polytypo 1.17.4 → 1.17.5
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/lib/polytypo/data/VERSION +1 -1
- data/lib/polytypo/data/fixtures/bg.json +1 -1
- data/lib/polytypo/data/fixtures/cs.json +1 -1
- data/lib/polytypo/data/fixtures/de-CH.json +1 -1
- data/lib/polytypo/data/fixtures/de-DE.json +1 -1
- data/lib/polytypo/data/fixtures/el.json +1 -1
- data/lib/polytypo/data/fixtures/en-GB.json +64 -1
- data/lib/polytypo/data/fixtures/en-US.json +64 -1
- data/lib/polytypo/data/fixtures/es.json +1 -1
- data/lib/polytypo/data/fixtures/fi.json +1 -1
- data/lib/polytypo/data/fixtures/fr-CA.json +1 -1
- data/lib/polytypo/data/fixtures/fr.json +1 -1
- data/lib/polytypo/data/fixtures/hr.json +1 -1
- data/lib/polytypo/data/fixtures/hu.json +1 -1
- data/lib/polytypo/data/fixtures/it.json +1 -1
- data/lib/polytypo/data/fixtures/locale-resolution.json +1 -1
- data/lib/polytypo/data/fixtures/lt.json +1 -1
- data/lib/polytypo/data/fixtures/nl.json +1 -1
- data/lib/polytypo/data/fixtures/pl.json +1 -1
- data/lib/polytypo/data/fixtures/pt-BR.json +1 -1
- data/lib/polytypo/data/fixtures/pt-PT.json +1 -1
- data/lib/polytypo/data/fixtures/ro.json +1 -1
- data/lib/polytypo/data/fixtures/ru.json +1 -1
- data/lib/polytypo/data/fixtures/sk.json +1 -1
- data/lib/polytypo/data/fixtures/sl.json +1 -1
- data/lib/polytypo/data/fixtures/sr.json +1 -1
- data/lib/polytypo/data/fixtures/sv.json +1 -1
- data/lib/polytypo/data/fixtures/tr.json +1 -1
- data/lib/polytypo/data/fixtures/uk.json +1 -1
- data/lib/polytypo/data/locales/registry.json +1 -1
- data/lib/polytypo/data/rules/modes.md +49 -4
- data/lib/polytypo/data/rules/order.json +1 -1
- data/lib/polytypo/modes/spans.rb +53 -2
- data/lib/polytypo/version.rb +1 -1
- metadata +1 -1
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: 1ada78393268412e145c286f3a52ca2b1529637472b413e2ad96533763ca4feb
|
|
4
|
+
data.tar.gz: ca2eab4d74188ca6f57e71a573722e8b51454621e2d648fcd4df6417d4a2701b
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 72da361c3230aa58f7c06cb595685d53e6a949858eae318ae207fc5819924c598d99246393afdc917c3abe1431a6cbe88d548133e758668ef3d23ad382bfc6f2
|
|
7
|
+
data.tar.gz: b7b2f9646e6f326010604a6097409d34e3c80b3ffabf4a28582946832c4705da343437bd538188363b32b97393889f6ed204d06de47cf315623e80765e61830e
|
data/lib/polytypo/data/VERSION
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
1.17.
|
|
1
|
+
1.17.5
|
|
@@ -1,5 +1,5 @@
|
|
|
1
1
|
{
|
|
2
|
-
"spec": "1.17.
|
|
2
|
+
"spec": "1.17.5",
|
|
3
3
|
"locale": "en-GB",
|
|
4
4
|
"cases": [
|
|
5
5
|
{
|
|
@@ -1871,6 +1871,69 @@
|
|
|
1871
1871
|
"in": "Note[^a's] and it's.\n\n[^a's]: it's text\n",
|
|
1872
1872
|
"out": "Note[^a's] and it’s.\n\n[^a's]: it’s text\n",
|
|
1873
1873
|
"note": "modes.md §3.7.3: footnote labels are skipped in both dialects."
