Added doc regarding get_entry_by_title behavior

As get_entry_by_title behavior changes based on how the ZIM was created, it's important
user is informed of this.

As a result, it is clear that example.zim in tests cannot be searched by title at all
so this test was removed

Author: renaud gaudin

libzim/libzim.pyx

@@ -759,12 +759,19 @@ cdef class Archive:
         return Entry.from_entry(move(entry))
 
     def has_entry_by_title(self, title: str) -> pybool:
-        """Whether Archive has en entry with this title"""
+        """Whether Archive has en entry with this title
+
+            Uses get_entry_by_title() so it's specificities apply as well"""
         return self.c_archive.hasEntryByTitle(<string>title.encode('UTF-8'))
 
     def get_entry_by_title(self, title: str) -> Entry:
         """Entry from a title -> Entry
 
+            If ZIM doesn't contain a listing/titleOrdered/v1 entry (most likely
+            because if was created without any FRONT_ARTICLE) then this yields results
+            for matching path if the title was not set at creation time.
+            Otherwise raises KeyError.
+
             Parameters
             ----------
             title : str

tests/test_libzim_reader.py

@@ -469,7 +469,8 @@ def test_reader_get_entries(
     with pytest.raises(KeyError):
         zim.get_entry_by_title("___missing")
 
-    # FIXME: unable to retrieve entry by title on example.zim
+    # example.zim cannot be queried by title as all its entries have been created
+    # with empty titles but the ZIM contains a v1 title listing.
     if test_title and filename != "example.zim":
         assert zim.has_entry_by_title(test_title)
         assert zim.get_entry_by_title(test_title).path == entry.path