: # then any trailing count, in ANY encoding, a plain digit, or a repeat index: (?: [₁₂₃₄₅₆₇₈₉][₀₁₂₃₄₅₆₇₈₉]* # subscript glyphs H₂ |[¹²³⁴⁵⁶⁷⁸⁹][⁰¹²³⁴⁵⁶⁷⁸⁹]* # superscript glyphs x² |\\u208[1-9](?:\\u208[0-9])* # \u escape, subscript H\u2082 (unparsed JSON) |(?:\\u00[bB][239]|\\u207[4-9])(?:\\u2070|\\u00[bB][239]|\\u207[4-9])* # \u escape, superscript |&\#(?:0*832[1-9]|[xX]0*208[1-9]);(?:&\#(?:0*832[0-9]|[xX]0*208[0-9]);)* # HTML entity, subscript H₂ / H₂ |&\#(?:0*(?:178|179|185|830[89]|831[0-3])|[xX]0*(?:[bB][239]|207[4-9]));(?:&\#(?:0*(?:178|179|185|8304|830[89]|831[0-3])|[xX]0*(?:2070|[bB][239]|207[4-9]));)* # HTML entity, superscript |<su[bp]>[1-9][0-9]*<\/su[bp]> # / tag H2 ) |[1-9][0-9]* # ...or a plain inline integer |[ₙₘₓ] # ...or a repeat-index glyph (…)ₙ ← NEW |<su[bp]>[nmx]<\/su[bp]> # ...or a repeat index in a tag (…)n ← NEW )* )+ # one or more units (a lone valid element is allowed) ) (?:[+-](?![A-Za-z0-9]))? # optional ionic charge on the head (K+, …F₃-) (?: # ============ zero or more salt / hydrate components ============ (?:\s*[·•‧∙⋅・･*]\s*|\.(?=[A-Za-z(\[])) # separator: a spaced dot variant, OR a tight "." right before a component (?: # optional coefficient: a sub/sup number, an integer/fraction, or x/n [₁₂₃₄₅₆₇₈₉][₀₁₂₃₄₅₆₇₈₉]*|[¹²³⁴⁵⁶⁷⁸⁹][⁰¹²³⁴⁵⁶⁷⁸⁹]*|\\u208[1-9](?:\\u208[0-9])*|(?:\\u00[bB][239]|\\u207[4-9])(?:\\u2070|\\u00[bB][239]|\\u207[4-9])*|&\#(?:0*832[1-9]|[xX]0*208[1-9]);(?:&\#(?:0*832[0-9]|[xX]0*208[0-9]);)*|&\#(?:0*(?:178|179|185|830[89]|831[0-3])|[xX]0*(?:[bB][239]|207[4-9]));(?:&\#(?:0*(?:178|179|185|8304|830[89]|831[0-3])|[xX]0*(?:2070|[bB][239]|207[4-9]));)*|<su[bp]>[1-9][0-9]*<\/su[bp]>|[1-9][0-9]*(?:\/[1-9][0-9]*)?|[xn] )? (?: # the component's own units (elements/brackets + counts), one or more (?: (?:H[eogsf]?|L[iavru]|B[eahkri]?|C[arofmusenld]?|N[eiahopdb]?|O[sg]?|F[rlem]?|M[godtcn]|A[lrsgutmc]|S[icerngmb]?|P[uabotmrd]?|Kr?|T[icebmsalh]|V|Z[nr]|G[ade]|R[buhenagf]|Yb?|I[nr]?|Xe|E[urs]|D[ysb]|W|U) |[()\[\]] )+ (?: (?:[₁₂₃₄₅₆₇₈₉][₀₁₂₃₄₅₆₇₈₉]*|[¹²³⁴⁵⁶⁷⁸⁹][⁰¹²³⁴⁵⁶⁷⁸⁹]*|\\u208[1-9](?:\\u208[0-9])*|(?:\\u00[bB][239]|\\u207[4-9])(?:\\u2070|\\u00[bB][239]|\\u207[4-9])*|&\#(?:0*832[1-9]|[xX]0*208[1-9]);(?:&\#(?:0*832[0-9]|[xX]0*208[0-9]);)*|&\#(?:0*(?:178|179|185|830[89]|831[0-3])|[xX]0*(?:[bB][239]|207[4-9]));(?:&\#(?:0*(?:178|179|185|8304|830[89]|831[0-3])|[xX]0*(?:2070|[bB][239]|207[4-9]));)*|<su[bp]>[1-9][0-9]*<\/su[bp]>)|[1-9][0-9]*|[ₙₘₓ]|<su[bp]>[nmx]<\/su[bp]> # ← NEW: repeat index also allowed on components )* )+ (?:[+-](?![A-Za-z0-9]))? # optional charge on the component )* )

