com.google.googlejavaformat.DocBuilder Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of google-java-format Show documentation
A Java source code formatter that follows Google Java Style.
There is a newer version: 1.25.0
/*
 * Copyright 2015 Google Inc.
 *
 * 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.googlejavaformat;

import com.google.common.base.MoreObjects;
import java.util.ArrayDeque;
import java.util.List;

/** A {@code DocBuilder} converts a sequence of {@link Op}s into a {@link Doc}. */
public final class DocBuilder {
  private final Doc.Level base = Doc.Level.make(Indent.Const.ZERO);
  private final ArrayDeque stack = new ArrayDeque<>();

  /**
   * A possibly earlier {@link Doc.Level} for appending text, à la Philip Wadler.
   *
   * Processing {@link Doc}s presents a subtle problem. Suppose we have a {@link Doc} for to an
   * assignment node, {@code a = b}, with an optional {@link Doc.Break} following the {@code =}.
   * Suppose we have 5 characters to write it, so that we think we don't need the break.
   * Unfortunately, this {@link Doc} lies in an expression statement {@link Doc} for the statement
   * {@code a = b;} and this statement does not fit in 3 characters. This is why many formatters
   * sometimes emit lines that are too long, or cheat by using a narrower line length to avoid such
   * problems.
   *
   * One solution to this problem is not to decide whether a {@link Doc.Level} should be broken
   * until later (in this case, after the semicolon has been seen). A simpler approach is to rewrite
   * the {@link Doc} as here, so that the semicolon moves inside the inner {@link Doc}, and we can
   * decide whether to break that {@link Doc} without seeing later text.
   */
  private Doc.Level appendLevel = base;

  /** Start to build a {@code DocBuilder}. */
  public DocBuilder() {
    stack.addLast(base);
  }

  /**
   * Add a list of {@link Op}s to the {@link OpsBuilder}.
   *
   * @param ops the {@link Op}s
   * @return the {@link OpsBuilder}
   */
  public DocBuilder withOps(List ops) {
    for (Op op : ops) {
      op.add(this); // These operations call the operations below to build the doc.
    }
    return this;
  }

  /**
   * Open a new {@link Doc.Level}.
   *
   * @param plusIndent the extra indent for the {@link Doc.Level}
   */
  void open(Indent plusIndent) {
    Doc.Level level = Doc.Level.make(plusIndent);
    stack.addLast(level);
  }

  /** Close the current {@link Doc.Level}. */
  void close() {
    Doc.Level top = stack.removeLast();
    stack.peekLast().add(top);
  }

  /**
   * Add a {@link Doc} to the current {@link Doc.Level}.
   *
   * @param doc the {@link Doc}
   */
  void add(Doc doc) {
    appendLevel.add(doc);
  }

  /**
   * Add a {@link Doc.Break} to the current {@link Doc.Level}.
   *
   * @param breakDoc the {@link Doc.Break}
   */
  void breakDoc(Doc.Break breakDoc) {
    appendLevel = stack.peekLast();
    appendLevel.add(breakDoc);
  }

  /**
   * Return the {@link Doc}.
   *
   * @return the {@link Doc}
   */
  public Doc build() {
    return base;
  }

  @Override
  public String toString() {
    return MoreObjects.toStringHelper(this)
        .add("base", base)
        .add("stack", stack)
        .add("appendLevel", appendLevel)
        .toString();
  }
}