|
|
1874
|
+
},
|
|
1875
|
+
{
|
|
1876
|
+
"id": "en-gb-markdown-commonmark-bracket-task-marker-space-after-open",
|
|
1877
|
+
"rule": "spaces",
|
|
1878
|
+
"mode": "markdown",
|
|
1879
|
+
"dialect": "commonmark",
|
|
1880
|
+
"in": "- [ x] item and a , b.\n",
|
|
1881
|
+
"out": "- [ x] item and a, b.\n",
|
|
1882
|
+
"note": "modes.md §3.4 bracket-shape rule: deleting the space after the opening bracket would make a task-list marker for a parser with GFM extensions; that edit is discarded. The space in a , b is deleted as usual."
|
|
1883
|
+
},
|
|
1884
|
+
{
|
|
1885
|
+
"id": "en-gb-markdown-commonmark-bracket-task-marker-space-before-close",
|
|
1886
|
+
"rule": "spaces",
|
|
1887
|
+
"mode": "markdown",
|
|
1888
|
+
"dialect": "commonmark",
|
|
1889
|
+
"in": "- [x ] item and a , b.\n",
|
|
1890
|
+
"out": "- [x ] item and a, b.\n",
|
|
1891
|
+
"note": "modes.md §3.4 bracket-shape rule: the same marker from the other side: deleting the space before the closing bracket is discarded."
|
|
1892
|
+
},
|
|
1893
|
+
{
|
|
1894
|
+
"id": "en-gb-markdown-commonmark-bracket-task-marker-both-spaces",
|
|
1895
|
+
"rule": "spaces",
|
|
1896
|
+
"mode": "markdown",
|
|
1897
|
+
"dialect": "commonmark",
|
|
1898
|
+
"in": "- [ X ] item , y\n",
|
|
1899
|
+
"out": "- [ X ] item, y\n",
|
|
1900
|
+
"note": "modes.md §3.4 bracket-shape rule: the skip over spaces keeps both deletions from being evaluated as safe on their own; the two are discarded together, and the space before the comma is deleted."
|
|
1901
|
+
},
|
|
1902
|
+
{
|
|
1903
|
+
"id": "en-gb-markdown-commonmark-bracket-footnote-space-after-open",
|
|
1904
|
+
"rule": "spaces",
|
|
1905
|
+
"mode": "markdown",
|
|
1906
|
+
"dialect": "commonmark",
|
|
1907
|
+
"in": "a [ ^1] b , c\n\n[^1]: n\n",
|
|
1908
|
+
"out": "a [ ^1] b, c\n\n[^1]: n\n",
|
|
1909
|
+
"note": "modes.md §3.4 bracket-shape rule: deleting the space after the opening bracket would make a footnote reference; that edit is discarded."
|
|
1910
|
+
},
|
|
1911
|
+
{
|
|
1912
|
+
"id": "en-gb-markdown-commonmark-bracket-footnote-space-before-close",
|
|
1913
|
+
"rule": "spaces",
|
|
1914
|
+
"mode": "markdown",
|
|
1915
|
+
"dialect": "commonmark",
|
|
1916
|
+
"in": "a [^1 ] b , c\n\n[^1]: n\n",
|
|
1917
|
+
"out": "a [^1 ] b, c\n\n[^1]: n\n",
|
|
1918
|
+
"note": "modes.md §3.4 bracket-shape rule: deleting the space before the closing bracket would make a footnote reference; that edit is discarded."
|
|
1919
|
+
},
|
|
1920
|
+
{
|
|
1921
|
+
"id": "en-gb-markdown-commonmark-bracket-ordinary-text-still-tightened",
|
|
1922
|
+
"rule": "spaces",
|
|
1923
|
+
"mode": "markdown",
|
|
1924
|
+
"dialect": "commonmark",
|
|
1925
|
+
"in": "a [ word ] b and ( x ) c.\n",
|
|
1926
|
+
"out": "a [word] b and (x) c.\n",
|
|
1927
|
+
"note": "modes.md §3.4 bracket-shape rule: the rule tests a shape, not a bracket: a bracket around an ordinary word, and a parenthesis around a single letter, are tightened as before."