Comment: capture 1: the whole formula

Negative Lookahead(?![^<>]*>)

Negated Character Class[^<>]*

*matches any number of characters outside the set below:

<the literal < (60₁₀ / 74₈ / 3C₁₆)

>the literal > (62₁₀ / 76₈ / 3E₁₆)

Comment: guard: don't start inside an HTML tag's attribute list

Non-Capturing Group(?: # ============ HEAD — one or more element/bracket "units" ============ (?: # a "unit": an element/bracket run, then any trailing counts (?: (?:H[eogsf]?|L[iavru]|B[eahkri]?|C[arofmusenld]?|N[eiahopdb]?|O[sg]?|F[rlem]?|M[godtcn]|A[lrsgutmc]|S[icerngmb]?|P[uabotmrd]?|Kr?|T[icebmsalh]|V|Z[nr]|G[ade]|R[buhenagf]|Yb?|I[nr]?|Xe|E[urs]|D[ysb]|W|U) # any element symbol (all 118) |[()\[\]] # ...or a bracket )+ (?: # then any trailing count, in ANY encoding, a plain digit, or a repeat index: (?: [₁₂₃₄₅₆₇₈₉][₀₁₂₃₄₅₆₇₈₉]* # subscript glyphs H₂ |[¹²³⁴⁵⁶⁷⁸⁹][⁰¹²³⁴⁵⁶⁷⁸⁹]* # superscript glyphs x² |\\u208[1-9](?:\\u208[0-9])* # \u escape, subscript H\u2082 (unparsed JSON) |(?:\\u00[bB][239]|\\u207[4-9])(?:\\u2070|\\u00[bB][239]|\\u207[4-9])* # \u escape, superscript |&\#(?:0*832[1-9]|[xX]0*208[1-9]);(?:&\#(?:0*832[0-9]|[xX]0*208[0-9]);)* # HTML entity, subscript H₂ / H₂ |&\#(?:0*(?:178|179|185|830[89]|831[0-3])|[xX]0*(?:[bB][239]|207[4-9]));(?:&\#(?:0*(?:178|179|185|8304|830[89]|831[0-3])|[xX]0*(?:2070|[bB][239]|207[4-9]));)* # HTML entity, superscript |<su[bp]>[1-9][0-9]*<\/su[bp]> # / tag H2 ) |[1-9][0-9]* # ...or a plain inline integer |[ₙₘₓ] # ...or a repeat-index glyph (…)ₙ ← NEW |<su[bp]>[nmx]<\/su[bp]> # ...or a repeat index in a tag (…)n ← NEW )* )+ # one or more units (a lone valid element is allowed) )

