Supply a Git pre-commit hook for tools/check-typo

Automatically runs tools/check-typo and rejects the commit if they don't
pass.

.gitattributes

@@ -135,6 +135,7 @@ tools/make-package-macosx text eol=lf
 tools/ocaml-objcopy-macosx text eol=lf
 tools/ocamlmktop.tpl text eol=lf
 tools/ocamlsize text eol=lf
+tools/pre-commit-githook text eol=lf
 
 # These two are cat scripts, so may not actually require this
 config/auto-aux/sharpbang text eol=lf

CONTRIBUTING.md

@@ -68,6 +68,9 @@ columns, not use tab characters (spaces only), and not use non-ASCII
 characters. These typographical rules can be checked with the script
 `tools/check-typo`.
 
+If you are working from a Git clone, you can automate this process by
+copying the file `tools/pre-commit-githook` to `.git/hooks/pre-commit`.
+
 Otherwise, there are no strongly enforced guidelines specific to the
 compiler -- and, as a result, the style may differ in the different
 parts of the compiler. The general [OCaml Programming

tools/check-typo

@@ -91,7 +91,7 @@ export LC_ALL=C
 # Special case for recursive call from the find command (see IGNORE_DIRS).
 case "$1" in
   --check-prune)
-    case `git check-attr ocaml-typo "$2" 2>/dev/null` in
+    case `git check-attr $OCAML_CT_CA_FLAG ocaml-typo "$2" 2>/dev/null` in
       *prune*) echo "INFO: pruned directory $2 (ocaml-typo=prune)" >&2; exit 0;;
       *) exit 3;;
     esac;;
@@ -113,6 +113,8 @@ while : ; do
   esac
 done
 
+OCAML_CT_CAT=${OCAML_CT_CAT:-cat}
+
 IGNORE_DIRS="
   -name .git -prune -o
   -type d -exec $0 --check-prune {} ; -prune -o
@@ -149,7 +151,8 @@ IGNORE_DIRS="
       EMPTY=`git hash-object -t tree /dev/null`
       git diff-tree --numstat $EMPTY HEAD -- "$f" | grep -q "^-[[:blank:]]-" \
         && continue
-      svnrules=`git check-attr ocaml-typo "$f" | sed -e 's/.*: //'`
+      svnrules=`git check-attr $OCAML_CT_CA_FLAG ocaml-typo "$f" \
+                  | sed -e 's/.*: //'`
       case $svnrules in unspecified) svnrules= ;; esac
     fi
     rules="$userrules"
@@ -169,14 +172,15 @@ IGNORE_DIRS="
         # requires the entire line to match. This specifically detects the
         # presence of lines containing malformed UTF-8. It may be tested using
         # https://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-test.txt
-        if LC_ALL=en_US.UTF8 grep -qaxv '.*' "$f" ; then
+        if $OCAML_CT_CAT "$OCAML_CT_PREFIX$f" \
+             | LC_ALL=en_US.UTF8 grep -qaxv '.*' ; then
           echo File "$f" is not correctly encoded in UTF-8
           exit 2
         fi
         ;;
     esac
 
-    (cat "$f" | tr -d '\r'; echo) \
+    ($OCAML_CT_CAT "$OCAML_CT_PREFIX$f" | tr -d '\r'; echo) \
     | awk -v rules="$rules" -v svnrules="$svnrules" -v file="$f" \
       '
         function is_err(name) {

tools/pre-commit-githook

@@ -0,0 +1,49 @@
+#!/bin/sh
+#**************************************************************************
+#*                                                                        *
+#*                                 OCaml                                  *
+#*                                                                        *
+#*                David Allsopp, MetaStack Solutions Ltd.                 *
+#*                                                                        *
+#*   Copyright 2017 MetaStack Solutions Ltd.                              *
+#*                                                                        *
+#*   All rights reserved.  This file is distributed under the terms of    *
+#*   the GNU Lesser General Public License version 2.1, with the          *
+#*   special exception on linking described in the file LICENSE.          *
+#*                                                                        *
+#**************************************************************************
+
+# For what it's worth, allow for empty trees!
+if git rev-parse --verify HEAD >/dev/null 2>&1
+then
+  against=HEAD
+else
+  # Initial commit: diff against an empty tree object
+  against=4b825dc642cb6eb9a060e54bf8d69288fbee4904
+fi
+
+# Redirect output to stderr.
+exec 1>&2
+
+# Git's built-in mechanism for whitespace is neater than ours, so do it first.
+# The strange construction below creates a list of files which have either
+# white-at-eol or white-at-eof included in ocaml-typo in .gitattributes and by
+# prefixing the names with :! causes git diff-index to skip over them.
+FILES=$(git diff-index --cached --name-only $against \
+          | xargs git check-attr --cached ocaml-typo \
+          | sed -ne 's/\(.*\): ocaml-typo:.*[ ,]white-at-eo[fl]\(,\|$\)/:!\1/p')
+if ! git diff-index --check --cached $against -- $FILES ; then
+  exit 1
+fi
+
+# Now run check-typo over all the files in the index
+ERRORS=0
+export OCAML_CT_PREFIX=:
+export OCAML_CT_CAT="git cat-file --textconv"
+export OCAML_CT_CA_FLAG=--cached
+git diff --staged --name-only | (while IFS= read -r path
+do
+  if ! tools/check-typo $path ; then
+    ERRORS=1
+  fi
+done; exit $ERRORS)