ellipt 0.2.1__tar.gz

ellipt-0.2.1/.gitignore

@@ -0,0 +1,16 @@
+__pycache__/
+*.py[cod]
+.venv/
+venv/
+build/
+dist/
+*.egg-info/
+.pytest_cache/
+.mypy_cache/
+.ruff_cache/
+.coverage
+htmlcov/
+
+# Unicode data files, fetched locally by scripts/fetch_upstream.py; only the manifest is tracked.
+data/ucd/*
+!data/ucd/MANIFEST.json

ellipt-0.2.1/CHANGELOG.md

@@ -0,0 +1,53 @@
+# Changelog
+
+All notable changes to this project are documented here. The format follows
+[Keep a Changelog](https://keepachangelog.com/en/1.1.0/) and the project uses
+[Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [Unreleased]
+
+## [0.2.1]
+
+### Changed
+
+- The Unicode Character Database files are no longer committed. `data/ucd/MANIFEST.json`
+  pins them to Unicode 18.0.0 (URL, size, sha256, licence) and `scripts/fetch_upstream.py`
+  fetches them; the generator check and the conformance tests skip until then. The
+  generated tables stay in the package and the Unicode licence notice ships in the wheel.
+- The sdist ships only the sources, tests, scripts and docs.
+
+## [0.2.0]
+
+### Added
+
+- `truncate` returns a frozen `Truncation` with `text`, `head`, `marker`, `tail`,
+  `original`, `unit`, `limit` and `size`, plus `truncated`, `removed` and
+  `removed_span`.
+- `unit=` on `shorten` and `truncate`: `"chars"`, `"graphemes"`, `"columns"` or
+  `"bytes"` (UTF-8). `measure(text, unit)` exposes the same measurement.
+- `words=True` cuts at whitespace when it can.
+- `shorten_path` drops whole directories from the middle and keeps the root and the
+  file name, then shortens the stem and keeps the extension.
+- `ambiguous=` (1 or 2) sets the width of East Asian Ambiguous characters.
+- `ELLIPSIS` (U+2026), `UNITS`, `UNICODE_VERSION` and `iter_graphemes`.
+- The `ellipt` command truncates every line of standard input or files.
+- Unicode 18.0.0 tables generated from a committed UCD snapshot, with the official
+  GraphemeBreakTest.txt in the test suite.
+
+### Changed
+
+- The cut never splits an extended grapheme cluster (UAX #29): accents, flags, emoji
+  ZWJ sequences and CR LF stay whole.
+- `width_aware=True` now measures grapheme clusters, so emoji sequences count as one
+  two-column glyph and the width data follows Unicode 18.0.0.
+- In the middle position, budget the left side cannot use goes to the right side.
+- Wrong argument types now raise `TypeError`: a non-str `text` or `marker`, and a
+  `width` that is not an int (a float or a bool).
+- Error messages say "limit" instead of "width" and name the unit.
+- With `width_aware=True`, control characters such as tab count as zero columns
+  (0.1.0 counted them as one), following wcwidth. Expand tabs first if they matter.
+
+## [0.1.0]
+
+- First release: `shorten` truncates a string with an ellipsis at the start,
+  middle or end, by character count or display width, with a configurable marker.

ellipt-0.2.1/CONTRIBUTING.md

@@ -0,0 +1,25 @@
+# Contributing
+
+Thanks for taking the time. This project keeps a small, predictable shape.
+
+## Setup
+
+```
+pip install -r requirements-dev.txt
+python scripts/fetch_upstream.py
+```
+
+The fetch step is optional: it downloads the Unicode data files listed in
+`data/ucd/MANIFEST.json` (checked against their sha256) into `data/ucd/`, which
+git ignores. The generator check and the tests that need them skip until they are there.
+
+## Before you open a pull request
+
+- Keep public API in `src/ellipt/__init__.py`; internals live in `_`-prefixed modules.
+- Add a test for the behaviour you change.
+- Never edit `src/ellipt/_ucd_*.py` by hand: they are generated from the fetched `data/ucd/` files by
+  `python scripts/gen_ucd_tables.py` (see GUIDE.md, "Unicode data").
+- Run `scripts/check.ps1` (format, lint, types, tests) and make sure it passes.
+- One logical change per commit, imperative subject line.
+
+Open an issue first if the change is large or shifts the public API.

