stopslop 0.1.0__tar.gz

stopslop-0.1.0/.gitignore

@@ -0,0 +1,24 @@
+# Crawled PDFs are large (325 MB for four venue-years) and re-fetchable: the
+# crawler is idempotent, so `stopslop crawl <venue> <year>` rebuilds them from the
+# ids recorded in stats.json. The measurements are what matter and they ARE
+# committed -- stats.json is the artifact a report can be audited against.
+corpus/**/*.pdf
+# Measurements are generated artifacts and live in the package, under
+# src/stopslop/data/<extractor>/. corpus/ is now purely a PDF cache.
+corpus/
+__pycache__/
+
+# Build artifacts
+*.egg-info/
+dist/
+build/
+.venv*/
+.pytest_cache/
+
+# Belt and braces against the one accident that would hurt: a crawl pointed at
+# the packaged tree. Only stats.json is ever meant to live there, and
+# tests/test_packaging.py fails if a PDF does. This stops it reaching a commit.
+src/stopslop/data/**/*.pdf
+
+# Per-extractor remeasure logs
+remeasure_*.log

stopslop-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Dmitrii Khizbullin
+
+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.

stopslop-0.1.0/PKG-INFO

@@ -0,0 +1,177 @@
+Metadata-Version: 2.5
+Name: stopslop
+Version: 0.1.0
+Summary: Measure a paper's prose against how a venue actually wrote, before and after LLMs existed
+Project-URL: Homepage, https://github.com/dmitrii-khizbullin/stopslop
+Project-URL: Source, https://github.com/dmitrii-khizbullin/stopslop
+Author: Dmitrii Khizbullin
+License: MIT License
+        
+        Copyright (c) 2026 Dmitrii Khizbullin
+        
+        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.
+License-File: LICENSE
+Keywords: academic-writing,corpus,llm,prose,stylometry
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Science/Research
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Topic :: Scientific/Engineering
+Classifier: Topic :: Text Processing :: Linguistic
+Requires-Python: >=3.9
+Requires-Dist: pypdf>=5
+Provides-Extra: pdfminer
+Requires-Dist: pdfminer-six>=20240706; extra == 'pdfminer'
+Provides-Extra: pymupdf
+Requires-Dist: pymupdf>=1.24.3; extra == 'pymupdf'
+Provides-Extra: pypdfium2
+Requires-Dist: pypdfium2>=4; extra == 'pypdfium2'
+Description-Content-Type: text/markdown
+
+# stopslop
+
+Measure a paper's prose against how a venue actually writes — before and after
+LLMs existed.
+
+Two baselines, always both: **fossil** (published up to 2022, so it cannot have
+been machine-assisted) and **current** (the last full year plus this one). The
+fossil window is the target; the current window is context, never an excuse.
+See [UNSLOP.md](UNSLOP.md).
+
+## Install
+
+The wheel carries the baselines for all seven venues — 105 venue-years of
+measurements — so there is nothing to crawl and nothing to configure.
+
+```bash
+pip install stopslop              # pypdf, BSD-3-Clause, no copyleft
+pip install "stopslop[pymupdf]"   # adds PyMuPDF (AGPL-3.0, opt-in)
+pip install -e .                  # editable, for hacking on it
+```
+
+Then, from any directory, in any project:
+
+```bash
+stopslop check paper.pdf --venue tmlr --out report.md
+```
+
+No system tools required. Everything else is stdlib `re` and `statistics`.
+
+## Extractors
+
+**The extractor is part of the measurement.** Across 24 papers the four
+supported extractors agree to within ~1% on prose metrics — first person
+18.1–18.3, -ly adverbs 13.2–13.4, passives 10.3–10.5 — and disagree sharply on
+layout-sensitive ones: `numerals` runs **43.4** under pymupdf against **61.4**
+under pypdf. So measurements are stored per extractor and never mixed, and
+`stopslop check` refuses rather than compare across them.
