working consistent list and remove with filtering

archivebox/cli/archivebox_list.py

@@ -5,12 +5,11 @@ __command__ = 'archivebox list'
 __description__ = 'List all the URLs currently in the archive.'
 
 import sys
-import json
 import argparse
 
 
-from ..legacy.util import reject_stdin, ExtendedEncoder
-from ..legacy.main import list_archive_data, csv_format
+from ..legacy.util import reject_stdin, to_json, to_csv
+from ..legacy.main import list_archive_data
 
 
 def main(args=None):
@@ -33,16 +32,10 @@ def main(args=None):
         action='store_true',
         help="Print the output in JSON format with all columns included.",
     )
-    parser.add_argument(
-        '--filter', #'-f',
-        type=str,
-        help="List only URLs matching the given regex pattern.",
-        default=None,
-    )
     parser.add_argument(
         '--sort', #'-s',
         type=str,
-        help="List the links sorted using the given key, e.g. timestamp or updated",
+        help="List the links sorted using the given key, e.g. timestamp or updated.",
         default=None,
     )
     parser.add_argument(
@@ -57,11 +50,26 @@ def main(args=None):
         help="List only URLs bookmarked after the given timestamp.",
         default=None,
     )
+    parser.add_argument(
+        '--filter-type',
+        type=str,
+        choices=('exact', 'substring', 'domain', 'regex'),
+        default='exact',
+        help='Type of pattern matching to use when filtering URLs',
+    )
+    parser.add_argument(
+        'patterns',
+        nargs='*',
+        type=str,
+        default=None,
+        help='List only URLs matching these filter patterns.'
+    )
     command = parser.parse_args(args)
     reject_stdin(__command__)
 
     links = list_archive_data(
-        filter_regex=command.filter,
+        filter_patterns=command.patterns,
+        filter_type=command.filter_type,
         before=command.before,
         after=command.after,
     )
@@ -69,10 +77,9 @@ def main(args=None):
         links = sorted(links, key=lambda link: getattr(link, command.sort))
 
     if command.csv:
-        print(command.csv)
-        print('\n'.join(csv_format(link, command.csv) for link in links))
+        print(to_csv(links, csv_cols=command.csv.split(','), header=True))
     elif command.json:
-        print(json.dumps(list(links), indent=4, cls=ExtendedEncoder))
+        print(to_json(links, indent=4, sort_keys=True))
     else:
         print('\n'.join(link.url for link in links))
     

archivebox/cli/archivebox_remove.py

@@ -0,0 +1,87 @@
+#!/usr/bin/env python3
+
+__package__ = 'archivebox.cli'
+__command__ = 'archivebox remove'
+__description__ = 'Remove the specified URLs from the archive.'
+
+import sys
+import argparse
+
+
+from ..legacy.main import list_archive_data, remove_archive_links
+from ..legacy.util import reject_stdin, to_csv, TimedProgress
+from ..legacy.config import ANSI
+
+
+def main(args=None):
+    args = sys.argv[1:] if args is None else args
+
+    parser = argparse.ArgumentParser(
+        prog=__command__,
+        description=__description__,
+        add_help=True,
+    )
+    parser.add_argument(
+        '--yes', # '-y',
+        action='store_true',
+        help='Remove links instantly without prompting to confirm.',
+    )
+    parser.add_argument(
+        '--delete', # '-r',
+        action='store_true',
+        help=(
+            "In addition to removing the link from the index, "
+            "also delete its archived content and metadata folder."
+        ),
+    )
+    parser.add_argument(
+        '--before', #'-b',
+        type=float,
+        help="List only URLs bookmarked before the given timestamp.",
+        default=None,
+    )
+    parser.add_argument(
+        '--after', #'-a',
+        type=float,
+        help="List only URLs bookmarked after the given timestamp.",
+        default=None,
+    )
+    parser.add_argument(
+        '--filter-type',
+        type=str,
+        choices=('exact', 'substring', 'domain', 'regex'),
+        default='exact',
+        help='Type of pattern matching to use when filtering URLs',
+    )
+    parser.add_argument(
+        'pattern',
+        nargs='?',
+        type=str,
+        default=None,
+        help='URLs matching this filter pattern will be removed from the index.'
+    )
+    command = parser.parse_args(args)
+    reject_stdin(__command__)
+
+    if not sys.stdin.isatty():
+        stdin_raw_text = sys.stdin.read()
+        if stdin_raw_text and command.url:
+            print(
+                '[X] You should pass either a pattern as an argument, '
+                'or pass a list of patterns via stdin, but not both.\n'
+            )
+            raise SystemExit(1)
+
+        patterns = [pattern.strip() for pattern in stdin_raw_text.split('\n')]
+    else:
+        patterns = [command.pattern]
+
+    remove_archive_links(
+        filter_patterns=patterns, filter_type=command.filter_type,
+        before=command.before, after=command.after,
+        yes=command.yes, delete=command.delete,
+    )
+    
+
+if __name__ == '__main__':
+    main()

