/*
- * Sone - FreenetLinkParser.java - Copyright © 2010 David Roden
+ * Sone - SoneTextParser.java - Copyright © 2010–2013 David Roden
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
package net.pterodactylus.sone.text;
+import static java.util.logging.Logger.getLogger;
+
import java.io.BufferedReader;
import java.io.IOException;
import java.io.Reader;
import java.util.regex.Matcher;
import java.util.regex.Pattern;
-import net.pterodactylus.sone.core.PostProvider;
-import net.pterodactylus.sone.core.SoneProvider;
import net.pterodactylus.sone.data.Post;
import net.pterodactylus.sone.data.Sone;
-import net.pterodactylus.util.logging.Logging;
+import net.pterodactylus.sone.data.impl.IdOnlySone;
+import net.pterodactylus.sone.database.PostProvider;
+import net.pterodactylus.sone.database.SoneProvider;
+import net.pterodactylus.util.io.Closer;
+
+import com.google.common.base.Optional;
+
import freenet.keys.FreenetURI;
/**
public class SoneTextParser implements Parser<SoneTextParserContext> {
/** The logger. */
- private static final Logger logger = Logging.getLogger(SoneTextParser.class);
+ private static final Logger logger = getLogger(SoneTextParser.class.getName());
/** Pattern to detect whitespace. */
private static final Pattern whitespacePattern = Pattern.compile("[\\u000a\u0020\u00a0\u1680\u180e\u2000\u2001\u2002\u2003\u2004\u2005\u2006\u2007\u2008\u2009\u200a\u200b\u200c\u200d\u202f\u205f\u2060\u2800\u3000]");
private enum LinkType {
/** Link is a KSK. */
- KSK,
+ KSK("KSK@"),
/** Link is a CHK. */
- CHK,
+ CHK("CHK@"),
/** Link is an SSK. */
- SSK,
+ SSK("SSK@"),
/** Link is a USK. */
- USK,
+ USK("USK@"),
/** Link is HTTP. */
- HTTP,
+ HTTP("http://"),
/** Link is HTTPS. */
- HTTPS,
+ HTTPS("https://"),
/** Link is a Sone. */
- SONE,
+ SONE("sone://"),
/** Link is a post. */
- POST,
+ POST("post://");
+
+ /** The scheme identifying this link type. */
+ private final String scheme;
+
+ /**
+ * Creates a new link type identified by the given scheme.
+ *
+ * @param scheme
+ * The scheme of the link type
+ */
+ private LinkType(String scheme) {
+ this.scheme = scheme;
+ }
+
+ /**
+ * Returns the scheme of this link type.
+ *
+ * @return The scheme of this link type
+ */
+ public String getScheme() {
+ return scheme;
+ }
}
public Iterable<Part> parse(SoneTextParserContext context, Reader source) throws IOException {
PartContainer parts = new PartContainer();
BufferedReader bufferedReader = (source instanceof BufferedReader) ? (BufferedReader) source : new BufferedReader(source);
- String line;
- boolean lastLineEmpty = true;
- int emptyLines = 0;
- while ((line = bufferedReader.readLine()) != null) {
- if (line.trim().length() == 0) {
- if (lastLineEmpty) {
- continue;
- }
- parts.add(new PlainTextPart("\n"));
- ++emptyLines;
- lastLineEmpty = emptyLines == 2;
- continue;
- }
- emptyLines = 0;
- boolean lineComplete = true;
- while (line.length() > 0) {
- int nextKsk = line.indexOf("KSK@");
- int nextChk = line.indexOf("CHK@");
- int nextSsk = line.indexOf("SSK@");
- int nextUsk = line.indexOf("USK@");
- int nextHttp = line.indexOf("http://");
- int nextHttps = line.indexOf("https://");
- int nextSone = line.indexOf("sone://");
- int nextPost = line.indexOf("post://");
- if ((nextKsk == -1) && (nextChk == -1) && (nextSsk == -1) && (nextUsk == -1) && (nextHttp == -1) && (nextHttps == -1) && (nextSone == -1) && (nextPost == -1)) {
- if (lineComplete && !lastLineEmpty) {
- parts.add(new PlainTextPart("\n" + line));
- } else {
- parts.add(new PlainTextPart(line));
+ try {
+ String line;
+ boolean lastLineEmpty = true;
+ int emptyLines = 0;
+ while ((line = bufferedReader.readLine()) != null) {
