buku/buku
2016-06-12 16:00:54 +05:30

1695 lines
56 KiB
Python
Executable File

#!/usr/bin/env python3
#
# Bookmark management utility
#
# Copyright (C) 2015-2016 Arun Prakash Jana <engineerarun@gmail.com>
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with buku. If not, see <http://www.gnu.org/licenses/>.
import sys
import os
import sqlite3
import argparse
import webbrowser
import html.parser as HTMLParser
from http.client import HTTPConnection, HTTPSConnection
from urllib.parse import urljoin, quote, unquote
import gzip
import io
import signal
# Globals
update = False # Update a bookmark in DB
tagManual = None # Tags for update command
titleManual = None # Manually add a title offline
description = None # Description of the bookmark
tagsearch = False # Search bookmarks by tag
titleData = None # Title fetched from a page
jsonOutput = False # Output json formatted result
showOpt = 0 # Modify show. 1: show only URL, 2: show URL and tag
debug = False # Enable debug logs
pipeargs = [] # Holds arguments piped to the program
noninteractive = False # Show the prompt or not
interrupted = False # Received SIGINT
DELIMITER = ',' # Delimiter used to store tags in DB
_VERSION_ = '2.2' # Program version
# Crypto globals
BLOCKSIZE = 65536
SALT_SIZE = 32
CHUNKSIZE = 0x80000 # Read/write 512 KB chunks
class BMHTMLParser(HTMLParser.HTMLParser):
"""Class to parse and fetch the title from a HTML page, if available"""
def __init__(self):
HTMLParser.HTMLParser.__init__(self)
self.inTitle = False
self.data = ''
self.lasttag = None
def handle_starttag(self, tag, attrs):
self.inTitle = False
if tag == 'title':
self.inTitle = True
self.lasttag = tag
def handle_endtag(self, tag):
global titleData
if tag == 'title':
self.inTitle = False
if self.data != '':
titleData = self.data
self.reset() # We have received title data, exit parsing
def handle_data(self, data):
if self.lasttag == 'title' and self.inTitle:
self.data = '%s%s' % (self.data, data)
def error(self, message):
pass
class BukuCrypt:
""" Class to handle encryption and decryption
of the database file. Functionally a separate entity.
Involves late imports in the static functions but it
saves ~100ms each time. Given that encrypt/decrypt are
not done automatically and any one should be called at
a time, this doesn't seem to be an outrageous approach.
"""
@staticmethod
def get_filehash(filepath):
"""Get the SHA256 hash of a file
Params: path to the file
"""
from hashlib import sha256
with open(filepath, 'rb') as f:
hasher = sha256()
buf = f.read(BLOCKSIZE)
while len(buf) > 0:
hasher.update(buf)
buf = f.read(BLOCKSIZE)
return hasher.digest()
@staticmethod
def encrypt_file(iterations):
"""Encrypt the bookmarks database file"""
try:
from getpass import getpass
import struct
from hashlib import sha256
from cryptography.hazmat.backends import default_backend
from cryptography.hazmat.primitives.ciphers import Cipher, algorithms, modes
except Exception:
print('cryptography missing')
sys.exit(1)
if iterations < 1:
printmsg('Iterations must be >= 1', 'ERROR')
sys.exit(1)
dbpath = os.path.join(BukuDb.get_dbdir_path(), 'bookmarks.db')
encpath = '%s.enc' % dbpath
if not os.path.exists(dbpath):
print('%s missing. Already encrypted?' % dbpath)
sys.exit(1)
# If both encrypted file and flat file exist, error out
if os.path.exists(dbpath) and os.path.exists(encpath):
printmsg('Both encrypted and flat DB files exist!', 'ERROR')
sys.exit(1)
password = ''
password = getpass()
passconfirm = getpass()
if password == '':
print('Empty password')
sys.exit(1)
if password != passconfirm:
print("Passwords don't match")
sys.exit(1)
# Get SHA256 hash of DB file
dbhash = BukuCrypt.get_filehash(dbpath)
# Generate random 256-bit salt and key
salt = os.urandom(SALT_SIZE)
key = ('%s%s' % (password, salt.decode('utf-8', 'replace'))).encode('utf-8')
for _ in range(iterations):
key = sha256(key).digest()
iv = os.urandom(16)
encryptor = Cipher(
algorithms.AES(key),
modes.CBC(iv),
backend=default_backend()
).encryptor()
filesize = os.path.getsize(dbpath)
with open(dbpath, 'rb') as infile:
with open(encpath, 'wb') as outfile:
outfile.write(struct.pack('<Q', filesize))
outfile.write(salt)
outfile.write(iv)
# Embed DB file hash in encrypted file
outfile.write(dbhash)
while True:
chunk = infile.read(CHUNKSIZE)
if len(chunk) == 0:
break
elif len(chunk) % 16 != 0:
chunk = '%s%s' % (chunk, ' ' * (16 - len(chunk) % 16))
outfile.write(encryptor.update(chunk) + encryptor.finalize())
os.remove(dbpath)
print('File encrypted')
sys.exit(0)
@staticmethod
def decrypt_file(iterations):
"""Decrypt the bookmarks database file"""
try:
from getpass import getpass
import struct
from hashlib import sha256
from cryptography.hazmat.backends import default_backend
from cryptography.hazmat.primitives.ciphers import Cipher, algorithms, modes
except Exception:
print('cryptography missing')
sys.exit(1)
if iterations < 1:
printmsg('Decryption failed', 'ERROR')
sys.exit(1)
dbpath = os.path.join(BukuDb.get_dbdir_path(), 'bookmarks.db')
encpath = '%s.enc' % dbpath
if not os.path.exists(encpath):
printmsg('%s missing' % encpath, 'ERROR')
sys.exit(1)
# If both encrypted file and flat file exist, error out
if os.path.exists(dbpath) and os.path.exists(encpath):
printmsg('Both encrypted and flat DB files exist!', 'ERROR')
sys.exit(1)
password = ''
password = getpass()
if password == '':
printmsg('Decryption failed', 'ERROR')
sys.exit(1)
with open(encpath, 'rb') as infile:
origsize = struct.unpack('<Q', infile.read(struct.calcsize('Q')))[0]
# Read 256-bit salt and generate key
salt = infile.read(32)
key = ('%s%s' % (password, salt.decode('utf-8', 'replace'))).encode('utf-8')
for _ in range(iterations):
key = sha256(key).digest()
iv = infile.read(16)
decryptor = Cipher(
algorithms.AES(key),
modes.CBC(iv),
backend=default_backend(),
).decryptor()
# Get original DB file's SHA256 hash from encrypted file
enchash = infile.read(32)
with open(dbpath, 'wb') as outfile:
while True:
chunk = infile.read(CHUNKSIZE)
if len(chunk) == 0:
break
outfile.write(decryptor.update(chunk) + decryptor.finalize())
outfile.truncate(origsize)
# Match hash of generated file with that of original DB file
dbhash = BukuCrypt.get_filehash(dbpath)
if dbhash != enchash:
os.remove(dbpath)
printmsg('Decryption failed', 'ERROR')
sys.exit(1)
else:
os.remove(encpath)
print('File decrypted')
class BukuDb:
def __init__(self, *args, **kwargs):
conn, cur = BukuDb.initdb()
self.conn = conn
self.cur = cur
self.noninteractive = noninteractive
@staticmethod
def get_dbdir_path():
"""Determine the directory path where dbfile will be stored:
if $XDG_DATA_HOME is defined, use it
else if $HOME exists, use it
else use the current directory
"""
data_home = os.environ.get('XDG_DATA_HOME')
if data_home is None:
if os.environ.get('HOME') is None:
data_home = '.'
