Merge pull request #171 from kiwix/remoteContentManager

Remote content manager
This commit is contained in:
Matthieu Gautier 2018-10-24 16:48:52 +02:00 committed by GitHub
commit 50b7e5664a
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
8 changed files with 82 additions and 32 deletions

View File

@ -45,7 +45,7 @@ class Book
bool update(const Book& other);
void update(const Reader& reader);
void updateFromXml(const pugi::xml_node& node, const std::string& baseDir);
void updateFromOpds(const pugi::xml_node& node);
void updateFromOpds(const pugi::xml_node& node, const std::string& urlHost);
std::string getHumanReadableIdFromPath();
bool readOnly() const { return m_readOnly; }
@ -67,7 +67,7 @@ class Book
const uint64_t& getArticleCount() const { return m_articleCount; }
const uint64_t& getMediaCount() const { return m_mediaCount; }
const uint64_t& getSize() const { return m_size; }
const std::string& getFavicon() const { return m_favicon; }
const std::string& getFavicon() const;
const std::string& getFaviconMimeType() const { return m_faviconMimeType; }
const std::string& getDownloadId() const { return m_downloadId; }
@ -115,7 +115,8 @@ class Book
uint64_t m_mediaCount;
bool m_readOnly;
uint64_t m_size;
std::string m_favicon;
mutable std::string m_favicon;
std::string m_faviconUrl;
std::string m_faviconMimeType;
};

View File

@ -33,7 +33,15 @@ class Book;
class OPDSDumper;
enum supportedListSortBy { UNSORTED, TITLE, SIZE, DATE, CREATOR, PUBLISHER };
enum supportedListMode { ALL, REMOTE, LOCAL };
enum supportedListMode {
ALL = 0,
LOCAL = 1,
REMOTE = 1 << 1,
NOLOCAL = 1 << 2,
NOREMOTE = 1 << 3,
VALID = 1 << 4,
NOVALID = 1 << 5
};
/**
* A Library store several books.
*/
@ -127,10 +135,15 @@ class Library
* List books in the library.
*
* @param mode The mode of listing :
* - ALL list all books.
* (LOCAL and REMOTE. Other filters are applied).
* - LOCAL list only local books.
* - REMOTE list only remote books.
* - LOCAL  : list only local books (with a path).
* - REMOTE : list only remote books (with an url).
* - VALID  : list only valid books (without a path or with a
* path pointing to a valid zim file).
* - NOLOCAL : list only books without valid path.
* - NOREMOTE : list only books without url.
* - NOVALID : list only books not valid.
* - ALL : Do not do any filter (LOCAL or REMOTE)
* - Flags can be combined.
* @param sortBy Attribute to sort by the book list.
* @param search List only books with search in the title, description.
* @param language List only books in this language.
@ -141,7 +154,7 @@ class Library
* @return The list of bookIds corresponding to the query.
*/
std::vector<std::string> listBooksIds(
supportedListMode = ALL,
int supportedListMode = ALL,
supportedListSortBy sortBy = UNSORTED,
const std::string& search = "",
const std::string& language = "",

View File

@ -84,6 +84,15 @@ class OPDSDumper
*/
void setSearchDescriptionUrl(const std::string& searchDescriptionUrl) { this->searchDescriptionUrl = searchDescriptionUrl; }
/**
* Set some informations about the search results.
*
* @param totalResult the total number of results of the search.
* @param startIndex the start index of the result.
* @param count the number of result of the current set (or page).
*/
void setOpenSearchInfo(int totalResult, int startIndex, int count);
/**
* Set the library to dump.
*
@ -98,6 +107,10 @@ class OPDSDumper
std::string date;
std::string rootLocation;
std::string searchDescriptionUrl;
int m_totalResults;
int m_startIndex;
int m_count;
bool m_isSearchResult = false;
private:
pugi::xml_node handleBook(Book book, pugi::xml_node root_node);

View File

@ -22,6 +22,7 @@
#include "common/base64.h"
#include "common/regexTools.h"
#include "common/networkTools.h"
#include <pugixml.hpp>
@ -131,7 +132,7 @@ void Book::updateFromXml(const pugi::xml_node& node, const std::string& baseDir)
#define VALUE(name) node.child(name).child_value()
void Book::updateFromOpds(const pugi::xml_node& node)
void Book::updateFromOpds(const pugi::xml_node& node, const std::string& urlHost)
{
m_id = VALUE("id");
if (!m_id.compare(0, 9, "urn:uuid:")) {
@ -149,7 +150,10 @@ void Book::updateFromOpds(const pugi::xml_node& node)
if (rel == "http://opds-spec.org/acquisition/open-access") {
m_url = linkNode.attribute("href").value();
m_size = strtoull(linkNode.attribute("length").value(), 0, 0);
break;
}
if (rel == "http://opds-spec.org/image/thumbnail") {
m_faviconUrl = urlHost + linkNode.attribute("href").value();
m_faviconMimeType = linkNode.attribute("type").value();
}
}
@ -189,4 +193,15 @@ void Book::setIndexPath(const std::string& indexPath)
: indexPath;
}
const std::string& Book::getFavicon() const {
if (m_favicon.empty() && !m_faviconUrl.empty()) {
try {
m_favicon = download(m_faviconUrl);
} catch(...) {
std::cerr << "Cannot download favicon from " << m_faviconUrl;
}
}
return m_favicon;
}
}

