@ -244,7 +244,7 @@ database_adlist_number() {
return ;
fi
output = $( { printf ".timeout 30000\\nUPDATE adlist SET number = %i, invalid_domains = %i WHERE id = %i;\\n" " ${ num_ source_lines} " " ${ num_invalid } " " ${ 1 } " | pihole-FTL sqlite3 " ${ gravityDBfile } " ; } 2>& 1 )
output = $( { printf ".timeout 30000\\nUPDATE adlist SET number = %i, invalid_domains = %i WHERE id = %i;\\n" " ${ num_ domains} " " ${ num_non_domains } " " ${ 1 } " | pihole-FTL sqlite3 " ${ gravityDBfile } " ; } 2>& 1 )
status = " $? "
if [ [ " ${ status } " -ne 0 ] ] ; then
@ -519,12 +519,12 @@ gravity_DownloadBlocklists() {
gravity_Blackbody = true
}
# num_t arget_lines does increase for every correctly added domain in pareseList()
num_t arget_line s= 0
num_ source_line s= 0
num_ invalid = 0
# num_t otal_imported_domains increases for each list processed
num_t otal_imported_domain s= 0
num_ domain s= 0
num_ non_domains = 0
parseList( ) {
local adlistID = " ${ 1 } " src = " ${ 2 } " target = " ${ 3 } " incorrect_lines sample_incorrect_lines
local adlistID = " ${ 1 } " src = " ${ 2 } " target = " ${ 3 } " non_domains sample_non_domains tmp_non_domains_str false_positive
# This sed does the following things:
# 1. Remove all lines containing no domains
# 2. Remove all domains containing invalid characters. Valid are: a-z, A-Z, 0-9, dot (.), minus (-), underscore (_)
@ -534,36 +534,65 @@ parseList() {
sed -r " /([^\.]+\.)+[^\.]{2,}/!d;/[^a-zA-Z0-9.\_-]/d;s/\. $//;s/ $/, ${ adlistID } /;/. $/a\\ " " ${ src } " >> " ${ target } "
# 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
# Get number of lines in source file
num_source_lines = " $( grep -c "^" " ${ src } " ) "
# Get the new number of lines in destination file
num_target_lines_new = " $( grep -c "^" " ${ target } " ) "
# Number of new correctly added lines
num_correct_lines = " $(( num_target_lines_new-num_target_lines )) "
# Update number of lines in target file
num_target_lines = " $num_target_lines_new "
num_invalid = " $(( num_source_lines-num_correct_lines )) "
if [ [ " ${ num_invalid } " -eq 0 ] ] ; then
echo " ${ INFO } Analyzed ${ num_source_lines } domains "
else
echo " ${ INFO } Analyzed ${ num_source_lines } domains, ${ num_invalid } domains invalid! "
fi
# Display sample of invalid lines if we found some
if [ ${# sample_incorrect_lines [@] } -ne 0 ] ; then
echo " Sample of invalid domains:"
for each in " ${ sample_incorrect_lines [@] } "
# Remove duplicates from the list
mapfile -t non_domains <<< " $( sed -r "/([^\.]+\.)+[^\.]{2,}/d" < " ${ src } " ) "
mapfile -t -O " ${# non_domains [@] } " non_domains <<< " $( sed -r "/[^a-zA-Z0-9.\_-]/!d" < " ${ src } " ) "
IFS = " " read -r -a non_domains <<< " $( tr ' ' '\n' <<< " ${ non_domains [@] } " | sort -u | tr '\n' ' ' ) "
# A list of items of common local hostnames not to report as unusable
# Some lists (i.e StevenBlack's) contain these as they are supposed to be used as HOST files
# but flagging them as unusable causes more confusion than it's worth - so we suppress them from the output
false_positives = (
"localhost"
"localhost.localdomain"
"local"
"broadcasthost"
"localhost"
"ip6-localhost"
"ip6-loopback"
"lo0 localhost"
"ip6-localnet"
"ip6-mcastprefix"
"ip6-allnodes"
"ip6-allrouters"
"ip6-allhosts"
)
# Read the unusable lines into a string
tmp_non_domains_str = " ${ non_domains [*] } "
for false_positive in " ${ false_positives [@] } " ; do
# Remove false positives from tmp_non_domains_str
tmp_non_domains_str = " ${ tmp_non_domains_str / ${ false_positive } / } "
done
# Read the string back into an array
IFS = " " read -r -a non_domains <<< " ${ tmp_non_domains_str } "
# Get a sample of non-domain entries, limited to 5 (the list should already have been de-duplicated)
IFS = " " read -r -a sample_non_domains <<< " $( tr ' ' '\n' <<< " ${ non_domains [@] } " | head -n 5 | tr '\n' ' ' ) "
local tmp_new_imported_total
# Get the new number of domains in destination file
tmp_new_imported_total = " $( grep -c "^" " ${ target } " ) "
# Number of imported lines for this file is the difference between the new total and the old total. (Or, the number of domains we just added.)
num_domains = " $(( tmp_new_imported_total-num_total_imported_domains )) "
# Replace the running total with the new total.
num_total_imported_domains = " $tmp_new_imported_total "
# Get the number of non_domains (this is the number of entries left after stripping the source of comments/duplicates/false positives/domains)
num_non_domains = " ${# non_domains [@] } "
# If there are unusable lines, we display some information about them. This is not error or major cause for concern.
if [ [ " ${ num_non_domains } " -ne 0 ] ] ; then
echo " ${ INFO } Imported ${ num_domains } domains, ignoring ${ num_non_domains } non-domain entries "
echo " Sample of non-domain entries:"
for each in " ${ sample_non_domains [@] } "
do
echo " - ${ each } "
echo " - ${ each } "
done
else
echo " ${ INFO } Imported ${ num_domains } domains "
fi
}
compareLists( ) {
local adlistID = " ${ 1 } " target = " ${ 2 } "
@ -716,8 +745,8 @@ gravity_DownloadBlocklistFromUrl() {
else
echo -e " ${ CROSS } List download failed: ${ COL_LIGHT_RED } no cached list available ${ COL_NC } "
# Manually reset these two numbers because we do not call parseList here
num_ source_line s= 0
num_ invalid = 0
num_ domain s= 0
num_ non_domains = 0
database_adlist_number " ${ adlistID } "
database_adlist_status " ${ adlistID } " "4"
fi
Adam Warner
1 year ago
committed by