turbodbc
Turbodbc is a Python component to
accessibility relational data sources through the Open Data Source Connection (ODBC) user interface.
For optimal efficiency, turbodbc uses integrated NumPy and also Apache Arrowhead assistance
and also inside depends on batched information transfer rather than single-record
interaction as various other preferred ODBC components do.
Structure turbodbc with pyarrow assistance has some cautions as it has construct
time discovery if pyarrow is set up and also requires pybind and also numerous debian
dev bundles to obtain the C++ collection.
By utilizing docker multistage develops
we can natively construct turbodbc with pyarrow assistance without obtaining the dev bundles.
right into the last photo.
Very first step is the base photo that has all needed debian bundles to run turbodbc later:
# phrase structure= docker/dockerfile:1
FROM debian: bullseye as base
# Develop customer, have to not be origin and also UID need to be more than 1000
RUN useradd -- uid 1100 application -- produce- house
RUN appropriate- obtain upgrade
RUN -- place = kind = cache, target=/ var/ cache/ appropriate appropriate- obtain mount -- yes python3 python3- venv git
RUN -- place = kind = cache, target=/ var/ cache/ appropriate appropriate- obtain mount -- yes libodbc1 odbcinst odbcinst1debian2 binutils- x86- 64- linux- gnu
RUN python3 - m venv / choose/ venv
ENV COURSE ="/ opt/venv/bin:$ {COURSE} "
WORKDIR / application/
ENV PYTHONPATH=/ application/
In the 2nd phase we mount the construct demands that are just required to.
put together turbodbc with arrowhead assistance. There are 2 essential notes:
First of all pyarrow needs to be set up prior to turbodbc is construct as the.
turbodbc construct procedure instantly finds if pyarrow is readily available.
To make the discovery job you require to pass -- no-build-isolation to.
the turbodbc mount and also ensure the arrowhead collections are connected appropriately.
FROM base as home builder
RUN -- place = kind = cache, target=/ var/ cache/ appropriate appropriate- obtain - yq mount
construct- necessary
gdb
lcov
libbz2- dev
libffi- dev
libgdbm- dev
liblzma- dev
libboost- dev
libncurses5- dev
libreadline6- dev
libsqlite3- dev
libssl- dev
lzma
lzma- dev
python3- dev
tk- dev
unixodbc- dev
uuid- dev
xvfb
zlib1g- dev
RUN pip3 mount - U pip= = 22.0 4 setuptools= = 45.2 0 wheel= = 0.37 1
RUN pip3 mount - U pybind11= = 2.10 1 numpy= = 1.23 5 pandas= = 1.5 2 6= = 1.16 0 pyarrow= = 5.0 0
RUN python3 - c " import pyarrow; pyarrow.create _ library_symlinks()"
&&& & CPPFLAGS ="- D_GLIBCXX_USE_CXX11_ABI= 0" pip3 mount -- no- construct- seclusion turbodbc= = 4.5 5
In the 3rd phase we produce a fresh phase and also just reuse venv with the turbodbc construct bundles
FROM base as jogger
DUPLICATE -- from = home builder / choose/ venv / choose/ venv
DUPLICATE demands txt / application/ demands txt
RUN -- place = kind = cache, target=/ origin/. cache pip mount -- demand / application/ demands txt
# Establish the Customer we produced over
CUSTOMER 1100
CMD []
