package be.nikiroo.fanfix.supported;
-import java.awt.image.BufferedImage;
-import java.io.BufferedReader;
-import java.io.ByteArrayInputStream;
-import java.io.File;
import java.io.IOException;
import java.io.InputStream;
-import java.io.InputStreamReader;
-import java.net.MalformedURLException;
import java.net.URL;
import java.util.ArrayList;
import java.util.Date;
import java.util.List;
import java.util.Map;
import java.util.Map.Entry;
-import java.util.Scanner;
+
+import org.jsoup.helper.DataUtil;
+import org.jsoup.nodes.Document;
+import org.jsoup.nodes.Element;
+import org.jsoup.nodes.Node;
import be.nikiroo.fanfix.Instance;
-import be.nikiroo.fanfix.bundles.Config;
import be.nikiroo.fanfix.bundles.StringId;
import be.nikiroo.fanfix.data.Chapter;
import be.nikiroo.fanfix.data.MetaData;
-import be.nikiroo.fanfix.data.Paragraph;
-import be.nikiroo.fanfix.data.Paragraph.ParagraphType;
import be.nikiroo.fanfix.data.Story;
-import be.nikiroo.utils.IOUtils;
import be.nikiroo.utils.Progress;
import be.nikiroo.utils.StringUtils;
* @author niki
*/
public abstract class BasicSupport {
- /**
- * The supported input types for which we can get a {@link BasicSupport}
- * object.
- *
- * @author niki
- */
- public enum SupportType {
- /** EPUB files created with this program */
- EPUB,
- /** Pure text file with some rules */
- TEXT,
- /** TEXT but with associated .info file */
- INFO_TEXT,
- /** My Little Pony fanfictions */
- FIMFICTION,
- /** Fanfictions from a lot of different universes */
- FANFICTION,
- /** Website with lots of Mangas */
- MANGAFOX,
- /** Furry website with comics support */
- E621,
- /** Furry website with stories */
- YIFFSTAR,
- /** CBZ files */
- CBZ,
- /** HTML files */
- HTML;
-
- /**
- * A description of this support type (more information than the
- * {@link BasicSupport#getSourceName()}).
- *
- * @return the description
- */
- public String getDesc() {
- String desc = Instance.getTrans().getStringX(StringId.INPUT_DESC,
- this.name());
-
- if (desc == null) {
- desc = Instance.getTrans().getString(StringId.INPUT_DESC, this);
- }
-
- return desc;
- }
-
- /**
- * The name of this support type (a short version).
- *
- * @return the name
- */
- public String getSourceName() {
- BasicSupport support = BasicSupport.getSupport(this);
- if (support != null) {
- return support.getSourceName();
- }
-
- return null;
- }
-
- @Override
- public String toString() {
- return super.toString().toLowerCase();
- }
-
- /**
- * Call {@link SupportType#valueOf(String.toUpperCase())}.
- *
- * @param typeName
- * the possible type name
- *
- * @return NULL or the type
- */
- public static SupportType valueOfUC(String typeName) {
- return SupportType.valueOf(typeName == null ? null : typeName
- .toUpperCase());
- }
-
- /**
- * Call {@link SupportType#valueOf(String.toUpperCase())} but return
- * NULL for NULL instead of raising exception.
- *
- * @param typeName
- * the possible type name
- *
- * @return NULL or the type
- */
- public static SupportType valueOfNullOkUC(String typeName) {
- if (typeName == null) {
- return null;
- }
-
- return SupportType.valueOfUC(typeName);
- }
-
- /**
- * Call {@link SupportType#valueOf(String.toUpperCase())} but return
- * NULL in case of error instead of raising an exception.
- *
- * @param typeName
- * the possible type name
- *
- * @return NULL or the type
- */
- public static SupportType valueOfAllOkUC(String typeName) {
- try {
- return SupportType.valueOfUC(typeName);
- } catch (Exception e) {
- return null;
- }
- }
- }
-
- private InputStream in;
+ private Document sourceNode;
+ private URL source;
private SupportType type;
private URL currentReferer; // with only one 'r', as in 'HTTP'...
- // quote chars
- private char openQuote = Instance.getTrans().getChar(
- StringId.OPEN_SINGLE_QUOTE);
- private char closeQuote = Instance.getTrans().getChar(
- StringId.CLOSE_SINGLE_QUOTE);
- private char openDoubleQuote = Instance.getTrans().getChar(
- StringId.OPEN_DOUBLE_QUOTE);
- private char closeDoubleQuote = Instance.getTrans().getChar(
- StringId.CLOSE_DOUBLE_QUOTE);
-
/**
* The name of this support class.
*
*/
protected abstract boolean isHtml();
- protected abstract MetaData getMeta(URL source, InputStream in)
- throws IOException;
+ /**
+ * Return the {@link MetaData} of this story.
+ *
+ * @return the associated {@link MetaData}, never NULL
+ *
+ * @throws IOException
+ * in case of I/O error
+ */
+ protected abstract MetaData getMeta() throws IOException;
/**
* Return the story description.
