normalize text by sorting lines (#279)

Author: danvk

testdata/shuffle-add/left/words.txt

@@ -0,0 +1,40 @@
+boilersuit
+indictments
+dexes
+swacked
+remorid
+radioallergosorbent
+prelaw
+deploring
+morgens
+thriftlessness
+polygons
+overpromise
+militarizations
+straddler
+stealings
+gimmickry
+unabbreviated
+hexagram
+rumbly
+lightheartednesses
+largeness
+monumentalizing
+strongbox
+waif
+derivers
+stubborn
+disenthral
+sorites
+cactoid
+leukocytic
+flabelliform
+dockage
+necrologist
+ridicules
+renouncements
+composedly
+undergraduate
+nonphotographic
+unendurably
+sniffier

testdata/shuffle-add/right/words.txt

@@ -0,0 +1,41 @@
+strongbox
+boilersuit
+prelaw
+sorites
+swacked
+dockage
+remorid
+composedly
+sniffier
+largeness
+necrologist
+polygons
+stubborn
+disenthral
+undergraduate
+waif
+overpromise
+morgens
+unabbreviated
+indictments
+lightheartednesses
+dexes
+ridicules
+deploring
+webdiff
+renouncements
+gimmickry
+radioallergosorbent
+monumentalizing
+militarizations
+nonphotographic
+thriftlessness
+flabelliform
+derivers
+rumbly
+leukocytic
+straddler
+unendurably
+stealings
+hexagram
+cactoid

ts/codediff/NormalizeJSONOption.tsx

@@ -11,7 +11,7 @@ export interface NormalizeJSONOptionProps {
 export function NormalizeJSONOption(props: NormalizeJSONOptionProps) {
   const {filePair} = props;
   const language = guessLanguageUsingFileName(filePair.a || filePair.b);
-  if (language !== 'json') {
+  if (language !== 'json' && language !== 'txt') {
     return null;
   }
 
@@ -27,7 +27,11 @@ export function NormalizeJSONOption(props: NormalizeJSONOptionProps) {
         onChange={toggleNormalizeJSON}
         id="normalize-json"
       />{' '}
-      <label htmlFor="normalize-json">Normalize JSON (z): indent, sort keys</label>
+      <label htmlFor="normalize-json">
+        {language === 'json'
+          ? 'Normalize JSON (z): indent, sort keys'
+          : 'Normalize text (z): sort lines'}
+      </label>
     </div>
   );
 }

webdiff/util.py

@@ -179,15 +179,29 @@ def get_pdiff_bbox(diff_path):
     }
 
 
+def normalize_text(in_path: str):
+    data = [*open(in_path)]
+    data.sort()
+    _, norm_path = tempfile.mkstemp(suffix='.txt')
+    with open(norm_path, 'w') as out:
+        out.writelines(data)
+    logging.debug(f'Normalized text file {in_path} -> {norm_path}')
+    return norm_path
+
+
 @memoize
 def normalize_json(in_path: str):
+    if in_path.lower().endswith('.txt'):
+        return normalize_text(in_path)
+
     with open(in_path) as f:
         try:
             data = json.load(f)
         except json.JSONDecodeError:
             # This would be a good place to try parsing as JSON5/JSONC.
             logging.debug(f'Unable to parse {in_path} as JSON')
             return in_path
+
     _, norm_path = tempfile.mkstemp(suffix='.json')
     with open(norm_path, 'w') as out:
         json.dump(data, out, indent=2, sort_keys=True)