Comment: ============ HEAD — one or more element/bracket "units" ============

Non-Capturing Group(?: # a "unit": an element/bracket run, then any trailing counts (?: (?:H[eogsf]?|L[iavru]|B[eahkri]?|C[arofmusenld]?|N[eiahopdb]?|O[sg]?|F[rlem]?|M[godtcn]|A[lrsgutmc]|S[icerngmb]?|P[uabotmrd]?|Kr?|T[icebmsalh]|V|Z[nr]|G[ade]|R[buhenagf]|Yb?|I[nr]?|Xe|E[urs]|D[ysb]|W|U) # any element symbol (all 118) |[()\[\]] # ...or a bracket )+ (?: # then any trailing count, in ANY encoding, a plain digit, or a repeat index: (?: [₁₂₃₄₅₆₇₈₉][₀₁₂₃₄₅₆₇₈₉]* # subscript glyphs H₂ |[¹²³⁴⁵⁶⁷⁸⁹][⁰¹²³⁴⁵⁶⁷⁸⁹]* # superscript glyphs x² |\\u208[1-9](?:\\u208[0-9])* # \u escape, subscript H\u2082 (unparsed JSON) |(?:\\u00[bB][239]|\\u207[4-9])(?:\\u2070|\\u00[bB][239]|\\u207[4-9])* # \u escape, superscript |&\#(?:0*832[1-9]|[xX]0*208[1-9]);(?:&\#(?:0*832[0-9]|[xX]0*208[0-9]);)* # HTML entity, subscript H₂ / H₂ |&\#(?:0*(?:178|179|185|830[89]|831[0-3])|[xX]0*(?:[bB][239]|207[4-9]));(?:&\#(?:0*(?:178|179|185|8304|830[89]|831[0-3])|[xX]0*(?:2070|[bB][239]|207[4-9]));)* # HTML entity, superscript |<su[bp]>[1-9][0-9]*<\/su[bp]> # / tag H2 ) |[1-9][0-9]* # ...or a plain inline integer |[ₙₘₓ] # ...or a repeat-index glyph (…)ₙ ← NEW |<su[bp]>[nmx]<\/su[bp]> # ...or a repeat index in a tag (…)n ← NEW )* )+

+repeats the group contents below at least once:

Comment: a "unit": an element/bracket run, then any trailing counts

+repeats the group contents below at least once:

1st AlternativeH[eogsf]?

Hthe literal H (72₁₀ / 110₈ / 48₁₆) (case sensitive)

Character Class[eogsf]?

?matches at most one character from the set below:

ethe literal e (101₁₀ / 145₈ / 65₁₆) (case sensitive)

othe literal o (111₁₀ / 157₈ / 6F₁₆) (case sensitive)

gthe literal g (103₁₀ / 147₈ / 67₁₆) (case sensitive)

sthe literal s (115₁₀ / 163₈ / 73₁₆) (case sensitive)

fthe literal f (102₁₀ / 146₈ / 66₁₆) (case sensitive)

2nd AlternativeL[iavru]

Lthe literal L (76₁₀ / 114₈ / 4C₁₆) (case sensitive)

Character Class[iavru]

ithe literal i (105₁₀ / 151₈ / 69₁₆) (case sensitive)

athe literal a (97₁₀ / 141₈ / 61₁₆) (case sensitive)

vthe literal v (118₁₀ / 166₈ / 76₁₆) (case sensitive)

rthe literal r (114₁₀ / 162₈ / 72₁₆) (case sensitive)

uthe literal u (117₁₀ / 165₈ / 75₁₆) (case sensitive)

3rd AlternativeB[eahkri]?

Bthe literal B (66₁₀ / 102₈ / 42₁₆) (case sensitive)

Character Class[eahkri]?

?matches at most one character from the set below:

ethe literal e (101₁₀ / 145₈ / 65₁₆) (case sensitive)

athe literal a (97₁₀ / 141₈ / 61₁₆) (case sensitive)

hthe literal h (104₁₀ / 150₈ / 68₁₆) (case sensitive)

kthe literal k (107₁₀ / 153₈ / 6B₁₆) (case sensitive)

rthe literal r (114₁₀ / 162₈ / 72₁₆) (case sensitive)

ithe literal i (105₁₀ / 151₈ / 69₁₆) (case sensitive)

4th AlternativeC[arofmusenld]?