*
- * @param source
- * the source of the story
- * @param in
- * the input (the main resource)
- *
* @return the description
*
* @throws IOException
* in case of I/O error
*/
- protected abstract String getDesc(URL source, InputStream in)
- throws IOException;
+ protected abstract String getDesc() throws IOException;
/**
- * Return the list of chapters (name and resource).
+ * Return the list of chapters (name and resource). *
+ * <p>
+ * Can be NULL if this {@link BasicSupport} do no use chapters.
*
- * @param source
- * the source of the story
- * @param in
- * the input (the main resource)
+ * @param pg
+ * the optional progress reporter
*
- * @return the chapters
+ * @return the chapters or NULL
*
* @throws IOException
* in case of I/O error
*/
- protected abstract List<Entry<String, URL>> getChapters(URL source,
- InputStream in) throws IOException;
+ protected abstract List<Entry<String, URL>> getChapters(Progress pg)
+ throws IOException;
/**
* Return the content of the chapter (possibly HTML encoded, if
* {@link BasicSupport#isHtml()} is TRUE).
*
- * @param source
- * the source of the story
- * @param in
- * the input (the main resource)
+ * @param chapUrl
+ * the chapter {@link URL}
* @param number
* the chapter number
+ * @param pg
+ * the optional progress reporter
*
* @return the content
*
* @throws IOException
* in case of I/O error
*/
- protected abstract String getChapterContent(URL source, InputStream in,
- int number) throws IOException;
-
- /**
- * Log into the support (can be a no-op depending upon the support).
- *
- * @throws IOException
- * in case of I/O error
- */
- public void login() throws IOException {
-
- }
+ protected abstract String getChapterContent(URL chapUrl, int number,
+ Progress pg) throws IOException;
/**
* Return the list of cookies (values included) that must be used to
* it.
*
* @return the cookies
- *
- * @throws IOException
- * in case of I/O error
*/
- public Map<String, String> getCookies() throws IOException {
+ public Map<String, String> getCookies() {
return new HashMap<String, String>();
}
/**
- * Return the canonical form of the main {@link URL}.
- *
- * @param source
- * the source {@link URL}
- *
- * @return the canonical form of this {@link URL}
- *
- * @throws IOException
- * in case of I/O error
- */
- public URL getCanonicalUrl(URL source) throws IOException {
- return source;
- }
-
- /**
- * Process the given story resource into a partially filled {@link Story}
- * object containing the name and metadata, except for the description.
+ * OAuth authorisation (aka, "bearer XXXXXXX").
*
- * @param url
- * the story resource
- *
- * @return the {@link Story}
- *
- * @throws IOException
- * in case of I/O error
+ * @return the OAuth string
*/
- public Story processMeta(URL url) throws IOException {
- return processMeta(url, true, false);
+ public String getOAuth() {
+ return null;
}
/**
- * Process the given story resource into a partially filled {@link Story}
- * object containing the name and metadata.
- *
- * @param url
- * the story resource
- *
- * @param close
- * close "this" and "in" when done
+ * Return the canonical form of the main {@link URL}.
*
- * @return the {@link Story}
+ * @param source
+ * the source {@link URL}, which can be NULL
*
- * @throws IOException
- * in case of I/O error
+ * @return the canonical form of this {@link URL} or NULL if the source was
+ * NULL
*/
- protected Story processMeta(URL url, boolean close, boolean getDesc)
- throws IOException {
- login();
-
- url = getCanonicalUrl(url);
-
- setCurrentReferer(url);
-
- in = openInput(url);
- if (in == null) {
- return null;
- }
-
- try {
- preprocess(url, getInput());
-
- Story story = new Story();
- MetaData meta = getMeta(url, getInput());
- if (meta.getCreationDate() == null
- || meta.getCreationDate().isEmpty()) {
- meta.setCreationDate(StringUtils.fromTime(new Date().getTime()));
- }
- story.setMeta(meta);
-
- if (meta != null && meta.getCover() == null) {
- meta.setCover(getDefaultCover(meta.getSubject()));
- }
-
- if (getDesc) {
- String descChapterName = Instance.getTrans().getString(
- StringId.DESCRIPTION);
- story.getMeta().setResume(
- makeChapter(url, 0, descChapterName,
- getDesc(url, getInput())));
- }
-
- return story;
- } finally {
- if (close) {
- try {
- close();
- } catch (IOException e) {
- Instance.syserr(e);
- }
-
- if (in != null) {
- in.close();
- }
- }
-
- setCurrentReferer(null);
- }
+ protected URL getCanonicalUrl(URL source) {
+ return source;
}
/**
- * Process the given story resource into a fully filled {@link Story}
- * object.
- *
- * @param url
- * the story resource
- * @param pg
- * the optional progress reporter
+ * The main {@link Node} for this {@link Story}.
