draft initial commit

.github/workflows/spelling.yml

@@ -0,0 +1,17 @@
+name: Spell checking
+on: [pull_request, push]
+
+jobs:
+  build:
+    name: Spell checker
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v2.0.0
+      with:
+        fetch-depth: 2
+    - uses: ./
+      env:
+        bucket: ssh://git@github.com/check-spelling/check-spelling.git
+        project: spelling-data
+        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        DEBUG: 1

Dockerfile

@@ -0,0 +1,26 @@
+FROM debian:9.5-slim
+
+RUN DEBIAN_FRONTEND=noninteractive apt-get -qq update < /dev/null > /dev/null
+RUN DEBIAN_FRONTEND=noninteractive apt-get install -qq curl git jq < /dev/null > /dev/null
+WORKDIR /app
+COPY docker-setup setup
+COPY reporter.json reporter.json
+COPY reporter.pl reporter.pl
+COPY w spelling-unknown-word-splitter.pl
+RUN ./setup
+RUN rm setup
+
+LABEL "com.github.actions.name"="Spell Checker"
+LABEL "com.github.actions.description"="Check repository for spelling errors"
+LABEL "com.github.actions.icon"="edit-3"
+LABEL "com.github.actions.color"="red"
+
+LABEL "repository"="http://github.com/jsoref/spelling-action"
+LABEL "homepage"="http://github.com/jsoref/spelling-action/tree/master/README.md"
+LABEL "maintainer"="Josh Soref <jsoref@noreply.users.github.com>"
+
+COPY test-spelling-unknown-words test-spelling-unknown-words.sh
+COPY exclude exclude.pl
+
+RUN chmod +x test-spelling-unknown-words.sh exclude.pl spelling-unknown-word-splitter.pl
+ENTRYPOINT ["/app/test-spelling-unknown-words.sh"]

README.md

@@ -0,0 +1,55 @@
+# Spelling tools
+
+## Overview
+
+Everyone makes typos. This includes people writing documentation and comments,
+but it also includes programmers naming variables, functions, apis, classes,
+and filenames.
+
+Often, programmers will use `InitialCapitalization`, `camelCase`,
+`ALL_CAPS`, or `IDLCase` when naming their things. When they do this, it makes
+it much harder for naive spelling tools to recognize misspellings, and as such,
+with a really high false-positive rate, people don't tend to enable spellchecking
+at all.
+
+This repository's tools are capable of tolerating all of those variations.
+Specifically, [w](https://github.com/jsoref/spelling/blob/master/w) understands
+enough about how programmers name things that it can split the above conventions
+into word-like things for checking against a dictionary.
+
+## Spell Checker GitHub Actions
+
+[![Spell checking](https://github.com/check-spelling/check-spelling/workflows/Spell%20checking/badge.svg?branch=master)](https://github.com/check-spelling/check-spelling/actions?query=workflow:"Spell+checking"+branch:master)
+
+[More information](https://github.com/jsoref/spelling#overview)
+
+### Required Configuration Variables
+
+
+| Variable | Description |
+| ------------- | ------------- |
+| bucket | a `gsutil` or `curl` compatible url for which the tool has read access to a couple of files. |
+| project      | a folder within `bucket`. This allows you to share common items across projects. |
+| GITHUB_TOKEN | Secret used to retrieve your code. |
+
+### Optional Configuration Variables
+
+| Variable | Description |
+| ------------- | ------------- |
+| VERBOSE | `1` if you want to be reminded of how many words are in your whitelist for each run. |
+
+## Behavior
+
+* This action will automatically comment on PRs / commits with its opinion.
+* It will try to identify a limited number of lines containing the words it
+doesn't recognize.
+
+## Limitations
+
+* It will not add comments to the first version of a file - if you want
+to ensure it assigns blame, insert a commit with an empty file before you
+fill the file.
+
+# License
+
+MIT

action.yml

@@ -0,0 +1,24 @@
+name: 'Spell checking'
+description: 'Spell check commits'
+author: 'jsoref'
+branding:
+  icon: 'edit-3'
+  color: 'red'
+inputs:
+  repo-token:
+    description: 'The GITHUB_TOKEN secret'
+  bucket:
+    description: 'Container for spelling exclusions and whitelist'
+    required: true
+  project:
+    description: 'Folder/Branch containing exclusions/whitelist'
+    required: true
+
+  debug:
+    description: 'Debug'
+runs:
+  using: 'docker'
+  image: 'Dockerfile'
+  env:
+    bucket: ${{ inputs.bucket }}
+    project: ${{ inputs.project }}

