traildb / traildb-python Goto Github PK

This project forked from bgyss/traildb-python

Python bindings for TrailDB

Dockerfile 2.36% Python 96.00% Shell 1.64%

traildb-python's Introduction

TrailDB

TrailDB is an efficient tool for storing and querying series of events. This repository contains the core C library and the tdb command line tool.

Learn more at traildb.io.

Quick start

For detailed installation instructions, see Getting Started guide.

Installing binaries

On OSX, TrailDB is available through homebrew:

$ brew install traildb

Linux binaries are not available yet.

Compiling and installing from source

Install Dependencies

$ apt-get install libarchive-dev libjudy-dev pkg-config

For RPM-based distros:

$ yum install judy-devel libarchive-devel pkg-config

For OSX:

$ brew install traildb/judy/judy libarchive pkg-config

For FreeBSD:

$ sudo pkg install python libarchive Judy pkgconf gcc

Note that your systems package manager may have too old of libjudy. You may also require a patch if you are using gcc 4.9.

Build TrailDB

$ ./waf configure
$ ./waf install

Alternatively you may use autotools

$ ./autogen.sh
$ ./configure
$ make
$ make install

Run Tests

$ ./waf test

traildb-python's People

Contributors

Stargazers

Watchers

Forkers

alexk307 ebottabi dfparker2002 michisarm

traildb-python's Issues

Docker container is broken

In the terminal:

docker run -p 8080:8888 -it c3h3/traildb-ipynb

Immediately after, in the first notebook in the browser:

In [1]:
from traildb import TrailDB
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-1-5a527c6b20d5> in <module>()
----> 1 from traildb import TrailDB

/root/.pyenv/versions/anaconda-2.3.0/lib/python2.7/site-packages/traildb/__init__.py in <module>()
----> 1 from .traildb import TrailDBError, TrailDBConstructor, TrailDB, TrailDBCursor, TrailDBEventFilter, tdb_item_field, tdb_item_val

/root/.pyenv/versions/anaconda-2.3.0/lib/python2.7/site-packages/traildb/traildb.py in <module>()
     90 api(lib.tdb_event_filter_new, [], tdb_event_filter)
     91 api(lib.tdb_event_filter_add_term, [tdb_event_filter, tdb_item, c_int], tdb_error)
---> 92 api(lib.tdb_event_filter_add_time_range, [c_uint64, c_uint64], tdb_error)
     93 api(lib.tdb_event_filter_new_clause, [tdb_event_filter], tdb_error)
     94 api(lib.tdb_event_filter_new_match_none, [], tdb_event_filter)

/root/.pyenv/versions/anaconda-2.3.0/lib/python2.7/ctypes/__init__.pyc in __getattr__(self, name)
    376         if name.startswith('__') and name.endswith('__'):
    377             raise AttributeError(name)
--> 378         func = self.__getitem__(name)
    379         setattr(self, name, func)
    380         return func

/root/.pyenv/versions/anaconda-2.3.0/lib/python2.7/ctypes/__init__.pyc in __getitem__(self, name_or_ordinal)
    381 
    382     def __getitem__(self, name_or_ordinal):
--> 383         func = self._FuncPtr((name_or_ordinal, self))
    384         if not isinstance(name_or_ordinal, (int, long)):
    385             func.__name__ = name_or_ordinal

AttributeError: /usr/local/lib/libtraildb.so: undefined symbol: tdb_event_filter_add_time_range

CFFI will allow us to use traildb with PyPy, leading to better performancce of the program in general.
Moreover, CFFI is faster even on CPython and it releases the GIL which allows for true parallelism as one might expect from a C library that performs I/O.
The only thing that is faster than CFFI on CPython is a C extension. We can provide both or keep the CFFI version only.

Python3 not supported

$ python setup.py install
running install
running build
running build_py
creating build
creating build/lib
creating build/lib/traildb
copying traildb/init.py -> build/lib/traildb
copying traildb/main.py -> build/lib/traildb
copying traildb/traildb.py -> build/lib/traildb
running install_lib
copying build/lib/traildb/init.py -> /Users/jobonilla/anaconda/lib/python3.5/site-packages/traildb
copying build/lib/traildb/main.py -> /Users/jobonilla/anaconda/lib/python3.5/site-packages/traildb
copying build/lib/traildb/traildb.py -> /Users/jobonilla/anaconda/lib/python3.5/site-packages/traildb
byte-compiling /Users/jobonilla/anaconda/lib/python3.5/site-packages/traildb/init.py to init.cpython-35.pyc
byte-compiling /Users/jobonilla/anaconda/lib/python3.5/site-packages/traildb/main.py to main.cpython-35.pyc
File "/Users/jobonilla/anaconda/lib/python3.5/site-packages/traildb/main.py", line 5
print cookie, trail
^
SyntaxError: Missing parentheses in call to 'print'

byte-compiling /Users/jobonilla/anaconda/lib/python3.5/site-packages/traildb/traildb.py to traildb.cpython-35.pyc
File "/Users/jobonilla/anaconda/lib/python3.5/site-packages/traildb/traildb.py", line 109
def tdb_item_val32(item): return (item >> 8) & 4294967295L # UINT32_MAX
^
SyntaxError: invalid syntax

running install_egg_info
Removing /Users/jobonilla/anaconda/lib/python3.5/site-packages/traildb-0.0.1-py3.5.egg-info
Writing /Users/jobonilla/anaconda/lib/python3.5/site-packages/traildb-0.0.1-py3.5.egg-info

Works if I install with python2.
Using Mac OS

Add support for multi cursor

TrailDBConstructor crashes if the first field is called username

I don't know why but it seems to me that TrailDBConstructor crashes if the first fieldname is username:

Python 2.7.13 (default, May 2 2017, 15:58:16)
[GCC 4.2.1 Compatible Apple LLVM 8.1.0 (clang-802.0.42)] on darwin
Type "help", "copyright", "credits" or "license" for more information.

from traildb import TrailDBConstructor, TrailDB
cons = TrailDBConstructor('foobar', ['username'])
[1] 88728 abort python

I noticed it because I was going through the tutorial and the first example didn't work http://traildb.io/docs/tutorial/#part-i-create-a-simple-traildb

I noticed that it doesn't happen if the field called "username" is not the first item in the array:

from traildb import TrailDBConstructor
TrailDBConstructor('foobar', ['alice', 'username'])
<traildb.traildb.TrailDBConstructor object at 0x10f609450>