*
- * @return the {@link Story}
- *
- * @throws IOException
- * in case of I/O error
+ * @return the node
*/
- public Story process(URL url, Progress pg) throws IOException {
- if (pg == null) {
- pg = new Progress();
- } else {
- pg.setMinMax(0, 100);
- }
-
- url = getCanonicalUrl(url);
- pg.setProgress(1);
- try {
- Story story = processMeta(url, false, true);
- pg.setProgress(10);
- if (story == null) {
- pg.setProgress(100);
- return null;
- }
-
- setCurrentReferer(url);
-
- story.setChapters(new ArrayList<Chapter>());
-
- List<Entry<String, URL>> chapters = getChapters(url, getInput());
- pg.setProgress(20);
-
- int i = 1;
- if (chapters != null) {
- Progress pgChaps = new Progress(0, chapters.size());
- pg.addProgress(pgChaps, 80);
-
- long words = 0;
- for (Entry<String, URL> chap : chapters) {
- setCurrentReferer(chap.getValue());
- InputStream chapIn = Instance.getCache().open(
- chap.getValue(), this, true);
- try {
- Chapter cc = makeChapter(url, i, chap.getKey(),
- getChapterContent(url, chapIn, i));
- words += cc.getWords();
- story.getChapters().add(cc);
- if (story.getMeta() != null) {
- story.getMeta().setWords(words);
- }
- } finally {
- chapIn.close();
- }
-
- pgChaps.setProgress(i++);
- }
- } else {
- pg.setProgress(100);
- }
-
- return story;
-
- } finally {
- try {
- close();
- } catch (IOException e) {
- Instance.syserr(e);
- }
-
- if (in != null) {
- in.close();
- }
-
- setCurrentReferer(null);
- }
+ protected Element getSourceNode() {
+ return sourceNode;
}
/**
- * The support type.
+ * The main {@link URL} for this {@link Story}.
*
- * @return the type
+ * @return the URL
*/
- public SupportType getType() {
- return type;
+ protected URL getSource() {
+ return source;
}
/**
this.currentReferer = currentReferer;
}
+ /**
+ * The support type.
+ *
+ * @return the type
+ */
+ public SupportType getType() {
+ return type;
+ }
+
/**
* The support type.
*
* @param type
* the new type
- *
- * @return this
*/
- protected BasicSupport setType(SupportType type) {
+ protected void setType(SupportType type) {
this.type = type;
- return this;
}
/**
- * Prepare the support if needed before processing.
+ * Open an input link that will be used for the support.
+ * <p>
+ * Can return NULL, in which case you are supposed to work without a source
+ * node.
*
* @param source
- * the source of the story
- * @param in
- * the input (the main resource)
+ * the source {@link URL}
+ *
+ * @return the {@link InputStream}
*
* @throws IOException
- * on I/O error
+ * in case of I/O error
*/
- protected void preprocess(URL source, InputStream in) throws IOException {
+ protected Document loadDocument(URL source) throws IOException {
+ String url = getCanonicalUrl(source).toString();
+ return DataUtil.load(Instance.getCache().open(source, this, false),
+ "UTF-8", url.toString());
}
/**
- * Now that we have processed the {@link Story}, close the resources if any.
+ * Log into the support (can be a no-op depending upon the support).
*
* @throws IOException
- * on I/O error
+ * in case of I/O error
*/
- protected void close() throws IOException {
+ protected void login() throws IOException {
}
/**
- * Create a {@link Chapter} object from the given information, formatting
- * the content as it should be.
- *
- * @param number
- * the chapter number
- * @param name
- * the chapter name
- * @param content
- * the chapter content
- *
- * @return the {@link Chapter}
- *
- * @throws IOException
- * in case of I/O error
+ * Now that we have processed the {@link Story}, close the resources if any.
*/
- protected Chapter makeChapter(URL source, int number, String name,
- String content) throws IOException {
- // Chapter name: process it correctly, then remove the possible
- // redundant "Chapter x: " in front of it
- String chapterName = processPara(name).getContent().trim();
- for (String lang : Instance.getConfig().getString(Config.CHAPTER)
- .split(",")) {
- String chapterWord = Instance.getConfig().getStringX(
- Config.CHAPTER, lang);
- if (chapterName.startsWith(chapterWord)) {
- chapterName = chapterName.substring(chapterWord.length())
- .trim();
- break;
- }
- }
-
- if (chapterName.startsWith(Integer.toString(number))) {
- chapterName = chapterName.substring(
- Integer.toString(number).length()).trim();
- }
-
- if (chapterName.startsWith(":")) {
- chapterName = chapterName.substring(1).trim();
- }
- //
-
- Chapter chap = new Chapter(number, chapterName);
-
- if (content != null) {
- List<Paragraph> paras = makeParagraphs(source, content);
- long words = 0;
- for (Paragraph para : paras) {
- words += para.getWords();
- }
- chap.setParagraphs(paras);
- chap.setWords(words);
- }
-
- return chap;
-
+ protected void close() {
+ setCurrentReferer(null);
}
/**
- * Convert the given content into {@link Paragraph}s.