+
+| extra | licence | speed | notes |
+|---|---|---|---|
+| *(default)* `pypdf` | BSD-3-Clause | 0.82 s/paper | pure Python, installs anywhere |
+| `[pymupdf]` | **AGPL-3.0** or commercial | 0.12 s/paper | best quality; opt-in, never a default |
+| `[pypdfium2]` | BSD-3 / Apache-2.0 | **0.10 s/paper** | fastest; platform binary wheels |
+| `[pdfminer]` | MIT | 2.13 s/paper | slowest |
+
+Baselines for **all** of them ship in the base wheel, so an extra only decides
+what you can *measure* with, never what you can *read*. Pick one with
+`--extractor` or `$UNSLOP_EXTRACTOR`.
+
+`pip install stopslop` carries no copyleft: PyMuPDF is AGPL and is opt-in, so
+taking on that licence is always your deliberate choice.
+
+## Use
+
+```bash
+stopslop check paper.pdf --venue tmlr --out reports/paper.md   # score one paper
+stopslop trend tmlr --out reports/tmlr_trend.md                # one venue's drift
+stopslop crosstrend --out reports/total_trend_points.md        # venues kept apart
+stopslop pooltrend  --out reports/total_trend.md               # venues pooled
+stopslop stats tmlr 2022                                       # one crawled year
+stopslop crawl tmlr 2022 2025 2026 --limit 20                  # fetch and measure
+```
+
+`crawl` is the only command that touches the network, and it needs no key or
+account. Everything else is offline.
+
+### Where the baselines come from
+
+Resolved at call time, most specific first:
+
+1. `--root DIR` on any command
+2. `$UNSLOP_MEASUREMENTS`
+3. `./data` in the working directory, if it exists
+4. the measurements bundled in the package
+
+So a project that measures its own papers into `./data` uses them
+automatically, and everyone else gets the shipped ones. PDFs are cached
+separately, under `--pdf-root` / `$UNSLOP_PDFS` / `./corpus`.
+
+## Plots
+
+`--plot` needs matplotlib, which is **not a dependency and not an extra**.
+Plotting is an analytics concern and the measurement path has no use for it, so
+it goes in a separate environment:
+
+```bash
+python3 -m venv .venv-analytics
+.venv-analytics/bin/pip install -e . -r analytics-requirements.txt
+.venv-analytics/bin/stopslop trend tmlr --plot --out reports/tmlr_trend.md
+```
+
+Every other command works without it.
+
+## Development
+
+```bash
+pip install -e ".[pymupdf,pypdfium2,pdfminer]" --group dev
+pytest
+stopslop remeasure --extractor pypdf --root src/stopslop/data   # offline, no network
+```
+
+Measurements are generated artifacts and are written **straight into the
+package**, so there is no snapshot to keep in sync. `corpus/` is a pure PDF
+cache: large, re-fetchable, and entirely gitignored. `remeasure` re-reads those
+PDFs under a chosen extractor without touching the network — `crawl` would
+re-query arXiv for a paper list we already have.
+
+## Layout
+
+```
+src/stopslop/
+  extractors/        pypdf | pymupdf | pypdfium2 | pdfminer
+  data/<extractor>/<venue>/<year>/stats.json
+                     the measurements, shipped in the wheel
+corpus/<venue>/<year>/*.pdf
+                     PDF cache: 3.7 GB, gitignored, re-fetchable
+reports/             generated trend reports and checked papers
+tests/
+```
+
+## Status
+
+Pre-release. Published to PyPI as **`stopslop`**: the obvious names, `unslop`
+and `deslop`, are both taken by humanizers, which do the opposite of what this
+tool is for --- it measures and locates, it never edits your paper. Every
+candidate that was checked is in "The rename" in [UNSLOP.md](UNSLOP.md), and the
+remaining release blockers are in the Decision section of
+[COMPETITION.md](COMPETITION.md).
+
+MIT licensed.