archivebox/legacy/index.py

@@ -15,6 +15,8 @@ from .config import (
     FOOTER_INFO,
     TIMEOUT,
     URL_BLACKLIST_PTN,
+    ANSI,
+    stderr,
 )
 from .util import (
     scheme,

archivebox/legacy/main.py

@@ -1,10 +1,10 @@
 import re
-import json
+import shutil
 
 from typing import List, Optional, Iterable
 
 from .schema import Link
-from .util import enforce_types, ExtendedEncoder
+from .util import enforce_types, TimedProgress, to_csv
 from .index import (
     links_after_timestamp,
     load_links_index,
@@ -12,6 +12,7 @@ from .index import (
 )
 from .archive_methods import archive_link
 from .config import (
+    ANSI,
     ONLY_NEW,
     OUTPUT_DIR,
     check_dependencies,
@@ -61,23 +62,91 @@ def update_archive_data(import_path: Optional[str]=None, resume: Optional[float]
     return all_links
 
 
+LINK_FILTERS = {
+    'exact': lambda link, pattern: (link.url == pattern) or (link.base_url == pattern),
+    'substring': lambda link, pattern: pattern in link.url,
+    'regex': lambda link, pattern: bool(re.match(pattern, link.url)),
+    'domain': lambda link, pattern: link.domain == pattern,
+}
+
+def link_matches_filter(link: Link, filter_patterns: List[str], filter_type: str='exact') -> bool:
+    for pattern in filter_patterns:
+        if LINK_FILTERS[filter_type](link, pattern):
+            return True
+
+    return False
+
+
 @enforce_types
-def list_archive_data(filter_regex: Optional[str]=None, after: Optional[float]=None, before: Optional[float]=None) -> Iterable[Link]:
+def list_archive_data(filter_patterns: Optional[List[str]]=None, filter_type: str='exact',
+                      after: Optional[float]=None, before: Optional[float]=None) -> Iterable[Link]:
     
     all_links, _ = load_links_index(out_dir=OUTPUT_DIR)
 
-    pattern = re.compile(filter_regex, re.IGNORECASE) if filter_regex else None
-
     for link in all_links:
-        if pattern and not pattern.match(link.url):
-            continue
         if after is not None and float(link.timestamp) < after:
             continue
         if before is not None and float(link.timestamp) > before:
             continue
+        
+        if filter_patterns:
+            if link_matches_filter(link, filter_patterns, filter_type):
+                yield link
+        else:
+            yield link
 
-        yield link
 
+@enforce_types
+def remove_archive_links(filter_patterns: List[str], filter_type: str='exact',
+                         after: Optional[float]=None, before: Optional[float]=None,
+                         yes: bool=False, delete: bool=False):
+    
+    check_dependencies()
+
+    print('[*] Finding links in the archive index matching these {} patterns:'.format(filter_type))
+    print('    {}'.format(' '.join(filter_patterns)))
+    timer = TimedProgress(360, prefix='      ')
+    try:
+        links = list(list_archive_data(
+            filter_patterns=filter_patterns,
+            filter_type=filter_type,
+            after=after,
+            before=before,
+        ))
+    finally:
+        timer.end()
+    if not len(links):
+        print()
+        print('{red}[X] No matching links found.{reset}'.format(**ANSI))
+        raise SystemExit(1)
+
+    print()
+    print('-------------------------------------------------------------------')
+    print(to_csv(links, csv_cols=['link_dir', 'url', 'is_archived', 'num_outputs']))
+    print('-------------------------------------------------------------------')
+    print()
+    if not yes:
+        resp = input('{lightyellow}[?] Are you sure you want to permanently remove these {} archived links? N/y: {reset}'.format(len(links), **ANSI))
+        
+        if not resp.lower() == 'y':
+            raise SystemExit(0)
 
