Add local libc database provider for libcdb (#2356)

* Add local libc database provider for libcdb

* Remove unnecessary assert

* Add docstring for local_libcdb

* Suppress warning output while `context.libdb` sets default

* Testing the local system's libc first

* Set falsely `context.lcoal_libcdb` to turn off local libc-database integration

* Fix docstring

* Make path check in validator

* Fix doctests

* Add CHANGELOG

Author: the-soloist

CHANGELOG.md

@@ -71,7 +71,9 @@ The table below shows which release corresponds to each branch, and what date th
 
 ## 4.14.0 (`dev`)
 
+- [#2356][2356] Add local libc database provider for libcdb
 
+[2356]: https://github.com/Gallopsled/pwntools/pull/2356
 
 ## 4.13.0 (`beta`)
 

pwnlib/args.py

@@ -159,6 +159,11 @@ def STDERR(v):
     """Sends logging to ``stderr`` by default, instead of ``stdout``"""
     context.log_console = sys.stderr
 
+def LOCAL_LIBCDB(v):
+    """Sets path to local libc-database via ``context.local_libcdb``, e.g. 
+    ``LOCAL_LIBCDB='/path/to/libc-databse'``"""
+    context.local_libcdb = v
+
 hooks = {
     'LOG_LEVEL': LOG_LEVEL,
     'LOG_FILE': LOG_FILE,
@@ -170,6 +175,7 @@ def STDERR(v):
     'NOASLR': NOASLR,
     'NOPTRACE': NOPTRACE,
     'STDERR': STDERR,
+    'LOCAL_LIBCDB': LOCAL_LIBCDB,
 }
 
 def initialize():

pwnlib/context/__init__.py

@@ -360,6 +360,7 @@ class ContextType(object):
         'endian': 'little',
         'gdbinit': "",
         'kernel': None,
+        'local_libcdb': "/var/lib/libc-database",
         'log_level': logging.INFO,
         'log_file': _devnull(),
         'log_console': sys.stdout,
@@ -1071,6 +1072,32 @@ def log_console(self, stream):
             stream = open(stream, 'wt')
         return stream
 
+    @_validator
+    def local_libcdb(self, path):
+        """ 
+        Sets path to local libc-database, get more information for libc-database:
+        https://github.com/niklasb/libc-database
+
+        Works in :attr:`pwnlib.libcdb` when searching by local database provider.
+
+        The default value is ``/var/lib/libc-database``.
+
+        Sets `context.local_libcdb` to empty string or `None` will turn off local libc-database integration.
+
+        Examples:
+
+            >>> context.local_libcdb = pwnlib.data.elf.path
+            >>> context.local_libcdb = 'foobar'
+            Traceback (most recent call last):
+            ...
+            AttributeError: 'foobar' does not exist, please download libc-database first
+        """
+
+        if not os.path.isdir(path):
+            raise AttributeError("'%s' does not exist, please download libc-database first" % path)
+
+        return path
+
     @property
     def mask(self):
         return (1 << self.bits) - 1

pwnlib/libcdb.py

@@ -11,6 +11,7 @@
 
 from pwnlib.context import context
 from pwnlib.elf import ELF
+from pwnlib.filesystem.path import Path
 from pwnlib.log import getLogger
 from pwnlib.tubes.process import process
 from pwnlib.util.fiddling import enhex
@@ -126,7 +127,23 @@ def provider_local_system(hex_encoded_id, hash_type):
         return local_libc.data
     return None
 
-PROVIDERS = [provider_local_system, provider_libcdb, provider_libc_rip]
+# Offline search https://github.com/niklasb/libc-database for hash type
+def provider_local_database(hex_encoded_id, hash_type):
+    if not context.local_libcdb:
+        return None
+
+    localdb = Path(context.local_libcdb)
+    if not localdb.is_dir():
+        return None
+
+    log.debug("Searching local libc database, %s: %s", hash_type, hex_encoded_id)
+    for libc_path in localdb.rglob("*.so"):
+        if hex_encoded_id == HASHES[hash_type](libc_path):
+            return read(libc_path)
+
+    return None
+
+PROVIDERS = [provider_local_system, provider_local_database, provider_libcdb, provider_libc_rip]
 
 def search_by_hash(hex_encoded_id, hash_type='build_id', unstrip=True):
     assert hash_type in HASHES, hash_type