stopslop-0.1.0/README.md

@@ -0,0 +1,130 @@
+# stopslop
+
+Measure a paper's prose against how a venue actually writes — before and after
+LLMs existed.
+
+Two baselines, always both: **fossil** (published up to 2022, so it cannot have
+been machine-assisted) and **current** (the last full year plus this one). The
+fossil window is the target; the current window is context, never an excuse.
+See [UNSLOP.md](UNSLOP.md).
+
+## Install
+
+The wheel carries the baselines for all seven venues — 105 venue-years of
+measurements — so there is nothing to crawl and nothing to configure.
+
+```bash
+pip install stopslop              # pypdf, BSD-3-Clause, no copyleft
+pip install "stopslop[pymupdf]"   # adds PyMuPDF (AGPL-3.0, opt-in)
+pip install -e .                  # editable, for hacking on it
+```
+
+Then, from any directory, in any project:
+
+```bash
+stopslop check paper.pdf --venue tmlr --out report.md
+```
+
+No system tools required. Everything else is stdlib `re` and `statistics`.
+
+## Extractors
+
+**The extractor is part of the measurement.** Across 24 papers the four
+supported extractors agree to within ~1% on prose metrics — first person
+18.1–18.3, -ly adverbs 13.2–13.4, passives 10.3–10.5 — and disagree sharply on
+layout-sensitive ones: `numerals` runs **43.4** under pymupdf against **61.4**
+under pypdf. So measurements are stored per extractor and never mixed, and
+`stopslop check` refuses rather than compare across them.
+
+| extra | licence | speed | notes |
+|---|---|---|---|
+| *(default)* `pypdf` | BSD-3-Clause | 0.82 s/paper | pure Python, installs anywhere |
+| `[pymupdf]` | **AGPL-3.0** or commercial | 0.12 s/paper | best quality; opt-in, never a default |
+| `[pypdfium2]` | BSD-3 / Apache-2.0 | **0.10 s/paper** | fastest; platform binary wheels |
+| `[pdfminer]` | MIT | 2.13 s/paper | slowest |
+
+Baselines for **all** of them ship in the base wheel, so an extra only decides
+what you can *measure* with, never what you can *read*. Pick one with
+`--extractor` or `$UNSLOP_EXTRACTOR`.
+
+`pip install stopslop` carries no copyleft: PyMuPDF is AGPL and is opt-in, so
+taking on that licence is always your deliberate choice.
+
+## Use
+
+```bash
+stopslop check paper.pdf --venue tmlr --out reports/paper.md   # score one paper
+stopslop trend tmlr --out reports/tmlr_trend.md                # one venue's drift
+stopslop crosstrend --out reports/total_trend_points.md        # venues kept apart
+stopslop pooltrend  --out reports/total_trend.md               # venues pooled
+stopslop stats tmlr 2022                                       # one crawled year
+stopslop crawl tmlr 2022 2025 2026 --limit 20                  # fetch and measure
+```
+
+`crawl` is the only command that touches the network, and it needs no key or
+account. Everything else is offline.
+
+### Where the baselines come from
+
+Resolved at call time, most specific first:
+
+1. `--root DIR` on any command
+2. `$UNSLOP_MEASUREMENTS`
+3. `./data` in the working directory, if it exists
+4. the measurements bundled in the package
+
+So a project that measures its own papers into `./data` uses them
+automatically, and everyone else gets the shipped ones. PDFs are cached
+separately, under `--pdf-root` / `$UNSLOP_PDFS` / `./corpus`.
+
+## Plots
+
+`--plot` needs matplotlib, which is **not a dependency and not an extra**.
+Plotting is an analytics concern and the measurement path has no use for it, so
+it goes in a separate environment:
+
+```bash
+python3 -m venv .venv-analytics
+.venv-analytics/bin/pip install -e . -r analytics-requirements.txt
+.venv-analytics/bin/stopslop trend tmlr --plot --out reports/tmlr_trend.md
+```
+
+Every other command works without it.
+
+## Development
+
+```bash
+pip install -e ".[pymupdf,pypdfium2,pdfminer]" --group dev
+pytest
+stopslop remeasure --extractor pypdf --root src/stopslop/data   # offline, no network
+```
+
+Measurements are generated artifacts and are written **straight into the
+package**, so there is no snapshot to keep in sync. `corpus/` is a pure PDF
+cache: large, re-fetchable, and entirely gitignored. `remeasure` re-reads those
+PDFs under a chosen extractor without touching the network — `crawl` would
+re-query arXiv for a paper list we already have.
+
+## Layout
+
+```
+src/stopslop/
+  extractors/        pypdf | pymupdf | pypdfium2 | pdfminer
+  data/<extractor>/<venue>/<year>/stats.json
+                     the measurements, shipped in the wheel
+corpus/<venue>/<year>/*.pdf
+                     PDF cache: 3.7 GB, gitignored, re-fetchable
+reports/             generated trend reports and checked papers
+tests/
+```
+
+## Status
+
+Pre-release. Published to PyPI as **`stopslop`**: the obvious names, `unslop`
+and `deslop`, are both taken by humanizers, which do the opposite of what this
+tool is for --- it measures and locates, it never edits your paper. Every
+candidate that was checked is in "The rename" in [UNSLOP.md](UNSLOP.md), and the
+remaining release blockers are in the Decision section of
+[COMPETITION.md](COMPETITION.md).
+
+MIT licensed.