else:
data_home = os.path.join(os.environ.get('HOME'), '.local', 'share')
return os.path.join(data_home, 'buku')
@staticmethod
def move_legacy_dbfile():
"""Move database file from earlier path used in versions <= 1.8
to new path. Errors out if both the old and new DB files exist.
"""
olddbpath = os.path.join(os.environ.get('HOME'), '.cache', 'buku')
olddbfile = os.path.join(olddbpath, 'bookmarks.db')
if not os.path.exists(olddbfile):
return
newdbpath = BukuDb.get_dbdir_path()
newdbfile = os.path.join(newdbpath, 'bookmarks.db')
if os.path.exists(newdbfile):
print('Both old (%s) and new (%s) databases exist, need manual action' % (olddbfile, newdbfile))
sys.exit(1)
if not os.path.exists(newdbpath):
os.makedirs(newdbpath)
os.rename(olddbfile, newdbfile)
print('Database was moved from old (%s) to new (%s) location.\n' % (olddbfile, newdbfile))
os.rmdir(olddbpath)
@staticmethod
def initdb():
"""Initialize the database connection. Create DB file and/or bookmarks table
if they don't exist. Alert on encryption options on first execution.
Returns: connection, cursor
"""
dbpath = BukuDb.get_dbdir_path()
if not os.path.exists(dbpath):
os.makedirs(dbpath)
dbfile = os.path.join(dbpath, 'bookmarks.db')
encpath = os.path.join(dbpath, 'bookmarks.db.enc')
# Notify if DB file needs to be decrypted first
if os.path.exists(encpath) and not os.path.exists(dbfile):
print('Unlock database first')
sys.exit(1)
# Show info on first creation
if not os.path.exists(dbfile):
print('DB file is being created. You may want to encrypt it later.')
try:
# Create a connection
conn = sqlite3.connect(dbfile)
cur = conn.cursor()
# Create table if it doesn't exist
cur.execute("CREATE TABLE if not exists bookmarks \
(id integer PRIMARY KEY, URL text NOT NULL UNIQUE, metadata text default \'\', tags text default \',\', desc text default \'\')")
conn.commit()
except Exception as e:
print('\x1b[1mEXCEPTION\x1b[21m [initdb]: (%s) %s' % (type(e).__name__, e))
sys.exit(1)
# Add description column in existing DB (from version 2.1)
try:
cur.execute("ALTER TABLE bookmarks ADD COLUMN desc text default \'\'")
conn.commit()
except Exception:
pass
return (conn, cur)
def get_bookmark_by_index(self, index):
"""Get a bookmark from database by its ID.
Return data as a tuple. Return None if index not found.
"""
self.cur.execute('SELECT * FROM bookmarks WHERE id = ?', (index,))
results = self.cur.fetchall()
if len(results) == 0:
return None
else:
return results[0]
def get_bookmark_index(self, url):
"""Check if URL already exists in DB
Params: URL to search
Returns: DB index if URL found, else -1
"""
self.cur.execute('SELECT id FROM bookmarks WHERE URL = ?', (url,))
resultset = self.cur.fetchall()
if len(resultset) == 0:
return -1
return resultset[0][0]
def add_bookmark(self, url, title_manual=None, tag_manual=None, desc=None, delayed_commit=False):
"""Add a new bookmark
:param url: url to bookmark
:param tag_manual: string of comma-separated tags to add manually
:param title_manual: string title to add manually
:param desc: string description
"""
# Ensure that the URL does not exist in DB already
id = self.get_bookmark_index(url)
if id != -1:
print('URL [%s] already exists at index %d' % (url, id))
return
# Process title
if title_manual is not None:
meta = title_manual
else:
meta = network_handler(url)
if meta == '':
print('\x1B[91mTitle: []\x1B[0m\n')
elif debug:
print('Title: [%s]\n' % meta)
# Process tags
if tag_manual is None:
tag_manual = DELIMITER
else:
if tag_manual[0] != DELIMITER:
tag_manual = '%s%s' % (DELIMITER, tag_manual)
if tag_manual[-1] != DELIMITER:
tag_manual = '%s%s' % (tag_manual, DELIMITER)
# Process description
if desc is None:
desc = ''
try:
self.cur.execute('INSERT INTO bookmarks(URL, metadata, tags, desc) VALUES (?, ?, ?, ?)', (url, meta, tag_manual, desc))
if not delayed_commit:
self.conn.commit()
self.print_bookmark(self.cur.lastrowid)
except Exception as e:
print('\x1b[1mEXCEPTION\x1b[21m [add_bookmark]: (%s) %s' % (type(e).__name__, e))
def append_tag_at_index(self, index, tag_manual):
""" Append tags for bookmark at index
:param index: int position of record, 0 for all
:tag_manual: string of comma-separated tags to add manually
"""
if index == 0:
resp = input('Tags will be appended for ALL bookmarks. Enter \x1b[1my\x1b[21m to confirm: ')
if resp != 'y':
return
self.cur.execute('SELECT id, tags FROM bookmarks ORDER BY id ASC')
else:
self.cur.execute('SELECT id, tags FROM bookmarks WHERE id = ?', (index,))
resultset = self.cur.fetchall()
for row in resultset:
tags = '%s%s' % (row[1], tag_manual[1:])
tags = parse_tags([tags])
self.cur.execute('UPDATE bookmarks SET tags = ? WHERE id = ?', (tags, row[0],))
self.conn.commit()
def update_bookmark(self, index, url='', title_manual=None,
tag_manual=None, desc=None, append_tag=False):
""" Update an existing record at index
Update all records if index is 0 and url is not specified.
