[refactor] Use the bzip2 backend that is now integrated with rapidgzi…

…p 0.11.0+

.github/workflows/tests.yml

@@ -136,7 +136,6 @@ jobs:
           python3 ratarmount.py README.md || [ $? != 139 ]
           python3 -m pip uninstall --yes "$module"
         done
-        python3 -m pip install --upgrade 'git+https://github.com/mxmlnkn/indexed_bzip2.git@master#egginfo=indexed_bzip2&subdirectory=python/indexed_bzip2'
         python3 -m pip install --upgrade 'git+https://github.com/mxmlnkn/indexed_bzip2.git@master#egginfo=rapidgzip&subdirectory=python/rapidgzip'
 
     - name: Test ratarmountcore Installation From Tarball

AppImage/build-ratarmount-appimage.sh

@@ -46,12 +46,9 @@ function installAppImagePythonPackages()
     # The compilation call will contain this: -I/opt/_internal/cpython-3.9.15/include/python3.9
     # Even though it should be like this: -I<path-to-appdir>/opt/_internal/cpython-3.9.15/include/python3.9
     if [[ -n "$USE_CUTTING_EDGE_BACKENDS" ]]; then
-        "$APP_PYTHON_BIN" -I -m pip install --no-cache-dir \
-            'git+https://github.com/mxmlnkn/indexed_bzip2.git@master#egginfo=indexed_bzip2&subdirectory=python/indexed_bzip2'
         "$APP_PYTHON_BIN" -I -m pip install --no-cache-dir \
             'git+https://github.com/mxmlnkn/indexed_bzip2.git@master#egginfo=rapidgzip&subdirectory=python/rapidgzip'
     else
-        "$APP_PYTHON_BIN" -I -m pip install --no-cache-dir indexed_bzip2
         "$APP_PYTHON_BIN" -I -m pip install --no-cache-dir rapidgzip
     fi
     "$APP_PYTHON_BIN" -I -m pip install --no-cache-dir ../core

core/ratarmountcore/SQLiteIndexedTar.py

@@ -18,11 +18,6 @@
 from timeit import default_timer as timer
 from typing import Any, Callable, cast, Dict, Generator, IO, Iterable, List, Optional, Tuple, Union
 
-try:
-    import indexed_bzip2
-except ImportError:
-    pass
-
 try:
     import indexed_gzip
 except ImportError:
@@ -896,9 +891,10 @@ def _updateProgressBar(self, progressBar, fileobj: Any) -> None:
             return
 
         try:
-            if hasattr(fileobj, 'tell_compressed') and (
-                ('indexed_bzip2' in sys.modules and isinstance(fileobj, indexed_bzip2.IndexedBzip2File))
-                or ('rapidgzip' in sys.modules and isinstance(fileobj, rapidgzip.RapidgzipFile))
+            if (
+                hasattr(fileobj, 'tell_compressed')
+                and 'rapidgzip' in sys.modules
+                and (isinstance(fileobj, rapidgzip.IndexedBzip2File) or isinstance(fileobj, rapidgzip.RapidgzipFile))
             ):
                 # Note that because bz2 works on a bitstream the tell_compressed returns the offset in bits
                 progressBar.update(fileobj.tell_compressed() // 8)
@@ -1254,7 +1250,7 @@ def _tryToMarkAsAppended(self, storedStats: Dict[str, Any], archiveStats: os.sta
         # of this size should not take more than 10s, so pretty negligible in my opinion.
         #
         # For compressed archives, detecting appended archives does not help much because the bottleneck is
-        # the decompression not the indexing of files. And because indexed_bzip2 and indexed_gzip probably
+        # the decompression not the indexing of files. And because rapidgzip and indexed_gzip probably
         # assume that the index is complete once import_index has been called, we have to recreate the full
         # block offsets anyway.
         if self.compression:
@@ -1534,7 +1530,7 @@ def _openCompressedFile(
                     fileobj=fileobj, drop_handles=False, spacing=gzipSeekPointSpacing, buffer_size=bufferSize
                 )
         elif compression == 'bz2':
-            tar_file = indexed_bzip2.open(fileobj, parallelization=parallelization)  # type: ignore
+            tar_file = rapidgzip.IndexedBzip2File(fileobj, parallelization=parallelization)  # type: ignore
         elif (
             compression == 'xz'
             and xz

core/ratarmountcore/compressions.py

@@ -10,11 +10,6 @@
 
 from .utils import isLatinAlpha, isLatinDigit, isLatinHexAlpha, formatNumber, ALPHA, DIGITS, HEX
 
