dirsql 0.4.98__tar.gz → 0.4.99__tar.gz

{dirsql-0.4.98 → dirsql-0.4.99}/Cargo.lock

@@ -573,7 +573,7 @@ dependencies = [
 
 [[package]]
 name = "dirsql"
-version = "0.4.98"
+version = "0.4.99"
 dependencies = [
  "assert_cmd",
  "axum",
@@ -625,7 +625,7 @@ dependencies = [
 
 [[package]]
 name = "dirsql-py-ext"
-version = "0.4.98"
+version = "0.4.99"
 dependencies = [
  "dirsql",
  "pyo3",

{dirsql-0.4.98 → dirsql-0.4.99}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dirsql
-Version: 0.4.98
+Version: 0.4.99
 Requires-Dist: bin-shim>=0.1
 Summary: Ephemeral SQL index over a local directory
 Keywords: sql,filesystem,directory,sqlite,index

{dirsql-0.4.98/packages/python → dirsql-0.4.99}/docs/reference/glob.md

@@ -49,6 +49,13 @@ docs/nested/deep.md
 docs/nested/.cache/x.md
 node_modules/pkg/index.js
 real/r.md
+uni/1.md
+uni/a.md
+uni/é.md
+uni/É.md
+uni/中.md
+uni/٣.md
+uni/¡.md
 linkdir -> real
 link.md -> top.md
 broken -> nowhere
@@ -69,13 +76,21 @@ A test runs every row against dirsql as a path-table and, with the leading
 | Pattern | Rows | Divergence |
 | --- | --- | --- |
 | `./*.md` | `1.md`, `2.md`, `3.md`, `4.md`, `[a].md`, `[z.md`, `a.md`, `b.md`, `c.md`, `link.md`, `my file.md`, `top.md`, `{q}.md` |  |
-| `./**/*.md` | `1.md`, `2.md`, `3.md`, `4.md`, `[a].md`, `[z.md`, `a.md`, `b.md`, `c.md`, `docs/api.md`, `docs/guide.md`, `docs/nested/deep.md`, `link.md`, `linkdir/r.md`, `my file.md`, `real/r.md`, `top.md`, `{q}.md` |  |
-| `./**` | `1.md`, `2.md`, `3.md`, `4.md`, `[a].md`, `[z.md`, `a.md`, `b.md`, `c.md`, `docs/api.md`, `docs/guide.md`, `docs/nested/deep.md`, `link.md`, `my file.md`, `node_modules/pkg/index.js`, `real/r.md`, `top.md`, `x.txt`, `{q}.md` |  |
+| `./**/*.md` | `1.md`, `2.md`, `3.md`, `4.md`, `[a].md`, `[z.md`, `a.md`, `b.md`, `c.md`, `docs/api.md`, `docs/guide.md`, `docs/nested/deep.md`, `link.md`, `linkdir/r.md`, `my file.md`, `real/r.md`, `top.md`, `uni/1.md`, `uni/a.md`, `uni/¡.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md`, `{q}.md` |  |
+| `./**` | `1.md`, `2.md`, `3.md`, `4.md`, `[a].md`, `[z.md`, `a.md`, `b.md`, `c.md`, `docs/api.md`, `docs/guide.md`, `docs/nested/deep.md`, `link.md`, `my file.md`, `node_modules/pkg/index.js`, `real/r.md`, `top.md`, `uni/1.md`, `uni/a.md`, `uni/¡.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md`, `x.txt`, `{q}.md` |  |
 | `./?.md` | `1.md`, `2.md`, `3.md`, `4.md`, `a.md`, `b.md`, `c.md` |  |
 | `./[ab].md` | `a.md`, `b.md` |  |
 | `./[!a].md` | `1.md`, `2.md`, `3.md`, `4.md`, `b.md`, `c.md` |  |
 | `./\[a\].md` | `[a].md` |  |
 | `./[z.md` | `[z.md` |  |
+| `./uni/[[:alpha:]].md` | `uni/a.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md` |  |
+| `./uni/[[:alnum:]].md` | `uni/1.md`, `uni/a.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md` |  |
+| `./uni/[[:lower:]].md` | `uni/a.md`, `uni/é.md` |  |
+| `./uni/[[:upper:]].md` | `uni/É.md` |  |
+| `./uni/[[:digit:]].md` | `uni/1.md` |  |
+| `./uni/[[:punct:]].md` | `uni/¡.md` |  |
+| `./uni/[[:word:]].md` | `uni/1.md`, `uni/a.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md` |  |
+| `./uni/[![:alpha:]].md` | `uni/1.md`, `uni/¡.md` |  |
 | `./[d/x]ocs/api.md` | none |  |
 | `./docs[/]api.md` | none |  |
 | `./[!/]*/api.md` | none |  |
@@ -106,9 +121,11 @@ A test runs every row against dirsql as a path-table and, with the leading
 | `./docs/*/../api.md` | `docs/nested/../api.md` |  |
 | `./docs` | `docs/api.md`, `docs/guide.md` | directory-name |
 | `./docs/` | `docs/api.md`, `docs/guide.md` | directory-name |
-| `./*/` | `docs/api.md`, `docs/guide.md`, `linkdir/r.md`, `real/r.md` | directory-name |
+| `./*/` | `docs/api.md`, `docs/guide.md`, `linkdir/r.md`, `real/r.md`, `uni/1.md`, `uni/a.md`, `uni/¡.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md` | directory-name |
 | `./d*` | none | files-only |
 | `./*.log` | none | gitignore |
+| `./ignored.log` | `ignored.log` |  |
+| `./{ignored,x}.log` | `ignored.log` |  |
 | `./**/*.js` | `node_modules/pkg/index.js` |  |
 | `./node_modules/*/index.js` | `node_modules/pkg/index.js` |  |
 | `./docs/**/*.md` | `docs/api.md`, `docs/guide.md`, `docs/nested/deep.md` |  |
@@ -121,7 +138,7 @@ A test runs every row against dirsql as a path-table and, with the leading
 
