gist-1b471e5f5215c368fd78d9.../SCRAPE_LONDON.SH
Lillian Skinner f7824989f9
2025-05-16 20:03:35 -04:00

195 lines
12 KiB
Bash

#/bash
echo -e "\n-========================================================================-"
echo -e "-=- -=-"
echo -e "-=- SCRAPE_LONDON.SH: Downloads committee videos and agendas -=-"
echo -e "-=- -=-"
echo -e "-=- https://gist.github.com/rvtr/1b471e5f5215c368fd78d9aba05f8dc2 -=-"
echo -e "-=- Lillian Skinner -=-"
echo -e "-=- -=-"
echo -e "-========================================================================-"
# Warning to all who read this script:
# It is bad. I know it is bad, but I am tired okay, and sometimes sloppy just works.
# London seems to have recently blocked unusual user agents. Can't use wget or even ping. Thankfully pretend to be a real person!
WGET_UA="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/125.0.0.0 Safari/537.36 Viewer/99.9.8782.87"
TEMP_DIR="./tmp/"
SEARCH_PAGE="./tmp/index.html"
AGENDA_DIR="./Agenda/"
AGENDA_HTML="./tmp/work.html"
VIDEO_DIR="./Video/"
#VIDEO_TIMESTAMP_JSON="./tmp/time.json"
if [ -d "$TEMP_DIR" ]; then
rm -r $TEMP_DIR
fi
if [ -d "$AGENDA_DIR" ]; then
rm -r $AGENDA_DIR
fi
if [ -d "$VIDEO_DIR" ]; then
rm -r $VIDEO_DIR
fi
mkdir $TEMP_DIR
mkdir $AGENDA_DIR
mkdir $VIDEO_DIR
SEARCH_URL="https://london.ca/government/council-civic-administration/council-committee-meetings/meetings"
# Need to confirm. When stacking params does the date need to be f[1]?
SEARCH_PARAM_COMMITTEE="f[0]=meeting_type%3A"
SEARCH_PARAM_DATE="f[0]=meeting_date%3A"
SEARCH_PARAM_QUERY="search=query&sort_by=field_meeting_date"
i=0
SEARCH_END="FALSE"
while [[ $SEARCH_END == "FALSE" ]]; do
echo "Downloading search results... Page: $i"
wget --user-agent="$WGET_UA" $SEARCH_URL"?page=$i" -O $SEARCH_PAGE -q #--show-progress
if [ $? -ne 8 ]; then
FOUNDMEETING="FALSE"
GREP404=$(cat $SEARCH_PAGE | grep "No results found.")
if [[ "$GREP404" == "" ]]; then
while IFS= read -r LINE; do
# All meeting items in the search results are formatted like so:
# - One line with the name
# - Second line with all other info including links
#
# We can find the first line by the class "views-field-field-meeting-notes"
# FOUNDMEETING=TRUE will show that the first line has been found, and so the next line read will be "confirmed" as line 2 of the meeting info
# The first two links of every second line are (in order) the PDF and HTML agendas
if [[ "TRUE" == $FOUNDMEETING ]]; then
FOUNDMEETING="FALSE"
echo "-========================================================================-"
echo " Working on $MEETING_NAME"
echo "-========================================================================-"
# Grab meeting item links
echo $LINE | sed 's/href=./\nhref="/g' | grep 'href="https' | sed 's/.*href="\([^"]*\)".*/\1/p' | uniq > "./tmp/meeting_urls"
# Grab meeting item types
echo $LINE | sed 's/rel=.noreferrer.>/\nrel="noreferrer">/g' | grep 'rel="noreferrer">' | sed 's/.*rel="noreferrer">\([^<]*\)<.*/\1/p' | uniq > "./tmp/meeting_types"
AGENDA_HTML_URL=""
AGENDA_PDF_URL=""
AGENDA_REVISE_HTML_URL=""
AGENDA_REVISE_PDF_URL=""
MINUTES_HTML_URL=""
MINUTES_PDF_URL=""
MINUTES_ATTACH_PDF_URL=""
echo "Found the following documents:"
while IFS= read -r LINEA1 && IFS= read -r LINEA2 <&3; do
echo " - $LINEA2"
case "$LINEA2" in
"Agenda (HTML) ")
AGENDA_HTML_URL="$LINEA1" ;;
"Agenda (PDF) ")
AGENDA_PDF_URL="$LINEA1" ;;
"Revised Agenda (HTML) ")
AGENDA_REVISE_HTML_URL="$LINEA1" ;;
"Revised Agenda (PDF) ")
AGENDA_REVISE_PDF_URL="$LINEA1" ;;
"Minutes (HTML) ")
MINUTES_HTML_URL="$LINEA1" ;;
"Minutes (PDF) ")
MINUTES_PDF_URL="$LINEA1" ;;
"Minutes with Attachments (PDF) ")
MINUTES_ATTACH_PDF_URL="$LINEA1" ;;
esac
done < ./tmp/meeting_urls 3< ./tmp/meeting_types
# Always prefer Revised Agendas
echo "Downloading agenda HTML..."