-try:
-    import indexed_bzip2
-except ImportError:
-    indexed_bzip2 = None  # type: ignore
-
 try:
     import indexed_gzip
 except ImportError:
@@ -74,7 +69,7 @@
     'bz2': CompressionInfo(
         ['bz2', 'bzip2'],
         ['tb2', 'tbz', 'tbz2', 'tz2'],
-        [CompressionModuleInfo('indexed_bzip2', lambda x: indexed_bzip2.open(x))],  # type: ignore
+        [CompressionModuleInfo('rapidgzip', lambda x: rapidgzip.IndexedBzip2File(x))],  # type: ignore
         lambda x: (x.read(4)[:3] == b'BZh' and x.read(6) == (0x314159265359).to_bytes(6, 'big')),
     ),
     'gz': CompressionInfo(

core/setup.cfg

@@ -41,15 +41,14 @@ install_requires =
 
 [options.extras_require]
 full =
-    indexed_bzip2 >= 1.3.1, < 2.0
     indexed_gzip >= 1.6.3, < 2.0
     indexed_zstd >= 1.3.1, < 2.0; sys_platform=="darwin"
     indexed_zstd >= 1.2.2, < 2.0; platform_system!="Windows"
     python-xz ~= 0.4.0
-    rapidgzip >= 0.10.0
+    rapidgzip >= 0.11.1
     rarfile ~= 4.0
 
-bzip2 = indexed_bzip2 >= 1.3.1, < 2.0
+bzip2 = rapidgzip >= 0.11.1
 gzip  = indexed_gzip >= 1.6.3, < 2.0
 rar   = rarfile ~= 4.0
 xz    = python-xz ~= 0.4.0

core/tests/test_SQLiteIndexedTar.py

@@ -15,7 +15,7 @@
 import tarfile
 import tempfile
 
-import indexed_bzip2
+import rapidgzip
 
 sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), '..')))
 
@@ -48,7 +48,9 @@ def _makeFolder(tarArchive, name):
 
     @staticmethod
     def test_context_manager(parallelization):
-        with SQLiteIndexedTar(findTestFile('single-file.tar'), writeIndex=False, parallelization=parallelization) as indexedTar:
+        with SQLiteIndexedTar(
+            findTestFile('single-file.tar'), writeIndex=False, parallelization=parallelization
+        ) as indexedTar:
             assert indexedTar.listDir('/')
 
     @staticmethod
@@ -438,9 +440,9 @@ def test_appending_to_large_archive(parallelization, tmpdir):
 
         # Create a TAR large in size as well as file count
         tarPath = os.path.join(tmpdir, "foo.tar")
-        with indexed_bzip2.open(findTestFile("tar-with-300-folders-with-1000-files-0B-files.tar.bz2")) as file, open(
-            tarPath, 'wb'
-        ) as extracted:
+        with rapidgzip.IndexedBzip2File(
+            findTestFile("tar-with-300-folders-with-1000-files-0B-files.tar.bz2")
+        ) as file, open(tarPath, 'wb') as extracted:
             while True:
                 data = file.read(1024 * 1024)
                 if not data:

ratarmount.py

@@ -949,9 +949,9 @@ def __call__(self, parser, args, values, option_string=None):
             ("zstd", "/facebook/zstd/dev/LICENSE"),
             ("zlib", "/madler/zlib/master/LICENSE"),
             ("ratarmountcore", "/mxmlnkn/ratarmount/master/core/LICENSE"),
-            ("indexed_bzip2", "/mxmlnkn/indexed_bzip2/master/LICENSE"),
             ("indexed_gzip", "/mxmlnkn/indexed_gzip/master/LICENSE"),
             ("indexed_zstd", "/martinellimarco/indexed_zstd/master/LICENSE"),
+            ("rapidgzip", "/mxmlnkn/rapidgzip/master/LICENSE-MIT"),
         ]:
             licenseUrl = "https://raw.githubusercontent.com" + githubPath
             licenseContents = urllib.request.urlopen(licenseUrl).read().decode()

setup.cfg

@@ -35,13 +35,12 @@ py_modules = ratarmount
 install_requires =
     ratarmountcore ~= 0.6.0
     fusepy
-    indexed_bzip2 >= 1.3.1, < 2.0
     indexed_gzip >= 1.6.3, < 2.0
     indexed_zstd >= 1.3.1, < 2.0; sys_platform=="darwin"
     indexed_zstd >= 1.2.2, < 2.0; platform_system!="Windows"
     dataclasses; python_version < "3.7.0"
     python-xz ~= 0.4.0
-    rapidgzip ~= 0.10.0
+    rapidgzip >= 0.11.1
     rarfile ~= 4.0
 
 [options.extras_require]

tests/runtests.sh

@@ -822,9 +822,9 @@ benchmarkDecoderBackends()
         printf '% 5s : ' "$compression"
         case "$compression" in
             bzip2)
-                python3 -m timeit 'from indexed_bzip2 import IndexedBzip2File as IBF; IBF( '"'$file'"' ).read();'
+                python3 -m timeit 'from rapidgzip import IndexedBzip2File as IBF; IBF( '"'$file'"' ).read();'
                 printf '% 5s : ' "pbz2"
-                python3 -m timeit 'from indexed_bzip2 import IndexedBzip2File as IBF; IBF( '"'$file'"', parallelization = 0 ).read();'
+                python3 -m timeit 'from rapidgzip import IndexedBzip2File as IBF; IBF( '"'$file'"', parallelization = 0 ).read();'
                 ;;
             gzip)
                 python3 -m timeit 'from indexed_gzip import IndexedGzipFile as IBF; IBF( '"'$file'"' ).read();'