 | Tag | Rule | Why |
 | --- | --- | --- |
-| `gitignore` | Inside a git repo, `.gitignore` files hide matching files and directories. `--no-ignore` turns it off. | A query over a repo should not drown in build output. |
+| `gitignore` | Inside a git repo, `.gitignore` files hide files and directories a wildcard matches, never one the pattern spells out. `--no-ignore` turns it off. | A query over a repo should not drown in build output. |
 | `directory-name` | A pattern's last component that names a directory lists the files directly inside it, like `ls docs`. A trailing `/` is `*` appended. | A table is rows of files, and `ls` is the shell habit. |
 | `files-only` | A pattern that matches a directory but no file returns nothing. | Every row is a file. |
 | `absolute` | `/`, `~/` and `../` patterns report absolute `path` values, with `..` folded out textually. | The scanned root is derived from the pattern, so there is no relative base. |
@@ -144,6 +161,9 @@ stderr; the exit code is unchanged. Path-tables never warn.
   `{a..c}` expand, as do zero-padded
   `{01..10}` and stepped `{1..10..2}` ranges. Groups nest.
 - `\[` is a literal bracket.
+- POSIX classes (`[[:alpha:]]`) match Unicode characters, as bash does in a UTF-8
+  locale: `[[:alpha:]]` matches `é` and `中`. `[[:digit:]]`, `[[:xdigit:]]` and
+  `[[:ascii:]]` stay ASCII.
 - A dot-named file or directory is listed only where a component spells the
   dot at that depth: `./**/.*` finds `docs/.draft.md` but does not enter
   `.config/`.

{dirsql-0.4.98/packages/rust → dirsql-0.4.99}/docs/reference/path-tables.md

@@ -262,6 +262,9 @@ directory, deeper files override shallower ones, `!pattern` re-includes, and
 an ignored directory is pruned rather than walked. In a typical repo this
 excludes build output, virtualenvs, and caches with zero ceremony.
 
+A `.gitignore` never hides a path the pattern spells out: `'./debug.log'` lists
+that file even when `*.log` is ignored, while `'./*.log'` does not.
+
 A `.gitignore` is in force only when a directory holding `.git` encloses it.
 Outside a repo none applies, at the scan's start, above it or below it: a
 `~/.claude` whose `.gitignore` is `*` still lists its files when `~/.claude`

{dirsql-0.4.98 → dirsql-0.4.99}/packages/python/Cargo.toml

@@ -4,7 +4,7 @@ name = "dirsql-py-ext"
 # pypi/maturin handler can rewrite it via `write-version` before
 # `maturin build`. `pyproject.toml` declares `dynamic = ["version"]`
 # and maturin reads this field. Mirrors `packages/rust/Cargo.toml`.
-version = "0.4.98"
+version = "0.4.99"
 # abi3 note: the `abi3-py310` pyo3 feature (below) builds ONE stable-ABI
 # `cp310-abi3` wheel per platform that loads on every CPython >= 3.10
 # (matching `pyproject.toml`'s `requires-python = ">=3.10"`), instead of

dirsql-0.4.99/packages/python/changelog.d/2026-10-07-literal-gitignore.md

@@ -0,0 +1,3 @@
+**Changed**
+
+- A pattern with no wildcard, such as ./debug.log, lists the file even when .gitignore ignores it; wildcard matches are still filtered.

dirsql-0.4.99/packages/python/changelog.d/2026-10-07-posix-classes-unicode.md

@@ -0,0 +1,3 @@
+**Changed**
+
+- POSIX classes such as `[[:alpha:]]`, `[[:upper:]]` and `[[:punct:]]` match Unicode characters instead of ASCII only: `./[[:alpha:]].md` now matches `é.md`. `digit`, `xdigit` and `ascii` stay ASCII.