|
|
1928
|
+
},
|
|
1929
|
+
{
|
|
1930
|
+
"id": "en-gb-markdown-commonmark-bracket-task-marker-two-spaces",
|
|
1931
|
+
"rule": "spaces",
|
|
1932
|
+
"mode": "markdown",
|
|
1933
|
+
"dialect": "commonmark",
|
|
1934
|
+
"in": "- [ ] item and a , b.\n",
|
|
1935
|
+
"out": "- [ ] item and a, b.\n",
|
|
1936
|
+
"note": "modes.md §3.4 bracket-shape rule: collapsing two spaces between the brackets to one would make an unchecked task-list marker out of literal text; that edit is discarded."
|
|
1874
1937
|
}
|
|
1875
1938
|
]
|
|
1876
1939
|
}
|
|
@@ -1,5 +1,5 @@
|
|
|
1
1
|
{
|
|
2
|
-
"spec": "1.17.
|
|
2
|
+
"spec": "1.17.5",
|
|
3
3
|
"locale": "en-US",
|
|
4
4
|
"cases": [
|
|
5
5
|
{
|
|
@@ -4284,6 +4284,69 @@
|
|
|
4284
4284
|
"in": "Read https://example.com/a -- fine and b -- c\n",
|
|
4285
4285
|
"out": "Read https://example.com/a -- fine and b—c\n",
|
|
4286
4286
|
"note": "modes.md §3.7.3 terminator guard: collapsing the spaced dash would delete the space that ends the URL, and the parser would read the dash and the next word as part of it; that edit is discarded. The same collapse away from a URL (b -- c) is applied."
|
|
4287
|
+
},
|
|
4288
|
+
{
|
|
4289
|
+
"id": "en-us-markdown-commonmark-bracket-task-marker-space-after-open",
|
|
4290
|
+
"rule": "spaces",
|
|
4291
|
+
"mode": "markdown",
|
|
4292
|
+
"dialect": "commonmark",
|
|
4293
|
+
"in": "- [ x] item and a , b.\n",
|
|
4294
|
+
"out": "- [ x] item and a, b.\n",
|
|
4295
|
+
"note": "modes.md §3.4 bracket-shape rule: deleting the space after the opening bracket would make a task-list marker for a parser with GFM extensions; that edit is discarded. The space in a , b is deleted as usual."
|
|
4296
|
+
},
|
|
4297
|
+
{
|
|
4298
|
+
"id": "en-us-markdown-commonmark-bracket-task-marker-space-before-close",
|
|
4299
|
+
"rule": "spaces",
|
|
4300
|
+
"mode": "markdown",
|
|
4301
|
+
"dialect": "commonmark",
|
|
4302
|
+
"in": "- [x ] item and a , b.\n",
|
|
4303
|
+
"out": "- [x ] item and a, b.\n",
|
|
4304
|
+
"note": "modes.md §3.4 bracket-shape rule: the same marker from the other side: deleting the space before the closing bracket is discarded."
|
|
4305
|
+
},
|
|
4306
|
+
{
|
|
4307
|
+
"id": "en-us-markdown-commonmark-bracket-task-marker-both-spaces",
|
|
4308
|
+
"rule": "spaces",
|
|
4309
|
+
"mode": "markdown",
|
|
4310
|
+
"dialect": "commonmark",
|
|
4311
|
+
"in": "- [ X ] item , y\n",
|
|
4312
|
+
"out": "- [ X ] item, y\n",
|
|
4313
|
+
"note": "modes.md §3.4 bracket-shape rule: the skip over spaces keeps both deletions from being evaluated as safe on their own; the two are discarded together, and the space before the comma is deleted."