ellipt-0.2.1/GUIDE.md

@@ -0,0 +1,142 @@
+# ellipt guide
+
+This guide states the exact rules `ellipt` follows, where its Unicode data comes from,
+and how it behaves at the edges.
+
+## The promise
+
+For `truncate(text, limit, where=..., unit=..., marker=..., words=..., ambiguous=...)`:
+
+1. If `text` already fits in `limit`, it comes back whole and `truncated` is false.
+2. Otherwise the result is `head + marker + tail`, where `head` is a prefix and `tail`
+   a suffix of `text`. With `where="end"` the tail is empty, with `where="start"` the
+   head is empty, and with `where="middle"` both ends are kept.
+3. `head` and `tail` end on extended grapheme cluster boundaries of `text` (UAX #29).
+4. The size of the result, measured in `unit`, never exceeds `limit`. The marker is
+   measured in the same unit and counted inside the limit.
+5. A marker larger than the limit raises `ValueError`, as do a negative limit and an
+   unknown `where`, `unit` or `ambiguous`. Wrong types raise `TypeError`.
+
+`shorten` returns `truncate(...).text`. `shorten_path` keeps promises 1, 3 (at
+separators and in the file name) and 4.
+
+## Units
+
+| unit | counts | use it for |
+| --- | --- | --- |
+| `chars` | code points, `len(s)` | the 0.1.0 default, Python-side limits |
+| `graphemes` | extended grapheme clusters | "characters" as a reader counts them |
+| `columns` | terminal display cells | tables, status lines, fixed-width output |
+| `bytes` | UTF-8 bytes | database columns, file names, protocol fields |
+
+Columns follow UAX #11 East Asian Width with Markus Kuhn's wcwidth rules: controls,
+nonspacing and enclosing marks, format characters (except SOFT HYPHEN) and Hangul medial
+vowels and final consonants are zero; Wide and Fullwidth are two; East Asian Ambiguous
+characters are `ambiguous` cells (1 by default, 2 for CJK terminals); everything else is
+one. A grapheme cluster that draws as one emoji (VS16, a skin tone, a ZWJ sequence, a
+tag sequence or a flag) is two cells, and an emoji with VS15 is one. These are the same
+rules and the same data as the sibling `cellstr` library.
+
+```python
+from ellipt import ELLIPSIS, measure, truncate
+
+measure("\u2500\u2500", "columns")  # -> 2
+measure("\u2500\u2500", "columns", ambiguous=2)  # -> 4
+measure("\u2764\ufe0f", "columns")  # -> 2
+measure("e\u0301", "graphemes")  # -> 1
+```
+
+## Choosing a marker
+
+U+2026 HORIZONTAL ELLIPSIS is one code point, one grapheme and one column, so it leaves
+two more units of text than `"..."` in those units. In UTF-8 both are three bytes. CSS
+`text-overflow: ellipsis` renders U+2026. The default stays `"..."` because it is plain
+ASCII and safe in every encoding and font; pass `marker=ELLIPSIS` for typographic output.
+Any string works, including `""` for a hard cut and `" [...] "` for prose.
+
+```python
+truncate("abcdefgh", 5).text  # -> 'ab...'
+truncate("abcdefgh", 5, marker=ELLIPSIS).text  # -> 'abcd\u2026'
+```
+
+## Where to cut
+
+- `end` keeps the start, like CSS `text-overflow: ellipsis`. Good for titles and
+  previews.
+- `start` keeps the end. Good for log lines, long ids and anything whose tail matters.
+- `middle` keeps both ends, the way macOS Finder and editors shorten file names. The
+  left side gets the larger half; when a wide cluster stops it early, the spare budget
+  goes to the right side, then back to the left.
+
+```python
+truncate("a\u4f60bcdefgh", 5, where="middle", unit="columns", marker=".").text  # -> 'a.fgh'
+```
+
+## Words
+
+`words=True` moves each cut to whitespace (any character for which `str.isspace` is
+true) and drops the whitespace next to the marker. In the middle position the budget a
+snap frees on one side is offered to the other side, again on word boundaries. It does
+not use dictionary-based word segmentation, so text without spaces (Chinese, Japanese,
+Thai) gets a hard cut at a grapheme boundary. If a single word fills the whole budget,
+the cut is hard too, rather than returning only the marker.
+
+```python
+truncate("Supercalifragilistic", 8, words=True).text  # -> 'Super...'
+truncate("one two three", 11, where="start", words=True).text  # -> '...three'
+truncate("a b c d e f", 7, where="middle", words=True).text  # -> 'a...e f'
+```
+
+## Paths
+
+`shorten_path` keeps the first component (`/`, `C:\`, `~/`, `\\server\share\`, or the top
+directory of a relative path) and the file name, and keeps as many directories next to the file name as
+fit. The order of fallbacks is:
+
+1. `root + marker + /dir/.../name`, dropping directories from the root side first;
+2. `marker + /name`;
+3. `name` alone, when it fits;
+4. the name shortened in the middle of its stem with the extension kept (`.tar.gz`,
+   `.tar.xz` and other `.tar.*` extensions stay together);
+5. the name shortened in the middle.
+
+Both `/` and `\` count as separators unless `sep` names one. A trailing separator stays
+with the last directory. An empty marker is rejected, because `a/b/c` cut to `a//c`
+would read as a different, real path.
+
+```python
+from ellipt import shorten_path
+
+shorten_path(r"C:\Users\me\AppData\Local\Temp\build.log", 24)  # -> 'C:\\...\\Temp\\build.log'
+shorten_path("/srv/www/site/public/", 16)  # -> '/.../public/'
+```
+
+## Edge cases
+
+- A marker that would merge with the neighbouring cluster (for example one that starts
+  with U+FE0F or a combining mark) can change the measured size. The result is measured
+  again after joining and the cut backs off until it fits.
+- Lone surrogates are counted as three bytes (the `surrogatepass` encoding) so strings
+  from `os.fsdecode` do not crash the byte unit.
+- CR LF is a single grapheme cluster and is never split.
+- Tabs and other control characters are zero columns, as in wcwidth. The library does
+  not know your tab stops, so expand tabs first (`s.expandtabs()`); the command line tool
+  does this for you with `--tabsize`.
+- ANSI escape sequences are measured as ordinary text. Strip them first.
+- Long inputs: an end cut in `chars` or `bytes` only segments the first `limit + 1` code
+  points. Other cuts segment the whole string once, which is linear in its length.
+
+## Unicode data
+
+The tables in `src/ellipt/_ucd_*.py` are generated by `scripts/gen_ucd_tables.py` from
+the Unicode 18.0.0 data files. Those files are not committed: `data/ucd/MANIFEST.json`
+pins their URLs, sizes and SHA-256 checksums, and `python scripts/fetch_upstream.py`
+fetches them into `data/ucd/`. With them in place the test suite regenerates the tables
+and compares them byte for byte, checks the checksums, and runs all 853 cases of the
+official `GraphemeBreakTest.txt` through both the segmenter and the truncation engine;
+without them those tests skip.
+
+To move to a new Unicode version, update the URLs, sizes and checksums in
+`data/ucd/MANIFEST.json`, run `python scripts/fetch_upstream.py` and
+`python scripts/gen_ucd_tables.py`, and update the expected case count in
+`tests/test_grapheme_conformance.py`.

