Better computation of getArticleCount.

It corresponds to what is made on libkiwix.
openzim · Nov 2, 2022 · 7a40cce · 7a40cce
1 parent 93d632c
commit 7a40cce
Show file tree

Hide file tree

Showing 3 changed files with 13 additions and 11 deletions.
diff --git a/include/zim/archive.h b/include/zim/archive.h
@@ -140,9 +140,7 @@ namespace zim
        *  The definition of "article" depends of the zim archive.
        *  On recent archives, this correspond to all entries marked as "FRONT_ARTICLE"
        *  at creaton time.
-       *  On old archives, this correspond to all entries in 'A' namespace.
-       *  Few archives may have been created without namespace but also without specific
-       *  article listing. In this case, articles are all user entries.
+       *  On old archives, this corresponds to all "text/html*" entries.
        *
        *  @return the number of articles in the archive.
        */

diff --git a/src/archive.cpp b/src/archive.cpp
@@ -70,15 +70,19 @@ namespace zim
   {
     if (m_impl->hasFrontArticlesIndex()) {
       return m_impl->getFrontEntryCount().v;
-    } else if (m_impl->hasNewNamespaceScheme()) {
-      return m_impl->getNamespaceEntryCount('C').v;
     } else {
-      return m_impl->getNamespaceEntryCount('A').v;
+      try {
+        return countMimeType(
+          getMetadata("Counter"),
+          [](const std::string& mimetype) { return mimetype.find("text/html") == 0; }
+        );
+      } catch(const EntryNotFound& e) {
+        const char articleNs = m_impl->hasNewNamespaceScheme() ? 'C' : 'A';
+        return m_impl->getNamespaceEntryCount(articleNs).v;
+      }
     }
   }
 
-
-
   entry_index_type Archive::getMediaCount() const
   {
     return countMimeType(

diff --git a/test/archive.cpp b/test/archive.cpp
@@ -322,9 +322,9 @@ TEST(ZimArchive, articleNumber)
      // Name                                          mediaCount,  withns                                           nons
      //                                                            {articles, userEntries, allEntries}, {articles, userEntries, allEntries}
     {"small.zim",                                     1,           { 1,       17,          17,       }, { 1,       2,           16        }},
-    {"wikibooks_be_all_nopic_2017-02.zim",            34,          { 70,      118,         118,      }, { 66,      109,         123       }},
-    {"wikibooks_be_all_nopic_2017-02_splitted.zim",   34,          { 70,      118,         118,      }, { 66,      109,         123       }},
-    {"wikipedia_en_climate_change_nopic_2020-01.zim", 333,         { 7253,    7646,        7646,     }, { 1837,    7633,        7649      }}
+    {"wikibooks_be_all_nopic_2017-02.zim",            34,          { 66,      118,         118,      }, { 66,      109,         123       }},
+    {"wikibooks_be_all_nopic_2017-02_splitted.zim",   34,          { 66,      118,         118,      }, { 66,      109,         123       }},
+    {"wikipedia_en_climate_change_nopic_2020-01.zim", 333,         { 1837,    7646,        7646,     }, { 1837,    7633,        7649      }}
   };
   // "withns" zim files have no notion of user entries, so EntryCount == allEntryCount.
   // for small.zim, there is always 1 article, whatever the article is in 'A' namespace or in specific index.