URL is an exception because URLs are unique in DB.
:param index: int position to update, 0 for all
:param url: address
:param tag_manual: string of comma-separated tags to add manually
:param title_manual: string title to add manually
:param desc: string description
:return:
"""
arguments = []
query = 'UPDATE bookmarks SET'
to_update = False
# Update URL if passed as argument
if url != '':
query = '%s URL = ?,' % query
arguments += (url,)
to_update = True
# Update tags if passed as argument
if tag_manual is not None:
if append_tag:
self.append_tag_at_index(index, tag_manual)
else:
query = '%s tags = ?,' % query
arguments += (tag_manual,)
to_update = True
# Update description if passed as an argument
if desc is not None:
query = '%s desc = ?,' % query
arguments += (desc,)
to_update = True
# Update title
#
# 1. if -t has no arguments, delete existing title
# 2. if -t has arguments, update existing title
# 3. if -t option is omitted at cmdline:
# if URL is passed, update the title from web using the URL
# 4. if no other argument (url, tag, comment) passed update title from web using DB URL
meta = None
if title_manual is not None:
meta = title_manual
elif url != '':
meta = network_handler(url)
if meta == '':
print('\x1B[91mTitle: []\x1B[0m')
elif debug:
print('Title: [%s]' % meta)
elif not to_update and not append_tag:
self.refreshdb(index)
if index > 0:
self.print_bookmark(index)
return
if meta is not None:
query = '%s metadata = ?,' % query
arguments += (meta,)
to_update = True
if not to_update: # Nothing to update
return
if index == 0: # Update all records
if url != '':
printmsg('All URLs cannot be same', 'ERROR')
return
resp = input('ALL bookmarks will be updated. Enter \x1b[1my\x1b[21m to confirm: ')
if resp != 'y':
return
query = query[:-1]
else:
query = '%s WHERE id = ?' % query[:-1]
arguments += (index,)
if debug:
print('query: [%s], args: [%s]' % (query, arguments))
try:
self.cur.execute(query, arguments)
self.conn.commit()
if self.cur.rowcount == 1:
self.print_bookmark(index)
elif self.cur.rowcount == 0:
print('No matching index')
except sqlite3.IntegrityError:
print('URL already exists')
def refreshdb(self, index):
"""Refresh ALL records in the database. Fetch title for each
bookmark from the web and update the records. Doesn't udpate
the record if title is empty.
This API doesn't change DB index, URL or tags of a bookmark.
:param index: index of record to update, or 0 for all records
"""
if index == 0:
self.cur.execute('SELECT id, url FROM bookmarks ORDER BY id ASC')
else:
self.cur.execute('SELECT id, url FROM bookmarks WHERE id = ?', (index,))
resultset = self.cur.fetchall()
for row in resultset:
title = network_handler(row[1])
if title == '':
print('\x1b[1mIndex %d: empty title\x1b[21m\x1B[0m\n' % row[0])
continue
else:
print('Title: [%s]' % title)
self.cur.execute('UPDATE bookmarks SET metadata = ? WHERE id = ?', (title, row[0],))
print('Index %d updated\n' % row[0])
if interrupted:
printmsg('Aborting refreshdb ...', 'WARNING')
break
self.conn.commit()
def searchdb(self, keywords, all_keywords=False, json=False):
"""Search the database for an entries with tags or URL
or title info matching keywords and list those.
:param keywords: keywords to search
:param all_keywords: search any or all keywords
:param json: json formatted output
"""
arguments = []
placeholder = "'%' || ? || '%'"
query = 'SELECT id, url, metadata, tags, desc FROM bookmarks WHERE'
if all_keywords: # Match all keywords in URL or Title
for token in keywords:
query = '%s (tags LIKE (%s) OR URL LIKE (%s) OR metadata LIKE (%s) OR desc LIKE (%s)) AND' % (query, placeholder, placeholder, placeholder, placeholder)
arguments += (token, token, token, token)
query = query[:-4]
else: # Match any keyword in URL or Title
for token in keywords:
query = '%s tags LIKE (%s) OR URL LIKE (%s) OR metadata LIKE (%s) OR desc LIKE (%s) OR' % (query, placeholder, placeholder, placeholder, placeholder)
arguments += (token, token, token, token)
query = query[:-3]
if debug:
print("\"%s\", (%s)" % (query, arguments))
self.cur.execute(query, arguments)
results = self.cur.fetchall()
if len(results) == 0:
return
if not json:
prompt(results, self.noninteractive)
else:
print(format_json(results))
def search_by_tag(self, tag, json=False):
"""Search and list bookmarks with a tag
:param tag: tag to search
:param json: print in json format
"""
self.cur.execute("SELECT id, url, metadata, tags, desc FROM bookmarks WHERE tags LIKE '%' || ? || '%'", (tag,))
results = self.cur.fetchall()
if len(results) == 0:
return
if not json:
prompt(results, self.noninteractive)
else:
print(format_json(results))
def compactdb(self, index):
"""When an entry at index is deleted, move the last
entry in DB to index, if index is lesser.