ellipt-0.2.1/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Javier Lianes
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

ellipt-0.2.1/LICENSE-UNICODE

@@ -0,0 +1,39 @@
+UNICODE LICENSE V3
+
+COPYRIGHT AND PERMISSION NOTICE
+
+Copyright © 1991-2026 Unicode, Inc.
+
+NOTICE TO USER: Carefully read the following legal agreement. BY
+DOWNLOADING, INSTALLING, COPYING OR OTHERWISE USING DATA FILES, AND/OR
+SOFTWARE, YOU UNEQUIVOCALLY ACCEPT, AND AGREE TO BE BOUND BY, ALL OF THE
+TERMS AND CONDITIONS OF THIS AGREEMENT. IF YOU DO NOT AGREE, DO NOT
+DOWNLOAD, INSTALL, COPY, DISTRIBUTE OR USE THE DATA FILES OR SOFTWARE.
+
+Permission is hereby granted, free of charge, to any person obtaining a
+copy of data files and any associated documentation (the "Data Files") or
+software and any associated documentation (the "Software") to deal in the
+Data Files or Software without restriction, including without limitation
+the rights to use, copy, modify, merge, publish, distribute, and/or sell
+copies of the Data Files or Software, and to permit persons to whom the
+Data Files or Software are furnished to do so, provided that either (a)
+this copyright and permission notice appear with all copies of the Data
+Files or Software, or (b) this copyright and permission notice appear in
+associated Documentation.
+
+THE DATA FILES AND SOFTWARE ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY
+KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT OF
+THIRD PARTY RIGHTS.
+
+IN NO EVENT SHALL THE COPYRIGHT HOLDER OR HOLDERS INCLUDED IN THIS NOTICE
+BE LIABLE FOR ANY CLAIM, OR ANY SPECIAL INDIRECT OR CONSEQUENTIAL DAMAGES,
+OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS,
+WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION,
+ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THE DATA
+FILES OR SOFTWARE.
+
+Except as contained in this notice, the name of a copyright holder shall
+not be used in advertising or otherwise to promote the sale, use or other
+dealings in these Data Files or Software without prior written
+authorization of the copyright holder.