Cthe literal C (67₁₀ / 103₈ / 43₁₆) (case sensitive)

Character Class[arofmusenld]?

?matches at most one character from the set below:

athe literal a (97₁₀ / 141₈ / 61₁₆) (case sensitive)

rthe literal r (114₁₀ / 162₈ / 72₁₆) (case sensitive)

othe literal o (111₁₀ / 157₈ / 6F₁₆) (case sensitive)

fthe literal f (102₁₀ / 146₈ / 66₁₆) (case sensitive)

mthe literal m (109₁₀ / 155₈ / 6D₁₆) (case sensitive)

uthe literal u (117₁₀ / 165₈ / 75₁₆) (case sensitive)

sthe literal s (115₁₀ / 163₈ / 73₁₆) (case sensitive)

ethe literal e (101₁₀ / 145₈ / 65₁₆) (case sensitive)

nthe literal n (110₁₀ / 156₈ / 6E₁₆) (case sensitive)

lthe literal l (108₁₀ / 154₈ / 6C₁₆) (case sensitive)

dthe literal d (100₁₀ / 144₈ / 64₁₆) (case sensitive)

5th AlternativeN[eiahopdb]?

Nthe literal N (78₁₀ / 116₈ / 4E₁₆) (case sensitive)

Character Class[eiahopdb]?

?matches at most one character from the set below:

ethe literal e (101₁₀ / 145₈ / 65₁₆) (case sensitive)

ithe literal i (105₁₀ / 151₈ / 69₁₆) (case sensitive)

athe literal a (97₁₀ / 141₈ / 61₁₆) (case sensitive)

hthe literal h (104₁₀ / 150₈ / 68₁₆) (case sensitive)

othe literal o (111₁₀ / 157₈ / 6F₁₆) (case sensitive)

pthe literal p (112₁₀ / 160₈ / 70₁₆) (case sensitive)

dthe literal d (100₁₀ / 144₈ / 64₁₆) (case sensitive)

bthe literal b (98₁₀ / 142₈ / 62₁₆) (case sensitive)

6th AlternativeO[sg]?

Othe literal O (79₁₀ / 117₈ / 4F₁₆) (case sensitive)

Character Class[sg]?

?matches at most one character from the set below:

sthe literal s (115₁₀ / 163₈ / 73₁₆) (case sensitive)

gthe literal g (103₁₀ / 147₈ / 67₁₆) (case sensitive)

7th AlternativeF[rlem]?

Fthe literal F (70₁₀ / 106₈ / 46₁₆) (case sensitive)

Character Class[rlem]?

?matches at most one character from the set below:

rthe literal r (114₁₀ / 162₈ / 72₁₆) (case sensitive)

lthe literal l (108₁₀ / 154₈ / 6C₁₆) (case sensitive)

ethe literal e (101₁₀ / 145₈ / 65₁₆) (case sensitive)

mthe literal m (109₁₀ / 155₈ / 6D₁₆) (case sensitive)

8th AlternativeM[godtcn]

Mthe literal M (77₁₀ / 115₈ / 4D₁₆) (case sensitive)

Character Class[godtcn]

9th AlternativeA[lrsgutmc]

10th AlternativeS[icerngmb]?

11th AlternativeP[uabotmrd]?

12th AlternativeKr?

13th AlternativeT[icebmsalh]

14th AlternativeV

15th AlternativeZ[nr]

16th AlternativeG[ade]

17th AlternativeR[buhenagf]

18th AlternativeYb?

19th AlternativeI[nr]?