Params: index of deleted entry
"""
self.cur.execute('SELECT MAX(id) from bookmarks')
results = self.cur.fetchall()
if len(results) == 1 and results[0][0] is None: # Return if the last index was just deleted
return
for row in results:
if row[0] > index:
self.cur.execute('SELECT id, URL, metadata, tags, desc FROM bookmarks WHERE id = ?', (row[0],))
results = self.cur.fetchall()
for row in results:
self.cur.execute('DELETE FROM bookmarks WHERE id = ?', (row[0],))
self.conn.commit()
self.cur.execute('INSERT INTO bookmarks(id, URL, metadata, tags, desc) VALUES (?, ?, ?, ?, ?)', (index, row[1], row[2], row[3], row[4],))
self.conn.commit()
print('Index %d moved to %d' % (row[0], index))
def delete_bookmark(self, index):
"""Delete a single record or remove the table if index is None
Params: index to delete
"""
if index == 0: # Remove the table
resp = input('ALL bookmarks will be removed. Enter \x1b[1my\x1b[21m to confirm: ')
if resp != 'y':
print('No bookmarks deleted')
return
self.cur.execute('DROP TABLE if exists bookmarks')
self.conn.commit()
print('All bookmarks deleted')
else: # Remove a single entry
try:
self.cur.execute('DELETE FROM bookmarks WHERE id = ?', (index,))
self.conn.commit()
if self.cur.rowcount == 1:
print('Removed index %d' % index)
self.compactdb(index)
else:
print('No matching index')
except IndexError:
print('Index out of bound')
def print_bookmark(self, index, empty=False):
"""Print bookmark details at index or all bookmarks if index is None
Print only bookmarks with blank title or tag if empty is True
Note: URL is printed on top because title may be blank
Params: index to print (0 for all)
empty flag to show only bookmarks with no title or tags
"""
if index == 0: # Show all entries
if not empty:
self.cur.execute('SELECT * FROM bookmarks')
resultset = self.cur.fetchall()
else:
self.cur.execute("SELECT * FROM bookmarks WHERE metadata = '' OR tags = ?", (DELIMITER,))
resultset = self.cur.fetchall()
print('\x1b[1m%s records found\x1b[21m\n' % len(resultset))
if not jsonOutput:
if showOpt == 0:
for row in resultset:
print_record(row)
elif showOpt == 1:
for row in resultset:
print('%s\t%s' % (row[0], row[1]))
elif showOpt == 2:
for row in resultset:
print('%s\t%s\t%s' % (row[0], row[1], row[3][1:-1]))
else:
print(format_json(resultset))
else: # Show record at index
try:
self.cur.execute('SELECT * FROM bookmarks WHERE id = ?', (index,))
results = self.cur.fetchall()
if len(results) == 0:
print('No matching index')
return
except IndexError:
print('Index out of bound')
return
if not jsonOutput:
for row in results:
if showOpt == 0:
print_record(row)
elif showOpt == 1:
print('%s\t%s' % (row[0], row[1]))
elif showOpt == 2:
print('%s\t%s\t%s' % (row[0], row[1], row[3][1:-1]))
else:
print(format_json(results, True))
def list_tags(self):
"""Print all unique tags ordered alphabetically"""
count = 1
Tags = []
uniqueTags = []
for row in self.cur.execute('SELECT DISTINCT tags FROM bookmarks ORDER BY tags'):
tagset = row[0].strip(DELIMITER).split(DELIMITER)
for tag in tagset:
if tag not in Tags:
Tags += (tag,)
if Tags[0] == '':
uniqueTags = sorted(Tags[1:], key=str.lower)
else:
uniqueTags = sorted(Tags, key=str.lower)
for tag in uniqueTags:
print('%6d. %s' % (count, tag))
count += 1
def replace_tag(self, orig, new=None):
"""Replace orig tags with new tags in DB for all records.
Remove orig tag if new tag is empty.
Params: original and new tags
"""
update = False
delete = False
newtags = DELIMITER
orig = '%s%s%s' % (DELIMITER, orig, DELIMITER)
if new is None:
delete = True
else:
newtags = parse_tags(new)
if newtags == DELIMITER:
delete = True
if orig == newtags:
print('Tags are same.')
return
self.cur.execute('SELECT id, tags FROM bookmarks WHERE tags LIKE ?', ('%' + orig + '%',))
results = self.cur.fetchall()
for row in results:
if not delete:
# Check if tag newtags is already added
if row[1].find(newtags) >= 0:
newtags = DELIMITER
tags = row[1].replace(orig, newtags)
tags = parse_tags([tags])
self.cur.execute('UPDATE bookmarks SET tags = ? WHERE id = ?', (tags, row[0],))
print('Index %d updated' % row[0])
update = True
if update:
self.conn.commit()
def browse_by_index(self, index):
"""Open URL at index in browser
Params: index
"""
try:
for row in self.cur.execute('SELECT URL FROM bookmarks WHERE id = ?', (index,)):
url = unquote(row[0])
browser_open(url)
return
print('No matching index')
except IndexError:
print('Index out of bound')
def export_bookmark(self, fp):
"""Export bookmarks to a Firefox
bookmarks formatted html file.
Params: Path to file to export to
"""
import time
if os.path.exists(fp):
resp = input('%s exists. Overwrite? Enter \x1b[1my\x1b[21m to confirm: ' % fp)
if resp != 'y':
return
try:
f = open(fp, mode='w', encoding='utf-8')
except Exception as e:
print('\x1b[1mEXCEPTION\x1b[21m [export_bookmark]: (%s) %s' % (type(e).__name__, e))
return
count = 0
timestamp = int(time.time())
self.cur.execute('SELECT * FROM bookmarks')
resultset = self.cur.fetchall()
f.write("""<!DOCTYPE NETSCAPE-Bookmark-file-1>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=UTF-8">
<TITLE>Bookmarks</TITLE>
<H1>Bookmarks</H1>
<DL><p>
<DT><H3 ADD_DATE="%s" LAST_MODIFIED="%s" PERSONAL_TOOLBAR_FOLDER="true">Buku bookmarks</H3>
<DL><p>
""" % (timestamp, timestamp))
for row in resultset:
entry = ' <DT><A HREF="%s" ADD_DATE="%s" LAST_MODIFIED="%s"' % (row[1], timestamp, timestamp)
if row[3] != DELIMITER:
entry = '%s TAGS="%s"' % (entry, row[3][1:-1])
entry = '%s>%s</A>\n' % (entry, row[2])
if row[4] != '':
entry = '%s <DD>%s\n' % (entry, row[4])
f.write(entry)
count += 1
f.write(" </DL><p>\n</DL><p>")
f.close()
print('%s bookmarks exported' % count)
def import_bookmark(self, fp):
"""Import bookmarks from a html file.