ellipt-0.2.1/PKG-INFO

@@ -0,0 +1,134 @@
+Metadata-Version: 2.5
+Name: ellipt
+Version: 0.2.1
+Summary: Truncate strings with an ellipsis at the start, middle or end, by characters, grapheme clusters, display columns or UTF-8 bytes, without splitting emoji or accents
+Project-URL: Homepage, https://lianes8studio.com
+Project-URL: Repository, https://github.com/JaviLianes8/ellipt
+Project-URL: Issues, https://github.com/JaviLianes8/ellipt/issues
+Project-URL: Changelog, https://github.com/JaviLianes8/ellipt/blob/main/CHANGELOG.md
+Project-URL: LinkedIn, https://www.linkedin.com/in/jlianes/
+Project-URL: Funding, https://buymeacoffee.com/jlianesglrs
+Author-email: Javier Lianes <javier@lianes8studio.com>
+License-Expression: MIT
+License-File: LICENSE
+License-File: LICENSE-UNICODE
+Keywords: cli,display-width,ellipsis,grapheme,path,shorten,string,truncate,unicode
+Classifier: Development Status :: 4 - Beta
+Classifier: Environment :: Console
+Classifier: Intended Audience :: Developers
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Topic :: Software Development :: Libraries
+Classifier: Topic :: Text Processing
+Classifier: Typing :: Typed
+Requires-Python: >=3.10
+Provides-Extra: dev
+Requires-Dist: build; extra == 'dev'
+Requires-Dist: mypy; extra == 'dev'
+Requires-Dist: pytest; extra == 'dev'
+Requires-Dist: ruff; extra == 'dev'
+Requires-Dist: twine; extra == 'dev'
+Description-Content-Type: text/markdown
+
+# ellipt
+
+Truncate strings with an ellipsis at the start, middle or end. The limit can count code
+points, grapheme clusters, terminal columns or UTF-8 bytes, the marker is counted inside
+the limit, and the cut never splits an accent, a flag or an emoji sequence. Typed, zero
+dependencies, Unicode 18.0.0 tables embedded.
+
+## Install
+
+```
+pip install ellipt
+```
+
+## Usage
+
+```python
+from ellipt import ELLIPSIS, measure, shorten, shorten_path, truncate
+
+shorten("hello world", 8)  # -> 'hello...'
+shorten("hello world", 8, where="start")  # -> '...world'
+shorten("hello world", 9, where="middle")  # -> 'hel...rld'
+shorten("hello world", 6, marker=ELLIPSIS)  # -> 'hello…'
+
+# Text that already fits is returned unchanged.
+shorten("short", 20)  # -> 'short'
+```
+
+### Pick the unit of the limit
+
+```python
+family = "\U0001f468\u200d\U0001f469\u200d\U0001f467"  # one emoji, five code points
+text = "ab" + family + "cd"
+
+measure(text, "chars")  # -> 9
+measure(text, "graphemes")  # -> 5
+measure(text, "columns")  # -> 6
+measure(text, "bytes")  # -> 22
+
+# The family is kept whole or dropped whole, never cut into pieces.
+shorten(text, 6, marker=ELLIPSIS)  # -> 'ab…'
+shorten(text, 4, unit="graphemes", marker=ELLIPSIS)  # -> 'ab👨\u200d👩\u200d👧…'
+shorten("你好世界", 5, unit="columns", marker=ELLIPSIS)  # -> '你好…'
+shorten("café crème", 8, unit="bytes", marker="")  # -> 'café cr'
+```
+
+`width_aware=True` from 0.1.0 still works and means `unit="columns"`.
+
+### Know what was removed
+
+```python
+r = truncate("The quick brown fox jumps", 14, words=True, marker=ELLIPSIS)
+r.text  # -> 'The quick…'
+r.removed  # -> ' brown fox jumps'
+r.truncated  # -> True
+r.size <= r.limit  # -> True
+```
+
+`words=True` backs the cut off to whitespace; with one long word it cuts hard instead.
+
+### Shorten paths the way file browsers do
+
+```python
+shorten_path("/home/me/projects/app/src/main.py", 20)  # -> '/.../app/src/main.py'
+shorten_path("/home/me/projects/app/src/main.py", 11)  # -> '.../main.py'
+shorten_path("photos/an_extremely_long_holiday_name.jpeg", 20)  # -> 'an_ext...y_name.jpeg'
+```
+
+## Command line
+
+```
+$ printf 'hello world\n' | ellipt 8
+hello...
+$ git ls-files | ellipt 28 --path
+$ ellipt 30 --where middle --ellipsis notes.txt
+$ ellipt 255 --unit bytes --marker "" < titles.txt
+```
+
+The command reads standard input (or the files given) as UTF-8, keeps each LF or CRLF
+line ending, and counts terminal columns by default, expanding tabs to 8-column stops
+first (`--tabsize`). `--ellipsis` uses U+2026 as the marker without having to type it.
+`ellipt --help` lists every option. Errors go to standard error with exit status 1, and
+a reader that stops early (`| head`) is not an error.
+
+## Why
+
+`textwrap.shorten` only cuts at the end, collapses whitespace and always cuts on words.
+Plain slicing (`s[:n] + "..."`) splits emoji and accents, overflows a terminal with wide
+characters and can exceed a byte limit. `ellipt` cuts where you ask, in the unit you
+need, and guarantees the result is never longer than the limit. See [GUIDE.md](GUIDE.md)
+for the exact rules, the Unicode sources and the edge cases.
+
+## License
+
+MIT. See [LICENSE](LICENSE). The embedded Unicode data is covered by the Unicode
+License v3, see [LICENSE-UNICODE](LICENSE-UNICODE).
+
+## Links
+
+- Website: https://lianes8studio.com
+- GitHub: https://github.com/JaviLianes8
+- LinkedIn: https://www.linkedin.com/in/jlianes/