|
|
4314
|
+
},
|
|
4315
|
+
{
|
|
4316
|
+
"id": "en-us-markdown-commonmark-bracket-footnote-space-after-open",
|
|
4317
|
+
"rule": "spaces",
|
|
4318
|
+
"mode": "markdown",
|
|
4319
|
+
"dialect": "commonmark",
|
|
4320
|
+
"in": "a [ ^1] b , c\n\n[^1]: n\n",
|
|
4321
|
+
"out": "a [ ^1] b, c\n\n[^1]: n\n",
|
|
4322
|
+
"note": "modes.md §3.4 bracket-shape rule: deleting the space after the opening bracket would make a footnote reference; that edit is discarded."
|
|
4323
|
+
},
|
|
4324
|
+
{
|
|
4325
|
+
"id": "en-us-markdown-commonmark-bracket-footnote-space-before-close",
|
|
4326
|
+
"rule": "spaces",
|
|
4327
|
+
"mode": "markdown",
|
|
4328
|
+
"dialect": "commonmark",
|
|
4329
|
+
"in": "a [^1 ] b , c\n\n[^1]: n\n",
|
|
4330
|
+
"out": "a [^1 ] b, c\n\n[^1]: n\n",
|
|
4331
|
+
"note": "modes.md §3.4 bracket-shape rule: deleting the space before the closing bracket would make a footnote reference; that edit is discarded."
|
|
4332
|
+
},
|
|
4333
|
+
{
|
|
4334
|
+
"id": "en-us-markdown-commonmark-bracket-ordinary-text-still-tightened",
|
|
4335
|
+
"rule": "spaces",
|
|
4336
|
+
"mode": "markdown",
|
|
4337
|
+
"dialect": "commonmark",
|
|
4338
|
+
"in": "a [ word ] b and ( x ) c.\n",
|
|
4339
|
+
"out": "a [word] b and (x) c.\n",
|
|
4340
|
+
"note": "modes.md §3.4 bracket-shape rule: the rule tests a shape, not a bracket: a bracket around an ordinary word, and a parenthesis around a single letter, are tightened as before."
|
|
4341
|
+
},
|
|
4342
|
+
{
|
|
4343
|
+
"id": "en-us-markdown-commonmark-bracket-task-marker-two-spaces",
|
|
4344
|
+
"rule": "spaces",
|
|
4345
|
+
"mode": "markdown",
|
|
4346
|
+
"dialect": "commonmark",
|
|
4347
|
+
"in": "- [ ] item and a , b.\n",
|
|
4348
|
+
"out": "- [ ] item and a, b.\n",
|
|
4349
|
+
"note": "modes.md §3.4 bracket-shape rule: collapsing two spaces between the brackets to one would make an unchecked task-list marker out of literal text; that edit is discarded."
|
|
4287
4350
|
}
|
|
4288
4351
|
]
|
|
4289
4352
|
}
|
|
@@ -1,5 +1,5 @@
|
|
|
1
1
|
{
|
|
2
|
-
"spec": "1.17.
|
|
2
|
+
"spec": "1.17.5",
|
|
3
3
|
"$comment": "Locale resolution input. Algorithm is specified in spec/rules/locale-resolution.md and is identical in every runtime — never delegate it to a platform locale-negotiation library. \"spec\" here must track spec/VERSION exactly — it is not itself the global version source; scripts/validate-spec.mjs enforces the match.",
|
|
4
4
|
"locales": [
|
|
5
5
|
"en-US",
|
|
@@ -19,7 +19,7 @@ step 9 with §3.8.5's content-line condition, added in 1.10.0, §3.8.4's carriag
|
|
|
19
19
|
continuation line with the bails of §3.8.6 that use it, added in 1.11.0, and §3.8.4's node lines,
|
|
20
20
|
declined keys, node markers and own-line region, with the changes to steps 1, 3, 4, 5, 6, 7 and 8
|
|
21
21
|
that go with them, added in 1.11.0, and §3.9 with the amendments it carries to §3.1, §3.5 step 2,
|
|
22
|
-
§4, §5, §7 item 7 and §8, added in 1.13.0, and §3.4's delimiter-neighbour rule with §3.7.3's autolink-literal tail, terminator guard and reference labels, and the amendments they carry to §3.4's deletion paragraph, §5 items 2 and 3, its fifth-part note and testing obligation, and §6, added in 1.17.2, and §3.4's escape-and-reference rule with the amendments it carries to §3.6, §3.7.3, §5 item 2's fifth-part note, §5 item 3 and the testing obligation, added in 1.17.4).
