Add dependencies so the pytesseract can run within docker

This commit is contained in:
Kuba Orlik 2022-02-27 22:27:21 +01:00
parent f6f2d713db
commit c2680ddca5
3 changed files with 71 additions and 7 deletions

View File

@ -1,7 +1,41 @@
FROM alpine:3.15
FROM python:3.11.0a5-alpine3.15
RUN apk add firefox xvfb scrot vips-tools xterm xdotool fluxbox xprop imagemagick bash tesseract-ocr
# inspired by https://github.com/darktohka/pytesseract-docker/blob/master/Dockerfile
RUN apk add terminus-font ttf-inconsolata ttf-dejavu font-noto font-noto ttf-font-awesome font-noto-extra
RUN apk update
RUN apk add firefox xvfb scrot vips-tools xterm xdotool fluxbox xprop imagemagick bash tesseract-ocr terminus-font ttf-inconsolata ttf-dejavu font-noto font-noto ttf-font-awesome font-noto-extra
ENV SHELL /bin/sh
ENV CC /usr/bin/clang
ENV CXX /usr/bin/clang++
ENV LANG C.UTF-8
ENV PYTHONUNBUFFERED 1
ENV PIP_DISABLE_PIP_VERSION_CHECK 1
ENV PIP_NO_CACHE_DIR 0
ENV TESSDATA_PREFIX /usr/local/share/tessdata
WORKDIR /tmp
RUN apk add --no-cache openssl leptonica openjpeg tiff libpng zlib freetype libgcc libstdc++
# Install development tools
RUN apk add --no-cache --virtual .dev-deps file linux-headers git make automake autoconf libtool pkgconfig clang g++ openssl-dev leptonica-dev openjpeg-dev tiff-dev libpng-dev zlib-dev freetype-dev
# Install Tesseract from master
RUN mkdir /usr/local/share/tessdata
RUN mkdir tesseract
WORKDIR /tmp/tesseract
RUN wget https://github.com/tesseract-ocr/tessdata_fast/raw/main/eng.traineddata -P "$TESSDATA_PREFIX"
RUN git clone --depth 1 https://github.com/tesseract-ocr/tesseract.git .
RUN ./autogen.sh
RUN ./configure
RUN make -j$(nproc)
RUN make install
# Install Python dependencies
RUN pip install -U --no-cache-dir pytesseract
# Cleanup
RUN apk del .dev-deps
RUN rm -f /usr/local/lib/*.a
RUN rm -rf /tmp/* /var/cache/apk/*
WORKDIR /opt

15
get-text-position.py Normal file
View File

@ -0,0 +1,15 @@
import pytesseract
from pytesseract import Output
# from PIL import Image
img = 'cropped.png'
d = pytesseract.image_to_data(img, output_type=Output.DICT)
# print(d)
n_boxes = len(d['level'])
for i in range(n_boxes):
(x, y, w, h, text) = (d['left'][i], d['top'][i], d['width'][i], d['height'][i], d['text'][i])
if text=="":
continue
print(x, y, repr(text))

View File

@ -48,6 +48,20 @@ keycombo(){
sleep 0.5
}
annotate_header(){
header=$1
d=$(date "+%Y-%m-%d__%H_%M_%S")
filename="$d__$header.png"
cropped_filename="$filename__cropped.png"
left=2056
top=330
width=824
height=1260
scrot $filename
vips extract_area "$filename" "$cropped_filename" $left $top $width $height
}
rm -rf /root/.mozilla/firefox/bifup8k5.docker/sessionstore-backups
#echo 'user_pref("layout.css.devPixelsPerPx", "1.5");' >> /root/.mozilla/firefox/bifup8k5.docker/prefs.js
@ -131,10 +145,11 @@ do
xdotool key Tab
sleep 0.05
xdotool key Down
sleep 0.1
scrot
echo "########## EXTRACTED TEXT: "
extract_text 2056 330 824 1260 | grep cookie
sleep 0.2
annotate_header set-cookie
# scrot
# echo "########## EXTRACTED TEXT: "
# extract_text 2056 330 824 1260 | grep cookie
done