Use phutil_split_lines() in Differential

Summary:
  - We currently treat "\r" as a newline, but should not because VCSes do not.
  - We get an extra empty line at the end of diffs created after D3442 because we now retain newlines.
  - Historically we've converted tab pre-cache, but do it post-cache instead so we can add prefs about it, as we should handle it better than we do (e.g., let the user set it to a different width, infer width from comments in the file, expand it to actual tab stops, or show it visually in some way).

Test Plan:
  - Verified diffs no longer have an empty line at the end.
  - Created a diff of a "\r" file and verified it displayed somewhat reasonably. All browsers treat "\r" as a real newline so it's not necessarily perfect, but we can clean that up later. Hopefully these files are exceedingly rare.
  - Created a file with tabs and verified it came out reasonably.

Reviewers: vrana, btrahan

Reviewed By: vrana

CC: aran

Maniphest Tasks: T1857

Differential Revision: https://secure.phabricator.com/D3759

src/applications/differential/parser/DifferentialChangesetParser.php

@@ -60,7 +60,7 @@ final class DifferentialChangesetParser {
   private $markupEngine;
   private $highlightErrors;
 
-  const CACHE_VERSION = 6;
+  const CACHE_VERSION = 7;
   const CACHE_MAX_SIZE = 8e6;
 
   const ATTR_GENERATED  = 'attr:generated';
@@ -132,8 +132,8 @@ final class DifferentialChangesetParser {
     foreach ($changeset->getHunks() as $hunk) {
       $n_old = $hunk->getOldOffset();
       $n_new = $hunk->getNewOffset();
-      $changes = rtrim($hunk->getChanges(), "\n");
+      $changes = phutil_split_lines($hunk->getChanges());
-      foreach (explode("\n", $changes) as $line) {
+      foreach ($changes as $line) {
         $diff_type = $line[0]; // Change type in diff of diffs.
         $orig_type = $line[1]; // Change type in the original diff.
         if ($diff_type == ' ') {
@@ -267,12 +267,7 @@ final class DifferentialChangesetParser {
 
   public function parseHunk(DifferentialHunk $hunk) {
     $lines = $hunk->getChanges();
-
+    $lines = phutil_split_lines($lines);
-    $lines = str_replace(
-      array("\t", "\r\n", "\r"),
-      array('  ', "\n",   "\n"),
-      $lines);
-    $lines = explode("\n", $lines);
 
     $types = array();
     foreach ($lines as $line_index => $line) {
@@ -573,7 +568,7 @@ final class DifferentialChangesetParser {
         $old_corpus[] = $o['text'];
       }
     }
-    $old_corpus_block = implode("\n", $old_corpus);
+    $old_corpus_block = implode('', $old_corpus);
 
     $new_corpus = array();
     foreach ($this->new as $n) {
@@ -581,7 +576,7 @@ final class DifferentialChangesetParser {
         $new_corpus[] = $n['text'];
       }
     }
-    $new_corpus_block = implode("\n", $new_corpus);
+    $new_corpus_block = implode('', $new_corpus);
 
     $this->markGenerated($new_corpus_block);
 
@@ -600,6 +595,7 @@ final class DifferentialChangesetParser {
       'old' => $old_corpus_block,
       'new' => $new_corpus_block,
     );
+
     $this->highlightErrors = false;
     foreach (Futures($futures) as $key => $future) {
       try {
@@ -820,6 +816,14 @@ final class DifferentialChangesetParser {
   }
 
   protected function getHighlightFuture($corpus) {
+    if (preg_match('/\r(?!\n)/', $corpus)) {
+      // TODO: Pygments converts "\r" newlines into "\n" newlines, so we can't
+      // use it on files with "\r" newlines. If we have "\r" not followed by
+      // "\n" in the file, skip highlighting.
+      $result = phutil_escape_html($corpus);
+      return new ImmediateFuture($result);
+    }
+
     return $this->highlightEngine->getHighlightFuture(
       $this->highlightEngine->getLanguageFromFilename($this->filename),
       $corpus);
@@ -827,7 +831,7 @@ final class DifferentialChangesetParser {
 
   protected function processHighlightedSource($data, $result) {
 
-    $result_lines = explode("\n", $result);
+    $result_lines = phutil_split_lines($result);
     foreach ($data as $key => $info) {
       if (!$info) {
         unset($result_lines[$key]);
@@ -1774,13 +1778,20 @@ final class DifferentialChangesetParser {
       $notice = $this->renderChangeTypeHeader($this->changeset, false);
     }
 
-    return implode(
+    $result = implode(
       "\n",
       array(
         $notice,
         $props,
         $table,
       ));
+
+    // TODO: Let the user customize their tab width / display style.
+    $result = str_replace("\t", '  ', $result);
+
+    // TODO: We should possibly post-process "\r" as well.
+
+    return $result;
   }
 
   protected function renderChangeTypeHeader($changeset, $force) {