|
|
22
|
+
§4, §5, §7 item 7 and §8, added in 1.13.0, and §3.4's delimiter-neighbour rule with §3.7.3's autolink-literal tail, terminator guard and reference labels, and the amendments they carry to §3.4's deletion paragraph, §5 items 2 and 3, its fifth-part note and testing obligation, and §6, added in 1.17.2, and §3.4's escape-and-reference rule with the amendments it carries to §3.6, §3.7.3, §5 item 2's fifth-part note, §5 item 3 and the testing obligation, added in 1.17.4, and §3.4's bracket-shape rule, added in 1.17.5).
|
|
23
23
|
|
|
24
24
|
---
|
|
25
25
|
|
|
@@ -460,6 +460,48 @@ in five runtimes; it is therefore stricter than a parser: `AT&T ;x` (en-US) keep
|
|
|
460
460
|
though `&T;` is no reference. **That is the accepted cost**, together with a literal backslash
|
|
461
461
|
before a space and punctuation, which is rare in prose.
|
|
462
462
|
|
|
463
|
+
**The bracket-shape rule (`markdown` only, spec 1.17.5).** Three more shapes are made by joining
|
|
464
|
+
what a deleted space kept apart, and both dialects, which enable GFM (§3.7), read the joined form
|
|
465
|
+
as structure where the spaced form is literal text: a task-list marker, `[x]`, `[X]` or `[ ]` at the
|
|
466
|
+
start of a list item, and a footnote reference or definition, `[^label]`. Measured, with the
|
|
467
|
+
pre-1.17.5 behaviour:
|
|
468
|
+
|
|
469
|
+
| Mode | Input | Locale | Output before 1.17.5 | What changed |
|
|
470
|
+
| ---------- | ------------- | ------- | -------------------- | --------------------------------------------------------------- |
|
|
471
|
+
| `markdown` | `- [ x] item` | `en-GB` | `- [x] item` | the text of a literal bracket becomes a checked task-list item |
|
|
472
|
+
| `markdown` | `- [x ] item` | `en-GB` | `- [x] item` | the same item, from the other side |
|
|
473
|
+
| `markdown` | `- [ ] item` | `en-GB` | `- [ ] item` | two spaces become one, and the text becomes an unchecked item |
|
|
474
|
+
| `markdown` | `a [ ^1] b` | `en-GB` | `a [^1] b` | a footnote reference, if a definition `[^1]: …` exists |
|
|
475
|
+
|
|
476
|
+
Write `SKIP(i, +1)` for the smallest index greater than `i` whose code point is not U+0020, and
|
|
477
|
+
`SKIP(i, −1)` for the largest index smaller than `i` whose code point is not U+0020; a marker, or a
|
|
478
|
+
position outside the array, belongs to no class below. With `p`, `q` and `r` as before, and
|
|
479
|
+
`a = SKIP(p, −1)` and `b = SKIP(q, +1)` (so that `a` and `b` are the nearest code points that are
|
|
480
|
+
not U+0020 on either side of the deleted run):
|
|
481
|
+
|
|
482
|
+
> **A deletion (`r = 0`) is discarded if either of these holds, evaluated on the array the rule read.**
|
|
483
|
+
>
|
|
484
|
+
> 1. `cp[a]` is U+005B, and either `cp[b]` is U+005E or U+005D, or `cp[b]` is U+0078 or U+0058 and
|
|
485
|
+
> `cp[SKIP(b, +1)]` is U+005D.