{dirsql-0.4.98 → dirsql-0.4.99/packages/python}/docs/reference/glob.md

@@ -49,6 +49,13 @@ docs/nested/deep.md
 docs/nested/.cache/x.md
 node_modules/pkg/index.js
 real/r.md
+uni/1.md
+uni/a.md
+uni/é.md
+uni/É.md
+uni/中.md
+uni/٣.md
+uni/¡.md
 linkdir -> real
 link.md -> top.md
 broken -> nowhere
@@ -69,13 +76,21 @@ A test runs every row against dirsql as a path-table and, with the leading
 | Pattern | Rows | Divergence |
 | --- | --- | --- |
 | `./*.md` | `1.md`, `2.md`, `3.md`, `4.md`, `[a].md`, `[z.md`, `a.md`, `b.md`, `c.md`, `link.md`, `my file.md`, `top.md`, `{q}.md` |  |
-| `./**/*.md` | `1.md`, `2.md`, `3.md`, `4.md`, `[a].md`, `[z.md`, `a.md`, `b.md`, `c.md`, `docs/api.md`, `docs/guide.md`, `docs/nested/deep.md`, `link.md`, `linkdir/r.md`, `my file.md`, `real/r.md`, `top.md`, `{q}.md` |  |
-| `./**` | `1.md`, `2.md`, `3.md`, `4.md`, `[a].md`, `[z.md`, `a.md`, `b.md`, `c.md`, `docs/api.md`, `docs/guide.md`, `docs/nested/deep.md`, `link.md`, `my file.md`, `node_modules/pkg/index.js`, `real/r.md`, `top.md`, `x.txt`, `{q}.md` |  |
+| `./**/*.md` | `1.md`, `2.md`, `3.md`, `4.md`, `[a].md`, `[z.md`, `a.md`, `b.md`, `c.md`, `docs/api.md`, `docs/guide.md`, `docs/nested/deep.md`, `link.md`, `linkdir/r.md`, `my file.md`, `real/r.md`, `top.md`, `uni/1.md`, `uni/a.md`, `uni/¡.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md`, `{q}.md` |  |
+| `./**` | `1.md`, `2.md`, `3.md`, `4.md`, `[a].md`, `[z.md`, `a.md`, `b.md`, `c.md`, `docs/api.md`, `docs/guide.md`, `docs/nested/deep.md`, `link.md`, `my file.md`, `node_modules/pkg/index.js`, `real/r.md`, `top.md`, `uni/1.md`, `uni/a.md`, `uni/¡.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md`, `x.txt`, `{q}.md` |  |
 | `./?.md` | `1.md`, `2.md`, `3.md`, `4.md`, `a.md`, `b.md`, `c.md` |  |
 | `./[ab].md` | `a.md`, `b.md` |  |
 | `./[!a].md` | `1.md`, `2.md`, `3.md`, `4.md`, `b.md`, `c.md` |  |
 | `./\[a\].md` | `[a].md` |  |
 | `./[z.md` | `[z.md` |  |
+| `./uni/[[:alpha:]].md` | `uni/a.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md` |  |
+| `./uni/[[:alnum:]].md` | `uni/1.md`, `uni/a.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md` |  |
+| `./uni/[[:lower:]].md` | `uni/a.md`, `uni/é.md` |  |
+| `./uni/[[:upper:]].md` | `uni/É.md` |  |
+| `./uni/[[:digit:]].md` | `uni/1.md` |  |
+| `./uni/[[:punct:]].md` | `uni/¡.md` |  |
+| `./uni/[[:word:]].md` | `uni/1.md`, `uni/a.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md` |  |
+| `./uni/[![:alpha:]].md` | `uni/1.md`, `uni/¡.md` |  |
 | `./[d/x]ocs/api.md` | none |  |
 | `./docs[/]api.md` | none |  |
 | `./[!/]*/api.md` | none |  |
@@ -106,9 +121,11 @@ A test runs every row against dirsql as a path-table and, with the leading
 | `./docs/*/../api.md` | `docs/nested/../api.md` |  |
 | `./docs` | `docs/api.md`, `docs/guide.md` | directory-name |
 | `./docs/` | `docs/api.md`, `docs/guide.md` | directory-name |
-| `./*/` | `docs/api.md`, `docs/guide.md`, `linkdir/r.md`, `real/r.md` | directory-name |
+| `./*/` | `docs/api.md`, `docs/guide.md`, `linkdir/r.md`, `real/r.md`, `uni/1.md`, `uni/a.md`, `uni/¡.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md` | directory-name |
 | `./d*` | none | files-only |
 | `./*.log` | none | gitignore |
+| `./ignored.log` | `ignored.log` |  |
+| `./{ignored,x}.log` | `ignored.log` |  |
 | `./**/*.js` | `node_modules/pkg/index.js` |  |
 | `./node_modules/*/index.js` | `node_modules/pkg/index.js` |  |
 | `./docs/**/*.md` | `docs/api.md`, `docs/guide.md`, `docs/nested/deep.md` |  |
@@ -121,7 +138,7 @@ A test runs every row against dirsql as a path-table and, with the leading
 