stopslop-0.1.0/analytics-requirements.txt

@@ -0,0 +1,11 @@
+# Analytics only. NOT a dependency of the package, and deliberately not an
+# optional extra either -- see note 1 in pyproject.toml.
+#
+# Install into a SEPARATE environment, so that the measurement path stays
+# stdlib-only and nobody installing the checker pays for a plotting stack:
+#
+#     python3 -m venv .venv-analytics
+#     .venv-analytics/bin/pip install -e . -r analytics-requirements.txt
+#
+# Only `--plot` needs this. Every other command works without it.
+matplotlib>=3.5

stopslop-0.1.0/pyproject.toml

@@ -0,0 +1,101 @@
+# Build configuration.
+#
+# Two properties are load-bearing and easy to lose, so they are asserted here
+# rather than left to convention:
+#
+#   1. THE DEFAULT INSTALL IS PERMISSIVELY LICENSED. `pip install stopslop` pulls
+#      pypdf (BSD-3-Clause, pure Python), so the wheel carries no copyleft and
+#      needs no platform binary. PyMuPDF is the best extractor and is AGPL-3.0,
+#      so it is an OPT-IN EXTRA and never a default: `pip install
+#      "stopslop[pymupdf]"` makes that licence the user's deliberate choice.
+#
+#      matplotlib is neither a dependency nor an extra -- plotting belongs in a
+#      separate environment (see analytics-requirements.txt), because an extra
+#      is still a dependency that `pip install stopslop[...]` will happily pull.
+#
+#   2. NO PDFs IN THE WHEEL. `corpus/` at the repository root holds 3.7 GB of
+#      them. Only the baked JSON baselines under src/stopslop/data ship, and the
+#      exclude below is the belt to force-include's braces.
+
+[build-system]
+requires = ["hatchling>=1.18"]
+build-backend = "hatchling.build"
+
+[project]
+name = "stopslop"                    # Chosen 2026-09-08, after `unslop` and
+                                     # `deslop` were both found taken on PyPI by
+                                     # humanizers, which do the opposite of
+                                     # design rule 7. See "The rename" in
+                                     # UNSLOP.md for every candidate checked.
+dynamic = ["version"]
+description = "Measure a paper's prose against how a venue actually wrote, before and after LLMs existed"
+readme = "README.md"
+requires-python = ">=3.9"
+license = { file = "LICENSE" }
+authors = [{ name = "Dmitrii Khizbullin" }]
+keywords = ["stylometry", "academic-writing", "corpus", "llm", "prose"]
+classifiers = [
+    "Development Status :: 3 - Alpha",
+    "Intended Audience :: Science/Research",
+    "License :: OSI Approved :: MIT License",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3 :: Only",
+    "Topic :: Scientific/Engineering",
+    "Topic :: Text Processing :: Linguistic",
+]
+dependencies = [
+    # BSD-3-Clause and pure Python: installs on any platform, no binary wheel,
+    # no copyleft. Slower than the alternatives and that is an accepted trade.
+    "pypdf>=5",
+]
+
+[project.optional-dependencies]
+# Each extra adds one extractor. Measurements for all of them ship in the base
