Parse escaped sequence strictly in CSV parser

This change alters the CSV parser behavior. The parser only resolves
  * String resource reference. ["@string/res" -> "<content_of_res>"]
  * Other occurrence of escape sequence will be intact. ["\x" -> "\x"]

Before this change, escape sequence in moreKeys string is parsed three
times. At first in parsing string resource, next in CSV parser, and at
last in KeySpecParser. So that representing single escape character
itself is a bit annoying, "\\\\\\\\".

Now we can represent single escape character itself in string resource by "\\\\".

Change-Id: Ib978e17b779cc82585eed8241ac3857508b14bc7

1 files changed, 11 insertions, 43 deletions

diff --git a/java/src/com/android/inputmethod/keyboard/internal/KeySpecParser.java b/java/src/com/android/inputmethod/keyboard/internal/KeySpecParser.java
index ba12676ad..a84b469ea 100644
--- a/java/src/com/android/inputmethod/keyboard/internal/KeySpecParser.java
+++ b/java/src/com/android/inputmethod/keyboard/internal/KeySpecParser.java
@@ -48,7 +48,7 @@ public class KeySpecParser {
     private static final char ESCAPE_CHAR = '\\';
     private static final char PREFIX_AT = '@';
     private static final char SUFFIX_SLASH = '/';
-    private static final String PREFIX_STRING = PREFIX_AT + "string";
+    private static final String PREFIX_STRING = PREFIX_AT + "string" + SUFFIX_SLASH;
     private static final char LABEL_END = '|';
     private static final String PREFIX_ICON = PREFIX_AT + "icon" + SUFFIX_SLASH;
     private static final String PREFIX_CODE = PREFIX_AT + "integer" + SUFFIX_SLASH;
@@ -293,13 +293,11 @@ public class KeySpecParser {
                 sb.append(res.getString(resId));
                 pos = end - 1;
             } else if (c == ESCAPE_CHAR) {
-                pos++;
                 if (sb != null) {
-                    sb.append(c);
-                    if (pos < size) {
-                        sb.append(text.charAt(pos));
-                    }
+                    // Append both escape character and escaped character.
+                    sb.append(text.substring(pos, Math.min(pos + 2, size)));
                 }
+                pos++;
             } else if (sb != null) {
                 sb.append(c);
             }
@@ -309,10 +307,7 @@ public class KeySpecParser {
 
     private static int searchResourceNameEnd(String text, int start) {
         final int size = text.length();
-        if (start >= size || text.charAt(start) != SUFFIX_SLASH) {
-            throw new RuntimeException("Resource name not specified");
-        }
-        for (int pos = start + 1; pos < size; pos++) {
+        for (int pos = start; pos < size; pos++) {
             final char c = text.charAt(pos);
             // String resource name should be consisted of [a-z_0-9].
             if ((c >= 'a' && c <= 'z') || c == '_' || (c >= '0' && c <= '9')) {
@@ -333,7 +328,6 @@ public class KeySpecParser {
             return new String[] { text };
         }
 
-        final StringBuilder sb = new StringBuilder();
         ArrayList<String> list = null;
         int start = 0;
         for (int pos = 0; pos < size; pos++) {
@@ -342,44 +336,18 @@ public class KeySpecParser {
                 if (list == null) {
                     list = new ArrayList<String>();
                 }
-                if (sb.length() == 0) {
-                    list.add(text.substring(start, pos));
-                } else {
-                    list.add(sb.toString());
-                    sb.setLength(0);
-                }
+                list.add(text.substring(start, pos));
                 // Skip comma
                 start = pos + 1;
-                continue;
-            }
-            // TODO: Only parse escaped comma. Other escaped character should be passed through
-            // with escaped character prefixed.
-            // Skip escaped sequence.
-            if (c == ESCAPE_CHAR) {
-                if (start == pos) {
-                    // Skip escaping comma at the beginning of the text.
-                    start++;
-                    pos++;
-                } else {
-                    if (start < pos && sb.length() == 0) {
-                        sb.append(text.substring(start, pos));
-                    }
-                    // Skip comma
-                    pos++;
-                    if (pos < size) {
-                        sb.append(text.charAt(pos));
-                    }
-                }
-            } else if (sb.length() > 0) {
-                sb.append(c);
+            } else if (c == ESCAPE_CHAR) {
+                // Skip escape character and escaped character.
+                pos++;
             }
         }
         if (list == null) {
-            return new String[] {
-                    sb.length() > 0 ? sb.toString() : text.substring(start)
-            };
+            return new String[] { text.substring(start) };
         } else {
-            list.add(sb.length() > 0 ? sb.toString() : text.substring(start));
+            list.add(text.substring(start));
             return list.toArray(new String[list.size()]);
         }
     }