+ if (line.trim().length() == 0) {
+ if (lastLineEmpty) {
+ continue;
}
- break;
- }
- int next = Integer.MAX_VALUE;
- LinkType linkType = null;
- if ((nextKsk > -1) && (nextKsk < next)) {
- next = nextKsk;
- linkType = LinkType.KSK;
- }
- if ((nextChk > -1) && (nextChk < next)) {
- next = nextChk;
- linkType = LinkType.CHK;
- }
- if ((nextSsk > -1) && (nextSsk < next)) {
- next = nextSsk;
- linkType = LinkType.SSK;
- }
- if ((nextUsk > -1) && (nextUsk < next)) {
- next = nextUsk;
- linkType = LinkType.USK;
- }
- if ((nextHttp > -1) && (nextHttp < next)) {
- next = nextHttp;
- linkType = LinkType.HTTP;
- }
- if ((nextHttps > -1) && (nextHttps < next)) {
- next = nextHttps;
- linkType = LinkType.HTTPS;
- }
- if ((nextSone > -1) && (nextSone < next)) {
- next = nextSone;
- linkType = LinkType.SONE;
- }
- if ((nextPost > -1) && (nextPost < next)) {
- next = nextPost;
- linkType = LinkType.POST;
+ parts.add(new PlainTextPart("\n"));
+ ++emptyLines;
+ lastLineEmpty = emptyLines == 2;
+ continue;
}
- if (linkType == LinkType.SONE) {
- if (next > 0) {
- parts.add(new PlainTextPart(line.substring(0, next)));
- }
- if (line.length() >= (next + 7 + 43)) {
- String soneId = line.substring(next + 7, next + 50);
- Sone sone = soneProvider.getSone(soneId, false);
- if ((sone != null) && (sone.getName() != null)) {
- parts.add(new SonePart(sone));
+ emptyLines = 0;
+ /*
+ * lineComplete tracks whether the block you are parsing is the
+ * first block of the line. this is important because sometimes
+ * you have to add an additional line break.
+ */
+ boolean lineComplete = true;
+ while (line.length() > 0) {
+ int nextKsk = line.indexOf("KSK@");
+ int nextChk = line.indexOf("CHK@");
+ int nextSsk = line.indexOf("SSK@");
+ int nextUsk = line.indexOf("USK@");
+ int nextHttp = line.indexOf("http://");
+ int nextHttps = line.indexOf("https://");
+ int nextSone = line.indexOf("sone://");
+ int nextPost = line.indexOf("post://");
+ if ((nextKsk == -1) && (nextChk == -1) && (nextSsk == -1) && (nextUsk == -1) && (nextHttp == -1) && (nextHttps == -1) && (nextSone == -1) && (nextPost == -1)) {
+ if (lineComplete && !lastLineEmpty) {
+ parts.add(new PlainTextPart("\n" + line));
} else {
- parts.add(new PlainTextPart(line.substring(next, next + 50)));
+ parts.add(new PlainTextPart(line));
}
- line = line.substring(next + 50);
- } else {
- parts.add(new PlainTextPart(line.substring(next)));
- line = "";
+ break;
+ }
+ int next = Integer.MAX_VALUE;
+ LinkType linkType = null;
+ if ((nextKsk > -1) && (nextKsk < next)) {
+ next = nextKsk;
+ linkType = LinkType.KSK;
+ }
+ if ((nextChk > -1) && (nextChk < next)) {
+ next = nextChk;
+ linkType = LinkType.CHK;
+ }
+ if ((nextSsk > -1) && (nextSsk < next)) {
+ next = nextSsk;
+ linkType = LinkType.SSK;
+ }
+ if ((nextUsk > -1) && (nextUsk < next)) {
+ next = nextUsk;
+ linkType = LinkType.USK;
+ }
+ if ((nextHttp > -1) && (nextHttp < next)) {
+ next = nextHttp;
+ linkType = LinkType.HTTP;
+ }
+ if ((nextHttps > -1) && (nextHttps < next)) {
+ next = nextHttps;
+ linkType = LinkType.HTTPS;
+ }
+ if ((nextSone > -1) && (nextSone < next)) {
+ next = nextSone;
+ linkType = LinkType.SONE;
+ }
+ if ((nextPost > -1) && (nextPost < next)) {
+ next = nextPost;
+ linkType = LinkType.POST;
+ }
+
+ /* cut off “freenet:” from before keys. */
+ if (((linkType == LinkType.KSK) || (linkType == LinkType.CHK) || (linkType == LinkType.SSK) || (linkType == LinkType.USK)) && (next >= 8) && (line.substring(next - 8, next).equals("freenet:"))) {
+ next -= 8;
+ line = line.substring(0, next) + line.substring(next + 8);
+ }
+
+ /* if there is text before the next item, write it out. */
+ if (lineComplete && !lastLineEmpty) {