20th AlternativeXe

21st AlternativeE[urs]

22nd AlternativeD[ysb]

23rd AlternativeW

24th AlternativeU

Comment: any element symbol (all 118)

2nd Alternative[()\[\]] # ...or a bracket

Non-Capturing Group(?: # then any trailing count, in ANY encoding, a plain digit, or a repeat index: (?: [₁₂₃₄₅₆₇₈₉][₀₁₂₃₄₅₆₇₈₉]* # subscript glyphs H₂ |[¹²³⁴⁵⁶⁷⁸⁹][⁰¹²³⁴⁵⁶⁷⁸⁹]* # superscript glyphs x² |\\u208[1-9](?:\\u208[0-9])* # \u escape, subscript H\u2082 (unparsed JSON) |(?:\\u00[bB][239]|\\u207[4-9])(?:\\u2070|\\u00[bB][239]|\\u207[4-9])* # \u escape, superscript |&\#(?:0*832[1-9]|[xX]0*208[1-9]);(?:&\#(?:0*832[0-9]|[xX]0*208[0-9]);)* # HTML entity, subscript H₂ / H₂ |&\#(?:0*(?:178|179|185|830[89]|831[0-3])|[xX]0*(?:[bB][239]|207[4-9]));(?:&\#(?:0*(?:178|179|185|8304|830[89]|831[0-3])|[xX]0*(?:2070|[bB][239]|207[4-9]));)* # HTML entity, superscript |<su[bp]>[1-9][0-9]*<\/su[bp]> # / tag H2 ) |[1-9][0-9]* # ...or a plain inline integer |[ₙₘₓ] # ...or a repeat-index glyph (…)ₙ ← NEW |<su[bp]>[nmx]<\/su[bp]> # ...or a repeat index in a tag (…)n ← NEW )*

Comment: one or more units (a lone valid element is allowed)

Non-Capturing Group(?:[+-](?![A-Za-z0-9]))?

Comment: optional ionic charge on the head (K+, …F₃-)

Non-Capturing Group(?: # ============ zero or more salt / hydrate components ============ (?:\s*[·•‧∙⋅・･*]\s*|\.(?=[A-Za-z(\[])) # separator: a spaced dot variant, OR a tight "." right before a component (?: # optional coefficient: a sub/sup number, an integer/fraction, or x/n [₁₂₃₄₅₆₇₈₉][₀₁₂₃₄₅₆₇₈₉]*|[¹²³⁴⁵⁶⁷⁸⁹][⁰¹²³⁴⁵⁶⁷⁸⁹]*|\\u208[1-9](?:\\u208[0-9])*|(?:\\u00[bB][239]|\\u207[4-9])(?:\\u2070|\\u00[bB][239]|\\u207[4-9])*|&\#(?:0*832[1-9]|[xX]0*208[1-9]);(?:&\#(?:0*832[0-9]|[xX]0*208[0-9]);)*|&\#(?:0*(?:178|179|185|830[89]|831[0-3])|[xX]0*(?:[bB][239]|207[4-9]));(?:&\#(?:0*(?:178|179|185|8304|830[89]|831[0-3])|[xX]0*(?:2070|[bB][239]|207[4-9]));)*|<su[bp]>[1-9][0-9]*<\/su[bp]>|[1-9][0-9]*(?:\/[1-9][0-9]*)?|[xn] )? (?: # the component's own units (elements/brackets + counts), one or more (?: (?:H[eogsf]?|L[iavru]|B[eahkri]?|C[arofmusenld]?|N[eiahopdb]?|O[sg]?|F[rlem]?|M[godtcn]|A[lrsgutmc]|S[icerngmb]?|P[uabotmrd]?|Kr?|T[icebmsalh]|V|Z[nr]|G[ade]|R[buhenagf]|Yb?|I[nr]?|Xe|E[urs]|D[ysb]|W|U) |[()\[\]] )+ (?: (?:[₁₂₃₄₅₆₇₈₉][₀₁₂₃₄₅₆₇₈₉]*|[¹²³⁴⁵⁶⁷⁸⁹][⁰¹²³⁴⁵⁶⁷⁸⁹]*|\\u208[1-9](?:\\u208[0-9])*|(?:\\u00[bB][239]|\\u207[4-9])(?:\\u2070|\\u00[bB][239]|\\u207[4-9])*|&\#(?:0*832[1-9]|[xX]0*208[1-9]);(?:&\#(?:0*832[0-9]|[xX]0*208[0-9]);)*|&\#(?:0*(?:178|179|185|830[89]|831[0-3])|[xX]0*(?:[bB][239]|207[4-9]));(?:&\#(?:0*(?:178|179|185|8304|830[89]|831[0-3])|[xX]0*(?:2070|[bB][239]|207[4-9]));)*|<su[bp]>[1-9][0-9]*<\/su[bp]>)|[1-9][0-9]*|[ₙₘₓ]|<su[bp]>[nmx]<\/su[bp]> # ← NEW: repeat index also allowed on components )* )+ (?:[+-](?![A-Za-z0-9]))? # optional charge on the component )*