 | Tag | Rule | Why |
 | --- | --- | --- |
-| `gitignore` | Inside a git repo, `.gitignore` files hide matching files and directories. `--no-ignore` turns it off. | A query over a repo should not drown in build output. |
+| `gitignore` | Inside a git repo, `.gitignore` files hide files and directories a wildcard matches, never one the pattern spells out. `--no-ignore` turns it off. | A query over a repo should not drown in build output. |
 | `directory-name` | A pattern's last component that names a directory lists the files directly inside it, like `ls docs`. A trailing `/` is `*` appended. | A table is rows of files, and `ls` is the shell habit. |
 | `files-only` | A pattern that matches a directory but no file returns nothing. | Every row is a file. |
 | `absolute` | `/`, `~/` and `../` patterns report absolute `path` values, with `..` folded out textually. | The scanned root is derived from the pattern, so there is no relative base. |
@@ -144,6 +161,9 @@ stderr; the exit code is unchanged. Path-tables never warn.
   `{a..c}` expand, as do zero-padded
   `{01..10}` and stepped `{1..10..2}` ranges. Groups nest.
 - `\[` is a literal bracket.
+- POSIX classes (`[[:alpha:]]`) match Unicode characters, as bash does in a UTF-8
+  locale: `[[:alpha:]]` matches `é` and `中`. `[[:digit:]]`, `[[:xdigit:]]` and
+  `[[:ascii:]]` stay ASCII.
 - A dot-named file or directory is listed only where a component spells the
   dot at that depth: `./**/.*` finds `docs/.draft.md` but does not enter
   `.config/`.

{dirsql-0.4.98 → dirsql-0.4.99/packages/python}/docs/reference/path-tables.md

@@ -262,6 +262,9 @@ directory, deeper files override shallower ones, `!pattern` re-includes, and
 an ignored directory is pruned rather than walked. In a typical repo this
 excludes build output, virtualenvs, and caches with zero ceremony.
 
+A `.gitignore` never hides a path the pattern spells out: `'./debug.log'` lists
+that file even when `*.log` is ignored, while `'./*.log'` does not.
+
 A `.gitignore` is in force only when a directory holding `.git` encloses it.
 Outside a repo none applies, at the scan's start, above it or below it: a
 `~/.claude` whose `.gitignore` is `*` still lists its files when `~/.claude`

dirsql-0.4.99/packages/python/e2e-attestations/bot-1421-posix-unicode.json

@@ -0,0 +1,7 @@
+{
+  "command": "uv run python -m pytest tests/e2e/glob_speed_matchers_test.py -x -q",
+  "ran_at": 1791395236,
+  "exit_code": 0,
+  "commit": "b0d6654084056647e26fb061ad055ded511d5a2c",
+  "branch": "bot/1421-posix-unicode"
+}

dirsql-0.4.99/packages/python/e2e-attestations/bot-1422-literal-gitignore.json

@@ -0,0 +1,7 @@
+{
+  "command": "uv run python -m pytest tests/e2e/glob_speed_one_level_test.py -x -q",
+  "ran_at": 1791398612,
+  "exit_code": 0,
+  "commit": "575ebdcb3e11ea49aefcd52ec2bff9c9a4990257",
+  "branch": "bot/1422-literal-gitignore"
+}

dirsql-0.4.99/packages/python/migrations.d/2026-10-07-literal-gitignore.md

@@ -0,0 +1,23 @@
+### Named paths ignore .gitignore
+
+#### Summary
+
+A path the pattern spells out is listed even when a .gitignore ignores it.
+
+#### Required changes
+
+_None._
+
+#### Deprecations removed
+
+_None._
+
+#### Behavior changes without code changes
+
+- A pattern with no wildcard, such as ./debug.log, lists the file even when .gitignore ignores it; wildcard matches are still filtered.
+
+#### Verification
+
+```bash
+dirsql query "SELECT path FROM './debug.log'"
+```

dirsql-0.4.99/packages/python/migrations.d/2026-10-07-posix-classes-unicode.md

@@ -0,0 +1,23 @@
+### POSIX classes match Unicode characters
+
+#### Summary
+
+POSIX classes in globs match Unicode characters, as bash does in a UTF-8 locale.
+
+#### Required changes
+
+_None._
+
+#### Deprecations removed
+
+_None._
+
+#### Behavior changes without code changes
+
+- POSIX classes such as `[[:alpha:]]`, `[[:upper:]]` and `[[:punct:]]` match Unicode characters instead of ASCII only: `./[[:alpha:]].md` now matches `é.md`. `digit`, `xdigit` and `ascii` stay ASCII.
+
+#### Verification
+
+```bash
+dirsql query "SELECT path FROM './[[:alpha:]].md'"
+```

{dirsql-0.4.98 → dirsql-0.4.99}/packages/rust/Cargo.toml

@@ -6,7 +6,7 @@ description = "Ephemeral SQL index over a local directory"
 # version stays for the binding crates (publish = false). The literal
 # intentionally lags the published version: putitoutthere rewrites it to
 # the planned release version at build time and never commits it back.
-version = "0.4.98"
+version = "0.4.99"
 edition.workspace = true
 # Literal `license` (not `license.workspace = true`) because
 # putitoutthere's preflight check reads `[package].license` directly