+# wheel regardless, so an extra only decides what you can MEASURE with, never
+# what you can READ. Extra names normalise to lowercase (PEP 685), so
+# `stopslop[PyMuPDF]` and `stopslop[pymupdf]` are the same request.
+pymupdf = ["pymupdf>=1.24.3"]        # AGPL-3.0 or commercial; opt-in only
+pypdfium2 = ["pypdfium2>=4"]         # BSD-3-Clause / Apache-2.0, fastest
+pdfminer = ["pdfminer.six>=20240706"]  # MIT, slowest
+
+[project.scripts]
+stopslop = "stopslop.cli:main"
+
+[project.urls]
+# The PUBLIC repository, which does not exist yet: this repo is the development
+# one and is not what gets released (see "The rename" in UNSLOP.md). These links
+# 404 until it is created, and a version's metadata can never be re-uploaded, so
+# they point forward deliberately rather than at the dev repo.
+Homepage = "https://github.com/dmitrii-khizbullin/stopslop"
+Source = "https://github.com/dmitrii-khizbullin/stopslop"
+
+[dependency-groups]
+dev = ["pytest>=7", "build>=1.0", "twine>=5"]
+
+[tool.hatch.version]
+path = "src/stopslop/__init__.py"
+
+[tool.hatch.build.targets.wheel]
+packages = ["src/stopslop"]
+
+# No force-include needed: src/stopslop/data is inside the package directory, so
+# hatchling ships it with the package. Forcing it as well adds every file twice
+# and the build fails on the duplicate path.
+
+[tool.hatch.build]
+exclude = ["*.pdf", "**/__pycache__"]
+
+[tool.hatch.build.targets.sdist]
+# The sdist carries the code, the baked baselines, the tests and the README.
+#
+# IT DOES NOT CARRY THE PROSE. `*.md` used to be included, which swept in
+# UNSLOP.md, COMPETITION.md and reports/ -- the strategy journal, the competitor
+# analysis and an audit of an unpublished submission. This repository is the
+# development repository; the public one is a separate, cleaned repository under
+# the released name. A PyPI upload is permanent and mirrored, so the split has
+# to be enforced here rather than remembered at upload time.
+include = ["src/stopslop", "tests", "README.md", "LICENSE", "analytics-requirements.txt"]
+exclude = ["*.pdf", "*.png", "**/__pycache__"]
+
+[tool.pytest.ini_options]
+testpaths = ["tests"]

stopslop-0.1.0/src/stopslop/__init__.py

@@ -0,0 +1,14 @@
+"""stopslop -- measure a paper's prose against how a venue actually writes.
+
+Two baselines, always both: `fossil` (papers published up to and including 2022,
+which cannot have been machine-assisted) and `current` (the last full year plus
+the year in progress). A metric that has drifted across the whole field is not
+evidence about one paper, and only the second baseline can show that.
+"""
+__version__ = "0.1.0"
+
+FOSSIL_MAX_YEAR = 2022      # ChatGPT: 2022-11-30
+
+#: Sent on every outbound request. Defined once: it was hardcoded as
+#: "stopslop/0.1" in two files and was already wrong.
+USER_AGENT = f"stopslop/{__version__}"

stopslop-0.1.0/src/stopslop/__main__.py

@@ -0,0 +1,2 @@
+from .cli import main
+raise SystemExit(main())