Negative Lookahead(?![^\s<])

Comment: RIGHT boundary: following char is whitespace, '<' (start of a tag), or end-of-string

Global pattern flags

g modifier: global. Finds all matches instead of stopping after the first

m modifier: multiline. Causes ^ and $ to match the start and end of each line, not only the start and end of the string

x modifier: extended. Ignores unescaped whitespace and comments outside character classes

25 tokens

Regular Expression

(?<![^\s>])                                                               # LEFT boundary: preceding char is whitespace, '>' (end of a tag), or start-of-string
(                                                                         # capture 1: the whole formula
  (?![^<>]*>)                                                             # guard: don't start inside an HTML tag's attribute list
  (?:                                                                     # ============ HEAD — one or more element/bracket "units" ============
    (?:                                                                   #   a "unit": an element/bracket run, then any trailing counts
      (?:
        (?:H[eogsf]?|L[iavru]|B[eahkri]?|C[arofmusenld]?|N[eiahopdb]?|O[sg]?|F[rlem]?|M[godtcn]|A[lrsgutmc]|S[icerngmb]?|P[uabotmrd]?|Kr?|T[icebmsalh]|V|Z[nr]|G[ade]|R[buhenagf]|Yb?|I[nr]?|Xe|E[urs]|D[ysb]|W|U)  # any element symbol (all 118)
        |[()\[\]]                                                         #   ...or a bracket
      )+
      (?:                                                                 #   then any trailing count, in ANY encoding, a plain digit, or a repeat index:
        (?:
          [₁₂₃₄₅₆₇₈₉][₀₁₂₃₄₅₆₇₈₉]*                                        #     subscript glyphs        H₂
          |[¹²³⁴⁵⁶⁷⁸⁹][⁰¹²³⁴⁵⁶⁷⁸⁹]*                                       #     superscript glyphs      x²
          |\\u208[1-9](?:\\u208[0-9])*                                     #     \u escape, subscript    H\u2082  (unparsed JSON)
          |(?:\\u00[bB][239]|\\u207[4-9])(?:\\u2070|\\u00[bB][239]|\\u207[4-9])*   #  \u escape, superscript
          |&\#(?:0*832[1-9]|[xX]0*208[1-9]);(?:&\#(?:0*832[0-9]|[xX]0*208[0-9]);)*   # HTML entity, subscript   H&#8322; / H&#x2082;
          |&\#(?:0*(?:178|179|185|830[89]|831[0-3])|[xX]0*(?:[bB][239]|207[4-9]));(?:&\#(?:0*(?:178|179|185|8304|830[89]|831[0-3])|[xX]0*(?:2070|[bB][239]|207[4-9]));)*  # HTML entity, superscript
          |<su[bp]>[1-9][0-9]*<\/su[bp]>                                    #     <sub>/<sup> tag          H<sub>2</sub>
        )
        |[1-9][0-9]*                                                       #     ...or a plain inline integer
