[SPARK-2912] [Spark QA] Include commit hash in Spark QA messages

You can find the [discussion that motivated this PR here](http://mail-archives.apache.org/mod_mbox/spark-dev/201408.mbox/%3CCABPQxssy0ri2QAz=cc9Tx+EXYWARm7pNcVm8apqCwc-esLbO4Qmail.gmail.com%3E).

As described in [SPARK-2912](https://issues.apache.org/jira/browse/SPARK-2912), the goal of this PR (and related ones to come) is to include useful detail in Spark QA's messages that are intended to make a committer's job easier to do.

Since this work depends on Jenkins, I cannot test this locally. Hence, I will be iterating via this PR.

Notes:
* This is a duplicate of a [previous PR](https://github.com/apache/spark/pull/1811), without the extraneous commits.
* This PR also resolves an issue targeted by [another open PR](https://github.com/apache/spark/pull/1809).

Closes #1809.

Author: Nicholas Chammas <nicholas.chammas@gmail.com>
Author: nchammas <nicholas.chammas@gmail.com>

Closes #1816 from nchammas/master and squashes the following commits:

c1be644 [Nicholas Chammas] [SPARK-2912] include commit hash in messages
8f641ac [nchammas] Merge pull request #7 from apache/master

1 files changed, 138 insertions, 49 deletions

diff --git a/dev/run-tests-jenkins b/dev/run-tests-jenkins
index 3076eb847b..721f09be5b 100755
--- a/dev/run-tests-jenkins
+++ b/dev/run-tests-jenkins
@@ -19,67 +19,156 @@
 
 # Wrapper script that runs the Spark tests then reports QA results
 # to github via its API.
+# Environment variables are populated by the code here:
+#+ https://github.com/jenkinsci/ghprb-plugin/blob/master/src/main/java/org/jenkinsci/plugins/ghprb/GhprbTrigger.java#L139
 
 # Go to the Spark project root directory
 FWDIR="$(cd `dirname $0`/..; pwd)"
 cd "$FWDIR"
 
+function get_jq () {
+  # Get jq so we can parse some JSON, man.
+  # Essential if we want to do anything with the GitHub API responses.
+  local JQ_EXECUTABLE_URL="http://stedolan.github.io/jq/download/linux64/jq"
+
+  echo "Fetching jq from ${JQ_EXECUTABLE_URL}"
+  
+  curl --silent --output "$FWDIR/dev/jq" "$JQ_EXECUTABLE_URL"
+  local curl_status=$?
+
+  if [ $curl_status -ne 0 ]; then
+      echo "Failed to get jq." >&2
+      return $curl_status
+  fi
+
+  chmod u+x "$FWDIR/dev/jq"
+}
+
 COMMENTS_URL="https://api.github.com/repos/apache/spark/issues/$ghprbPullId/comments"
+PULL_REQUEST_URL="https://github.com/apache/spark/pull/$ghprbPullId"
+
+function post_message () {
+  local message=$1
+  local data="{\"body\": \"$message\"}"
+  local HTTP_CODE_HEADER="HTTP Response Code: "
+  
+  echo "Attempting to post to Github..."
+  
+  local curl_output=$(
+    curl `#--dump-header -` \
+      --silent \
+      --user x-oauth-basic:$GITHUB_OAUTH_KEY \
+      --request POST \
+      --data "$data" \
+      --write-out "${HTTP_CODE_HEADER}%{http_code}\n" \
+      --header "Content-Type: application/json" \
+      "$COMMENTS_URL" #> /dev/null #| "$FWDIR/dev/jq" .id #| head -n 8
+  )
+  local curl_status=${PIPESTATUS[0]}
+
+  if [ "$curl_status" -ne 0 ]; then
+      echo "Failed to post message to GitHub." >&2
+      echo " > curl_status: ${curl_status}" >&2
+      echo " > curl_output: ${curl_output}" >&2
+      echo " > data: ${data}" >&2
+      # exit $curl_status
+  fi
+  
+  local api_response=$(
+    echo "${curl_output}" \
+    | grep -v -e "^${HTTP_CODE_HEADER}"
+  )
+  
+  local http_code=$(
+    echo "${curl_output}" \
+    | grep -e "^${HTTP_CODE_HEADER}" \
+    | sed -r -e "s/^${HTTP_CODE_HEADER}//g"
+  )
+
+  if [ -n "$http_code" ] && [ "$http_code" -ne "201" ]; then
+      echo " > http_code: ${http_code}." >&2
+      echo " > api_response: ${api_response}" >&2
+      echo " > data: ${data}" >&2
+  fi
+  
+  if [ "$curl_status" -eq 0 ] && [ "$http_code" -eq "201" ]; then
+    echo " > Post successful."
+  fi
+}
+
+COMMIT_URL="https://github.com/apache/spark/commit/${ghprbActualCommit}"
+# GitHub doesn't auto-link short hashes when submitted via the API, unfortunately. :(
+short_commit_hash=${ghprbActualCommit:0:7}
+
+# check PR merge-ability and check for new public classes
+{
+  if [ "$sha1" == "$ghprbActualCommit" ]; then
+    merge_note=" * This patch **does not** merge cleanly!"
+  else
+    merge_note=" * This patch merges cleanly."
+
+    non_test_files=$(git diff master --name-only | grep -v "\/test" | tr "\n" " ")
+    new_public_classes=$(
+        git diff master ${non_test_files}       `# diff this patch against master and...` \
+      | grep "^\+"                              `# filter in only added lines` \
+      | sed -r -e "s/^\+//g"                    `# remove the leading +` \
+      | grep -e "trait " -e "class "            `# filter in lines with these key words` \
+      | grep -e "{" -e "("                      `# filter in lines with these key words, too` \
+      | grep -v -e "\@\@" -e "private"          `# exclude lines with these words` \
+      | grep -v -e "^// " -e "^/\*" -e "^ \* "  `# exclude comment lines` \
+      | sed -r -e "s/\{.*//g"                   `# remove from the { onwards` \
+      | sed -r -e "s/\}//g"                     `# just in case, remove }; they mess the JSON` \
+      | sed -r -e "s/\"/\\\\\"/g"               `# escape double quotes; they mess the JSON` \
+      | sed -r -e "s/^(.*)$/\`\1\`/g"           `# surround with backticks for style` \
+      | sed -r -e "s/^/  \* /g"                 `# prepend '  *' to start of line` \
+      | sed -r -e "s/$/\\\n/g"                  `# append newline to end of line` \
+      | tr -d "\n"                              `# remove actual LF characters`
+    )
 
