abe.conf

# Config file for Abe.

# dbtype and connect-args are required.
# If not configured to run under FastCGI, host OR port is required.

# "NAME=VALUE" and "NAME VALUE" are equivalent.  Just "NAME" sets
# value to True.  Values of the form "..." or [...]  or {...} are
# parsed as extended JSON, as are the values true, false, and null.
# Abe extends JSON syntax to allow strings to contain control
# characters (tab, LF, and CR) and to allow comments (from # to end of
# line) between tokens.

# "NAME += VALUE" appends VALUE to an array or updates an object
# defined by NAME.  Hyphen in NAME is read as underscore, so
# "connect_args" and "connect-args" are synonymous.  Later values
# override earlier ones (unless using "+=").

# All values may be passed as command-line arguments by preceding NAME
# with "--".  Example: "python -m Abe.abe --upgrade --dbtype=sqlite3
# --connect-args=abe.sqlite --port 2750".

# dbtype is the name of a Python module.  connect-args are arguments
# to pass to the module's "connect" function.  Consult the module's
# documentation for values.  The user needs permission to create
# tables.  Consult your database documentation on how to configure a
# database.  See also README.txt.

# PostgreSQL example; see also README-POSTGRES.txt:
#dbtype = psycopg2
#connect-args = { "database": "abe" }

# MySQL example; see also README-MYSQL.txt:
#dbtype MySQLdb
#connect-args {"user":"root","db":"abe"}

# SQLite3 example:
#dbtype = sqlite3
#connect-args = abe.sqlite

# Oracle example:
#dbtype=cx_Oracle
#connect-args=user/pass@SID

# ODBC examples.  Don't use the same schema via ODBC and non-ODBC!
#dbtype = pyodbc
#connect-args = DSN=Abe;PWD=B!tc0iN;
#connect-args = DRIVER={ORACLE};DB=ORCL;UID=abe;PWD=Bitcoin;
#connect-args = DRIVER={PostgreSQL ANSI};Database=abe;

# Workaround for ODBC drivers that begin in autocommit mode:
#connect-args {"":"DRIVER={PostgreSQL ANSI};Database=abe;","autocommit":false}

# IBM DB2 example:
#dbtype = ibm_db_dbi
#connect-args {"dsn":"DATABASE=abe;UID=db2inst1;PWD=B!tCo1N","conn_options":{"102":0}}

# Specify port and/or host to serve HTTP instead of FastCGI:
#port 2750
#host localhost

# Specify no-serve to exit immediately after importing block files:
#no-serve

# Specify no-load to start abe server without ever loading the
# blockchain - this is useful if you have a dedicated instance loading
# blocks into your Abe database.
#no-load

# "upgrade" tells Abe to upgrade database objects automatically after
# code updates:
#upgrade
#
# Sometimes upgrades fail and leave the database unusable.  Sometimes
# upgrades take several minutes or longer.  You may get an idea of the
# time required by looking at comments near the bottom of
# Abe/upgrade.py.  For example, if your schema version is Abe18 and
# you are upgrading to schema version Abe19, the relevant lines are:
#
#    ('Abe18',   add_block_num_tx),       # Seconds
#    ('Abe18.1', add_block_ss_destroyed), # Seconds
#    ('Abe18.2', init_block_tx_sums),     # 5 minutes
#    ('Abe18.3', replace_chain_summary),  # Fast
#
# In this case, you could expect a 5-minute upgrade.  The following
# SQL shows the current schema version, provided that it is at least
# Abe13:
#
#    SELECT configvar_value FROM configvar
#     WHERE configvar_name = 'schema_version'
#
# "python -m Abe.abe -v" shows the schema version that the software
# requires.  Sometimes you can upgrade a live database by running
# "python -m Abe.abe --config YOUR_ABE_CONFIG --no-serve --upgrade"
# but this often results in server errors and may sometimes corrupt
# the data.  It is best if the upgrade process has exclusive access to
# the database.