- *
- * @param source
- * the source URL of the story
- * @param content
- * the textual content
+ * Process the given story resource into a partially filled {@link Story}
+ * object containing the name and metadata, except for the description.
*
- * @return the {@link Paragraph}s
+ * @return the {@link Story}
*
* @throws IOException
* in case of I/O error
*/
- protected List<Paragraph> makeParagraphs(URL source, String content)
- throws IOException {
- if (isHtml()) {
- // Special <HR> processing:
- content = content.replaceAll("(<hr [^>]*>)|(<hr/>)|(<hr>)",
- "\n* * *\n");
- }
+ public final Story processMeta() throws IOException {
+ Story story = null;
- List<Paragraph> paras = new ArrayList<Paragraph>();
- InputStream in = new ByteArrayInputStream(content.getBytes("UTF-8"));
try {
- BufferedReader buff = new BufferedReader(new InputStreamReader(in,
- "UTF-8"));
-
- for (String encodedLine = buff.readLine(); encodedLine != null; encodedLine = buff
- .readLine()) {
- String lines[];
- if (isHtml()) {
- lines = encodedLine.split("(<p>|</p>|<br>|<br/>|\\n)");
- } else {
- lines = new String[] { encodedLine };
- }
-
- for (String aline : lines) {
- String line = aline.trim();
-
- URL image = null;
- if (line.startsWith("[") && line.endsWith("]")) {
- image = getImageUrl(this, source,
- line.substring(1, line.length() - 1).trim());
- }
-
- if (image != null) {
- paras.add(new Paragraph(image));
- } else {
- paras.add(processPara(line));
- }
- }
- }
+ story = processMeta(false, null);
} finally {
- in.close();
- }
-
- // Check quotes for "bad" format
- List<Paragraph> newParas = new ArrayList<Paragraph>();
- for (Paragraph para : paras) {
- newParas.addAll(requotify(para));
- }
- paras = newParas;
-
- // Remove double blanks/brks
- fixBlanksBreaks(paras);
-
- return paras;
- }
-
- /**
- * Fix the {@link ParagraphType#BLANK}s and {@link ParagraphType#BREAK}s of
- * those {@link Paragraph}s.
- * <p>
- * The resulting list will not contain a starting or trailing blank/break
- * nor 2 blanks or breaks following each other.
- *
- * @param paras
- * the list of {@link Paragraph}s to fix
- */
- protected void fixBlanksBreaks(List<Paragraph> paras) {
- boolean space = false;
- boolean brk = true;
- for (int i = 0; i < paras.size(); i++) {
- Paragraph para = paras.get(i);
- boolean thisSpace = para.getType() == ParagraphType.BLANK;
- boolean thisBrk = para.getType() == ParagraphType.BREAK;
-
- if (i > 0 && space && thisBrk) {
- paras.remove(i - 1);
- i--;
- } else if ((space || brk) && (thisSpace || thisBrk)) {
- paras.remove(i);
- i--;
- }
-
- space = thisSpace;
- brk = thisBrk;
- }
-
- // Remove blank/brk at start
- if (paras.size() > 0
- && (paras.get(0).getType() == ParagraphType.BLANK || paras.get(
- 0).getType() == ParagraphType.BREAK)) {
- paras.remove(0);
+ close();
}
- // Remove blank/brk at end
- int last = paras.size() - 1;
- if (paras.size() > 0
- && (paras.get(last).getType() == ParagraphType.BLANK || paras
- .get(last).getType() == ParagraphType.BREAK)) {
- paras.remove(last);
- }
+ return story;
}
/**
- * Get the default cover related to this subject (see <tt>.info</tt> files).
- *
- * @param subject
- * the subject
+ * Process the given story resource into a partially filled {@link Story}
+ * object containing the name and metadata.
*
- * @return the cover if any, or NULL
- */
- static BufferedImage getDefaultCover(String subject) {
- if (subject != null && !subject.isEmpty()
- && Instance.getCoverDir() != null) {
- try {
- File fileCover = new File(Instance.getCoverDir(), subject);
- return getImage(null, fileCover.toURI().toURL(), subject);
- } catch (MalformedURLException e) {
- }
- }
-
- return null;
- }
-
- /**
- * Return the list of supported image extensions.
+ * @param getDesc
+ * retrieve the description of the story, or not
+ * @param pg
+ * the optional progress reporter
*
- * @param emptyAllowed
- * TRUE to allow an empty extension on first place, which can be
- * used when you may already have an extension in your input but
- * are not sure about it
+ * @return the {@link Story}, never NULL
*
- * @return the extensions
+ * @throws IOException
+ * in case of I/O error
*/
- static String[] getImageExt(boolean emptyAllowed) {
- if (emptyAllowed) {
- return new String[] { "", ".png", ".jpg", ".jpeg", ".gif", ".bmp" };
+ protected Story processMeta(boolean getDesc, Progress pg)
+ throws IOException {
+ if (pg == null) {
+ pg = new Progress();
} else {
- return new String[] { ".png", ".jpg", ".jpeg", ".gif", ".bmp" };
+ pg.setMinMax(0, 100);
}
- }
- /**
- * Check if the given resource can be a local image or a remote image, then
- * refresh the cache with it if it is.
