com.facebook.buck.zip.ZipOutputStreams.java Source code

Java tutorial

Introduction

Here is the source code for com.facebook.buck.zip.ZipOutputStreams.java

Source

/*
 * Copyright 2013-present Facebook, 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.facebook.buck.zip;

import com.facebook.buck.timing.Clock;
import com.facebook.buck.timing.DefaultClock;
import com.facebook.buck.util.HumanReadableException;
import com.google.common.base.Preconditions;

import java.io.BufferedOutputStream;
import java.io.File;
import java.io.FileNotFoundException;
import java.io.FileOutputStream;
import java.io.OutputStream;

public class ZipOutputStreams {

    private ZipOutputStreams() {
        // factory class
    }

    /**
     * Create a new {@link CustomZipOutputStream} that outputs to the given {@code zipFile}. Note that
     * the parent directory of the {@code zipFile} must exist already. The returned stream will throw
     * an exception should duplicate entries be added.
     *
     * @param zipFile The file to write to.
     */
    public static CustomZipOutputStream newOutputStream(File zipFile) {
        Preconditions.checkNotNull(zipFile);
        try {
            return newOutputStream(new BufferedOutputStream(new FileOutputStream(zipFile)));
        } catch (FileNotFoundException e) {
            throw new RuntimeException(e);
        }
    }

    /**
     * Create a new {@link CustomZipOutputStream} that will by default act in the same way as
     * {@link java.util.zip.ZipOutputStream}, notably by throwing an exception if duplicate entries
     * are added.
     *
     * @param out The output stream to write to.
     */
    public static CustomZipOutputStream newOutputStream(OutputStream out) {
        return newOutputStream(out, HandleDuplicates.THROW_EXCEPTION);
    }

    /**
     * Create a new {@link CustomZipOutputStream} that handles duplicate entries in the way dictated
     * by {@code mode}.
     *
     * @param zipFile The file to write to.
     * @param mode How to handle duplicate entries.
     */
    public static CustomZipOutputStream newOutputStream(File zipFile, HandleDuplicates mode)
            throws FileNotFoundException {
        Preconditions.checkNotNull(zipFile);

        return newOutputStream(new BufferedOutputStream(new FileOutputStream(zipFile)), mode);
    }

    /**
     * Create a new {@link CustomZipOutputStream} that handles duplicate entries in the way dictated
     * by {@code mode}.
     *
     * @param out The output stream to write to.
     * @param mode How to handle duplicate entries.
     */
    public static CustomZipOutputStream newOutputStream(OutputStream out, HandleDuplicates mode) {
        Clock clock = new DefaultClock();

        switch (mode) {
        case APPEND_TO_ZIP:
        case THROW_EXCEPTION:
            return new AppendingZipOutputStream(clock, out, mode == HandleDuplicates.THROW_EXCEPTION);

        case OVERWRITE_EXISTING:
            return new OverwritingZipOutputStream(clock, out);

        default:
            throw new HumanReadableException("Unable to determine which zip output mode to use: %s", mode);
        }
    }

    public static enum HandleDuplicates {
        /** Duplicate entries are simply appended to the zip. */
        APPEND_TO_ZIP,
        /** An exception should be thrown if a duplicate entry is added to a zip. */
        THROW_EXCEPTION,
        /** A duplicate entry overwrites an existing entry with the same name. */
        OVERWRITE_EXISTING;
    }
}