User:Hhorak/scripts/meetbot-fetch

Fetch Environment-and-Stacks meeting logs from meetbot logs archive pages.

#!/bin/bash

# Author: Honza Horak
# Purpose: Fetch Environment-and-Stacks meeting logs from meetbot logs web
#          archive.
# License: Public Domain

export LC_ALL=C

# some hard values
MEETINGROOM_ARCHIVE="http://meetbot.fedoraproject.org/fedora-meeting/"

START_DATE="2013-11-12"

TMP_PAGE=`mktemp /var/tmp/meetbot-fetch-XXXXXX.html`

LOG_SUFFIX="\-[[:digit:]]*\.[[:digit:]]*\.log\.html"

# loops from START_DATE to NOW by 7 days steps
working_date="$START_DATE"

while [ `date --date "$working_date" +%s` -le `date +%s` ] ; do

    # store the html page where all logs from particular date are stored
    log_url="${MEETINGROOM_ARCHIVE}${working_date}/"
    wget "$log_url" -q -O "$TMP_PAGE"

    # try several variants of the meeting subject
    for name_variant in "env_and_stacks" "env-and-stacks" \
                        "environment_and_stacks" "environment-and-stacks" ; do
        meet_url=`grep -ohe "${name_variant}\.${working_date}${LOG_SUFFIX}" \
                  "$TMP_PAGE" 2>/dev/null | tail -n 1`
        [ -n "$meet_url" ] && break
    done

    # print some nice output
    if [ -n "$meet_url" ] ; then
        full_url="${MEETINGROOM_ARCHIVE}${working_date}/${meet_url}"
        echo "$full_url"
        echo
    fi

    # add 7 days
    working_date=`date -d "$working_date 7 days" +%F`

done