{dirsql-0.4.98 → dirsql-0.4.99}/packages/rust/docs/reference/glob.md

@@ -49,6 +49,13 @@ docs/nested/deep.md
 docs/nested/.cache/x.md
 node_modules/pkg/index.js
 real/r.md
+uni/1.md
+uni/a.md
+uni/é.md
+uni/É.md
+uni/中.md
+uni/٣.md
+uni/¡.md
 linkdir -> real
 link.md -> top.md
 broken -> nowhere
@@ -69,13 +76,21 @@ A test runs every row against dirsql as a path-table and, with the leading
 | Pattern | Rows | Divergence |
 | --- | --- | --- |
 | `./*.md` | `1.md`, `2.md`, `3.md`, `4.md`, `[a].md`, `[z.md`, `a.md`, `b.md`, `c.md`, `link.md`, `my file.md`, `top.md`, `{q}.md` |  |
-| `./**/*.md` | `1.md`, `2.md`, `3.md`, `4.md`, `[a].md`, `[z.md`, `a.md`, `b.md`, `c.md`, `docs/api.md`, `docs/guide.md`, `docs/nested/deep.md`, `link.md`, `linkdir/r.md`, `my file.md`, `real/r.md`, `top.md`, `{q}.md` |  |
-| `./**` | `1.md`, `2.md`, `3.md`, `4.md`, `[a].md`, `[z.md`, `a.md`, `b.md`, `c.md`, `docs/api.md`, `docs/guide.md`, `docs/nested/deep.md`, `link.md`, `my file.md`, `node_modules/pkg/index.js`, `real/r.md`, `top.md`, `x.txt`, `{q}.md` |  |
+| `./**/*.md` | `1.md`, `2.md`, `3.md`, `4.md`, `[a].md`, `[z.md`, `a.md`, `b.md`, `c.md`, `docs/api.md`, `docs/guide.md`, `docs/nested/deep.md`, `link.md`, `linkdir/r.md`, `my file.md`, `real/r.md`, `top.md`, `uni/1.md`, `uni/a.md`, `uni/¡.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md`, `{q}.md` |  |
+| `./**` | `1.md`, `2.md`, `3.md`, `4.md`, `[a].md`, `[z.md`, `a.md`, `b.md`, `c.md`, `docs/api.md`, `docs/guide.md`, `docs/nested/deep.md`, `link.md`, `my file.md`, `node_modules/pkg/index.js`, `real/r.md`, `top.md`, `uni/1.md`, `uni/a.md`, `uni/¡.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md`, `x.txt`, `{q}.md` |  |
 | `./?.md` | `1.md`, `2.md`, `3.md`, `4.md`, `a.md`, `b.md`, `c.md` |  |
 | `./[ab].md` | `a.md`, `b.md` |  |
 | `./[!a].md` | `1.md`, `2.md`, `3.md`, `4.md`, `b.md`, `c.md` |  |
 | `./\[a\].md` | `[a].md` |  |
 | `./[z.md` | `[z.md` |  |
+| `./uni/[[:alpha:]].md` | `uni/a.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md` |  |
+| `./uni/[[:alnum:]].md` | `uni/1.md`, `uni/a.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md` |  |
+| `./uni/[[:lower:]].md` | `uni/a.md`, `uni/é.md` |  |
+| `./uni/[[:upper:]].md` | `uni/É.md` |  |
+| `./uni/[[:digit:]].md` | `uni/1.md` |  |
+| `./uni/[[:punct:]].md` | `uni/¡.md` |  |
+| `./uni/[[:word:]].md` | `uni/1.md`, `uni/a.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md` |  |
+| `./uni/[![:alpha:]].md` | `uni/1.md`, `uni/¡.md` |  |
 | `./[d/x]ocs/api.md` | none |  |
 | `./docs[/]api.md` | none |  |
 | `./[!/]*/api.md` | none |  |
@@ -106,9 +121,11 @@ A test runs every row against dirsql as a path-table and, with the leading
 | `./docs/*/../api.md` | `docs/nested/../api.md` |  |
 | `./docs` | `docs/api.md`, `docs/guide.md` | directory-name |
 | `./docs/` | `docs/api.md`, `docs/guide.md` | directory-name |
-| `./*/` | `docs/api.md`, `docs/guide.md`, `linkdir/r.md`, `real/r.md` | directory-name |
+| `./*/` | `docs/api.md`, `docs/guide.md`, `linkdir/r.md`, `real/r.md`, `uni/1.md`, `uni/a.md`, `uni/¡.md`, `uni/É.md`, `uni/é.md`, `uni/٣.md`, `uni/中.md` | directory-name |
 | `./d*` | none | files-only |
 | `./*.log` | none | gitignore |
+| `./ignored.log` | `ignored.log` |  |
+| `./{ignored,x}.log` | `ignored.log` |  |
 | `./**/*.js` | `node_modules/pkg/index.js` |  |
 | `./node_modules/*/index.js` | `node_modules/pkg/index.js` |  |
 | `./docs/**/*.md` | `docs/api.md`, `docs/guide.md`, `docs/nested/deep.md` |  |
@@ -121,7 +138,7 @@ A test runs every row against dirsql as a path-table and, with the leading
 