Supports Firefox, Google Chrome and IE imports
Params: Path to file to import
"""
try:
import bs4
with open(fp, mode='r', encoding='utf-8') as f:
soup = bs4.BeautifulSoup(f, 'html.parser')
except ImportError:
printmsg('Beautiful Soup not found', 'ERROR')
return
except Exception as e:
print('\x1b[1mEXCEPTION\x1b[21m [import_bookmark]: (%s) %s' % (type(e).__name__, e))
return
html_tags = soup.findAll('a')
for tag in html_tags:
# Extract comment from <dd> tag
desc = None
comment_tag = tag.findNextSibling('dd')
if comment_tag:
desc = comment_tag.text[0:comment_tag.text.find('\n')]
self.add_bookmark(tag['href'],
tag.string,
('%s%s%s' % (DELIMITER, tag['tags'], DELIMITER)) if tag.has_attr('tags') else None,
desc,
True)
self.conn.commit()
f.close()
def mergedb(self, fp):
"""Merge bookmarks from another Buku database file
Params: Path to file to merge
"""
try:
# Connect to input DB
if sys.version_info >= (3,4,4):
# Python 3.4.4 and above
connfp = sqlite3.connect('file:%s?mode=ro' % fp, uri=True)
else:
connfp = sqlite3.connect(fp)
curfp = connfp.cursor()
except Exception as e:
print('\x1b[1mEXCEPTION\x1b[21m [mergedb]: (%s) %s' % (type(e).__name__, e))
sys.exit(1)
curfp.execute('SELECT * FROM bookmarks')
resultset = curfp.fetchall()
for row in resultset:
self.add_bookmark(row[1], row[2], row[3], row[4], True)
self.conn.commit()
try:
curfp.close()
connfp.close()
except Exception:
pass
def close_quit(self, exitval=0):
"""Close a DB connection and exit"""
if self.conn is not None:
try:
self.cur.close()
self.conn.close()
except Exception: # we don't really care about errors, we're closing down anyway
pass
sys.exit(exitval)
# Generic functions
def connect_server(url, fullurl=False, forced=False):
"""Connect to a server and fetch the requested page data.
Supports gzip compression.
If forced is True, for URLs like http://www.domain.com
or http://www.domain.com/ path is www.domain.com or
www.domain.com/ correspondingly.
If fullurl is False, for URLs like http://www.domain.com/,
path is /, else www.domain.com/.
Params: URL to fetch, use complete url as path, force flag
Returns: connection, HTTP(S) GET response
"""
if url.find('%20') != -1:
url = unquote(url).replace(' ', '%20')
else:
url = unquote(url)
if debug:
print('unquoted: %s' % url)
if url.find('https://') >= 0: # Secure connection
server = url[8:]
marker = server.find('/')
if marker > 0:
if not fullurl and not forced:
url = server[marker:]
server = server[:marker]
elif not forced: # Handle domain name without trailing /
url = '/'
urlconn = HTTPSConnection(server, timeout=30)
elif url.find('http://') >= 0: # Insecure connection
server = url[7:]
marker = server.find('/')
if marker > 0:
if not fullurl and not forced:
url = server[marker:]
server = server[:marker]
elif not forced:
url = '/'
urlconn = HTTPConnection(server, timeout=30)
else:
printmsg('Not a valid HTTP(S) url', 'WARNING')
if url.find(':') == -1:
printmsg("Doesn't appear to be a valid url either", 'WARNING')
return (None, None)
if debug:
print('server [%s] url [%s]' % (server, url))
# Handle URLs passed with %xx escape
try:
url.encode('ascii')
except Exception:
url = quote(url)
urlconn.request('GET', url, None, {
'Accept-encoding': 'gzip',
'DNT': '1',
})
return (urlconn, urlconn.getresponse())
def get_page_title(resp):
"""Invoke HTML parser and extract title from HTTP response
Params: GET response and invoke HTML parser
"""
data = None
charset = resp.headers.get_content_charset()
if resp.headers.get('Content-Encoding') == 'gzip':
if debug:
print('gzip response')
data = gzip.GzipFile(fileobj=io.BytesIO(resp.read())).read()
else:
data = resp.read()
if charset is None:
charset = 'utf-8'
if debug:
printmsg('Charset missing in response', 'WARNING')
if debug:
print('charset: %s' % charset)
parser = BMHTMLParser()
try:
if charset == 'utf-8':
parser.feed(data.decode(charset, 'replace'))
else:
parser.feed(data.decode(charset))
except Exception as e:
if debug and str(e) != 'we should not get here!':
# Suppress Exception due to intentional self.reset() in HTMLParser
print('\x1b[1mEXCEPTION\x1b[21m [get_page_title]: (%s) %s' % (type(e).__name__, e))
def network_handler(url):
"""Handle server connection and redirections
Params: URL to fetch
Returns: page title or empty string, if not found
"""
global titleData
titleData = None
urlconn = None
retry = False
try:
urlconn, resp = connect_server(url, False)
while True:
if resp is None:
break
elif resp.status == 200:
get_page_title(resp)
break
elif resp.status in [301, 302]:
redirurl = urljoin(url, resp.getheader('location', ''))
if debug:
printmsg(redirurl, 'REDIRECTION')
retry = False # Reset retry, start fresh on redirection
if redirurl.find('sorry/IndexRedirect?') >= 0: # gracefully handle Google blocks
printmsg('Connection blocked due to unusual activity', 'ERROR')
break
marker = redirurl.find('redirectUrl=')
if marker != -1:
redirurl = redirurl[marker + 12:]
# break same URL redirection loop
if url == redirurl:
printmsg('Detected repeated redirection to same URL', 'ERROR')
break
url = redirurl
urlconn.close()
# Try with complete URL on redirection
urlconn, resp = connect_server(url, True)
elif resp.status == 403 and not retry:
"""Handle URLs of the form https://www.domain.com or
https://www.domain.com/ which fails when trying to fetch
resource '/', retry with full path.