-def csv_format(link: Link, csv_cols: str) -> str:
-    return ','.join(json.dumps(getattr(link, col), cls=ExtendedEncoder) for col in csv_cols.split(','))
+    all_links, _ = load_links_index(out_dir=OUTPUT_DIR)
+    to_keep = []
+
+    for link in all_links:
+        should_remove = (
+            (after is not None and float(link.timestamp) < after)
+            or (before is not None and float(link.timestamp) > before)
+            or link_matches_filter(link, filter_patterns, filter_type)
+        )
+        if not should_remove:
+            to_keep.append(link)
+        elif should_remove and delete:
+            shutil.rmtree(link.link_dir)
+
+    num_removed = len(all_links) - len(to_keep)
+    write_links_index(links=to_keep, out_dir=OUTPUT_DIR, finished=True)
+    print()
+    print('{red}[√] Removed {} out of {} links from the archive index.{reset}'.format(num_removed, len(all_links), **ANSI))
+    print('    Index now contains {} links.'.format(len(to_keep)))

archivebox/legacy/schema.py

@@ -50,16 +50,33 @@ class ArchiveResult:
     def from_json(cls, json_info):
         from .util import parse_date
 
-        allowed_fields = {f.name for f in fields(cls)}
         info = {
             key: val
             for key, val in json_info.items()
-            if key in allowed_fields
+            if key in cls.field_names()
         }
         info['start_ts'] = parse_date(info['start_ts'])
         info['end_ts'] = parse_date(info['end_ts'])
         return cls(**info)
 
+    def to_json(self, indent=4, sort_keys=True):
+        from .util import to_json
+
+        return to_json(self, indent=indent, sort_keys=sort_keys)
+
+    def to_csv(self, cols=None):
+        from .util import to_json
+
+        cols = cols or self.field_names()
+        return ','.join(
+            to_json(getattr(self, col), indent=False)
+            for col in cols
+        )
+    
+    @classmethod
+    def field_names(cls):
+        return [f.name for f in fields(cls)]
+
     @property
     def duration(self) -> int:
         return (self.end_ts - self.start_ts).seconds
@@ -145,11 +162,10 @@ class Link:
     def from_json(cls, json_info):
         from .util import parse_date
         
-        allowed_fields = {f.name for f in fields(cls)}
         info = {
             key: val
             for key, val in json_info.items()
-            if key in allowed_fields
+            if key in cls.field_names()
         }
         info['updated'] = parse_date(info['updated'])
 
@@ -166,6 +182,22 @@ class Link:
         info['history'] = cast_history
         return cls(**info)
 
+    def to_json(self, indent=4, sort_keys=True):
+        from .util import to_json
+
+        return to_json(self, indent=indent, sort_keys=sort_keys)
+
+    def to_csv(self, csv_cols: List[str]):
+        from .util import to_json
+
+        return ','.join(
+            to_json(getattr(self, col), indent=None)
+            for col in csv_cols
+        )
+
+    @classmethod
+    def field_names(cls):
+        return [f.name for f in fields(cls)]
 
     @property
     def link_dir(self) -> str:

archivebox/legacy/util.py

@@ -6,7 +6,7 @@ import time
 import shutil
 
 from json import JSONEncoder
-from typing import List, Optional, Any, Union
+from typing import List, Optional, Any, Union, IO
 from inspect import signature
 from functools import wraps
 from hashlib import sha256
@@ -616,13 +616,27 @@ class ExtendedEncoder(JSONEncoder):
         return JSONEncoder.default(self, obj)
 
 
+def to_json(obj: Any, file: IO=None, indent: Optional[int]=4, sort_keys: bool=True, cls=ExtendedEncoder) -> Optional[str]:
+    if file:
+        json.dump(obj, file, indent=indent, sort_keys=sort_keys, cls=ExtendedEncoder)
+        return None
+    else:
+        return json.dumps(obj, indent=indent, sort_keys=sort_keys, cls=ExtendedEncoder)
+
+
+def to_csv(links: List[Link], csv_cols: Optional[List[str]]=None, header: bool=True) -> str:
+    csv_cols = csv_cols or ['timestamp', 'is_archived', 'url']
+    header_str = '{}\n'.format(','.join(csv_cols)) if header else ''
+    return header_str + '\n'.join(link.to_csv(csv_cols=csv_cols) for link in links)
+
+
 def atomic_write(contents: Union[dict, str], path: str) -> None:
     """Safe atomic write to filesystem by writing to temp file + atomic rename"""
     try:
         tmp_file = '{}.tmp'.format(path)
         with open(tmp_file, 'w+', encoding='utf-8') as f:
             if isinstance(contents, dict):
-                json.dump(contents, f, indent=4, cls=ExtendedEncoder)
+                to_json(contents, file=f)
             else:
                 f.write(contents)