ellipt-0.2.1/README.md

@@ -0,0 +1,101 @@
+# ellipt
+
+Truncate strings with an ellipsis at the start, middle or end. The limit can count code
+points, grapheme clusters, terminal columns or UTF-8 bytes, the marker is counted inside
+the limit, and the cut never splits an accent, a flag or an emoji sequence. Typed, zero
+dependencies, Unicode 18.0.0 tables embedded.
+
+## Install
+
+```
+pip install ellipt
+```
+
+## Usage
+
+```python
+from ellipt import ELLIPSIS, measure, shorten, shorten_path, truncate
+
+shorten("hello world", 8)  # -> 'hello...'
+shorten("hello world", 8, where="start")  # -> '...world'
+shorten("hello world", 9, where="middle")  # -> 'hel...rld'
+shorten("hello world", 6, marker=ELLIPSIS)  # -> 'hello…'
+
+# Text that already fits is returned unchanged.
+shorten("short", 20)  # -> 'short'
+```
+
+### Pick the unit of the limit
+
+```python
+family = "\U0001f468\u200d\U0001f469\u200d\U0001f467"  # one emoji, five code points
+text = "ab" + family + "cd"
+
+measure(text, "chars")  # -> 9
+measure(text, "graphemes")  # -> 5
+measure(text, "columns")  # -> 6
+measure(text, "bytes")  # -> 22
+
+# The family is kept whole or dropped whole, never cut into pieces.
+shorten(text, 6, marker=ELLIPSIS)  # -> 'ab…'
+shorten(text, 4, unit="graphemes", marker=ELLIPSIS)  # -> 'ab👨\u200d👩\u200d👧…'
+shorten("你好世界", 5, unit="columns", marker=ELLIPSIS)  # -> '你好…'
+shorten("café crème", 8, unit="bytes", marker="")  # -> 'café cr'
+```
+
+`width_aware=True` from 0.1.0 still works and means `unit="columns"`.
+
+### Know what was removed
+
+```python
+r = truncate("The quick brown fox jumps", 14, words=True, marker=ELLIPSIS)
+r.text  # -> 'The quick…'
+r.removed  # -> ' brown fox jumps'
+r.truncated  # -> True
+r.size <= r.limit  # -> True
+```
+
+`words=True` backs the cut off to whitespace; with one long word it cuts hard instead.
+
+### Shorten paths the way file browsers do
+
+```python
+shorten_path("/home/me/projects/app/src/main.py", 20)  # -> '/.../app/src/main.py'
+shorten_path("/home/me/projects/app/src/main.py", 11)  # -> '.../main.py'
+shorten_path("photos/an_extremely_long_holiday_name.jpeg", 20)  # -> 'an_ext...y_name.jpeg'
+```
+
+## Command line
+
+```
+$ printf 'hello world\n' | ellipt 8
+hello...
+$ git ls-files | ellipt 28 --path
+$ ellipt 30 --where middle --ellipsis notes.txt
+$ ellipt 255 --unit bytes --marker "" < titles.txt
+```
+
+The command reads standard input (or the files given) as UTF-8, keeps each LF or CRLF
+line ending, and counts terminal columns by default, expanding tabs to 8-column stops
+first (`--tabsize`). `--ellipsis` uses U+2026 as the marker without having to type it.
+`ellipt --help` lists every option. Errors go to standard error with exit status 1, and
+a reader that stops early (`| head`) is not an error.
+
+## Why
+
+`textwrap.shorten` only cuts at the end, collapses whitespace and always cuts on words.
+Plain slicing (`s[:n] + "..."`) splits emoji and accents, overflows a terminal with wide
+characters and can exceed a byte limit. `ellipt` cuts where you ask, in the unit you
+need, and guarantees the result is never longer than the limit. See [GUIDE.md](GUIDE.md)
+for the exact rules, the Unicode sources and the edge cases.
+
+## License
+
+MIT. See [LICENSE](LICENSE). The embedded Unicode data is covered by the Unicode
+License v3, see [LICENSE-UNICODE](LICENSE-UNICODE).
+
+## Links
+
+- Website: https://lianes8studio.com
+- GitHub: https://github.com/JaviLianes8
+- LinkedIn: https://www.linkedin.com/in/jlianes/