-function post_message {
-  message=$1
-  data="{\"body\": \"$message\"}"
-  echo "Attempting to post to Github:"
-  echo "$data"
+    if [ "$new_public_classes" == "" ]; then
+      public_classes_note=" * This patch adds no public classes."
+    else
+      public_classes_note=" * This patch adds the following public classes _(experimental)_:"
+      public_classes_note="${public_classes_note}\n${new_public_classes}"
+    fi
+  fi
+}
 
-  curl -D- -u x-oauth-basic:$GITHUB_OAUTH_KEY -X POST --data "$data" -H \
-    "Content-Type: application/json" \
-    $COMMENTS_URL | head -n 8
+# post start message
+{
+  start_message="\
+  [QA tests have started](${BUILD_URL}consoleFull) for \
+  PR $ghprbPullId at commit [\`${short_commit_hash}\`](${COMMIT_URL})."
+  
+  start_message="${start_message}\n${merge_note}"
+  # start_message="${start_message}\n${public_classes_note}"
+  
+  post_message "$start_message"
 }
 
-start_message="QA tests have started for PR $ghprbPullId."
-if [ "$sha1" == "$ghprbActualCommit" ]; then
-  start_message="$start_message This patch DID NOT merge cleanly! "
-else
-  start_message="$start_message This patch merges cleanly. "
-fi
-start_message="$start_message<br>View progress: "
-start_message="$start_message${BUILD_URL}consoleFull"
-
-post_message "$start_message"
-
-./dev/run-tests
-test_result="$?"
-
-result_message="QA results for PR $ghprbPullId:<br>"
-
-if [ "$test_result" -eq "0" ]; then
-  result_message="$result_message- This patch PASSES unit tests.<br>"
-else
-  result_message="$result_message- This patch FAILED unit tests.<br>"
-fi
-
-if [ "$sha1" != "$ghprbActualCommit" ]; then
-  result_message="$result_message- This patch merges cleanly<br>"
-  non_test_files=$(git diff master --name-only | grep -v "\/test" | tr "\n" " ")
-  new_public_classes=$(git diff master $non_test_files \
-    | grep -e "trait " -e "class " \
-    | grep -e "{" -e "("  \
-    | grep -v -e \@\@ -e private \
-    | grep \+ \
-    | sed "s/\+ *//" \
-    | tr "\n" "~" \
-    | sed "s/~/<br>/g")
-  if [ "$new_public_classes" == "" ]; then
-    result_message="$result_message- This patch adds no public classes<br>"
+# run tests
+{
+  ./dev/run-tests
+  test_result="$?"
+
+  if [ "$test_result" -eq "0" ]; then
+    test_result_note=" * This patch **passes** unit tests."
   else
-    result_message="$result_message- This patch adds the following public classes (experimental):<br>"
-    result_message="$result_message$new_public_classes"
+    test_result_note=" * This patch **fails** unit tests."
   fi
-fi
-result_message="${result_message}<br>For more information see test ouptut:"
-result_message="${result_message}<br>${BUILD_URL}consoleFull"
+}
 
-post_message "$result_message"
+# post end message
+{
+  result_message="\
+  [QA tests have finished](${BUILD_URL}consoleFull) for \
+  PR $ghprbPullId at commit [\`${short_commit_hash}\`](${COMMIT_URL})."
+
+  result_message="${result_message}\n${test_result_note}"
+  result_message="${result_message}\n${merge_note}"
+  result_message="${result_message}\n${public_classes_note}"
+
+  post_message "$result_message"
+}
 
 exit $test_result