@@ -2,38 +2,48 @@ FROM ubuntu:focal as base
2
2
RUN apt-get update
3
3
4
4
ENV TIKA_VERSION 1.27
5
+ ENV TIKA_SERVER_JAR tika-server
6
+
5
7
6
8
9
+ # "random" uid/gid hopefully not used anywhere else
10
+ ARG UID_GID="35002:35002"
11
+
7
12
FROM base as dependencies
8
13
9
- RUN DEBIAN_FRONTEND=noninteractive apt-get -y install openjdk-14 -jre-headless gdal-bin tesseract-ocr \
14
+ RUN DEBIAN_FRONTEND=noninteractive apt-get -y install openjdk-17 -jre-headless gdal-bin tesseract-ocr \
10
15
tesseract-ocr-eng tesseract-ocr-ita tesseract-ocr-fra tesseract-ocr-spa tesseract-ocr-deu curl
11
16
12
17
RUN echo ttf-mscorefonts-installer msttcorefonts/accepted-mscorefonts-eula select true | debconf-set-selections \
13
18
&& DEBIAN_FRONTEND=noninteractive apt-get install -y xfonts-utils fonts-freefont-ttf fonts-liberation ttf-mscorefonts-installer wget cabextract
14
19
15
20
FROM dependencies as fetch_tika
16
21
17
- ENV NEAREST_TIKA_SERVER_URL="https://www.apache.org/dyn/closer.cgi/tika/tika-server-${TIKA_VERSION}.jar?filename=tika/${TIKA_VERSION}/tika-server-${TIKA_VERSION}.jar&action=download" \
18
- ARCHIVE_TIKA_SERVER_URL="https://archive.apache.org/dist/tika/${TIKA_VERSION}/tika-server-${TIKA_VERSION}.jar" \
19
- DEFAULT_TIKA_SERVER_ASC_URL="https://downloads.apache.org/tika/${TIKA_VERSION}/tika-server-${TIKA_VERSION}.jar.asc" \
20
- ARCHIVE_TIKA_SERVER_ASC_URL="https://archive.apache.org/dist/tika/${TIKA_VERSION}/tika-server-${TIKA_VERSION}.jar.asc" \
22
+ ENV NEAREST_TIKA_SERVER_URL="https://www.apache.org/dyn/closer.cgi/tika/${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar?filename=tika/${TIKA_VERSION}/${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar&action=download" \
23
+ NEAREST_TIKA_SERVER_URL_OLD="https://www.apache.org/dyn/closer.cgi/tika/${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar?filename=tika/${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar&action=download" \
24
+ ARCHIVE_TIKA_SERVER_URL="https://archive.apache.org/dist/tika/${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar" \
25
+ DEFAULT_TIKA_SERVER_ASC_URL="https://downloads.apache.org/tika/${TIKA_VERSION}/${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar.asc" \
26
+ DEFAULT_TIKA_SERVER_ASC_URL_OLD="https://downloads.apache.org/tika/${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar.asc" \
27
+ ARCHIVE_TIKA_SERVER_ASC_URL="https://archive.apache.org/dist/tika/${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar.asc" \
21
28
TIKA_VERSION=$TIKA_VERSION
22
29
23
- RUN DEBIAN_FRONTEND=noninteractive apt-get -y install gnupg2 \
30
+ RUN DEBIAN_FRONTEND=noninteractive apt-get -y install gnupg2 wget \
24
31
&& wget -t 10 --max-redirect 1 --retry-connrefused -qO- https://downloads.apache.org/tika/KEYS | gpg --import \
25
- && wget -t 10 --max-redirect 1 --retry-connrefused $NEAREST_TIKA_SERVER_URL -O /tika-server-${TIKA_VERSION}.jar || rm /tika-server-${TIKA_VERSION}.jar \
26
- && sh -c "[ -f /tika-server-${TIKA_VERSION}.jar ]" || wget $ARCHIVE_TIKA_SERVER_URL -O /tika-server-${TIKA_VERSION}.jar || rm /tika-server-${TIKA_VERSION}.jar \
27
- && sh -c "[ -f /tika-server-${TIKA_VERSION}.jar ]" || exit 1 \
28
- && wget -t 10 --max-redirect 1 --retry-connrefused $DEFAULT_TIKA_SERVER_ASC_URL -O /tika-server-${TIKA_VERSION}.jar.asc || rm /tika-server-${TIKA_VERSION}.jar.asc \
29
- && sh -c "[ -f /tika-server-${TIKA_VERSION}.jar.asc ]" || wget $ARCHIVE_TIKA_SERVER_ASC_URL -O /tika-server-${TIKA_VERSION}.jar.asc || rm /tika-server-${TIKA_VERSION}.jar.asc \
30
- && sh -c "[ -f /tika-server-${TIKA_VERSION}.jar.asc ]" || exit 1 \
31
- && gpg --verify /tika-server-${TIKA_VERSION}.jar.asc /tika-server-${TIKA_VERSION}.jar
32
+ && wget -t 10 --max-redirect 1 --retry-connrefused $NEAREST_TIKA_SERVER_URL -O /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar || rm /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar \
33
+ && sh -c "[ -f /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar ]" || wget $NEAREST_TIKA_SERVER_URL_OLD -O /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar || rm /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar \
34
+ && sh -c "[ -f /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar ]" || wget $ARCHIVE_TIKA_SERVER_URL -O /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar || rm /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar \
35
+ && sh -c "[ -f /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar ]" || exit 1 \
36
+ && wget -t 10 --max-redirect 1 --retry-connrefused $DEFAULT_TIKA_SERVER_ASC_URL -O /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar.asc || rm /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar.asc \
37
+ && sh -c "[ -f /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar.asc ]" || wget $DEFAULT_TIKA_SERVER_ASC_URL_OLD -O /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar.asc || rm /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar.asc \
38
+ && sh -c "[ -f /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar.asc ]" || wget $ARCHIVE_TIKA_SERVER_ASC_URL -O /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar.asc || rm /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar.asc \
39
+ && sh -c "[ -f /${TIKA_SERVER_JAR}-${TIKA_VERSION}.jar.asc ]" || exit 1;
40
+
32
41
33
42
FROM dependencies as runtime
34
43
RUN apt-get clean -y && rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
35
44
ENV TIKA_VERSION=$TIKA_VERSION
36
- COPY --from=fetch_tika /tika-server -${TIKA_VERSION}.jar /tika-server-${TIKA_VERSION}.jar
45
+ COPY --from=fetch_tika /${TIKA_SERVER_JAR} -${TIKA_VERSION}.jar /tika-server-${TIKA_VERSION}.jar
37
46
47
+ USER $UID_GID
38
48
EXPOSE 9998
39
- ENTRYPOINT [ "/bin/sh" , "-c" , "exec java -jar /tika-server -${TIKA_VERSION}.jar -h 0.0.0.0 $0 $@" ]
49
+ ENTRYPOINT [ "/bin/sh" , "-c" , "exec java -jar /${TIKA_SERVER_JAR} -${TIKA_VERSION}.jar -h 0.0.0.0 $0 $@" ]
0 commit comments