|
|
486
|
+
> 2. `cp[b]` is U+005D, and either `cp[a]` is U+0078 or U+0058 and `cp[SKIP(a, −1)]` is U+005B, or
|
|
487
|
+
> the **label run** is at least two code points long, its leftmost code point is U+005E, and
|
|
488
|
+
> the nearest code point left of the run that is not U+0020 is U+005B. The label run is the
|
|
489
|
+
> longest run ending at `a` of code points that are none of U+0020, U+005B, U+005D, and not a
|
|
490
|
+
> marker.
|
|
491
|
+
>
|
|
492
|
+
> **An edit whose replacement is the single code point U+0020 is discarded** when `cp[a]` is U+005B
|
|
493
|
+
> and `cp[b]` is U+005D: `spaces` collapses a run of spaces to one rather than deleting it, and
|
|
494
|
+
> `[ ]` becoming `[ ]` is the unchecked marker.
|
|
495
|
+
|
|
496
|
+
The skip over U+0020 is what keeps `[ x ]` whole: the two deletions are evaluated on the same
|
|
497
|
+
array, and each sees the other space as absent. The rule reads no parser state and no dialect, so it
|
|
498
|
+
fires identically on every run (§5 item 3), and it tests deletions only because `spaces` is the
|
|
499
|
+
only rule that deletes. It is a test of _shape_, not of whether a footnote definition or a list item
|
|
500
|
+
exists. **The accepted costs:** `a [ x] b` in the middle of a sentence keeps its space although no
|
|
501
|
+
list item is there, as do `[ x ]`, `[ ^1]` and `[ ]` in prose; `a [ ^ ] b` loses its second space only,
|
|
502
|
+
because a label run needs two code points; and a footnote label that contains an inline element
|
|
503
|
+
(`[^*a*b ]`) is missed, because a marker ends the label run.
|
|
504
|
+
|
|
463
505
|
### 3.5 Applying the result
|
|
464
506
|
|
|
465
507
|
1. Extract the span sequence from the source. Record each span's **source offsets**.
|
|
@@ -1830,7 +1872,8 @@ they do not carry it over for free either. Write `M` for the whole mode transfor
|
|
|
1830
1872
|
can also create a backslash escape or a character reference from literal text (`foo\ ,`
|
|
1831
1873
|
becomes `foo\,` and `AT& ;T` becomes `AT&T`): the parse tree keeps its shape and its
|
|
1832
1874
|
text changes, which a shape comparison does not see. That class was not closed by 1.17.2 and
|
|
1833
|
-
is closed by §3.4's escape-and-reference rule (spec 1.17.4).
|
|
1875
|
+
is closed by §3.4's escape-and-reference rule (spec 1.17.4). A deletion can also join
|
|
1876
|
+
`[` and `x` or `^` into a shape a GFM parser reads as structure, and §3.4's bracket-shape rule (spec 1.17.5) closes that.
|
|
1834
1877
|
- **Deletion** is only ever of a U+0020, and structural whitespace borders a `BREAK` — a real
|
|
1835
1878
|
one in `text` mode, a −2 marker in the others — where `spaces` refuses to act
|
|
1836
1879
|
(`spaces.md` §3.2 step 4). A deletion therefore cannot turn `- item` into `-item`, collapse
|
|
@@ -1904,7 +1947,7 @@ they do not carry it over for free either. Write `M` for the whole mode transfor
|
|
|
1904
1947
|
to a side. The discard is a pure function of the array and its span table (since 1.17.2: of
|
|
1905
1948
|
`(p, q, r, s₀, s₁)`, of the code points beside the edit for §3.4's delimiter-neighbour rule, of
|
|
1906
1949
|
the same code points and the run of letters and digits around the join for §3.4's
|
|
1907
|
-
escape-and-reference rule (since 1.17.4), and of whether the span begins at the end of a web address for §3.7.3's terminator guard), so it
|
|
1950
|
+
escape-and-reference rule (since 1.17.4), of the code points around the join for §3.4's bracket-shape rule (since 1.17.5), and of whether the span begins at the end of a web address for §3.7.3's terminator guard), so it
|
|
1908
1951
|
happens identically on every run: a rule that was declined at an edge once is declined there always, and the output
|
|
1909
1952
|
contains nothing for the next run to reconsider.