ellipt-0.2.1/data/ucd/MANIFEST.json

@@ -0,0 +1,56 @@
+{
+  "DerivedCoreProperties.txt": {
+    "title": "Unicode Character Database 18.0.0, DerivedCoreProperties.txt",
+    "url": "https://www.unicode.org/Public/18.0.0/ucd/DerivedCoreProperties.txt",
+    "revision": "Unicode 18.0.0",
+    "fetched": "2026-10-13",
+    "size": 1159889,
+    "sha256": "09c928886a178fcafd93c29e4bd59073a058e5a100b716d425cb563ab50f68c9",
+    "licence": "Unicode License v3 (https://www.unicode.org/license.txt), see LICENSE-UNICODE"
+  },
+  "DerivedGeneralCategory.txt": {
+    "title": "Unicode Character Database 18.0.0, DerivedGeneralCategory.txt",
+    "url": "https://www.unicode.org/Public/18.0.0/ucd/extracted/DerivedGeneralCategory.txt",
+    "revision": "Unicode 18.0.0",
+    "fetched": "2026-10-13",
+    "size": 282143,
+    "sha256": "d6b151d2d40ee9b1876d26f417980f45ffae47b6055ccf7203cb31f07a030f94",
+    "licence": "Unicode License v3 (https://www.unicode.org/license.txt), see LICENSE-UNICODE"
+  },
+  "EastAsianWidth.txt": {
+    "title": "Unicode Character Database 18.0.0, EastAsianWidth.txt",
+    "url": "https://www.unicode.org/Public/18.0.0/ucd/EastAsianWidth.txt",
+    "revision": "Unicode 18.0.0",
+    "fetched": "2026-10-13",
+    "size": 204705,
+    "sha256": "a0cf29eacd00cfcaec4381c6b7c281685f18dbb4e7ff82b4076ccb342ca839aa",
+    "licence": "Unicode License v3 (https://www.unicode.org/license.txt), see LICENSE-UNICODE"
+  },
+  "GraphemeBreakProperty.txt": {
+    "title": "Unicode Character Database 18.0.0, GraphemeBreakProperty.txt",
+    "url": "https://www.unicode.org/Public/18.0.0/ucd/auxiliary/GraphemeBreakProperty.txt",
+    "revision": "Unicode 18.0.0",
+    "fetched": "2026-10-13",
+    "size": 100009,
+    "sha256": "0839dcb79e4ac639ecd538b1abf7c9d22e3f9dd265b7e182d33627aa4d75b45a",
+    "licence": "Unicode License v3 (https://www.unicode.org/license.txt), see LICENSE-UNICODE"
+  },
+  "GraphemeBreakTest.txt": {
+    "title": "Unicode Character Database 18.0.0, GraphemeBreakTest.txt",
+    "url": "https://www.unicode.org/Public/18.0.0/ucd/auxiliary/GraphemeBreakTest.txt",
+    "revision": "Unicode 18.0.0",
+    "fetched": "2026-10-13",
+    "size": 137090,
+    "sha256": "b0cf047ee94485bbdc846de2b902f5f8a815f6b674f9d04223cddadd91c9df31",
+    "licence": "Unicode License v3 (https://www.unicode.org/license.txt), see LICENSE-UNICODE"
+  },
+  "emoji-data.txt": {
+    "title": "Unicode Character Database 18.0.0, emoji-data.txt",
+    "url": "https://www.unicode.org/Public/18.0.0/ucd/emoji/emoji-data.txt",
+    "revision": "Unicode 18.0.0",
+    "fetched": "2026-10-13",
+    "size": 108719,
+    "sha256": "80d00f8e616a0ef27fd6b8de3b758c06383b5d917e2977709578e68baf733bf1",
+    "licence": "Unicode License v3 (https://www.unicode.org/license.txt), see LICENSE-UNICODE"
+  }
+}

