com.palantir.javaformat.java.JavaOutput Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of palantir-java-format Show documentation
Show all versions of palantir-java-format Show documentation
Palantir open source project
/*
* 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.palantir.javaformat.java;
import com.google.common.base.CharMatcher;
import com.google.common.base.MoreObjects;
import com.google.common.collect.DiscreteDomain;
import com.google.common.collect.ImmutableList;
import com.google.common.collect.Range;
import com.google.common.collect.RangeSet;
import com.google.common.collect.TreeRangeSet;
import com.palantir.javaformat.Input;
import com.palantir.javaformat.Input.Token;
import com.palantir.javaformat.Newlines;
import com.palantir.javaformat.OpsBuilder.BlankLineWanted;
import com.palantir.javaformat.Output;
import com.palantir.javaformat.doc.State;
import java.util.ArrayList;
import java.util.List;
import java.util.Map;
/*
* Throughout this file, {@code i} is an index for input lines, {@code j} is an index for output
* lines, {@code ij} is an index into either input or output lines, and {@code k} is an index for
* toks.
*/
/**
* {@code JavaOutput} extends {@link Output Output} to represent a Java output document. It includes methods to emit the
* output document.
*/
public final class JavaOutput extends Output {
private final String lineSeparator;
private final JavaInput javaInput; // Used to follow along while emitting the output.
private final List mutableLines = new ArrayList<>();
private final int kN; // The number of tokens or comments in the input, excluding the EOF.
private final InputMetadata inputMetadata;
private int iLine = 0; // Closest corresponding line number on input.
private int lastK = -1; // Last {@link Tok} index output.
private int spacesPending = 0;
private int newlinesPending = 0;
private StringBuilder lineBuilder = new StringBuilder();
/**
* {@code JavaOutput} constructor.
*
* @param javaInput the {@link JavaInput}, used to match up blank lines in the output
*/
public JavaOutput(JavaInput javaInput, InputMetadata inputMetadata) {
this.lineSeparator = javaInput.getLineSeparator();
this.javaInput = javaInput;
kN = javaInput.getkN();
this.inputMetadata = inputMetadata;
}
// TODO(jdd): Add invariant.
@Override
public void append(State state, String text, Range range) {
if (!range.isEmpty()) {
boolean sawNewlines = false;
// Skip over input line we've passed.
int iN = javaInput.getLineCount();
while (iLine < iN
&& (javaInput.getRanges(iLine).isEmpty()
|| javaInput.getRanges(iLine).upperEndpoint() <= range.lowerEndpoint())) {
if (javaInput.getRanges(iLine).isEmpty()) {
// Skipped over a blank line.
sawNewlines = true;
}
++iLine;
}
/*
* Output blank line if we've called {@link OpsBuilder#blankLine}{@code (true)} here, or if
* there's a blank line here and it's a comment.
*/
BlankLineWanted wanted = inputMetadata.blankLines().getOrDefault(lastK, BlankLineWanted.NO);
if (isComment(text) ? sawNewlines : wanted.wanted(state).orElse(sawNewlines)) {
++newlinesPending;
}
}
if (Newlines.isNewline(text)) {
/*
* Don't update range information, and swallow extra newlines. The case below for '\n' is for
* block comments.
*/
if (newlinesPending == 0) {
++newlinesPending;
}
spacesPending = 0;
} else {
boolean rangesSet = false;
int textN = text.length();
for (int i = 0; i < textN; i++) {
char c = text.charAt(i);
switch (c) {
case ' ':
++spacesPending;
break;
case '\r':
if (i + 1 < text.length() && text.charAt(i + 1) == '\n') {
i++;
}
// falls through
case '\n':
spacesPending = 0;
++newlinesPending;
break;
default:
while (newlinesPending > 0) {
// drop leading blank lines
if (!mutableLines.isEmpty() || lineBuilder.length() > 0) {
mutableLines.add(lineBuilder.toString());
}
lineBuilder = new StringBuilder();
rangesSet = false;
--newlinesPending;
}
while (spacesPending > 0) {
lineBuilder.append(' ');
--spacesPending;
}
lineBuilder.append(c);
if (!range.isEmpty()) {
if (!rangesSet) {
while (ranges.size() <= mutableLines.size()) {
ranges.add(Formatter.EMPTY_RANGE);
}
ranges.set(mutableLines.size(), union(ranges.get(mutableLines.size()), range));
rangesSet = true;
}
}
}
}
}
if (!range.isEmpty()) {
lastK = range.upperEndpoint();
}
}
@Override
public void indent(int indent) {
spacesPending = indent;
}
/** Flush any incomplete last line, then add the EOF token into our data structures. */
void flush() {
String lastLine = lineBuilder.toString();
if (!CharMatcher.whitespace().matchesAllOf(lastLine)) {
mutableLines.add(lastLine);
}
int jN = mutableLines.size();
Range eofRange = Range.closedOpen(kN, kN + 1);
while (ranges.size() < jN) {
ranges.add(Formatter.EMPTY_RANGE);
}
ranges.add(eofRange);
setLines(ImmutableList.copyOf(mutableLines));
}
// The following methods can be used after the Output has been built.
/**
* Emit a list of {@link Replacement}s to convert from input to output.
*
* @return a list of {@link Replacement}s, sorted by start index, without overlaps
*/
public ImmutableList getFormatReplacements(RangeSet iRangeSet0) {
ImmutableList.Builder result = ImmutableList.builder();
Map> kToJ = JavaOutput.makeKToIJ(this);
// Expand the token ranges to align with re-formattable boundaries.