# Include another configuration file, relative to this one's directory.
#config ../secret/abe.conf

# Add or replace object fields.  This example assumes connect-args
# previously held an object such as { "user": "abe" }.
#connect-args += { "password": "B!tc0iN" }

# datadir lists directories to scan for block files.
#datadir += /home/bitcoin/.bitcoin/testnet
#datadir += /home/bitcoin/.namecoin

#datadir += [{
#        "dirname": "/home/bitcoin/.bitcoin",
#        "loader": "rpc"    # See the comments for default-loader below.
#      }]

# datadir can supply information about new currencies.
# Note that "address_version" is a byte string: the byte (or, perhaps
# someday, several bytes) preceding the public key hash before base-58
# conversion.  Example:
#
#datadir += [{
#        "dirname": "/home/weeds/testnet",
#        "chain":   "Weeds",
#        "code3":   "WDS",
#        "address_version": "\u00f3" }]
#datadir += [{
#        "dirname":"/home/namecoin/.namecoin/testnet",
#        "chain":"NCTestnet",
#        "code3":"NC0",
#        "address_version":"\u006f"}]

# ignore-bit8-chains: list of chains for which block version bit 8 does
# NOT indicate a Namecoin-style merge-mined auxiliary proof-of-work.
#ignore-bit8-chains = ["Bitcoin", "Testnet"]

# Set binary-type=hex if you want to store binary data (hashes,
# scripts, keys, etc.) as hexadecimal strings regardless of the
# database's binary data support.
#
# To use experimental support for PostgreSQL BYTEA storage, set
# binary-type=pg-bytea.  This may become the default on databases that
# support it.
#
# To use MySQL binary types, set binary-type=binary. This may become the
# default on databases that support it.
#
# The binary-type setting affects only the first run, where Abe
# creates its tables.
#binary-type hex

# Some databases have trouble with the large integers that Abe uses
# for statistics.  Setting int-type=str causes Abe to pass certain
# integers to the database as strings and cast them to floating point
# in SQL expressions that calculate statistics.  Try this if SQLite
# complains "OverflowError: long too big to convert".
#int-type str

# URL component for static content: logos, style sheets.
#static-path =

# Filesystem location of static content, if served by Abe.
#document-root = Abe/htdocs

# Uncomment "auto-agpl" to add a "Source" link to each page pointing
# to a "/download" URL that streams the directory containing abe.py
# and all subdirectories as a compressed TAR archive.  This exposes
# files outside of the htdocs directory to the client, so use it with
# caution.
#auto-agpl

# Directory name and tarfile name prefix for auto-agpl source
# download.
#download-name = abe

# Primitive template system until I separate view from controller.
#template_vars += {
#    "CONTENT_TYPE": "text/html; charset=utf-8",
#    "APPNAME": "Abe",
#    "DONATIONS_BTC": "1PWC7PNHL1SgvZaN7xEtygenKjWobWsCuf",
#    "DONATIONS_NMC": "NJ3MSELK1cWnqUa6xhF2wUYAnz3RSrWXcK"
#}
#template_vars += { "DONATIONS_BTC": "1YourBitcoinAddressHere" }
#template_vars += { "HOMEPAGE": "chain/Bitcoin" }
#template = "
#<!DOCTYPE html>
#<html lang=\"en\">
#<head>
#    <link rel=\"stylesheet\" type=\"text/css\"
#     href=\"%(dotdot)s%(STATIC_PATH)sabe.css\" />
#    <link rel=\"shortcut icon\" href=\"%(dotdot)s%(STATIC_PATH)sfavicon.ico\" />
#    <title>%(title)s</title>
#</head>
#<body>
#    <h1><a href=\"%(dotdot)s%(HOMEPAGE)s\"><img
#     src=\"%(dotdot)s%(STATIC_PATH)slogo32.png\" alt=\"Abe logo\" /></a> %(h1)s
#    </h1>
#    %(body)s
#    <p style=\"font-size: smaller\">
#        <span style=\"font-style: italic\">
#            Powered by <a href=\"%(ABE_URL)s\">%(APPNAME)s</a>
#        </span>
#        Tips appreciated!
#        <a href=\"%(dotdot)saddress/%(DONATIONS_BTC)s\">BTC</a>
#        <a href=\"%(dotdot)saddress/%(DONATIONS_NMC)s\">NMC</a>
#    </p>
#</body>
#</html>
#"