View File

@ -191,7 +191,6 @@ std::string kiwix::download(const std::string& url) {
auto curl = curl_easy_init();
std::stringstream ss;
curl_easy_setopt(curl, CURLOPT_URL, url.c_str());
curl_easy_setopt(curl, CURLOPT_PORT, 80);
curl_easy_setopt(curl, CURLOPT_HTTPGET, 1L);
curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, &write_callback_to_iss);
curl_easy_setopt(curl, CURLOPT_WRITEDATA, &ss);

View File

@ -303,7 +303,7 @@ std::string Comparator<PUBLISHER>::get_keys(const std::string& id)
std::vector<std::string> Library::listBooksIds(
supportedListMode mode,
int mode,
supportedListSortBy sortBy,
const std::string& search,
const std::string& language,
@ -314,9 +314,20 @@ std::vector<std::string> Library::listBooksIds(
std::vector<std::string> bookIds;
for(auto& pair:books) {
auto& book = pair.second;
if (mode == LOCAL && book.getPath().empty())
auto local = !book.getPath().empty();
if (mode & LOCAL && !local)
continue;
if (mode == REMOTE && (!book.getPath().empty() || book.getUrl().empty()))
if (mode & NOLOCAL && local)
continue;
auto valid = book.isPathValid();
if (mode & VALID && !valid)
continue;
if (mode & NOVALID && valid)
continue;
auto remote = !book.getUrl().empty();
if (mode & REMOTE && !remote)
continue;
if (mode & NOREMOTE && remote)
continue;
if (maxSize != 0 && book.getSize() > maxSize)
continue;

View File

@ -18,7 +18,6 @@
*/
#include "manager.h"
#include "common/networkTools.h"
#include <pugixml.hpp>
@ -101,22 +100,7 @@ bool Manager::parseOpdsDom(const pugi::xml_document& doc, const std::string& url
kiwix::Book book;
book.setReadOnly(false);
book.updateFromOpds(entryNode);
for(pugi::xml_node linkNode = entryNode.child("link"); linkNode;
linkNode = linkNode.next_sibling("link")) {
std::string rel = linkNode.attribute("rel").value();
if (rel == "http://opds-spec.org/image/thumbnail") {
auto faviconUrl = urlHost + linkNode.attribute("href").value();
try {
book.setFavicon(download(faviconUrl));
book.setFaviconMimeType(linkNode.attribute("type").value());
} catch (...) {
std::cerr << "Cannot get favicon content from " << faviconUrl << std::endl;
}
break;
}
}
book.updateFromOpds(entryNode, urlHost);
/* Update the book properties with the new importer */
manipulator->addBookToLibrary(book);

View File

@ -50,6 +50,14 @@ std::string gen_date_str()
return is.str();
}
void OPDSDumper::setOpenSearchInfo(int totalResults, int startIndex, int count)
{
m_totalResults = totalResults;
m_startIndex = startIndex,
m_count = count;
m_isSearchResult = true;
}
#define ADD_TEXT_ENTRY(node, child, value) (node).append_child((child)).append_child(pugi::node_pcdata).set_value((value).c_str())
pugi::xml_node OPDSDumper::handleBook(Book book, pugi::xml_node root_node) {
@ -98,6 +106,12 @@ string OPDSDumper::dumpOPDSFeed(const std::vector<std::string>& bookIds)
ADD_TEXT_ENTRY(root_node, "title", title);
ADD_TEXT_ENTRY(root_node, "updated", date);
if (m_isSearchResult) {
ADD_TEXT_ENTRY(root_node, "totalResults", to_string(m_totalResults));
ADD_TEXT_ENTRY(root_node, "startIndex", to_string(m_startIndex));
ADD_TEXT_ENTRY(root_node, "itemsPerPage", to_string(m_count));
}
auto self_link_node = root_node.append_child("link");
self_link_node.append_attribute("rel") = "self";
self_link_node.append_attribute("href") = "";