unstructured/scripts/docker-build.sh

#!/usr/bin/env bash

set -euo pipefail
DOCKER_REPOSITORY="${DOCKER_REPOSITORY:-quay.io/unstructured-io/unstructured}"
PIP_VERSION="${PIP_VERSION:-23.1.2}"
DOCKER_IMAGE="${DOCKER_IMAGE:-unstructured:dev}"

DOCKER_BUILD_CMD=(docker buildx build --load -f Dockerfile-amd64
  --build-arg PIP_VERSION="$PIP_VERSION"
  --build-arg BUILDKIT_INLINE_CACHE=1
  --progress plain
  --platform linux/amd64
  --cache-from "$DOCKER_REPOSITORY":latest
  -t "$DOCKER_IMAGE" .)

# only build for specific platform if DOCKER_BUILD_PLATFORM is set
if [ -n "${DOCKER_BUILD_PLATFORM:-}" ]; then
  DOCKER_BUILD_CMD+=("--platform=$DOCKER_BUILD_PLATFORM")
fi

DOCKER_BUILDKIT=1 "${DOCKER_BUILD_CMD[@]}"
chore: update all bash scripts to use shebang: /usr/bin/env bash (#779) 2023-06-20 16:00:55 -07:00			`#!/usr/bin/env bash`
adding Dockerfile (#359) 2023-03-14 13:40:01 -07:00
			`set -euo pipefail`
chore: build with registry as cache (#454) 2023-04-06 00:34:07 -07:00			`DOCKER_REPOSITORY="${DOCKER_REPOSITORY:-quay.io/unstructured-io/unstructured}"`
build(deps): bump pip (#558) 2023-05-08 23:08:10 -07:00			`PIP_VERSION="${PIP_VERSION:-23.1.2}"`
fix: docker smoke test on build (#457) 2023-04-06 10:03:42 -07:00			`DOCKER_IMAGE="${DOCKER_IMAGE:-unstructured:dev}"`
adding Dockerfile (#359) 2023-03-14 13:40:01 -07:00
fix: reenable arm64 builds for docker (#3045) ### Summary Closes #3034 and reenables ARM64 in the docker build and publish job. This was taken out in #3039 because we've only build `libreoffice` for AMD64 and not ARM64. If Chainguard publishes an `apk` for `libreoffice`, we can support a Chainguard image for both architectures. The smoke test now differs for both architectures, to reflect differences in the directory structure. ### Testing Build and publish ran successfully for ARM64 (job [here](https://github.com/Unstructured-IO/unstructured/actions/runs/9129712470/job/25104907497)) and AMD64 (job [here](https://github.com/Unstructured-IO/unstructured/actions/runs/9129712470/job/25104907826)). 2024-05-17 15:27:20 -04:00			`DOCKER_BUILD_CMD=(docker buildx build --load -f Dockerfile-amd64`
chore: shell scripts default indent of 2 instead of 4 (#2287) Given the tendency for shell scripts to easily enter into a few levels of indentation and long line lengths, update the default to 2 spaces. 2023-12-18 23:48:21 -08:00			`--build-arg PIP_VERSION="$PIP_VERSION"`
			`--build-arg BUILDKIT_INLINE_CACHE=1`
			`--progress plain`
build: wolfi base image for Dockerfile (#3016) ### Summary Updates the `Dockerfile` to use the Chainguard `wolfi-base` image to reduce CVEs. Also adds a step in the docker publish job that scans the images and checks for CVEs before publishing. The job will fail if there are high or critical vulnerabilities. ### Testing Run `make docker-run-dev` and then `python3.11` once you're in. And that point, you can try: ```python from unstructured.partition.auto import partition elements = partition(filename="example-docs/DA-1p.pdf", skip_infer_table_types=["pdf"]) elements ``` Stop the container once you're done. 2024-05-15 18:53:15 -04:00			`--platform linux/amd64`
chore: shell scripts default indent of 2 instead of 4 (#2287) Given the tendency for shell scripts to easily enter into a few levels of indentation and long line lengths, update the default to 2 spaces. 2023-12-18 23:48:21 -08:00			`--cache-from "$DOCKER_REPOSITORY":latest`
			`-t "$DOCKER_IMAGE" .)`
chore: build with registry as cache (#454) 2023-04-06 00:34:07 -07:00
			`# only build for specific platform if DOCKER_BUILD_PLATFORM is set`
			`if [ -n "${DOCKER_BUILD_PLATFORM:-}" ]; then`
chore: shell scripts default indent of 2 instead of 4 (#2287) Given the tendency for shell scripts to easily enter into a few levels of indentation and long line lengths, update the default to 2 spaces. 2023-12-18 23:48:21 -08:00			`DOCKER_BUILD_CMD+=("--platform=$DOCKER_BUILD_PLATFORM")`
chore: build with registry as cache (#454) 2023-04-06 00:34:07 -07:00			`fi`

			`DOCKER_BUILDKIT=1 "${DOCKER_BUILD_CMD[@]}"`