if [[ $AGENDA_REVISE_HTML_URL != "" ]]; then
wget --user-agent="$WGET_UA" "$AGENDA_REVISE_HTML_URL" -O $AGENDA_HTML -q #--show-progress
elif [[ $AGENDA_HTML_URL != "" ]]; then
wget --user-agent="$WGET_UA" "$AGENDA_HTML_URL" -O $AGENDA_HTML -q #--show-progress
else
ERROR="TRUE"
fi
if [[ ERROR="FALSE" ]]; then
mkdir "./tmp/$MEETING_NAME/"
mkdir "./tmp/$MEETING_NAME/Attachments/"
# Direct video links is always "video.isilive.ca/<REGION>/<NAME>"
# There are some eScribe ones, but those are in m3u8s and are really annoying to work with
# ...not annoying as more sed though.
VIDEO_URL=$(grep 'id="isi_player"' ./tmp/work.html | sed -n 's/.*data-stream_name="\([^"]*\)".*/\1/p' | sed 's/ /%20/g')
if [[ $VIDEO_URL != "" ]]; then
echo "Found meeting recording."
echo "https://video.isilive.ca/london/"$VIDEO_URL > "./tmp/$MEETING_NAME/RecordingLink.txt"
fi
# Get attachment links
cat $AGENDA_HTML | grep "AgendaItemAgendaItem1TitleHeader" | sed 's/href=.filestream\.ashx/\nhref="filestream\.ashx/g' | grep 'filestream.ashx' | sed 's/href=.\([^/]*\)".*/\1/p' | awk '!x[$0]++' > "./tmp/attachment_urls"
# Get attachment names
cat $AGENDA_HTML | grep "AgendaItemAgendaItem1TitleHeader" | sed 's/data-original-title=./\ndata-original-title='\''/g' | grep 'data-original-title' | sed 's/data-original-title=.\([^'\''/]*\)'\''.*/\1/p' | awk '!x[$0]++' > "./tmp/attachment_names"
# Download attachment and use the name grabbed above
echo "Found the following agenda attachments:"
while IFS= read -r LINEA1 && IFS= read -r LINEA2 <&3; do
echo " - $LINEA2"
wget --user-agent="$WGET_UA" "https://pub-london.escribemeetings.com/$LINEA1" -O "./tmp/$MEETING_NAME/Attachments/$LINEA2" -q #--show-progress
done < ./tmp/attachment_urls 3< ./tmp/attachment_names
echo "All attachments saved."
if [[ $AGENDA_REVISE_PDF_URL != "" ]] || [[ $AGENDA_PDF_URL != "" ]]; then
if [[ $AGENDA_REVISE_PDF_URL != "" ]]; then
echo "Saving revised agenda..."
wget --user-agent="$WGET_UA" "$AGENDA_REVISE_PDF_URL" -O "./tmp/$MEETING_NAME/Agenda_Revised.pdf" -q #--show-progress
fi
if [[ $AGENDA_PDF_URL != "" ]]; then
echo "Saving regular agenda..."
wget --user-agent="$WGET_UA" "$AGENDA_PDF_URL" -O "./tmp/$MEETING_NAME/Agenda.pdf" -q #--show-progress
fi
else
if [[ $AGENDA_REVISE_HTML_URL != "" ]]; then
echo "Saving revised agenda as HTML (no PDF found!)"
wget --user-agent="$WGET_UA" "$AGENDA_REVISE_HTML_URL" -O "./tmp/$MEETING_NAME/Agenda_Revised.html" -q #--show-progress
fi
if [[ $AGENDA_HTML_URL != "" ]]; then
echo "Saving regular agenda as HTML (no PDF found!)"
wget --user-agent="$WGET_UA" "$AGENDA_HTML_URL" -O "./tmp/$MEETING_NAME/Agenda.html" -q #--show-progress
fi
fi
if [[ $MINUTES_ATTACH_PDF_URL != "" ]] || [[ $MINUTES_PDF_URL != "" ]]; then
if [[ $MINUTES_ATTACH_PDF_URL != "" ]]; then
echo "Saving minutes with attachments..."
wget --user-agent="$WGET_UA" "$MINUTES_ATTACH_PDF_URL" -O "./tmp/$MEETING_NAME/Minutes_With_Attachments.pdf" -q #--show-progress
fi
if [[ $MINUTES_PDF_URL != "" ]]; then
echo "Saving minutes..."
wget --user-agent="$WGET_UA" "$MINUTES_PDF_URL" -O "./tmp/$MEETING_NAME/Minutes.pdf" -q #--show-progress
fi
else
if [[ $MINUTES_HTML_URL != "" ]]; then
echo "Saving minutes as HTML (no PDF found!)"
wget --user-agent="$WGET_UA" "$MINUTES_HTML_URL" -O "./tmp/$MEETING_NAME/Minutes.html" -q #--show-progress
fi
fi
fi
fi
GREPMEETING=$(echo $LINE | grep "views-field-field-meeting-notes")
if [[ "$GREPMEETING" != "" ]]; then
FOUNDMEETING="TRUE"
MEETING_NAME=$(echo $LINE | sed -n 's/.*<div class="meeting__date">\([^<]*\)<\/div>.*/\1/p')
fi
done < $SEARCH_PAGE
else
SEARCH_END="TRUE"
fi
else
SEARCH_END="TRUE"
fi
((i++))
done