GHA: detect "bad" words in documentation

As we do in the source repo

Closes #353

.github/scripts/badwords.ok

@@ -0,0 +1,4 @@
+# whitelisted uses of bad words
+# file:line:rule
+docs/CVE-2020-19909.md:32:\bwill\b
+docs/CVE-2023-38546.md:36:file name\b

.github/scripts/badwords.pl

@@ -0,0 +1,94 @@
+#!/usr/bin/perl
+# Copyright (C) Daniel Stenberg, <[email protected]>, et al.
+#
+# SPDX-License-Identifier: curl
+#
+# bad[:=]correct
+#
+# If separator is '=', the string will be compared case sensitively.
+# If separator is ':', the check is done case insensitively.
+#
+
+my %wl;
+if($ARGV[0] eq "-w") {
+    shift @ARGV;
+    my $file = shift @ARGV;
+    open(W, "<$file");
+    while(<W>) {
+        if(/^#/) {
+            # allow #-comments
+            next;
+        }
+        if(/^([^:]*):(\d+):(.*)/) {
+            $wl{"$1:$2:$3"}=1;
+            #print STDERR "whitelisted $1:$2:$3\n";
+        }
+    }
+    close(W);
+}
+
+my $w;
+while(<STDIN>) {
+    chomp;
+    if($_ =~ /^#/) {
+        next;
+    }
+    if($_ =~ /^([^:=]*)([:=])(.*)/) {
+        my ($bad, $sep, $better)=($1, $2, $3);
+        push @w, $bad;
+        $alt{$bad} = $better;
+        if($sep eq "=") {
+            $exactcase{$bad} = 1;
+        }
+    }
+}
+
+my $errors;
+
+sub file {
+    my ($f) = @_;
+    my $l = 0;
+    open(F, "<$f");
+    while(<F>) {
+        my $in = $_;
+        $l++;
+        chomp $in;
+        if($in =~ /^    /) {
+            next;
+        }
+        # remove the link part
+        $in =~ s/(\[.*\])\(.*\)/$1/g;
+        # remove backticked texts
+        $in =~ s/\`.*\`//g;
+        foreach my $w (@w) {
+            my $case = $exactcase{$w};
+            if(($in =~ /^(.*)$w/i && !$case) ||
+               ($in =~ /^(.*)$w/ && $case) ) {
+                my $p = $1;
+                my $c = length($p)+1;
+
+                my $ch = "$f:$l:$w";
+                if($wl{$ch}) {
+                    # whitelisted
+                    print STDERR "$ch found but whitelisted\n";
+                    next;
+                }
+
+                print STDERR  "$f:$l:$c: error: found bad word \"$w\"\n";
+                printf STDERR " %4d | $in\n", $l;
+                printf STDERR "      | %*s^%s\n", length($p), " ",
+                    "~" x (length($w)-1);
+                printf STDERR " maybe use \"%s\" instead?\n", $alt{$w};
+                $errors++;
+            }
+        }
+    }
+    close(F);
+}
+
+my @files = @ARGV;
+
+foreach my $each (@files) {
+    file($each);
+}
+exit $errors;

.github/scripts/badwords.txt

@@ -0,0 +1,50 @@
+# Copyright (C) Daniel Stenberg, <[email protected]>, et al.
+#
+# SPDX-License-Identifier: curl
+#
+back-end:backend
+e-mail:email
+run-time:runtime
+set-up:setup
+tool chain:toolchain
+tool-chain:toolchain
+wild-card:wildcard
+wild card:wildcard
+i'm:I am
+you've:You have
+they've:They have
+they're:They are
+should've:should have
+don't:do not
+could've:could have
+doesn't:does not
+isn't:is not
+ a html: an html
+ a http: an http
+ a ftp: an ftp
+ url =URL
+internet\b=Internet
+isation:ization
+it's:it is
+there's:there is
+[^.]\. And: Rewrite it somehow?
+^(And|So|But) = Rewrite it somehow?
+\. But: Rewrite it somehow?
+\. So : Rewrite without "so" ?
+ dir :directory
+you'd:you would
+you'll:you will
+can't:cannot
+that's:that is
+web page:webpage
+host name\b:hostname
+host names\b:hostnames
+file name\b:filename
+file names\b:filenames
+\buser name\b:username
+\buser names\b:usernames
+didn't:did not
+doesn't:does not
+won't:will not
+couldn't:could not
+\bwill\b:rewrite to present tense

.github/workflows/badwords.yml

@@ -0,0 +1,29 @@
+# Copyright (C) Daniel Stenberg, <[email protected]>, et al.
+#
+# SPDX-License-Identifier: curl
+
+name: badwords
+
+on:
+  # Trigger the workflow on push or pull requests, but only for the
+  # master branch
+  push:
+    branches:
+    - master
+    - '*/ci'
+  pull_request:
+    branches:
+    - master
+
+permissions: {}
+
+jobs:
+  check:
+
+    runs-on: ubuntu-latest
+
+    steps:
+    - uses: actions/checkout@a5ac7e51b41094c92402da3b24376905380afc29 # v4
+
+    - name: check
+      run: ./.github/scripts/badwords.pl -w ./.github/scripts/badwords.ok < .github/scripts/badwords.txt docs/*.md