value/src/main/java/com/google/auto/value/processor/Reformatter.java

/*
 * Copyright 2014 Google LLC
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */
package com.google.auto.value.processor;

import com.google.common.base.CharMatcher;

/**
 * Postprocessor that runs over the output of the template engine in order to make it look nicer.
 * Mostly, this involves removing surplus horizontal and vertical space.
 *
 * @author emcmanus@google.com (Éamonn McManus)
 */
class Reformatter {
  /**
   * Characters that might start a continuation line. Since Google Style requires splitting before
   * operators, we expect that a continuation line will begin with one of these (or be inside
   * parentheses). We've omitted {@code /} from this list for now since none of our templates splits
   * before one, and otherwise we'd have to handle {@code //} and {@code /*}.
   */
  private static final CharMatcher OPERATORS = CharMatcher.anyOf("+-*%&|^<>=?:.").precomputed();

  static String fixup(String s) {
    StringBuilder out = new StringBuilder();
    JavaScanner scanner = new JavaScanner(s);
    s = scanner.string();
    int len = s.length();
    for (int start = 0, previous = 0, braces = 0, parens = 0, end = 0;
        start < len;
        previous = start, start = end) {
      end = scanner.tokenEnd(start);
      // The tokenized string always ends with \n so we can usually look at s.charAt(end) without
      // worrying about going past the end of the string.
      switch (s.charAt(start)) {
        case '(':
          parens++;
          break;
        case ')':
          parens--;
          break;
        case '{':
          braces++;
          break;
        case '}':
          braces--;
          break;
        case ' ':
          // This token is a string of consecutive spaces that is not at the start of a line.
          // Consecutive spaces at the start of a line are attached to the previous newline, and
          // we delete spaces at the start of the first line. So we are going to compress this
          // into just one space, and we are going to delete it entirely if it follows '(' or
          // precedes a newline or one of the punctuation characters here.
          if (start > 0 && s.charAt(previous) != '(' && "\n.,;)".indexOf(s.charAt(end)) < 0) {
            out.append(' ');
          }
          continue;
        case '\n':
          // This token is a newline plus any following spaces (the indentation of the next line).
          // If it is followed by something other than a newline then we will output the
          // newline, and replace the following spaces by our computed indentation. Otherwise, the
          // token is part of a sequence of newlines but it is not the last one. If this is a
          // context where we delete blank lines, or if this is not the first new line in the
          // sequence, or if we are at the start of the file, we will delete this one. Otherwise we
          // will output a single newline with no following indentation. Contexts where we delete
          // blank lines are inside parentheses or inside more than one set of braces.
          if (end < len && s.charAt(end) != '\n') {
            // Omit newlines at the very start of the file. Also delete newline+indent between
            // ( and ), since that shows up in some places where we output one parameter per line,
            // when there are no parameters.
            char prev = s.charAt(previous);
            char next = s.charAt(end);
            if (out.length() > 0 && (prev != '(' || next != ')')) {
              out.append('\n');
              // Replace any space after the newline with our computed indentation. The algorithm
              // here is simplistic but works OK for our current templates.
              int indent = braces * 2;
              if (parens > 0 || OPERATORS.matches(next)) {
                indent += 4;
              } else if (next == '}') {
                indent -= 2;
              }
              for (int i = 0; i < indent; i++) {
                out.append(' ');
              }
            }
            continue;
          }
          if (parens == 0 && braces < 2 && s.charAt(previous) != '\n' && out.length() > 0) {
            out.append('\n');
          }
          continue;
        default:
          break;
      }
      out.append(s, start, end);
    }
    return out.toString();
  }

  private Reformatter() {}
}