+ parts.add(new PlainTextPart("\n"));
}
- continue;
- }
- if (linkType == LinkType.POST) {
if (next > 0) {
parts.add(new PlainTextPart(line.substring(0, next)));
+ line = line.substring(next);
+ next = 0;
+ }
+ lineComplete = false;
+
+ Matcher matcher = whitespacePattern.matcher(line);
+ int nextSpace = matcher.find(0) ? matcher.start() : line.length();
+ String link = line.substring(0, nextSpace);
+ String name = link;
+ logger.log(Level.FINER, String.format("Found link: %s", link));
+ logger.log(Level.FINEST, String.format("CHK: %d, SSK: %d, USK: %d", nextChk, nextSsk, nextUsk));
+
+ /* if there is no text after the scheme, it’s not a link! */
+ if (link.equals(linkType.getScheme())) {
+ parts.add(new PlainTextPart(linkType.getScheme()));
+ line = line.substring(linkType.getScheme().length());
+ continue;
}
- if (line.length() >= (next + 7 + 36)) {
- String postId = line.substring(next + 7, next + 43);
- Post post = postProvider.getPost(postId, false);
- if ((post != null) && (post.getSone() != null)) {
- String postText = post.getText();
- postText = postText.substring(0, Math.min(postText.length(), 20)) + "…";
- parts.add(new PostPart(post));
+
+ if (linkType == LinkType.SONE) {
+ if (line.length() >= (7 + 43)) {
+ String soneId = line.substring(7, 50);
+ Optional<Sone> sone = soneProvider.getSone(soneId);
+ if (!sone.isPresent()) {
+ /*
+ * don’t use create=true above, we don’t want
+ * the empty shell.
+ */
+ sone = Optional.<Sone>of(new IdOnlySone(soneId));
+ }
+ parts.add(new SonePart(sone.get()));
+ line = line.substring(50);
} else {
- parts.add(new PlainTextPart(line.substring(next, next + 43)));
+ parts.add(new PlainTextPart(line));
+ line = "";
}
- line = line.substring(next + 43);
- } else {
- parts.add(new PlainTextPart(line.substring(next)));
- line = "";
+ continue;
}
- continue;
- }
- if ((next >= 8) && (line.substring(next - 8, next).equals("freenet:"))) {
- next -= 8;
- line = line.substring(0, next) + line.substring(next + 8);
- }
- Matcher matcher = whitespacePattern.matcher(line);
- int nextSpace = matcher.find(next) ? matcher.start() : line.length();
- if (nextSpace > (next + 4)) {
- if (!lastLineEmpty && lineComplete) {
- parts.add(new PlainTextPart("\n" + line.substring(0, next)));
- } else {
- if (next > 0) {
- parts.add(new PlainTextPart(line.substring(0, next)));
+ if (linkType == LinkType.POST) {
+ if (line.length() >= (7 + 36)) {
+ String postId = line.substring(7, 43);
+ Optional<Post> post = postProvider.getPost(postId);
+ if (post.isPresent()) {
+ parts.add(new PostPart(post.get()));
+ } else {
+ parts.add(new PlainTextPart(line.substring(0, 43)));
+ }
+ line = line.substring(43);
+ } else {
+ parts.add(new PlainTextPart(line));
+ line = "";
}
+ continue;
}
- String link = line.substring(next, nextSpace);
- String name = link;
- logger.log(Level.FINER, "Found link: %s", link);
- logger.log(Level.FINEST, "Next: %d, CHK: %d, SSK: %d, USK: %d", new Object[] { next, nextChk, nextSsk, nextUsk });
if ((linkType == LinkType.KSK) || (linkType == LinkType.CHK) || (linkType == LinkType.SSK) || (linkType == LinkType.USK)) {
FreenetURI uri;
parts.add(new LinkPart(link, name));
}
line = line.substring(nextSpace);
- } else {
- if (!lastLineEmpty && lineComplete) {
- parts.add(new PlainTextPart("\n" + line.substring(0, next + 4)));
- } else {
- parts.add(new PlainTextPart(line.substring(0, next + 4)));
- }
- line = line.substring(next + 4);
}
- lineComplete = false;
+ lastLineEmpty = false;
+ }
+ } finally {
+ if (bufferedReader != source) {
+ Closer.close(bufferedReader);
}
- lastLineEmpty = false;
}
for (int partIndex = parts.size() - 1; partIndex >= 0; --partIndex) {
Part part = parts.getPart(partIndex);