        |[ₙₘₓ]                                                            #     ...or a repeat-index glyph        (…)ₙ   ← NEW
        |<su[bp]>[nmx]<\/su[bp]>                                           #     ...or a repeat index in a tag     (…)<sub>n</sub>   ← NEW
      )*
    )+                                                                    #   one or more units (a lone valid element is allowed)
  )
  (?:[+-](?![A-Za-z0-9]))?                                                 # optional ionic charge on the head    (K+, …F₃-)
  (?:                                                                      # ============ zero or more salt / hydrate components ============
    (?:\s*[·•‧∙⋅・･*]\s*|\.(?=[A-Za-z(\[]))                                 #   separator: a spaced dot variant, OR a tight "." right before a component
    (?:                                                                    #   optional coefficient: a sub/sup number, an integer/fraction, or x/n
      [₁₂₃₄₅₆₇₈₉][₀₁₂₃₄₅₆₇₈₉]*|[¹²³⁴⁵⁶⁷⁸⁹][⁰¹²³⁴⁵⁶⁷⁸⁹]*|\\u208[1-9](?:\\u208[0-9])*|(?:\\u00[bB][239]|\\u207[4-9])(?:\\u2070|\\u00[bB][239]|\\u207[4-9])*|&\#(?:0*832[1-9]|[xX]0*208[1-9]);(?:&\#(?:0*832[0-9]|[xX]0*208[0-9]);)*|&\#(?:0*(?:178|179|185|830[89]|831[0-3])|[xX]0*(?:[bB][239]|207[4-9]));(?:&\#(?:0*(?:178|179|185|8304|830[89]|831[0-3])|[xX]0*(?:2070|[bB][239]|207[4-9]));)*|<su[bp]>[1-9][0-9]*<\/su[bp]>|[1-9][0-9]*(?:\/[1-9][0-9]*)?|[xn]
    )?
    (?:                                                                    #   the component's own units (elements/brackets + counts), one or more
      (?:
        (?:H[eogsf]?|L[iavru]|B[eahkri]?|C[arofmusenld]?|N[eiahopdb]?|O[sg]?|F[rlem]?|M[godtcn]|A[lrsgutmc]|S[icerngmb]?|P[uabotmrd]?|Kr?|T[icebmsalh]|V|Z[nr]|G[ade]|R[buhenagf]|Yb?|I[nr]?|Xe|E[urs]|D[ysb]|W|U)
        |[()\[\]]
      )+
      (?:
        (?:[₁₂₃₄₅₆₇₈₉][₀₁₂₃₄₅₆₇₈₉]*|[¹²³⁴⁵⁶⁷⁸⁹][⁰¹²³⁴⁵⁶⁷⁸⁹]*|\\u208[1-9](?:\\u208[0-9])*|(?:\\u00[bB][239]|\\u207[4-9])(?:\\u2070|\\u00[bB][239]|\\u207[4-9])*|&\#(?:0*832[1-9]|[xX]0*208[1-9]);(?:&\#(?:0*832[0-9]|[xX]0*208[0-9]);)*|&\#(?:0*(?:178|179|185|830[89]|831[0-3])|[xX]0*(?:[bB][239]|207[4-9]));(?:&\#(?:0*(?:178|179|185|8304|830[89]|831[0-3])|[xX]0*(?:2070|[bB][239]|207[4-9]));)*|<su[bp]>[1-9][0-9]*<\/su[bp]>)|[1-9][0-9]*|[ₙₘₓ]|<su[bp]>[nmx]<\/su[bp]>   # ← NEW: repeat index also allowed on components
      )*
    )+
    (?:[+-](?![A-Za-z0-9]))?                                               #   optional charge on the component
  )*
)
(?![^\s<])                                                                # RIGHT boundary: following char is whitespace, '<' (start of a tag), or end-of-string

Unit Tests

There are no unit tests, click to add some.