"""
urlconn.close()
if debug:
print('Received status 403: retrying.')
# Remove trailing /
if url[-1] == '/':
url = url[:-1]
urlconn, resp = connect_server(url, False, True)
retry = True
elif resp.status == 500 and not retry:
"""Retry on status 500 (Internal Server Error) with truncated
URL. Some servers support truncated request URL on redirection.
"""
urlconn.close()
if debug:
print('Received status 500: retrying.')
urlconn, resp = connect_server(url, False)
retry = True
else:
printmsg('[%s] %s' % (resp.status, resp.reason), 'ERROR')
break
except Exception as e:
print('\x1b[1mEXCEPTION\x1b[21m [network_handler]: (%s) %s' % (type(e).__name__, e))
finally:
if urlconn is not None:
urlconn.close()
if titleData is None:
return ''
return titleData.strip().replace('\n', '')
def parse_tags(keywords=None):
"""Format and get tag string from tokens"""
if keywords is None:
return None
tags = DELIMITER
origTags = []
uniqueTags = []
# Cleanse and get the tags
tagstr = ' '.join(keywords)
marker = tagstr.find(DELIMITER)
while marker >= 0:
token = tagstr[0:marker]
tagstr = tagstr[marker + 1:]
marker = tagstr.find(DELIMITER)
token = token.strip()
if token == '':
continue
tags = '%s%s%s' % (tags, token, DELIMITER)
tagstr = tagstr.strip()
if tagstr != '':
tags = '%s%s%s' % (tags, tagstr, DELIMITER)
if debug:
print(keywords)
print('tags: [%s]' % tags)
if tags == DELIMITER:
return tags
origTags += tags.strip(DELIMITER).split(DELIMITER)
for tag in origTags:
if tag not in uniqueTags:
uniqueTags += (tag, ) # Select unique tags
# Sort the tags
sortedTags = sorted(uniqueTags, key=str.lower)
# Wrap with delimiter
return '%s%s%s' % (DELIMITER, DELIMITER.join(sortedTags), DELIMITER)
def prompt(results, noninteractive=False):
"""Show each matching result from a search and prompt"""
count = 0
for row in results:
count += 1
print_record(row, count)
if noninteractive:
return
while True:
try:
nav = input('Result number to open: ')
if not nav:
nav = input('Result number to open: ')
if not nav:
# Quit on double enter
break
except EOFError:
return
if is_int(nav):
index = int(nav) - 1
if index < 0 or index >= count:
print('Index out of bound')
continue
try:
browser_open(unquote(results[index][1]))
except Exception as e:
print('\x1b[1mEXCEPTION\x1b[21m [searchdb]: (%s) %s' % (type(e).__name__, e))
else:
break
def print_record(row, count=0):
"""Print a single DB record
Handles differently for search and print (count = 0)
"""
# Print index and URL
if count != 0:
printstr = '\x1B[1m\x1B[93m%d. \x1B[0m\x1B[92m%s\x1B[0m \x1B[1m[%s]\x1B[0m\n' % (count, row[1], row[0])
else:
printstr = '\x1B[1m\x1B[93m%d. \x1B[0m\x1B[92m%s\x1B[0m\n' % (row[0], row[1])
# Print title
if row[2] != '':
printstr = '%s \x1B[91m>\x1B[0m %s\n' % (printstr, row[2])
# Print description
if row[4] != '':
printstr = '%s \x1B[91m+\x1B[0m %s\n' % (printstr, row[4])
# Print tags IF not default (DELIMITER)
if row[3] != DELIMITER:
printstr = '%s \x1B[91m#\x1B[0m %s\n' % (printstr, row[3][1:-1])
print(printstr)
def format_json(resultset, single=False):
"""Return results in Json format"""
global showOpt
if not single:
marks = []
for row in resultset:
if showOpt == 1:
record = {'uri': row[1]}
elif showOpt == 2:
record = {'uri': row[1], 'tags': row[3][1:-1]}
else:
record = {'uri': row[1], 'title': row[2], 'description': row[4], 'tags': row[3][1:-1]}
marks.append(record)
else:
marks = {}
for row in resultset:
if showOpt == 1:
marks['uri'] = row[1]
elif showOpt == 2:
marks['uri'] = row[1]
marks['tags'] = row[3][1:-1]
else:
marks['uri'] = row[1]
marks['title'] = row[2]
marks['description'] = row[4]
marks['tags'] = row[3][1:-1]
return json.dumps(marks, sort_keys=True, indent=4)
def is_int(string):
"""Check if a string is a digit
Params: string
"""
try:
int(string)
return True
except Exception:
return False
def browser_open(url):
"""Duplicate stdin, stdout (to suppress showing errors
on the terminal) and open URL in default browser
Params: url to open
"""
url = url.replace('%22', "\"")
_stderr = os.dup(2)
os.close(2)
_stdout = os.dup(1)
os.close(1)
fd = os.open(os.devnull, os.O_RDWR)
os.dup2(fd, 2)
os.dup2(fd, 1)
try:
webbrowser.open(url)
except Exception as e:
print('\x1b[1mEXCEPTION\x1b[21m [browser_open]: (%s) %s' % (type(e).__name__, e))
finally:
os.close(fd)
os.dup2(_stderr, 2)
os.dup2(_stdout, 1)
def sigint_handler(signum, frame):
"""Custom SIGINT handler"""
global interrupted
interrupted = True
print('\nInterrupted.', file=sys.stderr)
sys.exit(1)
signal.signal(signal.SIGINT, sigint_handler)
def printmsg(msg, level=None):
"""Print a message in 2 parts, with the level in bold
Params: msg, level
"""
if level is not None:
print('\x1b[1m%s: \x1b[21m%s\x1b[0m' % (level, msg))
else:
print('%s' % msg)
# Custom Action classes for argparse
class CustomUpdateAction(argparse.Action):
"""Class to capture if optional param 'update'
is actually used, even if sans arguments
"""
def __call__(self, parser, args, values, option_string=None):
global update
update = True
# NOTE: the following converts a None argument to an empty array []
setattr(args, self.dest, values)
class CustomTagAction(argparse.Action):
"""Class to capture if optional param 'tag'
is actually used, even if sans arguments
"""
def __call__(self, parser, args, values, option_string=None):
global tagManual
tagManual = [DELIMITER, ]
setattr(args, self.dest, values)