|
|
1910
1953
|
|
|
@@ -1944,7 +1987,9 @@ must include an address with `_`, `*` and `~` inside it after `"`, `'` and `(`,
|
|
|
1944
1987
|
directly after a digit. **Since 1.17.4 the comparison also covers the text values of the tree**, not only its shape: the
|
|
1945
1988
|
defect behind the escape-and-reference rule left the shape unchanged. The alphabet must then include
|
|
1946
1989
|
U+005C, `&`, `#`, `;`, `,` and a letter, and the templates a literal backslash and a literal
|
|
1947
|
-
`&name` each followed by a space and by each punctuation mark. **
|
|
1990
|
+
`&name` each followed by a space and by each punctuation mark. **Since 1.17.5 the alphabet also includes** U+005B, U+005D, U+005E,
|
|
1991
|
+
`x` and `X`, and the templates a list item and a footnote definition each holding every
|
|
1992
|
+
combination of those with zero, one or two spaces inside the brackets. **Two classes are carved out because they are accepted, not because
|
|
1948
1993
|
they are untested**: a text that matched no definition in the input and matches one in the output
|
|
1949
1994
|
(§3.7.3's accepted case), and an e-mail address, whose end only the parser decides.
|
|
1950
1995
|
|
|
@@ -1,5 +1,5 @@
|
|
|
1
1
|
{
|
|
2
|
-
"spec": "1.17.
|
|
2
|
+
"spec": "1.17.5",
|
|
3
3
|
"$comment": "Single source of truth for pipeline order. Rules run in ascending `order`. Disabling a rule removes it from the sequence and never reorders the rest. Rule ids are public API (see docs/ARCHITECTURE.md section 5). \"spec\" here must track spec/VERSION exactly — it is not itself the global version source; scripts/validate-spec.mjs enforces the match.",
|
|
4
4
|
"rules": [
|
|
5
5
|
{
|
data/lib/polytypo/modes/spans.rb
CHANGED
|
@@ -189,6 +189,56 @@ module Polytypo
|
|
|
189
189
|
end
|
|
190
190
|
private_class_method :creates_reference?
|
|
191
191
|
|
|
192
|
+
# modes.md 3.4 bracket-shape rule (markdown only, spec 1.17.5): a deletion that joins a
|
|
193
|
+
# task-list marker or a footnote label. A marker or a position outside the array is in no
|
|
194
|
+
# class. Ruby indexes negatively, so every read goes through +code_point_at+.
|
|
195
|
+
def self.code_point_at(cp, i)
|
|
196
|
+
i.negative? ? nil : cp[i]
|
|
197
|
+
end
|
|
198
|
+
private_class_method :code_point_at
|
|
199
|
+
|
|
200
|
+
def self.skip_spaces(cp, i, step)
|
|
201
|
+
k = i
|
|
202
|
+
k += step while k >= 0 && k < cp.length && cp[k] == SPACE
|
|
203
|
+
k
|
|
204
|
+
end
|
|
205
|
+
private_class_method :skip_spaces
|
|
206
|
+
|
|
207
|
+
def self.x_letter?(c)
|
|
208
|
+
c == 0x78 || c == 0x58
|
|
209
|
+
end
|
|
210
|
+
private_class_method :x_letter?