- *
- * @param source
- * the story source
- * @param line
- * the resource to check
- *
- * @return the image if found, or NULL
- *
- */
- static BufferedImage getImage(BasicSupport support, URL source, String line) {
- URL url = getImageUrl(support, source, line);
- if (url != null) {
- InputStream in = null;
- try {
- in = Instance.getCache().open(url, getSupport(url), true);
- return IOUtils.toImage(in);
- } catch (IOException e) {
- } finally {
- if (in != null) {
- try {
- in.close();
- } catch (IOException e) {
- }
- }
- }
- }
+ pg.setProgress(30);
- return null;
- }
+ Story story = new Story();
+ MetaData meta = getMeta();
+ if (meta.getCreationDate() == null || meta.getCreationDate().isEmpty()) {
+ meta.setCreationDate(StringUtils.fromTime(new Date().getTime()));
+ }
+ story.setMeta(meta);
- /**
- * Check if the given resource can be a local image or a remote image, then
- * refresh the cache with it if it is.
- *
- * @param source
- * the story source
- * @param line
- * the resource to check
- *
- * @return the image URL if found, or NULL
- *
- */
- static URL getImageUrl(BasicSupport support, URL source, String line) {
- URL url = null;
+ pg.setProgress(50);
- if (line != null) {
- // try for files
- String path = null;
- if (source != null) {
- path = new File(source.getFile()).getParent();
- try {
- String basePath = new File(new File(path), line.trim())
- .getAbsolutePath();
- for (String ext : getImageExt(true)) {
- if (new File(basePath + ext).exists()) {
- url = new File(basePath + ext).toURI().toURL();
- }
- }
- } catch (Exception e) {
- // Nothing to do here
- }
- }
-
- if (url == null) {
- // try for URLs
- try {
- for (String ext : getImageExt(true)) {
- if (Instance.getCache().check(new URL(line + ext))) {
- url = new URL(line + ext);
- break;
- }
- }
+ if (meta.getCover() == null) {
+ meta.setCover(BasicSupportHelper.getDefaultCover(meta.getSubject()));
+ }
- // try out of cache
- if (url == null) {
- for (String ext : getImageExt(true)) {
- try {
- url = new URL(line + ext);
- Instance.getCache().refresh(url, support, true);
- break;
- } catch (IOException e) {
- // no image with this ext
- url = null;
- }
- }
- }
- } catch (MalformedURLException e) {
- // Not an url
- }
- }
+ pg.setProgress(60);
- // refresh the cached file
- if (url != null) {
- try {
- Instance.getCache().refresh(url, support, true);
- } catch (IOException e) {
- // woops, broken image
- url = null;
- }
- }
+ if (getDesc) {
+ String descChapterName = Instance.getTrans().getString(
+ StringId.DESCRIPTION);
+ story.getMeta().setResume(
+ BasicSupportPara.makeChapter(this, source, 0,
+ descChapterName, //
+ getDesc(), isHtml(), null));
}
- return url;
+ pg.setProgress(100);
+ return story;
}
/**
- * Open the input file that will be used through the support.
+ * Actual processing step, without the calls to other methods.
+ * <p>
+ * Will convert the story resource into a fully filled {@link Story} object.
*
- * @param source
- * the source {@link URL}
+ * @param pg
+ * the optional progress reporter
*
- * @return the {@link InputStream}
+ * @return the {@link Story}, never NULL
*
* @throws IOException
* in case of I/O error
*/
- protected InputStream openInput(URL source) throws IOException {
- return Instance.getCache().open(source, this, false);
- }
+ // TODO: add final
+ public Story process(Progress pg) throws IOException {
+ setCurrentReferer(source);
+ login();
+ sourceNode = loadDocument(source);
- /**
- * Reset the given {@link InputStream} and return it.
- *
- * @param in
- * the {@link InputStream} to reset
- *
- * @return the same {@link InputStream} after reset
- */
- protected InputStream reset(InputStream in) {
try {
- in.reset();
- } catch (IOException e) {
+ return doProcess(pg);
+ } finally {
+ close();
}
- return in;
- }
-
- /**
- * Reset then return {@link BasicSupport#in}.
- *
- * @return {@link BasicSupport#in}
- */
- protected InputStream getInput() {
- return reset(in);
}
/**
- * Fix the author name if it is prefixed with some "by" {@link String}.
- *
- * @param author
- * the author with a possible prefix
+ * Process the given story resource into a fully filled {@link Story}
+ * object.