ellipt-0.2.1/pyproject.toml

@@ -0,0 +1,77 @@
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[project]
+name = "ellipt"
+version = "0.2.1"
+description = "Truncate strings with an ellipsis at the start, middle or end, by characters, grapheme clusters, display columns or UTF-8 bytes, without splitting emoji or accents"
+readme = "README.md"
+requires-python = ">=3.10"
+license = "MIT"
+license-files = ["LICENSE", "LICENSE-UNICODE"]
+authors = [{ name = "Javier Lianes", email = "javier@lianes8studio.com" }]
+keywords = ["ellipsis", "truncate", "shorten", "string", "unicode", "grapheme", "display-width", "path", "cli"]
+classifiers = [
+    "Development Status :: 4 - Beta",
+    "Intended Audience :: Developers",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3 :: Only",
+    "Environment :: Console",
+    "Operating System :: OS Independent",
+    "Topic :: Text Processing",
+    "Topic :: Software Development :: Libraries",
+    "Typing :: Typed",
+]
+dependencies = []
+
+[project.scripts]
+ellipt = "ellipt._cli:main"
+
+[project.optional-dependencies]
+dev = ["ruff", "mypy", "pytest", "build", "twine"]
+
+[project.urls]
+Homepage = "https://lianes8studio.com"
+Repository = "https://github.com/JaviLianes8/ellipt"
+Issues = "https://github.com/JaviLianes8/ellipt/issues"
+Changelog = "https://github.com/JaviLianes8/ellipt/blob/main/CHANGELOG.md"
+LinkedIn = "https://www.linkedin.com/in/jlianes/"
+Funding = "https://buymeacoffee.com/jlianesglrs"
+
+[tool.hatch.build.targets.sdist]
+# The Unicode data files are fetched locally and never shipped; only their manifest is.
+include = [
+    "src",
+    "tests",
+    "scripts/gen_ucd_tables.py",
+    "scripts/fetch_upstream.py",
+    "data/ucd/MANIFEST.json",
+    "CHANGELOG.md",
+    "CONTRIBUTING.md",
+    "GUIDE.md",
+]
+exclude = ["**/__pycache__"]
+
+[tool.hatch.build.targets.wheel]
+packages = ["src/ellipt"]
+
+[tool.ruff]
+line-length = 100
+target-version = "py310"
+
+[tool.ruff.lint]
+select = ["E", "F", "I", "UP", "B", "SIM", "RUF"]
+
+[tool.ruff.lint.per-file-ignores]
+# Tests use look-alike characters (fullwidth, ellipsis, accents) on purpose.
+"tests/*" = ["RUF001", "RUF003"]
+
+[tool.mypy]
+python_version = "3.10"
+strict = true
+
+[tool.pytest.ini_options]
+testpaths = ["tests"]
+pythonpath = ["src"]
+addopts = "-q"