docker-setup

@@ -0,0 +1,26 @@
+#!/bin/bash
+# This CI acceptance test is based on:
+# https://github.com/jsoref/spelling/tree/04648bdc63723e5cdf5cbeaff2225a462807abc8
+# It is conceptually `f` which runs `w` (spelling-unknown-word-splitter)
+# plus `fchurn` which uses `dn` mostly rolled together.
+set -e
+
+spellchecker='/app'
+w_location="$spellchecker/w"
+temp='/tmp/spelling'
+dict="$spellchecker/words"
+word_splitter="$spellchecker/spelling-unknown-word-splitter.pl"
+run_output="$spellchecker/unknown.words.txt"
+
+wordlist=https://github.com/check-spelling/check-spelling/raw/dictionary/dict.txt
+
+mkdir -p "$temp"
+if [ ! -e "$dict" ]; then
+  echo "Retrieving cached $(basename "$wordlist")"
+  # english.words is taken from rpm:
+  # https://rpmfind.net/linux/fedora/linux/development/rawhide/Everything/aarch64/os/Packages/w/"
+  # "words-.*.noarch.rpm"
+  (
+    curl -L -s "$wordlist" -o "$dict"
+  ) >/dev/null 2>/dev/null
+fi

exclude

@@ -0,0 +1,20 @@
+#!/usr/bin/perl
+# This script takes null delimited files as input
+# it drops paths that match the listed exclusions
+# output is null delimited to match input
+use File::Basename;
+my $dirname = dirname(__FILE__);
+
+my @excludes;
+open EXCLUDES, '<', $dirname.'/excludes.txt';
+while (<EXCLUDES>) {
+  s/^\s*(.*)\s*$/$1/;
+  push @excludes, $_;
+}
+$/="\0";
+my $exclude = scalar @excludes ? join "|", @excludes : '^$';
+while (<>) {
+  chomp;
+  next if m{$exclude};
+  print "$_$/";
+}

reporter.json

@@ -0,0 +1,20 @@
+{
+    "problemMatcher": [
+        {
+            "owner": "jsoref-spelling",
+            "pattern": [
+                {
+                    "regexp": "^(.+):[\\s]line\\s(\\d+),[\\s]columns\\s(\\d+)-(\\d+),\\s(Error|Warning|Info)\\s-\\s(.+)\\s\\((.+)\\)$",
+                    "file": 1,
+                    "line": 2,
+                    "column": 3,
+                    "endColumn": 4,
+                    "severity": 5,
+                    "message": 6,
+                    "code": 7
+                }
+            ]
+        }
+    ]
+}
+

reporter.pl

@@ -0,0 +1,35 @@
+#!/usr/bin/env perl
+
+die 'Please set $tokens' unless defined $ENV{tokens};
+my $tokens=$ENV{tokens};
+exit 0 unless $tokens =~ /\w/;
+$tokens=~ s/\s+/|/g;
+my $re = "\\b($tokens)\\b";
+my $blame=defined $ENV{with_blame};
+
+my $previous='';
+my $first_line=0;
+while (<>) {
+  my $line;
+  if ($blame) {
+    next if /^ /;
+    s/^[0-9a-f^]+\s+(.*?)\s(\d+)\) //;
+    ($ARGV, $line) = ($1, $2);
+  } else {
+    if ($previous ne $ARGV) {
+      $previous=$ARGV;
+      $first_line = $. - 1;
+    }
+    $line = $. - $first_line;
+  }
+  if ($blame) {
+    next if /^ /;
+    s/^[0-9a-f^]+\s+\d+\) //;
+  }
+  next unless $_ =~ /$re/;
+  while (/$re/g) {
+    my ($start, $token) = (1 + length $`, $1);
+    my $stop = $start + (length $token) - 1;
+    print "$ARGV: line $line, columns $start-$stop, Warning - '$token' is not a recognized word. (unrecognized-spelling)\n";
+  }
+}