Pi-hole/pi-hole
Pi-hole/pi-hole
1
0
Fork 0
mirror of https://github.com/pi-hole/pi-hole.git synced 2024-12-25 14:20:18 +00:00
Code Issues Projects Releases Packages Wiki Activity

Improve detection of incorrect lines during gravity (#5009)

Browse source
This commit is contained in:
Adam Warner 2022-12-27 12:32:31 +00:00 committed by GitHub
commit 955e36a955
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 20 additions and 14 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

34
gravity.sh
View file

@ -524,15 +524,20 @@ num_target_lines=0
num_source_lines=0 num_source_lines=0
num_invalid=0 num_invalid=0
parseList() { parseList() {
local adlistID="${1}" src="${2}" target="${3}" incorrect_lines local adlistID="${1}" src="${2}" target="${3}" incorrect_lines sample_incorrect_lines
# This sed does the following things: # This sed does the following things:
# 1. Remove all domains containing invalid characters. Valid are: a-z, A-Z, 0-9, dot (.), minus (-), underscore (_) # 1. Remove all lines containing no domains
# 2. Append ,adlistID to every line # 2. Remove all domains containing invalid characters. Valid are: a-z, A-Z, 0-9, dot (.), minus (-), underscore (_)
# 3. Remove trailing period (see https://github.com/pi-hole/pi-hole/issues/4701) # 3. Append ,adlistID to every line
# 4. Ensures there is a newline on the last line # 4. Remove trailing period (see https://github.com/pi-hole/pi-hole/issues/4701)
sed -e "/[^a-zA-Z0-9.\_-]/d;s/\.$//;s/$/,${adlistID}/;/.$/a\\" "${src}" >> "${target}" # 5. Ensures there is a newline on the last line
# Find (up to) five domains containing invalid characters (see above) sed -r "/([^\.]+\.)+[^\.]{2,}/!d;/[^a-zA-Z0-9.\_-]/d;s/\.$//;s/$/,${adlistID}/;/.$/a\\" "${src}" >> "${target}"
incorrect_lines="$(sed -e "/[^a-zA-Z0-9.\_-]/!d" "${src}" | head -n 5)"
# Find lines containing no domains or with invalid characters (see above)
# Remove duplicates and limit to 5 domains
mapfile -t incorrect_lines <<< "$(sed -r "/([^\.]+\.)+[^\.]{2,}/d" < "${src}")"
mapfile -t -O "${#incorrect_lines[@]}" incorrect_lines <<< "$(sed -r "/[^a-zA-Z0-9.\_-]/!d" < "${src}")"
IFS=" " read -r -a sample_incorrect_lines <<< "$(tr ' ' '\n' <<< "${incorrect_lines[@]}" | sort -u | head -n 5| tr '\n' ' ')"
local num_target_lines_new num_correct_lines local num_target_lines_new num_correct_lines
# Get number of lines in source file # Get number of lines in source file
@ -551,11 +556,12 @@ parseList() {
fi fi
# Display sample of invalid lines if we found some # Display sample of invalid lines if we found some
if [[ -n "${incorrect_lines}" ]]; then if [ ${#sample_incorrect_lines[@]} -ne 0 ]; then
echo " Sample of invalid domains:" echo " Sample of invalid domains:"
while IFS= read -r line; do for each in "${sample_incorrect_lines[@]}"
echo " - ${line}" do
done <<< "${incorrect_lines}" echo " - ${each}"
done
fi fi
} }
compareLists() { compareLists() {
@ -731,13 +737,13 @@ gravity_ParseFileIntoDomains() {
# 3) Remove comments (text starting with "#", include possible spaces before the hash sign) # 3) Remove comments (text starting with "#", include possible spaces before the hash sign)
# 4) Remove lines containing "/" # 4) Remove lines containing "/"
# 5) Remove leading tabs, spaces, etc. # 5) Remove leading tabs, spaces, etc.
# 6) Delete lines not matching domain names # 6) Remove empty lines
< "${src}" tr -d '\r' | \ < "${src}" tr -d '\r' | \
tr '[:upper:]' '[:lower:]' | \ tr '[:upper:]' '[:lower:]' | \
sed 's/\s*#.*//g' | \ sed 's/\s*#.*//g' | \
sed -r '/(\/).*$/d' | \ sed -r '/(\/).*$/d' | \
sed -r 's/^.*\s+//g' | \ sed -r 's/^.*\s+//g' | \
sed -r '/([^\.]+\.)+[^\.]{2,}/!d' > "${destination}" sed '/^$/d'> "${destination}"
chmod 644 "${destination}" chmod 644 "${destination}"
} }