 | Tag | Rule | Why |
 | --- | --- | --- |
-| `gitignore` | Inside a git repo, `.gitignore` files hide matching files and directories. `--no-ignore` turns it off. | A query over a repo should not drown in build output. |
+| `gitignore` | Inside a git repo, `.gitignore` files hide files and directories a wildcard matches, never one the pattern spells out. `--no-ignore` turns it off. | A query over a repo should not drown in build output. |
 | `directory-name` | A pattern's last component that names a directory lists the files directly inside it, like `ls docs`. A trailing `/` is `*` appended. | A table is rows of files, and `ls` is the shell habit. |
 | `files-only` | A pattern that matches a directory but no file returns nothing. | Every row is a file. |
 | `absolute` | `/`, `~/` and `../` patterns report absolute `path` values, with `..` folded out textually. | The scanned root is derived from the pattern, so there is no relative base. |
@@ -144,6 +161,9 @@ stderr; the exit code is unchanged. Path-tables never warn.
   `{a..c}` expand, as do zero-padded
   `{01..10}` and stepped `{1..10..2}` ranges. Groups nest.
 - `\[` is a literal bracket.
+- POSIX classes (`[[:alpha:]]`) match Unicode characters, as bash does in a UTF-8
+  locale: `[[:alpha:]]` matches `é` and `中`. `[[:digit:]]`, `[[:xdigit:]]` and
+  `[[:ascii:]]` stay ASCII.
 - A dot-named file or directory is listed only where a component spells the
   dot at that depth: `./**/.*` finds `docs/.draft.md` but does not enter
   `.config/`.

{dirsql-0.4.98/packages/python → dirsql-0.4.99/packages/rust}/docs/reference/path-tables.md

@@ -262,6 +262,9 @@ directory, deeper files override shallower ones, `!pattern` re-includes, and
 an ignored directory is pruned rather than walked. In a typical repo this
 excludes build output, virtualenvs, and caches with zero ceremony.
 
+A `.gitignore` never hides a path the pattern spells out: `'./debug.log'` lists
+that file even when `*.log` is ignored, while `'./*.log'` does not.
+
 A `.gitignore` is in force only when a directory holding `.git` encloses it.
 Outside a repo none applies, at the scan's start, above it or below it: a
 `~/.claude` whose `.gitignore` is `*` still lists its files when `~/.claude`

{dirsql-0.4.98 → dirsql-0.4.99}/packages/rust/src/matcher.rs

@@ -1,4 +1,4 @@
-use crate::posix_class::expand_posix_classes;
+use crate::posix_class::{class_regex, expand_posix_classes};
 use crate::scanner::{PathGlob, compile_glob, compile_globs};
 use globset::GlobBuilder;
 use regex::{Regex, RegexBuilder};
@@ -63,7 +63,10 @@ fn unicode_regex(byte_regex: &str) -> String {
             [] => break,
         }
     }
-    String::from_utf8_lossy(&out).into_owned()
+    String::from_utf8_lossy(&out)
+        .chars()
+        .map(|c| class_regex(c).unwrap_or_else(|| c.to_string()))
+        .collect()
 }
 
