Merge pull request #316 from kiwix/tagging_system

Tagging system

include/book.h

@@ -60,6 +60,9 @@ class Book
   const std::string& getUrl() const { return m_url; }
   const std::string& getName() const { return m_name; }
   const std::string& getTags() const { return m_tags; }
+  std::string getTagStr(const std::string& tagName) const;
+  bool getTagBool(const std::string& tagName) const;
+  const std::string& getFlavour() const { return m_flavour; }
   const std::string& getOrigId() const { return m_origId; }
   const uint64_t& getArticleCount() const { return m_articleCount; }
   const uint64_t& getMediaCount() const { return m_mediaCount; }
@@ -81,6 +84,7 @@ class Book
   void setDate(const std::string& date) { m_date = date; }
   void setUrl(const std::string& url) { m_url = url; }
   void setName(const std::string& name) { m_name = name; }
+  void setFlavour(const std::string& flavour) { m_flavour = flavour; }
   void setTags(const std::string& tags) { m_tags = tags; }
   void setOrigId(const std::string& origId) { m_origId = origId; }
   void setArticleCount(uint64_t articleCount) { m_articleCount = articleCount; }
@@ -103,6 +107,7 @@ class Book
   std::string m_date;
   std::string m_url;
   std::string m_name;
+  std::string m_flavour;
   std::string m_tags;
   std::string m_origId;
   uint64_t m_articleCount;

include/tools/otherTools.h

@@ -21,6 +21,7 @@
 #define KIWIX_OTHERTOOLS_H
 
 #include <string>
+#include <vector>
 
 namespace pugi {
   class xml_node;
@@ -28,9 +29,18 @@ namespace pugi {
 
 namespace kiwix
 {
-void sleep(unsigned int milliseconds);
+  void sleep(unsigned int milliseconds);
-std::string nodeToString(const pugi::xml_node& node);
+  std::string nodeToString(const pugi::xml_node& node);
-std::string converta2toa3(const std::string& a2code);
+  std::string converta2toa3(const std::string& a2code);
+
+  /*
+   * Convert all format tag string to new format
+   */
+  std::vector<std::string> convertTags(const std::string& tags_str);
+  std::string getTagValueFromTagList(const std::vector<std::string>& tagList,
+                                     const std::string& tagName);
+  bool convertStrToBool(const std::string& value);
+
 }
 
 #endif

src/book.cpp

@@ -23,6 +23,7 @@
 #include "tools/base64.h"
 #include "tools/regexTools.h"
 #include "tools/networkTools.h"
+#include "tools/otherTools.h"
 
 #include <pugixml.hpp>
 
@@ -58,6 +59,7 @@ bool Book::update(const kiwix::Book& other)
   m_date = other.m_date;
   m_url = other.m_url;
   m_name = other.m_name;
+  m_flavour = other.m_flavour;
   m_tags = other.m_tags;
   m_origId = other.m_origId;
   m_articleCount = other.m_articleCount;
@@ -84,6 +86,7 @@ void Book::update(const kiwix::Reader& reader)
   m_publisher = reader.getPublisher();
   m_date = reader.getDate();
   m_name = reader.getName();
+  m_flavour = reader.getFlavour();
   m_tags = reader.getTags();
   m_origId = reader.getOrigId();
   m_articleCount = reader.getArticleCount();
@@ -111,6 +114,7 @@ void Book::updateFromXml(const pugi::xml_node& node, const std::string& baseDir)
   m_date = ATTR("date");
   m_url = ATTR("url");
   m_name = ATTR("name");
+  m_flavour = ATTR("flavour");
   m_tags = ATTR("tags");
   m_origId = ATTR("origId");
   m_articleCount = strtoull(ATTR("articleCount"), 0, 0);
@@ -141,6 +145,7 @@ void Book::updateFromOpds(const pugi::xml_node& node, const std::string& urlHost
   if (!m_id.compare(0, 9, "urn:uuid:")) {
     m_id.erase(0, 9);
   }
+  // No path on opds.
   m_title = VALUE("title");
   m_description = VALUE("summary");
   m_language = VALUE("language");
@@ -148,6 +153,7 @@ void Book::updateFromOpds(const pugi::xml_node& node, const std::string& urlHost
   m_publisher = node.child("publisher").child("name").child_value();
   m_date = fromOpdsDate(VALUE("updated"));
   m_name = VALUE("name");
+  m_flavour = VALUE("flavour");
   m_tags = VALUE("tags");
   for(auto linkNode = node.child("link"); linkNode;
            linkNode = linkNode.next_sibling("link")) {
@@ -203,4 +209,12 @@ const std::string& Book::getFavicon() const {
   return m_favicon;
 }
 