*
- * @return the author without prefixes
- */
- protected String fixAuthor(String author) {
- if (author != null) {
- for (String suffix : new String[] { " ", ":" }) {
- for (String byString : Instance.getConfig()
- .getString(Config.BYS).split(",")) {
- byString += suffix;
- if (author.toUpperCase().startsWith(byString.toUpperCase())) {
- author = author.substring(byString.length()).trim();
- }
- }
- }
-
- // Special case (without suffix):
- if (author.startsWith("©")) {
- author = author.substring(1);
- }
- }
-
- return author;
- }
-
- /**
- * Check quotes for bad format (i.e., quotes with normal paragraphs inside)
- * and requotify them (i.e., separate them into QUOTE paragraphs and other
- * paragraphs (quotes or not)).
+ * @param pg
+ * the optional progress reporter
*
- * @param para
- * the paragraph to requotify (not necessarily a quote)
+ * @return the {@link Story}, never NULL
*
- * @return the correctly (or so we hope) quotified paragraphs
+ * @throws IOException
+ * in case of I/O error
*/
- protected List<Paragraph> requotify(Paragraph para) {
- List<Paragraph> newParas = new ArrayList<Paragraph>();
-
- if (para.getType() == ParagraphType.QUOTE
- && para.getContent().length() > 2) {
- String line = para.getContent();
- boolean singleQ = line.startsWith("" + openQuote);
- boolean doubleQ = line.startsWith("" + openDoubleQuote);
-
- // Do not try when more than one quote at a time
- // (some stories are not easily readable if we do)
- if (singleQ
- && line.indexOf(closeQuote, 1) < line
- .lastIndexOf(closeQuote)) {
- newParas.add(para);
- return newParas;
- }
- if (doubleQ
- && line.indexOf(closeDoubleQuote, 1) < line
- .lastIndexOf(closeDoubleQuote)) {
- newParas.add(para);
- return newParas;
- }
- //
-
- if (!singleQ && !doubleQ) {
- line = openDoubleQuote + line + closeDoubleQuote;
- newParas.add(new Paragraph(ParagraphType.QUOTE, line, para
- .getWords()));
- } else {
- char open = singleQ ? openQuote : openDoubleQuote;
- char close = singleQ ? closeQuote : closeDoubleQuote;
-
- int posDot = -1;
- boolean inQuote = false;
- int i = 0;
- for (char car : line.toCharArray()) {
- if (car == open) {
- inQuote = true;
- } else if (car == close) {
- inQuote = false;
- } else if (car == '.' && !inQuote) {
- posDot = i;
- break;
- }
- i++;
- }
-
- if (posDot >= 0) {
- String rest = line.substring(posDot + 1).trim();
- line = line.substring(0, posDot + 1).trim();
- long words = 1;
- for (char car : line.toCharArray()) {
- if (car == ' ') {
- words++;
- }
- }
- newParas.add(new Paragraph(ParagraphType.QUOTE, line, words));
- if (!rest.isEmpty()) {
- newParas.addAll(requotify(processPara(rest)));
- }
- } else {
- newParas.add(para);
- }
- }
+ public Story doProcess(Progress pg) throws IOException {
+ if (pg == null) {
+ pg = new Progress();
} else {
- newParas.add(para);
+ pg.setMinMax(0, 100);
}
- return newParas;
- }
-
- /**
- * Process a {@link Paragraph} from a raw line of text.
- * <p>
- * Will also fix quotes and HTML encoding if needed.
- *
- * @param line
- * the raw line
- *
- * @return the processed {@link Paragraph}
- */
- protected Paragraph processPara(String line) {
- line = ifUnhtml(line).trim();
-
- boolean space = true;
- boolean brk = true;
- boolean quote = false;
- boolean tentativeCloseQuote = false;
- char prev = '\0';
- int dashCount = 0;
- long words = 1;
-
- StringBuilder builder = new StringBuilder();
- for (char car : line.toCharArray()) {
- if (car != '-') {
- if (dashCount > 0) {
- // dash, ndash and mdash: - – —
- // currently: always use mdash
- builder.append(dashCount == 1 ? '-' : '—');
- }
- dashCount = 0;
- }
-
- if (tentativeCloseQuote) {
- tentativeCloseQuote = false;
- if (Character.isLetterOrDigit(car)) {
- builder.append("'");
- } else {
- // handle double-single quotes as double quotes
- if (prev == car) {
- builder.append(closeDoubleQuote);
- continue;
- } else {
- builder.append(closeQuote);
- }
- }
- }
-
- switch (car) {
- case ' ': // note: unbreakable space
- case ' ':
- case '\t':
- case '\n': // just in case
- case '\r': // just in case
- if (builder.length() > 0
- && builder.charAt(builder.length() - 1) != ' ') {
- words++;
- }
- builder.append(' ');
- break;
-
- case '\'':
- if (space || (brk && quote)) {
- quote = true;
- // handle double-single quotes as double quotes
- if (prev == car) {
- builder.deleteCharAt(builder.length() - 1);
- builder.append(openDoubleQuote);
- } else {
- builder.append(openQuote);
- }
- } else if (prev == ' ' || prev == car) {
- // handle double-single quotes as double quotes
- if (prev == car) {
- builder.deleteCharAt(builder.length() - 1);
- builder.append(openDoubleQuote);
- } else {
- builder.append(openQuote);
- }
- } else {
- // it is a quote ("I'm off") or a 'quote' ("This
- // 'good' restaurant"...)