 fn hex_byte(digits: [u8; 2]) -> u8 {
@@ -459,4 +462,67 @@ mod tests {
         assert_eq!(names(&matcher, "notes/7.md"), vec!["n"]);
         assert!(matcher.match_all(Path::new("notes/x.md")).is_empty());
     }
+
+    #[test]
+    fn a_posix_class_matches_non_ascii_characters() {
+        let matcher = TableMatcher::new(
+            &[
+                ("a/[[:alpha:]].md", "alpha"),
+                ("l/[[:lower:]].md", "lower"),
+                ("u/[[:upper:]].md", "upper"),
+                ("p/[[:punct:]].md", "punct"),
+            ],
+            &[],
+        )
+        .unwrap();
+        assert_eq!(names(&matcher, "a/\u{e9}.md"), vec!["alpha"]);
+        assert_eq!(names(&matcher, "a/\u{4e2d}.md"), vec!["alpha"]);
+        assert_eq!(names(&matcher, "l/\u{e9}.md"), vec!["lower"]);
+        assert_eq!(names(&matcher, "u/\u{c9}.md"), vec!["upper"]);
+        assert_eq!(names(&matcher, "p/\u{a1}.md"), vec!["punct"]);
+        assert!(matcher.match_all(Path::new("l/\u{c9}.md")).is_empty());
+    }
+
+    #[test]
+    fn every_posix_class_follows_unicode() {
+        let cases = [
+            ("alnum", "\u{e9}", "\u{b2}"),
+            ("alpha", "\u{663}", "1"),
+            ("blank", "\u{3000}", "\u{a0}"),
+            ("cntrl", "\u{85}", "\u{ad}"),
+            ("graph", "\u{a0}", "\u{3000}"),
+            ("lower", "\u{1c5}", "\u{c9}"),
+            ("print", "\u{3000}", "\u{85}"),
+            ("punct", "\u{301}", "\u{e9}"),
+            ("space", "\u{2003}", "\u{a0}"),
+            ("upper", "\u{1c5}", "\u{e9}"),
+            ("word", "\u{4e2d}", "\u{b2}"),
+            ("digit", "7", "\u{663}"),
+            ("xdigit", "f", "\u{ff46}"),
+            ("ascii", "a", "\u{e9}"),
+        ];
+        for (class, member, outsider) in cases {
+            let matcher =
+                TableMatcher::new(&[(format!("[[:{class}:]]").as_str(), "c")], &[]).unwrap();
+            assert_eq!(names(&matcher, member), vec!["c"], "{class} {member}");
+            assert!(
+                matcher.match_all(Path::new(outsider)).is_empty(),
+                "{class} {outsider}"
+            );
+        }
+    }
+
+    #[test]
+    fn a_posix_class_never_matches_the_separator() {
+        let matcher = TableMatcher::new(&[("a[[:punct:]]b", "c")], &[]).unwrap();
+        assert_eq!(names(&matcher, "a.b"), vec!["c"]);
+        assert!(matcher.match_all(Path::new("a/b")).is_empty());
+    }
+
+    #[test]
+    fn a_negated_posix_class_matches_the_other_unicode_characters() {
+        let matcher = TableMatcher::new(&[("[![:alpha:]].md", "c")], &[]).unwrap();
+        assert_eq!(names(&matcher, "\u{a1}.md"), vec!["c"]);
+        assert!(matcher.match_all(Path::new("\u{e9}.md")).is_empty());
+    }
 }

{dirsql-0.4.98 → dirsql-0.4.99}/packages/rust/src/posix_class.rs

@@ -60,7 +60,7 @@ fn bracket(chars: &[char], start: usize) -> Option<(usize, Option<String>)> {
             break;
         }
         if let Some((name, next)) = class_at(chars, j) {
-            ranges.extend_from_slice(class_ranges(&name));
+            ranges.extend(class_ranges(&name));
             has_class = true;
             j = next;
             continue;
@@ -111,24 +111,58 @@ fn class_at(chars: &[char], j: usize) -> Option<(String, usize)> {
     }
 }
 
-/// The ASCII members of a class; an unknown name has none, as in bash.
-fn class_ranges(name: &str) -> &'static [(char, char)] {
+/// Classes whose members are Unicode characters, in glibc's `C.utf8`: each is
+/// a regex bracket expression, stood in for inside the rewritten bracket by
+/// one noncharacter that [`class_regex`] turns back into it.
+const UNICODE_CLASSES: [(&str, &str); 11] = [
+    ("alnum", r"[\p{Alphabetic}\p{Nd}]"),
+    ("alpha", r"[\p{Alphabetic}\p{Nd}--0-9]"),
+    ("blank", r"[\p{Zs}\t--\u{a0}\u{2007}\u{202f}]"),
+    ("cntrl", r"[\p{Cc}\u{2028}\u{2029}]"),
+    (
+        "graph",
+        r"[[^\p{White_Space}\p{Cc}\p{Unassigned}]\u{a0}\u{2007}\u{202f}]",
+    ),
+    ("lower", r"[\p{Lowercase}\u{1c5}\u{1c8}\u{1cb}\u{1f2}]"),
+    ("print", r"[^\p{Cc}\p{Unassigned}\u{2028}\u{2029}]"),
+    (
+        "punct",
+        r"[[[^\p{White_Space}\p{Cc}\p{Unassigned}]\u{a0}\u{2007}\u{202f}]--[\p{Alphabetic}\p{Nd}]]",
+    ),
+    ("space", r"[\p{White_Space}--\u{a0}\u{2007}\u{202f}\u{85}]"),
+    ("upper", r"[\p{Uppercase}\p{Lt}]"),
+    ("word", r"[\p{Alphabetic}\p{Nd}_]"),
+];
+
+/// The first noncharacter, which no path holds.
+const MARKER_BASE: u32 = 0xFDD0;
+
+fn marker(index: usize) -> char {
+    let offset = u32::try_from(index).expect("a class index fits in u32");
+    char::from_u32(MARKER_BASE + offset).expect("noncharacters are chars")
+}
+
+/// The regex a marker stands for, as a bracket expression that leaves out
+/// the separator, since a bracket never matches `/`.
+pub(crate) fn class_regex(marker_char: char) -> Option<String> {
+    UNICODE_CLASSES
+        .iter()
+        .enumerate()
+        .find(|&(index, _)| marker(index) == marker_char)
+        .map(|(_, (_, regex))| format!("[{regex}--/]"))
+}
+
+/// The members of a class; an unknown name has none, as in bash.
+fn class_ranges(name: &str) -> Vec<(char, char)> {
     match name {
-        "alnum" => &[('0', '9'), ('A', 'Z'), ('a', 'z')],
-        "alpha" => &[('A', 'Z'), ('a', 'z')],
-        "ascii" => &[('\0', '\x7f')],
-        "blank" => &[(' ', ' '), ('\t', '\t')],
-        "cntrl" => &[('\0', '\x1f'), ('\x7f', '\x7f')],
-        "digit" => &[('0', '9')],
-        "graph" => &[('!', '~')],
-        "lower" => &[('a', 'z')],
-        "print" => &[(' ', '~')],
-        "punct" => &[('!', '/'), (':', '@'), ('[', '`'), ('{', '~')],
-        "space" => &[('\t', '\r'), (' ', ' ')],
-        "upper" => &[('A', 'Z')],
-        "word" => &[('0', '9'), ('A', 'Z'), ('a', 'z'), ('_', '_')],
-        "xdigit" => &[('0', '9'), ('A', 'F'), ('a', 'f')],
-        _ => &[],
+        "ascii" => vec![('\0', '\x7f')],
+        "digit" => vec![('0', '9')],
+        "xdigit" => vec![('0', '9'), ('A', 'F'), ('a', 'f')],
+        _ => UNICODE_CLASSES
+            .iter()
+            .position(|(known, _)| *known == name)
+            .map(|index| vec![(marker(index), marker(index))])
+            .unwrap_or_default(),
     }
 }
 