+std::string Book::getTagStr(const std::string& tagName) const {
+  return getTagValueFromTagList(convertTags(m_tags), tagName);
+}
+
+bool Book::getTagBool(const std::string& tagName) const {
+  return convertStrToBool(getTagStr(tagName));
+}
+
 }

src/libxml_dumper.cpp

@@ -58,6 +58,7 @@ void LibXMLDumper::handleBook(Book book, pugi::xml_node root_node) {
     ADD_ATTR_NOT_EMPTY(entry_node, "creator", book.getCreator());
     ADD_ATTR_NOT_EMPTY(entry_node, "publisher", book.getPublisher());
     ADD_ATTR_NOT_EMPTY(entry_node, "name", book.getName());
+    ADD_ATTR_NOT_EMPTY(entry_node, "flavour", book.getFlavour());
     ADD_ATTR_NOT_EMPTY(entry_node, "tags", book.getTags());
     ADD_ATTR_NOT_EMPTY(entry_node, "faviconMimeType", book.getFaviconMimeType());
     ADD_ATTR_NOT_EMPTY(entry_node, "faviconUrl", book.getFaviconUrl());

src/opds_dumper.cpp

@@ -76,6 +76,7 @@ pugi::xml_node OPDSDumper::handleBook(Book book, pugi::xml_node root_node) {
   ADD_TEXT_ENTRY(entry_node, "language", book.getLanguage());
   ADD_TEXT_ENTRY(entry_node, "updated", gen_date_from_yyyy_mm_dd(book.getDate()));
   ADD_TEXT_ENTRY(entry_node, "name", book.getName());
+  ADD_TEXT_ENTRY(entry_node, "flavour", book.getFlavour());
   ADD_TEXT_ENTRY(entry_node, "tags", book.getTags());
   ADD_TEXT_ENTRY(entry_node, "icon", rootLocation + "/meta?name=favicon&content=" + book.getHumanReadableIdFromPath());
 

src/reader.cpp

@@ -22,6 +22,8 @@
 
 #include <zim/search.h>
 
+#include "tools/otherTools.h"
+
 inline char hi(char v)
 {
   char hex[] = "0123456789abcdef";
@@ -352,43 +354,6 @@ string Reader::getLicense() const
   METADATA("License")
 }
 
-std::vector<std::string> convertTags(const std::string& tags_str)
-{
-  auto tags = split(tags_str, ";");
-  std::vector<std::string> tagsList;
-  bool picSeen(false), vidSeen(false), detSeen(false), indexSeen(false);
-  for (auto tag: tags) {
-    picSeen |= (tag == "nopic" || startsWith(tag, "_pictures:"));
-    vidSeen |= (tag == "novid" || startsWith(tag, "_videos:"));
-    detSeen |= (tag == "nodet" || startsWith(tag, "_details:"));
-    indexSeen |= startsWith(tag, "_ftindex");
-    if (tag == "nopic") {
-      tagsList.push_back("_pictures:no");
-    } else if (tag == "novid") {
-      tagsList.push_back("_videos:no");
-    } else if (tag == "nodet") {
-      tagsList.push_back("_details:no");
-    } else if (tag == "_ftindex") {
-      tagsList.push_back("_ftindex:yes");
-    } else {
-      tagsList.push_back(tag);
-    }
-  }
-  if (!indexSeen) {
-    tagsList.push_back("_ftindex:no");
-  }
-  if (!picSeen) {
-    tagsList.push_back("_pictures:yes");
-  }
-  if (!vidSeen) {
-    tagsList.push_back("_videos:yes");
-  }
-  if (!detSeen) {
-    tagsList.push_back("_details:yes");
-  }
-  return tagsList;
-}
-
 string Reader::getTags(bool original) const
 {
   string tags_str;
@@ -400,26 +365,6 @@ string Reader::getTags(bool original) const
   return join(tags, ";");
 }
 
-string getTagValueFromTagList(const std::vector<std::string>& tagList, const std::string& tagName)
-{
-  for (auto tag: tagList) {
-    if (tag[0] == '_') {
-      auto delimPos = tag.find(':');
-      if (delimPos == string::npos) {
-        // No delimiter... what to do ?
-        continue;
-      }
-      auto cTagName = tag.substr(1, delimPos-1);
-      auto cTagValue = tag.substr(delimPos+1);
-      if (cTagName == tagName) {
-        return cTagValue;
-      }
-    }
-  }
-  std::stringstream ss;
-  ss << tagName << " cannot be found";
-  throw std::out_of_range(ss.str());
-}
 
 string Reader::getTagStr(const std::string& tagName) const
 {
@@ -430,16 +375,7 @@ string Reader::getTagStr(const std::string& tagName) const
 
 bool Reader::getTagBool(const std::string& tagName) const
 {
-  auto tagValue = getTagStr(tagName);
+  return convertStrToBool(getTagStr(tagName));
-  if (tagValue == "yes") {
-    return true;
-  } else if (tagValue == "no") {
-    return false;
-  } else {
-    std::stringstream ss;
-    ss << "Tag value '" << tagValue << "' for " << tagName << " cannot be converted to bool.";
-    throw std::domain_error(ss.str());
-  }
 }
 
 string Reader::getRelation() const

src/tools/otherTools.cpp

@@ -25,7 +25,10 @@
 #include <unistd.h>
 #endif
 
+#include "tools/stringTools.h"
+
 #include <map>
+#include <sstream>
 #include <pugixml.hpp>
 