- tentativeCloseQuote = true;
- }
- break;
+ pg.setProgress(1);
+ Progress pgMeta = new Progress();
+ pg.addProgress(pgMeta, 10);
+ Story story = processMeta(true, pgMeta);
+ if (!pgMeta.isDone()) {
+ pgMeta.setProgress(pgMeta.getMax()); // 10%
+ }
- case '"':
- if (space || (brk && quote)) {
- quote = true;
- builder.append(openDoubleQuote);
- } else if (prev == ' ') {
- builder.append(openDoubleQuote);
- } else {
- builder.append(closeDoubleQuote);
- }
- break;
+ pg.setName("Retrieving " + story.getMeta().getTitle());
- case '-':
- if (space) {
- quote = true;
- } else {
- dashCount++;
- }
- space = false;
- break;
+ Progress pgGetChapters = new Progress();
+ pg.addProgress(pgGetChapters, 10);
+ story.setChapters(new ArrayList<Chapter>());
+ List<Entry<String, URL>> chapters = getChapters(pgGetChapters);
+ if (!pgGetChapters.isDone()) {
+ pgGetChapters.setProgress(pgGetChapters.getMax()); // 20%
+ }
- case '*':
- case '~':
- case '/':
- case '\\':
- case '<':
- case '>':
- case '=':
- case '+':
- case '_':
- case '–':
- case '—':
- space = false;
- builder.append(car);
- break;
+ if (chapters != null) {
+ Progress pgChaps = new Progress("Extracting chapters", 0,
+ chapters.size() * 300);
+ pg.addProgress(pgChaps, 80);
- case '‘':
- case '`':
- case '‹':
- case '﹁':
- case '〈':
- case '「':
- if (space || (brk && quote)) {
- quote = true;
- builder.append(openQuote);
- } else {
- // handle double-single quotes as double quotes
- if (prev == car) {
- builder.deleteCharAt(builder.length() - 1);
- builder.append(openDoubleQuote);
- } else {
- builder.append(openQuote);
- }
+ long words = 0;
+ int i = 1;
+ for (Entry<String, URL> chap : chapters) {
+ pgChaps.setName("Extracting chapter " + i);
+ URL chapUrl = chap.getValue();
+ String chapName = chap.getKey();
+ if (chapUrl != null) {
+ setCurrentReferer(chapUrl);
}
- space = false;
- brk = false;
- break;
- case '’':
- case '›':
- case '﹂':
- case '〉':
- case '」':
- space = false;
- brk = false;
- // handle double-single quotes as double quotes
- if (prev == car) {
- builder.deleteCharAt(builder.length() - 1);
- builder.append(closeDoubleQuote);
- } else {
- builder.append(closeQuote);
+ pgChaps.setProgress(i * 100);
+ Progress pgGetChapterContent = new Progress();
+ Progress pgMakeChapter = new Progress();
+ pgChaps.addProgress(pgGetChapterContent, 100);
+ pgChaps.addProgress(pgMakeChapter, 100);
+
+ String content = getChapterContent(chapUrl, i,
+ pgGetChapterContent);
+ if (!pgGetChapterContent.isDone()) {
+ pgGetChapterContent.setProgress(pgGetChapterContent
+ .getMax());
}
- break;
- case '«':
- case '“':
- case '﹃':
- case '《':
- case '『':
- if (space || (brk && quote)) {
- quote = true;
- builder.append(openDoubleQuote);
- } else {
- builder.append(openDoubleQuote);
+ Chapter cc = BasicSupportPara.makeChapter(this, chapUrl, i,
+ chapName, content, isHtml(), pgMakeChapter);
+ if (!pgMakeChapter.isDone()) {
+ pgMakeChapter.setProgress(pgMakeChapter.getMax());
}
- space = false;
- brk = false;
- break;
- case '»':
- case '”':
- case '﹄':
- case '》':
- case '』':
- space = false;
- brk = false;
- builder.append(closeDoubleQuote);
- break;
+ words += cc.getWords();
+ story.getChapters().add(cc);
+ story.getMeta().setWords(words);
- default:
- space = false;
- brk = false;
- builder.append(car);
- break;
+ i++;
}
- prev = car;
- }
-
- if (tentativeCloseQuote) {
- tentativeCloseQuote = false;
- builder.append(closeQuote);
- }
-
- line = builder.toString().trim();
-
- ParagraphType type = ParagraphType.NORMAL;
- if (space) {
- type = ParagraphType.BLANK;
- } else if (brk) {
- type = ParagraphType.BREAK;
- } else if (quote) {
- type = ParagraphType.QUOTE;
- }
-
- return new Paragraph(type, line, words);
- }
-
- /**
- * Remove the HTML from the input <b>if</b> {@link BasicSupport#isHtml()} is
- * true.