class CustomTitleAction(argparse.Action):
"""Class to capture if optional param 'title'
is actually used, even if sans arguments
"""
def __call__(self, parser, args, values, option_string=None):
global titleManual
titleManual = ''
setattr(args, self.dest, values)
class CustomDescAction(argparse.Action):
"""Class to capture if optional param 'comment'
is actually used, even if sans arguments
"""
def __call__(self, parser, args, values, option_string=None):
global description
description = ''
setattr(args, self.dest, values)
class CustomTagSearchAction(argparse.Action):
"""Class to capture if optional param 'stag'
is actually used, even if sans arguments
"""
def __call__(self, parser, args, values, option_string=None):
global tagsearch
tagsearch = True
setattr(args, self.dest, values)
class ExtendedArgumentParser(argparse.ArgumentParser):
"""Extend classic argument parser"""
# Print additional help and info
@staticmethod
def print_extended_help(file=None):
file.write('''
prompt keys:
1-N open the Nth search result in web browser
double Enter exit buku
Version %s
Copyright (C) 2015-2016 Arun Prakash Jana <engineerarun@gmail.com>
License: GPLv3
Webpage: https://github.com/jarun/buku
''' % _VERSION_)
# Help
def print_help(self, file=None):
super(ExtendedArgumentParser, self).print_help(file)
self.print_extended_help(file)
"""main starts here"""
# Handle piped input
def main(argv):
if not sys.stdin.isatty():
pipeargs.extend(argv)
for s in sys.stdin.readlines():
pipeargs.extend(s.split())
if __name__ == '__main__':
try:
main(sys.argv)
except KeyboardInterrupt:
pass
# If piped input, set argument vector
if len(pipeargs) > 0:
sys.argv = pipeargs
# Setup custom argument parser
argparser = ExtendedArgumentParser(
description='A private command-line bookmark manager. Your mini web!',
formatter_class=argparse.RawTextHelpFormatter,
usage='''buku [OPTIONS] [KEYWORD [KEYWORD ...]]''',
add_help=False
)
# General options
general_group = argparser.add_argument_group(
title='general options',
description='''-a, --add URL [tags ...]
bookmark URL with comma-separated tags
-u, --update [N] update fields of bookmark at DB index N
refresh all titles, if no arguments
refresh title of bookmark at N, if only
N is specified without any edit options
-d, --delete [N] delete bookmark at DB index N
delete all bookmarks, if no arguments
-h, --help show this information and exit''')
general_group.add_argument('-a', '--add', nargs='+', dest='addurl', metavar=('URL', 'tags'), help=argparse.SUPPRESS)
general_group.add_argument('-u', '--update', nargs='*', dest='update', action=CustomUpdateAction, metavar=('N', 'URL tags'), help=argparse.SUPPRESS)
general_group.add_argument('-d', '--delete', nargs='?', dest='delete', type=int, const=0, metavar='N', help=argparse.SUPPRESS)
general_group.add_argument('-h', '--help', dest='help', action='store_true', help=argparse.SUPPRESS)
# Edit options
edit_group = argparser.add_argument_group(
title='edit options',
description='''--url keyword specify url, works with -u only
--tag [...] set comma-separated tags, works with -a, -u
clears tags, if no arguments
-t, --title [...] manually set title, works with -a, -u
if no arguments:
-a: do not set title, -u: clear title
-c, --comment [...] description of the bookmark, works with
-a, -u; clears comment, if no arguments''')
edit_group.add_argument('--url', nargs=1, dest='url', metavar='url', help=argparse.SUPPRESS)
edit_group.add_argument('--tag', nargs='*', dest='tag', action=CustomTagAction, metavar='tag', help=argparse.SUPPRESS)
edit_group.add_argument('-t', '--title', nargs='*', dest='title', action=CustomTitleAction, metavar='title', help=argparse.SUPPRESS)
edit_group.add_argument('-c', '--comment', nargs='*', dest='desc', type=str, action=CustomDescAction, metavar='desc', help=argparse.SUPPRESS)
# Search options
search_group = argparser.add_argument_group(
title='search options',
description='''-s, --sany keyword [...]
search bookmarks for ANY matching keyword
-S, --sall keyword [...]
search bookmarks with ALL keywords
special keyword -
"blank": list entries with empty title/tag
--st, --stag [...] search bookmarks by tag
list tags alphabetically, if no arguments''')
search_group.add_argument('-s', '--sany', nargs='+', metavar='keyword', help=argparse.SUPPRESS)
search_group.add_argument('-S', '--sall', nargs='+', metavar='keyword', help=argparse.SUPPRESS)
search_group.add_argument('--st', '--stag', nargs='*', dest='stag', action=CustomTagSearchAction, metavar='keyword', help=argparse.SUPPRESS)
# Encryption options
crypto_group = argparser.add_argument_group(
title='encryption options',
description='''-l, --lock [N] encrypt DB file with N (> 0, default 8)
hash iterations to generate key
-k, --unlock [N] decrypt DB file with N (> 0, default 8)
hash iterations to generate key''')
crypto_group.add_argument('-k', '--unlock', nargs='?', dest='decrypt', type=int, const=8, metavar='N', help=argparse.SUPPRESS)
crypto_group.add_argument('-l', '--lock', nargs='?', dest='encrypt', type=int, const=8, metavar='N', help=argparse.SUPPRESS)
# Power toys
power_group = argparser.add_argument_group(
title='power toys',
description='''-p, --print [N] show details of bookmark at DB index N
show all bookmarks, if no arguments
-f, --format N modify -p output
N=1: show only URL, N=2: show URL and tag
-r, --replace oldtag [newtag ...]