 
@@ -204,3 +207,76 @@ std::string kiwix::nodeToString(const pugi::xml_node& node)
 std::string kiwix::converta2toa3(const std::string& a2code){
   return codeisomapping.at(a2code);
 }
+
+std::vector<std::string> kiwix::convertTags(const std::string& tags_str)
+{
+  auto tags = kiwix::split(tags_str, ";");
+  std::vector<std::string> tagsList;
+  bool picSeen(false), vidSeen(false), detSeen(false), indexSeen(false);
+  for (auto tag: tags) {
+    picSeen |= (tag == "nopic" || startsWith(tag, "_pictures:"));
+    vidSeen |= (tag == "novid" || startsWith(tag, "_videos:"));
+    detSeen |= (tag == "nodet" || startsWith(tag, "_details:"));
+    indexSeen |= kiwix::startsWith(tag, "_ftindex");
+    if (tag == "nopic") {
+      tagsList.push_back("_pictures:no");
+    } else if (tag == "novid") {
+      tagsList.push_back("_videos:no");
+    } else if (tag == "nodet") {
+      tagsList.push_back("_details:no");
+    } else if (tag == "_ftindex") {
+      tagsList.push_back("_ftindex:yes");
+    } else {
+      tagsList.push_back(tag);
+    }
+  }
+  if (!indexSeen) {
+    tagsList.push_back("_ftindex:no");
+  }
+  if (!picSeen) {
+    tagsList.push_back("_pictures:yes");
+  }
+  if (!vidSeen) {
+    tagsList.push_back("_videos:yes");
+  }
+  if (!detSeen) {
+    tagsList.push_back("_details:yes");
+  }
+  return tagsList;
+}
+
+std::string kiwix::getTagValueFromTagList(
+  const std::vector<std::string>& tagList, const std::string& tagName)
+{
+  for (auto tag: tagList) {
+    if (tag[0] == '_') {
+      auto delimPos = tag.find(':');
+      if (delimPos == std::string::npos) {
+        // No delimiter... what to do ?
+        continue;
+      }
+      auto cTagName = tag.substr(1, delimPos-1);
+      auto cTagValue = tag.substr(delimPos+1);
+      if (cTagName == tagName) {
+        return cTagValue;
+      }
+    }
+  }
+  std::stringstream ss;
+  ss << tagName << " cannot be found";
+  throw std::out_of_range(ss.str());
+}
+
+bool kiwix::convertStrToBool(const std::string& value)
+{
+  if (value == "yes") {
+    return true;
+  } else if (value == "no") {
+    return false;
+  }
+
+  std::stringstream ss;
+  ss << "Tag value '" << value << "' cannot be converted to bool.";
+  throw std::domain_error(ss.str());
+}
+

src/wrapper/java/book.cpp

@@ -58,6 +58,7 @@ GETTER(jstring, getPublisher)
 GETTER(jstring, getDate)
 GETTER(jstring, getUrl)
 GETTER(jstring, getName)
+GETTER(jstring, getFlavour)
 GETTER(jstring, getTags)
 GETTER(jlong, getArticleCount)
 GETTER(jlong, getMediaCount)
@@ -66,4 +67,11 @@ GETTER(jstring, getFavicon)
 GETTER(jstring, getFaviconUrl)
 GETTER(jstring, getFaviconMimeType)
 
+METHOD(jstring, Book, getTagStr, jstring tagName) try {
+  auto cRet = Book->getTagStr(jni2c(tagName, env));
+  return c2jni(cRet, env);
+} catch(...) {
+  return c2jni<std::string>("", env);
+}
+
 #undef GETTER

src/wrapper/java/org/kiwix/kiwixlib/Book.java

@@ -19,7 +19,15 @@ public class Book
   public native String getDate();
   public native String getUrl();
   public native String getName();
+  public native String getFlavour();
   public native String getTags();
+  /**
+   * Return the value associated to the tag tagName
+   *
+   * @param tagName the tag name to search for.
+   * @return The value of the tag. If the tag is not found, return empty string.
+   */
+  public native String getTagStr(String tagName);
 
   public native long getArticleCount();
   public native long getMediaCount();

static/opensearchdescription.xml

@@ -4,6 +4,7 @@
   <Description>Search zim files in the catalog.</Description>
   <Url type="application/atom+xml;profile=opds-catalog"
        xmlns:atom="http://www.w3.org/2005/Atom"
+       xmlns:k="http://kiwix.org/opensearchextension/1.0"
        indexOffset="0"
-       template="/{{root}}/catalog/search?q={searchTerms}&lang={language}&count={count}&start={startIndex}"/>
+       template="/{{root}}/catalog/search?q={searchTerms?}&lang={language?}&name={k:name?}&tag={k:tag?}&notag={k:notag?}&maxsize={k:maxsize?}&count={count?}&start={startIndex?}"/>
 </OpenSearchDescription>