# Save blocks to the database after reading this many bytes from a
# file.  Abe may run faster with a higher value, for example 10000,
# while loading lots of data with a single process.  Non-zero values
# can lead to errors when two processes load transaction data
# simultaneously.
#commit-bytes = 0

# "rescan" causes Abe to search all block files for new blocks.  This
# can take several minutes on a large chain, longer if many of the
# blocks are not already in Abe's database.  You might want to do this
# if Abe becomes stuck at an old block number: new blocks are loaded
# but not attached to the chain.  You must run once with "rescan"
# after switching to another copy of the block files.
#rescan

# For displaying short links, base-url overrides the first part of the
# URL.  This does not affect the link target, only the visible text.
# The value must include the trailing slash (/) if applicable.
#base-url = http://abe.example.org/

# History pages of addresses with more than this many receipts will be
# not be shown.  This protects against denial of service.  Use -1 for
# no limit.  This also limits the total inputs shown by
# /unspent/ADDR|ADDR|...
#address-history-rows-max 1000

# Argument to logging.config.dictConfig.  Requires Python 2.7 or later.
# http://docs.python.org/library/logging.config.html#logging-config-dictschema
#logging = {
#    "version":1,
#    "handlers": {
#        "console": {
#            "class": "logging.StreamHandler",
#            "formatter": "full",
#            "level": "DEBUG"}},
#    "formatters": {
#        "full": {
#            "format": "%(asctime)s [%(process)d:%(threadName)s] %(name)s %(levelname)s - %(message)s"}},
#    "root": {
#        "handlers": ["console"],
#        "level": "DEBUG"}}

# --log-sql logs all queries.  This is quite verbose, so it is disabled
# by default.
#log-sql

# Create and use the abe_firstbits table.  This affects only the first
# run, where Abe creates its tables, or the first run after an upgrade
# to firstbits-enabled Abe.  This is disabled by default.
#use-firstbits

# shortlink-type=firstbits constructs address short links using
# firstbits.  This is the default when use-firstbits is true.
# Otherwise, shortlink-type is a number of address characters to use.
# Fewer characters result in more collisions, where the page shows a
# list of matches instead of redirecting to the address history.
#shortlink-type 10

# keep-scriptsig=false prevents storage and display of transaction
# input signature scripts (scriptSig) and sequence numbers.  This
# reduces the database size by about 20%.
#keep-scriptsig

# Add transactions to the database.  The genesis transaction is
# unavailable via RPC and must be specified to enable full loading
# over RPC.  See Abe/genesis_tx.py.
#import-tx += 01000000010000000000000000000000000000000000000000000000000000000000000000ffffffff4d04ffff001d0104455468652054696d65732030332f4a616e2f32303039204368616e63656c6c6f72206f6e206272696e6b206f66207365636f6e64206261696c6f757420666f722062616e6b73ffffffff0100f2052a01000000434104678afdb0fe5548271967f1a67130b7105cd6a828e03909a67962e0ea1f61deb649f6bc3f4cef38c4f35504e51ec112de5c384df7ba0b8d578a4c702b6bf11d5fac00000000

# Use default-loader=blkfile to force Abe to load data by scanning
# blockfiles.  Use default-loader=rpc to force the use of RPC
# with a running bitcoind (or *coin daemon).  This requires:
#
#     * A readable bitcoin.conf with the RPC username and password
#
#     * HTTP access to bitcoind (SSL is not yet supported.)
#
#     * Bitcoin 0.8 or newer with the -txindex option.  (Run bitcoind
#       once with -reindex if you previously ran it without -txindex.)
#
#     * A "datadir" option with "chain":"Bitcoin" (or other chain
#       name)
#
#     * The "import-tx" option specifying the genesis transaction, if
#       "bitcoid getrawtransaction" does not return it and it is not
#       in Abe/genesis_tx.py.  (Bitcoin's genesis transaction is in
#       Abe/genesis_tx.py.  Otherwise, it requires manually parsing
#       the first blockfile.)
#
# By default, Abe tries RPC and falls back to blkfile if that fails.
# Use "rpc,blkfile" to specify this behaviour across upgrades, or
# "default" to let Abe choose.  The "loader" specified in a "datadir"
# entry takes precedence over "default-loader".
#
#default-loader = default