config.py.template

# -*- coding: utf-8 -*-
"""
Configuration file used by the main korp.py script.

Copy this file to config.py and change the settings below.
"""

# Host and port for the WSGI server
WSGI_HOST = "0.0.0.0"
WSGI_PORT = 1234

# The absolute path to the CQP binaries
CQP_EXECUTABLE = ""
CWB_SCAN_EXECUTABLE = ""

# The absolute path to the CWB registry files
CWB_REGISTRY = ""

# The default encoding for the cqp binary
CQP_ENCODING = "UTF-8"

# Locale to use when sorting
LC_COLLATE = "sv_SE.UTF-8"

# The maximum number of search results that can be returned per query (0 = no limit)
MAX_KWIC_ROWS = 0

# Number of threads to use during parallel processing
PARALLEL_THREADS = 3

# Database host and port
DBHOST = "0.0.0.0"
DBPORT = 3306

# Database name
DBNAME = ""

# Database character set (use "utf8mb4" for full Unicode)
DBCHARSET = "utf8"

# Word Picture table prefix
DBWPTABLE = "relations"

# Username and password for database access
DBUSER = ""
DBPASSWORD = ""

# Database collation for lemgram data
DBCOLLATE_LEMGRAM = "utf8_bin"

# URL to authentication server
AUTH_SERVER = ""

# Secret string used when communicating with authentication server
AUTH_SECRET = ""

# A text file with names of corpora needing authentication, one per line
PROTECTED_FILE = ""

# Cache path (optional). Script must have read and write access.
CACHE_DIR = ""

# Disk cache lifespan in minutes
CACHE_LIFESPAN = 20

# List of Memcached servers or sockets (socket paths must start with slash)
MEMCACHED_SERVERS = []

# Size of Memcached client pool
MEMCACHED_POOL_SIZE = 25

# Max number of rows from count command to cache
CACHE_MAX_STATS = 5000

# Corpus configuration directory
CORPUS_CONFIG_DIR = ""

# Set to True to enable "lab mode", potentially enabling experimental features and access to lab-only corpora
LAB_MODE = False

# Whether /corpus_config should create new presets for inlined
# attribute lists and output attribute list presets referenced by name;
# if False, attribute list presets are inlined
CORPUS_CONFIG_ATTRLIST_PRESETS = False

# Possible extra preset types in corpus and attribute definitions:
# object type -> (key in object -> configuration subdirectory), where
# object type is "corpus" or "attribute" and configuration
# subdirectory is also the top-level key for these presets in
# /corpus_config output
CORPUS_CONFIG_EXTRA_PRESETS = {
    "corpus": {
        "context": "contexts",
        "within": "withins",
    },
    "attribute": {
        "opts": "options",
        "translation": "translations",
    },
}

# Base file name for corpus configuration template files in subdirectories
# of "corpora", without a ".yaml" extension
CORPUS_CONFIG_TEMPLATE_BASENAME = "_template"

# Whether corpus results should be sorted by corpus id by default (true) or
# output in the order specified in the corpus parameter (false); this can be
# overridden with the parameter sort_corpora=true|false
SORT_CORPORA_DEFAULT = True

# List of names of plugin modules to load
PLUGINS = []

# Show plugin information in the result of the /info command: "name" = plugin
# names only, "info" = plugin information in the PLUGIN_INFO of the plugin,
# None = nothing
INFO_SHOW_PLUGINS = "info"

# korppluginlib configuration: the values here override those in
# korppluginlib.config; comment these out to use the defaults. For
# more information, see korppluginlib.config.

PLUGINLIB_CONFIG = dict(
    # Packages which may contain plugins; "" for top-level modules
    PACKAGES = ["korpplugins"],
    # Directories to search for plugins (packages) in addition to default ones
    SEARCH_PATH = [],
    # What to do when a plugin is not found: "error", "warn" or "ignore"
    HANDLE_NOT_FOUND = "warn",
    # What to output when loading plugins: 0 = nothing, 1 = plugin names,
    # 2 = plugin and plugin function names
    LOAD_VERBOSITY = 1,
)