- *
- * @param input
- * the input
- *
- * @return the no html version if needed
- */
- private String ifUnhtml(String input) {
- if (isHtml() && input != null) {
- return StringUtils.unhtml(input);
+ pgChaps.setName("Extracting chapters");
+ } else {
+ pg.setProgress(80);
}
- return input;
+ return story;
}
/**
// TEXT and INFO_TEXT always support files (not URLs though)
for (SupportType type : SupportType.values()) {
if (type != SupportType.TEXT && type != SupportType.INFO_TEXT) {
- BasicSupport support = getSupport(type);
+ BasicSupport support = getSupport(type, url);
if (support != null && support.supports(url)) {
return support;
}
for (SupportType type : new SupportType[] { SupportType.INFO_TEXT,
SupportType.TEXT }) {
- BasicSupport support = getSupport(type);
+ BasicSupport support = getSupport(type, url);
if (support != null && support.supports(url)) {
return support;
}
*
* @param type
* the type
+ * @param url
+ * the {@link URL} to support (can be NULL to get an
+ * "abstract support")
*
* @return an implementation that supports it, or NULL
*/
- public static BasicSupport getSupport(SupportType type) {
+ public static BasicSupport getSupport(SupportType type, URL url) {
+ BasicSupport support = null;
+
switch (type) {
case EPUB:
- return new Epub().setType(type);
+ support = new Epub();
+ break;
case INFO_TEXT:
- return new InfoText().setType(type);
+ support = new InfoText();
+ break;
case FIMFICTION:
- return new Fimfiction().setType(type);
+ try {
+ // Can fail if no client key or NO in options
+ support = new FimfictionApi();
+ } catch (IOException e) {
+ support = new Fimfiction();
+ }
+ break;
case FANFICTION:
- return new Fanfiction().setType(type);
+ support = new Fanfiction();
+ break;
case TEXT:
- return new Text().setType(type);
+ support = new Text();
+ break;
case MANGAFOX:
- return new MangaFox().setType(type);
+ support = new MangaFox();
+ break;
case E621:
- return new E621().setType(type);
+ support = new E621();
+ break;
case YIFFSTAR:
- return new YiffStar().setType(type);
+ support = new YiffStar();
+ break;
+ case E_HENTAI:
+ support = new EHentai();
+ break;
+ case MANGA_LEL:
+ support = new MangaLel();
+ break;
case CBZ:
- return new Cbz().setType(type);
+ support = new Cbz();
+ break;
case HTML:
- return new Html().setType(type);
+ support = new Html();
+ break;
}
- return null;
- }
-
- /**
- * Return the first line from the given input which correspond to the given
- * selectors.
- *
- * @param in
- * the input
- * @param needle
- * a string that must be found inside the target line (also
- * supports "^" at start to say "only if it starts with" the
- * needle)
- * @param relativeLine
- * the line to return based upon the target line position (-1 =
- * the line before, 0 = the target line...)
- *
- * @return the line
- */
- static String getLine(InputStream in, String needle, int relativeLine) {
- return getLine(in, needle, relativeLine, true);
- }
-
- /**
- * Return a line from the given input which correspond to the given
- * selectors.
- *
- * @param in
- * the input
- * @param needle
- * a string that must be found inside the target line (also
- * supports "^" at start to say "only if it starts with" the
- * needle)
- * @param relativeLine
- * the line to return based upon the target line position (-1 =
- * the line before, 0 = the target line...)
- * @param first
- * takes the first result (as opposed to the last one, which will
- * also always spend the input)
- *
- * @return the line
- */
- static String getLine(InputStream in, String needle, int relativeLine,
- boolean first) {
- String rep = null;
-
- try {
- in.reset();
- } catch (IOException e) {
- Instance.syserr(e);
- }
-
- List<String> lines = new ArrayList<String>();
- @SuppressWarnings("resource")
- Scanner scan = new Scanner(in, "UTF-8");
- int index = -1;
- scan.useDelimiter("\\n");
- while (scan.hasNext()) {
- lines.add(scan.next());
-
- if (index == -1) {
- if (needle.startsWith("^")) {
- if (lines.get(lines.size() - 1).startsWith(
- needle.substring(1))) {
- index = lines.size() - 1;
- }
-
- } else {
- if (lines.get(lines.size() - 1).contains(needle)) {
- index = lines.size() - 1;
- }
- }
- }
-
- if (index >= 0 && index + relativeLine < lines.size()) {
- rep = lines.get(index + relativeLine);
- if (first) {
- break;
- }
- }
+ if (support != null) {
+ support.setType(type);
+ support.source = support.getCanonicalUrl(url);
}
- return rep;
+ return support;
}
}