RangeSet breakableRanges = TreeRangeSet.create();
RangeSet iRangeSet = iRangeSet0.subRangeSet(Range.closed(0, javaInput.getkN()));
for (Range iRange : iRangeSet.asRanges()) {
Range range = expandToBreakableRegions(iRange.canonical(DiscreteDomain.integers()));
if (range.equals(EMPTY_RANGE)) {
// the range contains only whitespace
continue;
}
breakableRanges.add(range);
}
// Construct replacements for each reformatted region.
for (Range range : breakableRanges.asRanges()) {
Input.Tok startTok = startTok(javaInput.getToken(range.lowerEndpoint()));
Input.Tok endTok = endTok(javaInput.getToken(range.upperEndpoint() - 1));
// Add all output lines in the given token range to the replacement.
StringBuilder replacement = new StringBuilder();
int replaceFrom = startTok.getPosition();
// Replace leading whitespace in the input with the whitespace from the formatted file
while (replaceFrom > 0) {
char previous = javaInput.getText().charAt(replaceFrom - 1);
if (!CharMatcher.whitespace().matches(previous)) {
break;
}
replaceFrom--;
}
int i = kToJ.get(startTok.getIndex()).lowerEndpoint();
// Include leading blank lines from the formatted output, unless the formatted range
// starts at the beginning of the file.
while (i > 0 && getLine(i - 1).isEmpty()) {
i--;
}
// Write out the formatted range.
for (; i < kToJ.get(endTok.getIndex()).upperEndpoint(); i++) {
// It's possible to run out of output lines (e.g. if the input ended with
// multiple trailing newlines).
if (i < getLineCount()) {
if (i > 0) {
replacement.append(lineSeparator);
}
replacement.append(getLine(i));
}
}
int replaceTo = Math.min(
endTok.getPosition() + endTok.length(), javaInput.getText().length());
// If the formatted ranged ended in the trailing trivia of the last token before EOF,
// format all the way up to EOF to deal with trailing whitespace correctly.
if (endTok.getIndex() == javaInput.getkN() - 1) {
replaceTo = javaInput.getText().length();
}
// Replace trailing whitespace in the input with the whitespace from the formatted file.
// If the trailing whitespace in the input includes one or more line breaks, preserve the
// whitespace after the last newline to avoid re-indenting the line following the formatted
// line.
int newline = -1;
while (replaceTo < javaInput.getText().length()) {
char next = javaInput.getText().charAt(replaceTo);
if (!CharMatcher.whitespace().matches(next)) {
break;
}
int newlineLength = Newlines.hasNewlineAt(javaInput.getText(), replaceTo);
if (newlineLength != -1) {
newline = replaceTo;
// Skip over the entire newline; don't count the second character of \r\n as a newline.
replaceTo += newlineLength;
} else {
replaceTo++;
}
}
if (newline != -1) {
replaceTo = newline;
}
if (newline == -1) {
// There wasn't an existing trailing newline; add one.
replacement.append(lineSeparator);
}
for (; i < getLineCount(); i++) {
String after = getLine(i);
int idx = CharMatcher.whitespace().negate().indexIn(after);
if (idx == -1) {
// Write out trailing empty lines from the formatted output.
replacement.append(lineSeparator);
} else {
if (newline == -1) {
// If there wasn't a trailing newline in the input, indent the next line.
replacement.append(after, 0, idx);
}
break;
}
}
result.add(Replacement.create(replaceFrom, replaceTo, replacement.toString()));
}
return result.build();
}
/**
* Expand a token range to start and end on acceptable boundaries for re-formatting.
*
* @param iRange the {@link Range} of tokens
* @return the expanded token range
*/
private Range expandToBreakableRegions(Range iRange) {
// The original line range.
int loTok = iRange.lowerEndpoint();
int hiTok = iRange.upperEndpoint() - 1;
// Expand the token indices to formattable boundaries (e.g. edges of statements).
if (!partialFormatRanges().contains(loTok) || !partialFormatRanges().contains(hiTok)) {
return EMPTY_RANGE;
}
loTok = partialFormatRanges().rangeContaining(loTok).lowerEndpoint();
hiTok = partialFormatRanges().rangeContaining(hiTok).upperEndpoint();
return Range.closedOpen(loTok, hiTok + 1);
}
private RangeSet partialFormatRanges() {
return inputMetadata.partialFormatRanges();
}
/** The earliest position of any Tok in the Token, including leading whitespace. */
public static int startPosition(Token token) {
int min = token.getTok().getPosition();
for (Input.Tok tok : token.getToksBefore()) {
min = Math.min(min, tok.getPosition());
}
return min;
}
/** The earliest non-whitespace Tok in the Token. */
public static Input.Tok startTok(Token token) {
for (Input.Tok tok : token.getToksBefore()) {
if (tok.getIndex() >= 0) {
return tok;
}
}
return token.getTok();
}
/** The last non-whitespace Tok in the Token. */
public static Input.Tok endTok(Token token) {
for (int i = token.getToksAfter().size() - 1; i >= 0; i--) {
Input.Tok tok = token.getToksAfter().get(i);
if (tok.getIndex() >= 0) {
return tok;
}
}
return token.getTok();
}
private boolean isComment(String text) {
return text.startsWith("//") || text.startsWith("/*");
}
private static Range union(Range x, Range y) {
return x.isEmpty() ? y : y.isEmpty() ? x : x.span(y).canonical(DiscreteDomain.integers());
}
@Override
public String toString() {
return MoreObjects.toStringHelper(this)
.add("iLine", iLine)
.add("lastK", lastK)
.add("spacesPending", spacesPending)
.add("newlinesPending", newlinesPending)
.add("inputMetadata", inputMetadata)
.add("super", super.toString())
.toString();
}
}
© 2015 - 2024 Weber Informatics LLC | Privacy Policy