general: use less explicit kompress boilerplate in modules

now get_files/kompress library can handle it transparently

my/core/_deprecated/kompress.py

@@ -257,4 +257,8 @@ class ZipPath(zipfile_Path):
         )
         return os.stat_result(tuple(params.values()))
 
+    @property
+    def suffix(self) -> str:
+        return Path(self.parts[-1]).suffix
+
 # fmt: on

my/core/structure.py

@@ -123,7 +123,8 @@ def match_structure(
 
             searchdir = Path(tempfile.mkdtemp(dir=tdir))
 
-            zf = zipfile.ZipFile(base)
+            # base might already be a ZipPath, and str(base) would end with /
+            zf = zipfile.ZipFile(str(base).rstrip('/'))
             zf.extractall(path=str(searchdir))
 
         else:

my/google/takeout/html.py

@@ -146,12 +146,11 @@ class TakeoutHTMLParser(HTMLParser):
 
 
 def read_html(tpath: Path, file: str) -> Iterable[Parsed]:
-    from ...core.kompress import kopen
     results: List[Parsed] = []
     def cb(dt: datetime, url: Url, title: Title) -> None:
         results.append((dt, url, title))
     parser = TakeoutHTMLParser(callback=cb)
-    with kopen(tpath, file) as fo:
+    with (tpath / file).open() as fo:
         data = fo.read()
         parser.feed(data)
     return results

my/google/takeout/parser.py

@@ -94,10 +94,9 @@ def events(disable_takeout_cache: bool = DISABLE_TAKEOUT_CACHE) -> CacheResults:
     for path in reversed(inputs()):
         with ExitStack() as exit_stack:
             if config._use_zippath:
-                from my.core.kompress import ZipPath
                 # for later takeouts it's just 'Takeout' dir,
                 # but for older (pre 2015) it contains email/date in the subdir name
-                results = tuple(cast(Sequence[Path], ZipPath(path).iterdir()))
+                results = tuple(cast(Sequence[Path], path.iterdir()))
             else:
                 results = exit_stack.enter_context(match_structure(path, expected=EXPECTED, partial=True))
             for m in results:

my/google/takeout/paths.py

@@ -23,8 +23,6 @@ config = make_config(google)
 from pathlib import Path
 from typing import Optional, Iterable
 
-from ...core.kompress import kexists
-
 
 def get_takeouts(*, path: Optional[str]=None) -> Iterable[Path]:
     """
@@ -33,7 +31,7 @@ def get_takeouts(*, path: Optional[str]=None) -> Iterable[Path]:
     # TODO FIXME zip is not great..
     # allow a lambda expression? that way the user could restrict it
     for takeout in get_files(config.takeout_path, glob='*.zip'):
-        if path is None or kexists(takeout, path):
+        if path is None or (takeout / path).exists():
             yield takeout
 
 

my/instagram/gdpr.py

@@ -17,7 +17,6 @@ from my.core import (
     assert_never,
     make_logger,
 )
-from my.core.kompress import ZipPath
 
 from my.config import instagram as user_config
 
@@ -70,7 +69,7 @@ def _decode(s: str) -> str:
 
 
 def _entities() -> Iterator[Res[Union[User, _Message]]]:
-    last = ZipPath(max(inputs()))
+    last = max(inputs())
     # TODO make sure it works both with plan directory
     # idelaly get_files should return the right thing, and we won't have to force ZipPath/match_structure here
     # e.g. possible options are:

my/location/google.py

@@ -21,7 +21,6 @@ import geopy # type: ignore
 
 from ..core.common import LazyLogger, mcachew
 from ..core.cachew import cache_dir
-from ..core import kompress
 
 from my.core.warnings import high
 
@@ -135,7 +134,7 @@ def _iter_locations(path: Path, start=0, stop=None) -> Iterable[Location]:
         ctx = path.open('r')
     else: # must be a takeout archive
         # todo CPath? although not sure if it can be iterative?
-        ctx = kompress.open(path, _LOCATION_JSON)
+        ctx = (path / _LOCATION_JSON).open()
 
     if USE_GREP:
         unzip = f'unzip -p "{path}" "{_LOCATION_JSON}"'

my/stackexchange/gdpr.py

@@ -6,7 +6,7 @@ Stackexchange data (uses [[https://stackoverflow.com/legal/gdpr/request][officia
 
 ### config
 from my.config import stackexchange as user_config
-from ..core import dataclass, PathIsh, make_config
+from ..core import dataclass, PathIsh, make_config, get_files
 @dataclass
 class stackexchange(user_config):
     gdpr_path: PathIsh  # path to GDPR zip file
@@ -61,12 +61,11 @@ class Vote(NamedTuple):
     # todo expose vote type?
 
 import json
-from ..core.kompress import ZipPath
 from ..core.error import Res
 def votes() -> Iterable[Res[Vote]]:
     # TODO there is also some site specific stuff in qa/ directory.. not sure if its' more detailed
     # todo should be defensive? not sure if present when user has no votes
-    path = ZipPath(config.gdpr_path)
+    path = max(get_files(config.gdpr_path))
     votes_path = path / 'analytics' /  'qa\\vote.submit.json'  # yes, it does contain a backslash...
     j = json.loads(votes_path.read_text(encoding='utf-8-sig'))  # not sure why, but this encoding seems necessary
     for r in reversed(j): # they seem to be in decreasing order by default

my/twitter/archive.py

@@ -26,7 +26,6 @@ from functools import cached_property
 import html
 from ..core.common import Paths, datetime_aware
 from ..core.error import Res
-from ..core.kompress import ZipPath
 
 @dataclass
 class twitter_archive(user_config):
@@ -164,9 +163,7 @@ class Like(NamedTuple):
 
 class ZipExport:
     def __init__(self, archive_path: Path) -> None:
-        # todo maybe this should be insude get_files instead, perhps covered with a flag?
-        self.zpath = ZipPath(archive_path)
-
+        self.zpath = archive_path
         if (self.zpath / 'tweets.csv').exists():
             from ..core.warnings import high
             high("NOTE: CSV format (pre ~Aug 2018) isn't supported yet, this is likely not going to work.")