Releasing v1.6.0

- Using pylibzim 1.1.0
- Updated tests to match new behavior (duplicates, title index)

Author: rgaudin

CHANGELOG.md

@@ -5,9 +5,17 @@ All notable changes to this project are documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html) (as of version 1.5.0).
 
-## [Unreleased]
+## [1.6.0] - 2022-05-23
 
-- n/a
+## Added
+
+  - `Archive.get_metadata_item()` to retrieve full item instead of just value
+
+### Changed
+
+- Using pylibzim v1.1.0 (using libzim 7.2.1)
+  - Adding duplicate entries now raises RuntimeError
+  - filesize is fixed for larger ZIMs
 
 ## [1.5.0] - 2022-05-09
 

requirements.txt

@@ -1,10 +1,10 @@
-iso-639==0.4.5
+ iso-639==0.4.5
 requests>=2.25.1,<3.0
 colorthief==0.2.1
 python-resize-image>=1.1.19,<1.2
 Babel>=2.9,<3.0
 file-magic>=0.4.0,<0.5
-libzim>=1.0.0,<1.1
+libzim>=1.1.0,<1.2
 beautifulsoup4>=4.9.3,<4.10
 lxml>=4.6.3,<4.7
 optimize-images>=1.3.6,<1.6

src/zimscraperlib/VERSION

@@ -1 +1 @@
-1.5.0.post0
+1.6.0

tests/zim/conftest.py

@@ -39,7 +39,7 @@ def build_data(tmp_path, png_image):
     redirects_file = tmp_path / "redirects.tsv"
     with open(redirects_file, "w") as fh:
         fh.write(" \tAccueil\tBienvenue !!\twelcome\n")
-        fh.write(" \tAccueil\t\tcommons.png\n")
+        fh.write(" \tAccueil2\t\tcommons.png\n")
         fh.write(" \timage\t\tcommons.png\n")
     build_dir = tmp_path / "build"
     return {

tests/zim/test_fs.py

@@ -78,13 +78,13 @@ def test_make_zim_file_working(build_data, png_image):
     assert build_data["fpath"].exists()
     reader = Archive(build_data["fpath"])
     # welcome (actual) and two redirs
-    assert reader.entry_count == 7  # includes redirect
+    assert reader.entry_count == 8  # includes redirect
 
     assert reader.get_item("style.css").mimetype == "text/css"
     assert reader.get_item("app.js").mimetype == "application/javascript"
-    assert reader.get_suggestions_count("bienvenue") == 1
-    assert reader.get_suggestions_count("coucou") == 0
-    assert "Accueil" in list(reader.get_suggestions("bienvenue"))
+    assert reader.get_suggestions_count("bienvenue") == 0
+    assert reader.get_suggestions_count("coucou") == 1
+    assert "welcome" in list(reader.get_suggestions("coucou"))
 
 
 def test_make_zim_file_exceptions_while_building(tmp_path, png_image, build_data):

tests/zim/test_zim_creator.py

@@ -83,12 +83,13 @@ def test_zim_creator(tmp_path, png_image, html_file, html_str):
     # make sure we have our redirects
     assert reader.get_entry_by_path("home2").is_redirect
     assert reader.get_entry_by_path("home2").get_redirect_entry().path == f"{main_path}"
-    # make sure titles were indexed (html with title for xapian)
+    # make sure titles were indexed (html with title for xapian ; redirects are not)
     # see https://github.com/openzim/python-libzim/issues/125
-    assert "home2" in list(reader.get_suggestions("Home !!"))
-    assert "home3" in list(reader.get_suggestions("Home !!"))
-    assert "home4" in list(reader.get_suggestions("Home !!"))
-    assert "images/yahoo.png" not in list(reader.get_suggestions("Home !!"))
+    # see https://github.com/openzim/libzim/issues/642
+    assert "home2" not in list(reader.get_suggestions("Home !!"))  # no is_front > False
+    assert "home3" in list(reader.get_suggestions("Home !!"))  # is_front=True
+    assert "home4" not in list(reader.get_suggestions("Home !!"))  # is_front=False
+    assert "images/yahoo.png" in list(reader.get_suggestions("Home !!"))  # is_frontTrue
     # make sure full text was indexed
     assert reader.get_search_results_count("PDF doc") >= 1
 
@@ -278,6 +279,8 @@ def test_urlitem_nonhtmlgzip(tmp_path, gzip_nonhtml_url):
     fpath = tmp_path / "test.zim"
     with Creator(fpath) as creator:
         creator.add_item(URLItem(url=gzip_nonhtml_url))
+
+    with Creator(fpath) as creator:
         creator.add_item(URLItem(url=gzip_nonhtml_url, use_disk=True))
 
     zim = Archive(fpath)