sporada/ocr
1
0
Fork 0
mirror of https://gitlab.ub.uni-bielefeld.de/sfb1288inf/ocr.git synced 2024-12-26 16:54:19 +00:00
Code Issues Projects Releases Wiki Activity

Bump versions

Browse Source
This commit is contained in:
Patrick Jentsch 2020-04-06 09:21:52 +02:00
parent 364e3d626d
commit a75b32ca1d
2 changed files with 13 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
Dockerfile
View File

@ -1,26 +1,25 @@
FROM debian:9-slim FROM debian:10-slim
# Define image metadata
LABEL maintainer="inf_sfb1288@lists.uni-bielefeld.de" LABEL maintainer="inf_sfb1288@lists.uni-bielefeld.de"
ENV LANG=C.UTF-8 ENV LANG=C.UTF-8
# Install prerequisites
RUN apt-get update \ RUN apt-get update \
&& apt-get install -y --no-install-recommends \ && apt-get install -y --no-install-recommends \
apt-transport-https \ apt-transport-https \
build-essential \
ca-certificates \ ca-certificates \
gnupg2 \ gnupg2 \
imagemagick \ imagemagick \
poppler-utils \ poppler-utils \
python2.7 \ python2.7 \
python3.5 \ python3.7 \
wget \ wget \
zip \ zip
&& rm -rf /var/lib/apt/lists/*
ENV OCROPY_VERSION 1.3.3 ENV OCROPY_VERSION 1.3.3
ADD "https://github.com/tmbdev/ocropy/archive/v${OCROPY_VERSION}.tar.gz" . ADD "https://github.com/tmbdev/ocropy/archive/v${OCROPY_VERSION}.tar.gz" .
@ -31,13 +30,13 @@ RUN tar -xzf "v${OCROPY_VERSION}.tar.gz" \
python-pil \ python-pil \
python-tk \ python-tk \
$(cat PACKAGES) \ $(cat PACKAGES) \
&& rm -rf /var/lib/apt/lists/* \
&& python2.7 setup.py install \ && python2.7 setup.py install \
&& cd .. \ && cd .. \
&& rm -rf \ && rm -rf \
"ocropy-${OCROPY_VERSION}" \ "ocropy-${OCROPY_VERSION}" \
"v${OCROPY_VERSION}.tar.gz" "v${OCROPY_VERSION}.tar.gz"
ENV PYFLOW_VERSION=1.1.20 ENV PYFLOW_VERSION=1.1.20
ADD "https://github.com/Illumina/pyflow/releases/download/v${PYFLOW_VERSION}/pyflow-${PYFLOW_VERSION}.tar.gz" . ADD "https://github.com/Illumina/pyflow/releases/download/v${PYFLOW_VERSION}/pyflow-${PYFLOW_VERSION}.tar.gz" .
RUN tar -xzf "pyflow-${PYFLOW_VERSION}.tar.gz" \ RUN tar -xzf "pyflow-${PYFLOW_VERSION}.tar.gz" \
@ -48,7 +47,8 @@ RUN tar -xzf "pyflow-${PYFLOW_VERSION}.tar.gz" \
"pyflow-${PYFLOW_VERSION}" \ "pyflow-${PYFLOW_VERSION}" \
"pyflow-${PYFLOW_VERSION}.tar.gz" "pyflow-${PYFLOW_VERSION}.tar.gz"
RUN echo "deb https://notesalexp.org/tesseract-ocr/stretch/ stretch main" >> /etc/apt/sources.list \
RUN echo "deb https://notesalexp.org/tesseract-ocr/buster/ buster main" >> /etc/apt/sources.list \
&& wget -O - https://notesalexp.org/debian/alexp_key.asc | apt-key add - \ && wget -O - https://notesalexp.org/debian/alexp_key.asc | apt-key add - \
&& apt-get update \ && apt-get update \
&& apt-get install -y --no-install-recommends \ && apt-get install -y --no-install-recommends \
@ -61,11 +61,12 @@ RUN echo "deb https://notesalexp.org/tesseract-ocr/stretch/ stretch main" >> /et
tesseract-ocr-frm \ tesseract-ocr-frm \
tesseract-ocr-ita \ tesseract-ocr-ita \
tesseract-ocr-por \ tesseract-ocr-por \
tesseract-ocr-spa \ tesseract-ocr-spa
&& rm -rf /var/lib/apt/lists/*
RUN rm -rf /var/lib/apt/lists/*
# Install OCR pipeline
COPY hocrtotei /usr/local/bin COPY hocrtotei /usr/local/bin
COPY ocr /usr/local/bin COPY ocr /usr/local/bin

2
hocrtotei
View File

@ -1,4 +1,4 @@
#!/usr/bin/env python3.5 #!/usr/bin/env python3.7
# coding=utf-8 # coding=utf-8
from xml.sax.saxutils import escape from xml.sax.saxutils import escape