@@ -192,30 +226,46 @@ fn without(set: Vec<(u32, u32)>, c: char) -> Vec<(u32, u32)> {
 
 #[cfg(test)]
 mod tests {
+    #[test]
+    fn markers_count_up_from_the_first_noncharacter() {
+        assert_eq!(marker(0), '\u{fdd0}');
+        assert_eq!(marker(2), '\u{fdd2}');
+    }
+
     use super::*;
 
+    fn marked(name: &str) -> char {
+        let index = UNICODE_CLASSES
+            .iter()
+            .position(|(known, _)| *known == name)
+            .unwrap();
+        marker(index)
+    }
+
     #[test]
     fn a_class_becomes_its_ranges() {
         assert_eq!(expand_posix_classes("[[:digit:]].md"), "[0-9].md");
-        assert_eq!(expand_posix_classes("[[:upper:]]*"), "[A-Z]*");
+        assert_eq!(
+            expand_posix_classes("[[:upper:]]*"),
+            format!("[{}]*", marked("upper"))
+        );
     }
 
     #[test]
     fn every_named_class_expands() {
+        for name in [
+            "alnum", "alpha", "blank", "cntrl", "graph", "lower", "print", "punct", "space",
+            "upper", "word",
+        ] {
+            assert_eq!(
+                expand_posix_classes(&format!("[[:{name}:]]")),
+                format!("[{}]", marked(name)),
+                "{name}"
+            );
+        }
         let cases = [
-            ("alnum", "[0-9A-Za-z]"),
-            ("alpha", "[A-Za-z]"),
             ("ascii", "[]\0- \"-,.0-\\_-\x7f!^-]"),
-            ("blank", "[\t ]"),
-            ("cntrl", "[\0-\x1f\x7f]"),
             ("digit", "[0-9]"),
-            ("graph", "[]\"-,.0-\\_-~!^-]"),
-            ("lower", "[a-z]"),
-            ("print", "[] \"-,.0-\\_-~!^-]"),
-            ("punct", "[]\"-,.:-@[-\\_-`{-~!^-]"),
-            ("space", "[\t-\r ]"),
-            ("upper", "[A-Z]"),
-            ("word", "[0-9A-Z_a-z]"),
             ("xdigit", "[0-9A-Fa-f]"),
         ];
         for (name, want) in cases {
@@ -227,6 +277,14 @@ mod tests {
         }
     }
 
+    #[test]
+    fn every_marker_names_its_class_regex() {
+        for (index, (_, regex)) in UNICODE_CLASSES.iter().enumerate() {
+            assert_eq!(class_regex(marker(index)), Some(format!("[{regex}--/]")));
+        }
+        assert_eq!(class_regex('a'), None);
+    }
+
     #[test]
     fn a_negated_class_also_excludes_the_separator() {
         assert_eq!(expand_posix_classes("[![:digit:]]"), "[!/0-9]");
@@ -236,7 +294,10 @@ mod tests {
     #[test]
     fn classes_and_characters_combine() {
         assert_eq!(expand_posix_classes("[a[:digit:]]"), "[0-9a]");
-        assert_eq!(expand_posix_classes("[[:digit:][:upper:]]"), "[0-9A-Z]");
+        assert_eq!(
+            expand_posix_classes("[[:digit:][:upper:]]"),
+            format!("[0-9{}]", marked("upper"))
+        );
         assert_eq!(expand_posix_classes("[[:digit:]x-z]"), "[0-9x-z]");
     }