badwords: re-sync with curl, fix fallouts (#510)

Author: vszakats

.github/scripts/badwords.pl

@@ -20,7 +20,12 @@
 my @whitelist;
 my %alt;
 my %exactcase;
+my $skip_indented = 1;
 
+if($ARGV[0] eq "-a") {
+    shift @ARGV;
+    $skip_indented = 0;
+}
 my %wl;
 if($ARGV[0] eq "-w") {
     shift @ARGV;
@@ -68,7 +73,7 @@ sub file {
         my $in = $_;
         $l++;
         chomp $in;
-        if($in =~ /^    /) {
+        if($skip_indented && $in =~ /^    /) {
             next;
         }
         # remove the link part
@@ -94,7 +99,7 @@ sub file {
                 }
 
                 print STDERR  "$f:$l:$c: error: found bad word \"$w\"\n";
-                printf STDERR " %4d | $in\n", $l;
+                printf STDERR " %4d | %s\n", $l, $in;
                 printf STDERR "      | %*s^%s\n", length($p), " ",
                     "~" x (length($w)-1);
                 printf STDERR " maybe use \"%s\" instead?\n", $alt{$w};

.github/scripts/badwords.txt

@@ -10,59 +10,69 @@ tool chain:toolchain
 tool-chain:toolchain
 wild-card:wildcard
 wild card:wildcard
+\bit's:it is
+aren't:are not
+can't:cannot
+could've:could have
+couldn't:could not
+didn't:did not
+doesn't:does not
+don't=do not
+haven't:have not
+i'd:I would
+i'll:I will
 i'm:I am
-you've:You have
-we've:we have
-we're:we are
-we'll:we will
-we'd:we would
-they've:They have
-they're:They are
-they'll:They will
+i've:I have
+isn't:is not
+it'd:it would
+it'll:it will
+might've:might have
+needn't:need not
+should've:should have
+shouldn't:should not
+that's:that is
+there's:there is
 they'd:They would
-you've:you have
+they'll:They will
+they're:They are
+they've:They have
+this'll:this will
+wasn't:was not
+we'd:we would
+we'll:we will
+we're:we are
+we've:we have
+weren't:were not
+won't:will not
+would've:would have
+wouldn't:would not
 you'd:you would
 you'll:you will
 you're:you are
-should've:should have
-don't=do not
-could've:could have
-doesn't:does not
-isn't:is not
-aren't:are not
+you've:you have
  a html: an html
  a http: an http
  a ftp: an ftp
  a IPv4: an IPv4
  a IPv6: an IPv6
- url =URL
+ url [^=]=URL
 [^/]internet\b=Internet
 isation:ization
-\bit's:it is
-it'd:it would
-there's:there is
 [^.]\. And: Rewrite it somehow?
 ^(And|So|But) = Rewrite it somehow?
 \. But: Rewrite it somehow?
 \. So : Rewrite without "so" ?
- dir :directory
+ dir [^=]=directory
+ Dir [^=]=Directory
 sub-director:subdirector
-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 name\b:filename
 file names\b:filenames
 \buser name\b:username
 \buser names\b:usernames
 \bpass phrase:passphrase
-didn't:did not
-doesn't:does not
-won't:will not
-couldn't:could not
 \bwill\b:rewrite to present tense
 \b32bit=32-bit
 \b64bit=64-bit
@@ -78,5 +88,5 @@ couldn't:could not
 ---WWW::Curl
 ---NET::Curl
 ---Curl Corporation
-\bmanpages[^./&:-]:man pages
-\bmanpage[^si./&:-]:man page
+\bmanpages[^./;=&{:-]:man pages
+\bmanpage[^si./;=&{:-]:man page

_changes.html

@@ -16825,7 +16825,7 @@
 THISBOX(2.7)
 <p> Changes:
 <ul class="changes">
- CHG Removed the -l option and introduced the -f option instead. Now I'll
+ CHG Removed the -l option and introduced the -f option instead. Now I
    rewrite the former -l kludge in an external script that uses urlget to
    fetch multipart files like that.
  CHG '-f' is introduced, it means Fail without output in case of HTTP server

docs/survey/_2014.html

@@ -104,10 +104,10 @@
 <p>
 <img align="left" src="charts/chart_003.png">
 <pre>
-I've filed bug report(s)				47	25%
-I've sent patch(es)					56	29%
-I've responded to questions on mailing list / forums	43	23%
-I've helped out in other ways				38	20%
+I have filed bug report(s)				47	25%
+I have sent patch(es)					56	29%
+I have responded to questions on mailing list / forums	43	23%
+I have helped out in other ways				38	20%
 I run tests or provide infrastructure			7	4%
 </pre>
 </div>
@@ -281,7 +281,7 @@
 
 <div style="clear:left;">
 SUBTITLE(If you miss support for any protocol tell us which!)
-<p> Since this was a plain text field, I've parsed and edited the responses
+<p> Since this was a plain text field, I have parsed and edited the responses
  and then counted the number of times the same one occurred. Only a small
  portion of the participants filled in this field.