Python wrapper around running geth
as a subprocess
This library requires the geth
executable to be present.
Installation
pip install py-geth
To run geth connected to the mainnet
>>> from geth import LiveGethProcess
>>> geth = LiveGethProcess()
>>> geth.start()
Or a private local chain for testing. These require you to give them a name.
>>> from geth import DevGethProcess
>>> geth = DevGethProcess('testing')
>>> geth.start()
By default the DevGethProcess
sets up test chains in the default datadir
used by geth
. If you would like to change the location for these test
chains, you can specify an alternative base_dir
.
>>> geth = DevGethProcess('testing', '/tmp/some-other-base-dir/')
>>> geth.start()
Each instance has a few convenient properties.
>>> geth.data_dir
"~/.ethereum"
>>> geth.rpc_port
8545
>>> geth.ipc_path
"~/.ethereum/geth.ipc"
>>> geth.accounts
['0xd3cda913deb6f67967b99d67acdfa1712c293601']
>>> geth.is_alive
False
>>> geth.is_running
False
>>> geth.is_stopped
False
>>> geth.start()
>>> geth.is_alive
True # indicates that the subprocess hasn't exited
>>> geth.is_running
True # indicates that `start()` has been called (but `stop()` hasn't)
>>> geth.is_stopped
False
>>> geth.stop()
>>> geth.is_alive
False
>>> geth.is_running
False
>>> geth.is_stopped
True
When testing it can be nice to see the logging output produced by the geth
process. py-geth
provides a mixin class that can be used to log the stdout
and stderr output to a logfile.
>>> from geth import LoggingMixin, DevGethProcess
>>> class MyGeth(LoggingMixin, DevGethProcess):
... pass
>>> geth = MyGeth()
>>> geth.start()
All logs will be written to logfiles in ./logs/
in the current directory.
The underlying geth
process can take additional time to open the RPC or IPC
connections, as well as to start mining if it needs to generate the DAG. You
can use the following interfaces to query whether these are ready.
>>> geth.is_rpc_ready
True
>>> geth.wait_for_rpc(timeout=30) # wait up to 30 seconds for the RPC connection to open
>>> geth.is_ipc_ready
True
>>> geth.wait_for_ipc(timeout=30) # wait up to 30 seconds for the IPC socket to open
>>> geth.is_dag_generated
True
>>> geth.is_mining
True
>>> geth.wait_for_dag(timeout=600) # wait up to 10 minutes for the DAG to generate.
The DAG functionality currently only applies to the DAG for epoch 0.
This feature is experimental and subject to breaking changes.
Versions of geth
dating back to v1.9.14 can be installed using py-geth
.
See install.py for
the current list of supported versions.
Installation can be done via the command line:
$ python -m geth.install v1.11.5
Or from python using the install_geth
function.
>>> from geth import install_geth
>>> install_geth('v1.11.5')
The installed binary can be found in the $HOME/.py-geth
directory, under your
home directory. The v1.11.5
binary would be located at
$HOME/.py-geth/geth-v1.11.5/bin/geth
.
The DevGethProcess
is designed to facilitate testing. In that regard, it is
preconfigured as follows.
- A single account is created and allocated 1 billion ether.
- All APIs are enabled on both
rpc
andipc
interfaces. - Account 0 is unlocked
- Networking is configured to not look for or connect to any peers.
- The
networkid
of1234
is used. - Verbosity is set to
5
(DEBUG) - Mining is enabled with a single thread.
- The RPC interface tries to bind to 8545 but will find an open port if this port is not available.
- The DevP2P interface tries to bind to 30303 but will find an open port if this port is not available.
If you are running with mining
enabled, which is default for DevGethProcess
,
then you will likely need to generate the DAG
manually. If you do not, then
it will auto-generate the first time you run the process and this takes a
while.
To generate it manually:
$ geth makedag 0 ~/.ethash
This is especially important in CI environments like Travis-CI where your process will likely timeout during generation.
Clone the repository:
$ git clone [email protected]:ethereum/py-geth.git
Next, run the following from the newly-created py-geth
directory:
$ pip install -e ".[dev]"
You can run the tests with:
pytest tests
Or you can install tox
to run the full test suite.
Pandoc is required for transforming the markdown README to the proper format to render correctly on pypi.
For Debian-like systems:
apt install pandoc
Or on OSX:
brew install pandoc
To release a new version:
make release bump=$$VERSION_PART_TO_BUMP$$
The version format for this repo is {major}.{minor}.{patch}
for stable, and
{major}.{minor}.{patch}-{stage}.{devnum}
for unstable (stage
can be alpha or beta).
To issue the next version in line, specify which part to bump,
like make release bump=minor
or make release bump=devnum
.
If you are in a beta version, make release bump=stage
will switch to a stable.
To issue an unstable version when the current version is stable, specify the
new version explicitly, like make release bump="--new-version 4.0.0-alpha.1 devnum"
There is an automation script to facilitate adding support for new geth versions: update_geth.py
To add support for a geth version, run the following line from the py-geth directory, substituting
the version for the one you wish to add support for. Note that the v
in the versioning is
optional.
$ python update_geth.py v1.10.9
To introduce support for more than one version, pass in the versions in increasing order, ending with the latest version.
$ python update_geth.py v1.10.7 v1.10.8 v1.10.9
Always review your changes before committing as something may cause this existing pattern to change at some point. It is best to compare the git difference with a previous commit that introduced support for a new geth version to make sure everything looks good.