replace oldtag with newtag everywhere
delete oldtag, if no newtag
-j, --json Json formatted output for -p, -s, -S, --st
-e, --export file export bookmarks to Firefox format html
-i, --import file import bookmarks from html file; Firefox,
Google Chrome and IE formats supported
-m, --merge file merge bookmarks from another buku database
--noprompt do not show the prompt, run and exit
-o, --open N open bookmark at DB index N in web browser
-z, --debug show debug information and additional logs''')
power_group.add_argument('-p', '--print', nargs='?', dest='printindex', type=int, const=0, metavar='N', help=argparse.SUPPRESS)
power_group.add_argument('-f', '--format', dest='showOpt', type=int, choices=[1, 2], metavar='N', help=argparse.SUPPRESS)
power_group.add_argument('-r', '--replace', nargs='+', dest='replace', metavar=('oldtag', 'newtag'), help=argparse.SUPPRESS)
power_group.add_argument('-j', '--json', dest='jsonOutput', action='store_true', help=argparse.SUPPRESS)
general_group.add_argument('-e', '--export', nargs=1, dest='export', metavar='file', help=argparse.SUPPRESS)
general_group.add_argument('-i', '--import', nargs=1, dest='imports', metavar='file', help=argparse.SUPPRESS)
general_group.add_argument('-m', '--merge', nargs=1, dest='merge', metavar='file', help=argparse.SUPPRESS)
power_group.add_argument('--noprompt', dest='noninteractive', action='store_true', help=argparse.SUPPRESS)
power_group.add_argument('-o', '--open', dest='openurl', type=int, metavar='N', help=argparse.SUPPRESS)
power_group.add_argument('-z', '--debug', dest='debug', action='store_true', help=argparse.SUPPRESS)
# Show help and exit if no arguments
if len(sys.argv) < 2:
argparser.print_help(sys.stderr)
sys.exit(1)
# Parse the arguments
args = argparser.parse_args()
# Show help and exit if help requested
if args.help:
argparser.print_help(sys.stderr)
sys.exit(0)
# Assign the values to globals
if args.showOpt is not None:
showOpt = args.showOpt
if tagManual is not None and len(args.tag) > 0:
tagManual = args.tag
if titleManual is not None and len(args.title) > 0:
titleManual = ' '.join(args.title)
if description is not None and len(args.desc) > 0:
description = ' '.join(args.desc)
if args.jsonOutput:
import json
jsonOutput = args.jsonOutput
noninteractive = args.noninteractive
debug = args.debug
# Show version in debug logs
if debug:
print('Version %s' % _VERSION_)
# Move pre-1.9 database to new location
BukuDb.move_legacy_dbfile()
# Handle encrypt/decrypt options at top priority
if args.encrypt is not None:
BukuCrypt.encrypt_file(args.encrypt)
if args.decrypt is not None:
BukuCrypt.decrypt_file(args.decrypt)
# Initialize the database and get handles
bdb = BukuDb()
# Add a record
if args.addurl is not None:
# Parse tags into a comma-separated string
tags = DELIMITER
keywords = args.addurl
if tagManual is not None:
if tagManual[0] == '+' and len(tagManual) == 1:
pass
elif tagManual[0] == '+':
tagManual = tagManual[1:]
# In case of add, args.addurl may have URL followed by tags
# Add DELIMITER as url+tags may not end with comma
keywords = args.addurl + [DELIMITER] + tagManual
else:
keywords = args.addurl + [DELIMITER] + tagManual
if len(keywords) > 1:
tags = parse_tags(keywords[1:])
bdb.add_bookmark(args.addurl[0], titleManual, tags, description)
# Update record
if update:
if len(args.update) == 0:
update_index = 0
elif not args.update[0].isdigit():
printmsg('Index must be a number >= 0', 'ERROR')
bdb.close_quit(1)
else:
update_index = int(args.update[0])
if args.url is not None:
new_url = args.url[0]
else:
new_url = ''
append = False
if tagManual is not None:
if tagManual[0] == '+' and len(tagManual) == 1:
tagManual = None
elif tagManual[0] == '+':
tagManual = tagManual[1:]
append = True
# Parse tags into a comma-separated string
tags = parse_tags(tagManual)
bdb.update_bookmark(update_index, new_url, titleManual, tags, description, append)
# Delete record(s)
if args.delete is not None:
if args.delete < 0:
printmsg('Index must be >= 0', 'ERROR')
bdb.close_quit(1)
bdb.delete_bookmark(args.delete)
# Search URLs, titles, tags for any keyword
if args.sany is not None:
bdb.searchdb(args.sany, False, jsonOutput)
# Search URLs, titles, tags with all keywords
if args.sall is not None:
if args.sall[0] == 'blank' and len(args.sall) == 1:
bdb.print_bookmark(0, True)
else:
bdb.searchdb(args.sall, True, jsonOutput)
# Search bookmarks by tag
if tagsearch:
if len(args.stag) > 0:
tag = '%s%s%s' % (DELIMITER, ' '.join(args.stag).strip(DELIMITER), DELIMITER)
bdb.search_by_tag(tag, jsonOutput)
else:
bdb.list_tags()
# Print all records
if args.printindex is not None:
if args.printindex < 0:
printmsg('Index must be >= 0', 'ERROR')
bdb.close_quit(1)
bdb.print_bookmark(args.printindex)
# Replace a tag in DB
if args.replace is not None:
if len(args.replace) == 1:
bdb.replace_tag(args.replace[0])
else:
bdb.replace_tag(args.replace[0], args.replace[1:])
# Export bookmarks
if args.export is not None:
bdb.export_bookmark(args.export[0])
# Import bookmarks
if args.imports is not None:
bdb.import_bookmark(args.imports[0])
# Merge a database file and exit
if args.merge is not None:
bdb.mergedb(args.merge[0])
# Open URL in browser
if args.openurl is not None:
if args.openurl < 1:
printmsg('Index must be >= 1', 'ERROR')
bdb.close_quit(1)
bdb.browse_by_index(args.openurl)
# Close DB connection and quit
bdb.close_quit(0)