The match across environments, messages not compensated
Not part of the analysis. This is large-subset-environment.md, section 11, with messages compared as before post-0.11: there messages count nowhere, wherever a redirection sends them; here a message that went to standard output (2>&1) or into a file (2>file) counted as a difference, so the shells' own prefixes and wordings (dash: 1: x: not found, bash: line 1: x: command not found, sh: x: not found) made differences, which rulings for messages then had to legalize. The same runs, tools/large-subset-matchreport.pl with -raw (the rule table's rows for messages apply only there). Kept because it shows what the compensation changed: 11 snippets where all three agree only with it (7 messages in a file or a pipe, 4 A5), 5 that are chance without it and A5 with it, and 1 (test 2>&1 > cow) not legalized without it and A5 with it. The runs of post-0.12 (2026-11-17, with IN1), over every environment a snippet ran in, step 3's included, as there. The words and the columns: section 11 there.
Tree 1
40,410 all snippets the parser accepts; each run by dash, bash and ours
│
│ ≠coreutil ¬coreutil rest
├── 38,582 ours = dash = bash, in every environment¹
├── 13 timeout 0 11 2
├── 7 dash = bash; ours ≠ them 2 4 1
└── 1,808 dash ≠ bash, in some or all environments
├── 1,254 ours = dash 14 1,192 48
├── 355 ours = bash 116 234 5
├── 143 ours ≠ dash, ours ≠ bash 8 117 18
└── 56 depends on the environment 1 51 4
├── 39 ours = dash in some, = bash in others 0 36 3
├── 11 ours = dash in some, ≠ both in others 0 10 1
├── 3 ours = bash in some, ≠ both in others 1 2 0
└── 3 ours = dash, = bash, ≠ both, by environment 0 3 0
───────── ───────── ────
141 1,609 78
¹ Their commands were not recorded: nothing in them needs a reason.
Tree 2
40,410 all snippets the parser accepts; each run by dash, bash and ours
│
│ ≠coreutil ¬coreutil rest
├── 38,582 ours = dash = bash, in every environment¹
├── 13 timeout 0 11 2
├── 7 dash = bash; ours ≠ them 2 4 1
│ ├── 3 legalized by rulings² 2 1 0
│ ├── 2 chance 0 2 0
│ └── 2 not legalized (2 already in TODO) 0 1 1
└── 1,808 dash ≠ bash, in some or all environments
├── 1,254 ours = dash 14 1,192 48
│ ├── 1,234 legalized by rulings³ 14 1,174 46
│ ├── 19 ~ with HOME empty: awaiting A12 0 17 2
│ └── 1 chance 0 1 0
├── 355 ours = bash 116 234 5
│ ├── 282 legalized by rulings⁴ 116 166 0
│ ├── 64 ~ with HOME empty: awaiting A12 0 64 0
│ ├── 1 chance 0 0 1
│ └── 8 not legalized 0 4 4
├── 143 ours ≠ dash, ours ≠ bash 8 117 18
│ ├── 100 legalized by rulings⁵ 8 78 14
│ ├── 1 ~ with HOME empty: awaiting A12 0 1 0
│ ├── 39 chance 0 36 3
│ └── 3 not legalized (2 already in TODO) 0 2 1
└── 56 depends on the environment 1 51 4
├── 39 ours = dash in some, = bash in others 0 36 3
│ ├── 5 legalized by rulings⁶ 0 3 2
│ ├── 2 ~ with HOME empty: awaiting A12 0 2 0
│ ├── 2 chance 0 1 1
│ └── 30 not legalized 0 30 0
├── 11 ours = dash in some, ≠ both in others 0 10 1
│ ├── 4 legalized by rulings⁷ 0 4 0
│ ├── 2 ~ with HOME empty: awaiting A12 0 1 1
│ └── 5 chance 0 5 0
├── 3 ours = bash in some, ≠ both in others 1 2 0
│ legalized by rulings⁸
└── 3 ours = dash, = bash, ≠ both, by environment 0 3 0
legalized by rulings⁹
───────── ───────── ────
141 1,609 78
¹ As above. ² Q2 2, A4 1. ³ A5 1,232, rule E 2. ⁴ G16 126, builtin-vs-coreutils 116, B2 31, rule E 10. ⁵ $0 36, B2 35, A4 8, builtin-vs-coreutils 8, messages 8, A5 4, the runner 1. ⁶ A5 4, rule E 3, builtin-vs-coreutils 2. ⁷ messages 3, A5 1, A5 + messages 1. ⁸ $0 2, builtin-vs-coreutils 1. ⁹ $0 3, A5 3.
The 1,815 where ours differs from at least one shell (7 where dash = bash, 1,808 where dash ≠ bash): legalized by rulings 1,634, awaiting A12 88, chance 50, not legalized 43 (4 marked open, as there).
| Ruling | Snippets | |---|---| | A5 | 1,303 | | G16 | 214 | | builtin-vs-coreutils | 128 | | B2 | 67 | | $0 | 41 | | rule E | 15 | | messages | 11 | | A4 | 9 | | Q2 | 2 | | A5 + messages | 1 | | the runner | 1 |