|
|
211
|
+
|
|
212
|
+
def self.label_character?(c)
|
|
213
|
+
!c.nil? && c >= 0 && c != SPACE && c != 0x5B && c != 0x5D
|
|
214
|
+
end
|
|
215
|
+
private_class_method :label_character?
|
|
216
|
+
|
|
217
|
+
def self.creates_bracket_shape?(cp, p, q, replacement)
|
|
218
|
+
is_collapse = replacement.length == 1 && replacement[0] == SPACE
|
|
219
|
+
return false unless replacement.empty? || is_collapse
|
|
220
|
+
|
|
221
|
+
a = skip_spaces(cp, p - 1, -1)
|
|
222
|
+
b = skip_spaces(cp, q + 1, 1)
|
|
223
|
+
at_a = code_point_at(cp, a)
|
|
224
|
+
at_b = code_point_at(cp, b)
|
|
225
|
+
return at_a == 0x5B && at_b == 0x5D if is_collapse
|
|
226
|
+
return true if at_a == 0x5B && at_b == 0x5D
|
|
227
|
+
if at_a == 0x5B
|
|
228
|
+
return true if at_b == 0x5E
|
|
229
|
+
|
|
230
|
+
return x_letter?(at_b) && code_point_at(cp, skip_spaces(cp, b + 1, 1)) == 0x5D
|
|
231
|
+
end
|
|
232
|
+
return false if at_b != 0x5D
|
|
233
|
+
return true if x_letter?(at_a) && code_point_at(cp, skip_spaces(cp, a - 1, -1)) == 0x5B
|
|
234
|
+
|
|
235
|
+
run = a
|
|
236
|
+
run -= 1 while run >= 0 && label_character?(cp[run])
|
|
237
|
+
first = run + 1
|
|
238
|
+
a - first >= 1 && cp[first] == 0x5E && code_point_at(cp, skip_spaces(cp, run, -1)) == 0x5B
|
|
239
|
+
end
|
|
240
|
+
private_class_method :creates_bracket_shape?
|
|
241
|
+
|
|
192
242
|
# modes.md 3.4, two safety nets, both pure functions of (p, q, r, s0, s1):
|
|
193
243
|
#
|
|
194
244
|
# 1. No edit may contain a marker -- one that does is a bug, discarded rather than
|
|
@@ -202,8 +252,8 @@ module Polytypo
|
|
|
202
252
|
# point any rule emits whose meaning comes from its position rather than from itself
|
|
203
253
|
# (modes.md 5 item 2).
|
|
204
254
|
#
|
|
205
|
-
# +mode+ selects the neighbour rules of modes.md 3.4 -- delimiter-neighbour
|
|
206
|
-
# for markdown, the reference rule for markdown and html -- and +guarded+ holds, per span,
|
|
255
|
+
# +mode+ selects the neighbour rules of modes.md 3.4 -- delimiter-neighbour, the escape rule
|
|
256
|
+
# and the bracket-shape rule for markdown, the reference rule for markdown and html -- and +guarded+ holds, per span,
|
|
207
257
|
# whether its first code point is the terminator of a bare web address (modes.md 3.7.3).
|
|
208
258
|
def self.filter_boundary_edits(cp, edits, ranges, mode: nil, guarded: [])
|
|
209
259
|
edits.select do |edit|
|
|
@@ -216,6 +266,7 @@ module Polytypo
|
|
|
216
266
|
markdown = mode == "markdown"
|
|
217
267
|
next false if markdown && changes_delimiter_flank?(cp, p, q, edit.replacement)
|
|
218
268
|
next false if markdown && creates_escape?(cp, p, q, edit.replacement)
|
|
269
|
+
next false if markdown && creates_bracket_shape?(cp, p, q, edit.replacement)
|
|
219
270
|
next false if (markdown || mode == "html") && creates_reference?(cp, p, q, edit.replacement)
|
|
220
271
|
|
|
221
272
|
span = span_containing(ranges, p)
|
data/lib/polytypo/version.rb
CHANGED