mirror of
https://github.com/xroche/httrack.git
synced 2026-07-17 06:10:42 +03:00
Compare commits
2 Commits
fix-lang-i
...
fix-argord
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
5083991f5a | ||
|
|
1a020911db |
@@ -930,6 +930,7 @@ Server terminated
|
||||
Palvelin lopetettu
|
||||
A fatal error has occurred during this mirror
|
||||
Tällä peilillä tapahtui vakava virhe
|
||||
|
||||
Proxy type:
|
||||
Välityspalvelimen tyyppi:
|
||||
Proxy protocol. HTTP: standard proxy. HTTP (CONNECT tunnel): sends every request through a CONNECT tunnel, for CONNECT-only proxies like Tor's HTTPTunnelPort. SOCKS5: default port 1080.
|
||||
|
||||
@@ -374,7 +374,7 @@ Create error logging and report files
|
||||
Crea file di log per segnalare errori e informazioni
|
||||
Generate DOS 8-3 filenames ONLY
|
||||
Genera solo nomi di file in formato 8.3
|
||||
Generate ISO9660 filenames ONLY for CDROM medias
|
||||
Generate ISO 9660 filenames ONLY for CDROM medias
|
||||
Genera nomi di file in formato ISO9660 per i CDROM
|
||||
Do not create HTML error pages
|
||||
Non generare pagine d'errore HTML
|
||||
|
||||
@@ -918,7 +918,7 @@ normal\nextended\ndebug
|
||||
normal\nextendido\ncorrigir
|
||||
Download web site(s)\nDownload web site(s) + questions\nGet individual files\nDownload all sites in pages (multiple mirror)\nTest links in pages (bookmark test)\n* Continue interrupted download\n* Update existing download
|
||||
Copiar site(s) da Web\nCopiar site(s) interativos da Web (perguntas)\nReceber arquivos específicos\nCopiar todas as páginas do site (alternação múltipla)\nTestar links nas páginas (testar indicador)\n* Retomar download interrompido\n* Atualizar download existente
|
||||
Relative URI / Absolute URL (default)\nAbsolute URL / Absolute URL\nAbsolute URI / Absolute URL\nOriginal URL / Original URL
|
||||
Relative URL / Absolute URL (default)\nAbsolute URL / Absolute URL\nAbsolute URI / Absolute URL\nOriginal URL / Original URL
|
||||
URI Relativa / URL Absoluta (padrão)\nURL Absoluta / URL Absoluta\nURI Absoluta / URL Absoluta\nURL Original / URL Original
|
||||
Open Source offline browser
|
||||
Abrir origem offline no navegador
|
||||
|
||||
@@ -113,6 +113,7 @@ HTSEXT_API int hts_main(int argc, char **argv) {
|
||||
}
|
||||
|
||||
static int hts_main_internal(int argc, char **argv, httrackp * opt);
|
||||
static hts_boolean cmdl_shortopt_has(const char *s, char c);
|
||||
|
||||
// Main, récupère les paramètres et appelle le robot
|
||||
HTSEXT_API int hts_main2(int argc, char **argv, httrackp * opt) {
|
||||
@@ -304,12 +305,12 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
|
||||
hts_get_version_info(opt));
|
||||
return 0;
|
||||
} else {
|
||||
if (strncmp(tmp_argv[0], "--", 2)) { /* pas */
|
||||
if ((strchr(tmp_argv[0], 'q') != NULL))
|
||||
opt->quiet = 1; // ne pas poser de questions! (nohup par exemple)
|
||||
if ((strchr(tmp_argv[0], 'i') != NULL)) { // doit.log!
|
||||
argv_url = -1; /* forcer */
|
||||
opt->quiet = 1;
|
||||
if (strncmp(tmp_argv[0], "--", 2)) { /* not a long option */
|
||||
if (cmdl_shortopt_has(tmp_argv[0], 'q'))
|
||||
opt->quiet = HTS_TRUE; // never ask questions (nohup)
|
||||
if (cmdl_shortopt_has(tmp_argv[0], 'i')) { // doit.log!
|
||||
argv_url = -1;
|
||||
opt->quiet = HTS_TRUE;
|
||||
}
|
||||
} else if (strcmp(tmp_argv[0] + 2, "quiet") == 0) {
|
||||
opt->quiet = 1; // ne pas poser de questions! (nohup par exemple)
|
||||
@@ -2807,6 +2808,32 @@ int check_path(String * s, char *defaultname) {
|
||||
return return_value;
|
||||
}
|
||||
|
||||
/* Does the short-option cluster s carry c from the main option set (-i, -iC2,
|
||||
-%Mi)? Walked as the parser does below: %, &, @ and # each take the letter
|
||||
after them into another set, so the i of -%i is not the main-set -i. */
|
||||
static hts_boolean cmdl_shortopt_has(const char *s, char c) {
|
||||
const char *com;
|
||||
|
||||
if (s[0] != '-' || s[1] == '-')
|
||||
return HTS_FALSE;
|
||||
for (com = s + 1; *com != '\0'; com++) {
|
||||
switch (*com) {
|
||||
case '%':
|
||||
case '&':
|
||||
case '@':
|
||||
case '#':
|
||||
if (*(com + 1) != '\0')
|
||||
com++; /* skip the other set's letter */
|
||||
break;
|
||||
default:
|
||||
if (*com == c)
|
||||
return HTS_TRUE;
|
||||
break;
|
||||
}
|
||||
}
|
||||
return HTS_FALSE;
|
||||
}
|
||||
|
||||
// détermine si l'argument est une option
|
||||
int cmdl_opt(char *s) {
|
||||
if (s[0] == '-') { // c'est peut être une option
|
||||
|
||||
@@ -54,6 +54,13 @@ refused() {
|
||||
! echo "FAIL: $1 (exit $RC)" || exit 1
|
||||
}
|
||||
|
||||
# assert continue mode was entered: it drops the URL list, so with no cache to
|
||||
# resume the run ends on the usage screen rather than on any other error
|
||||
continued() {
|
||||
{ test "$RC" -ne 0 && grep -q 'usage:' "$1/.log"; } ||
|
||||
! echo "FAIL: $2 (exit $RC)" || exit 1
|
||||
}
|
||||
|
||||
# a value past the old 126/256 caps but within the cap is accepted, on both the
|
||||
# short and long form of each option
|
||||
long=$(nchars 900)
|
||||
@@ -102,4 +109,50 @@ for bad in nan nan:5 5:nan inf 10:5 99999; do
|
||||
refused "#185: invalid --pause '$bad' not refused cleanly"
|
||||
done
|
||||
|
||||
# An option is not -i (continue) merely because its name contains an 'i' (#615).
|
||||
# These used to wipe the URL given before them and exit on the usage screen.
|
||||
run "$tmp/ord-bti" --build-top-index
|
||||
accepted "$tmp/ord-bti" "#615: --build-top-index after the URL wiped the URL list"
|
||||
run "$tmp/ord-bti-s" "-%i"
|
||||
accepted "$tmp/ord-bti-s" "#615: -%i after the URL wiped the URL list"
|
||||
run "$tmp/ord-proto" --protocol 2
|
||||
accepted "$tmp/ord-proto" "#615: --protocol after the URL wiped the URL list"
|
||||
run "$tmp/ord-proto-s" "-@i2"
|
||||
accepted "$tmp/ord-proto-s" "#615: -@i2 after the URL wiped the URL list"
|
||||
|
||||
# %, &, @ and # take the letter after them into another option set, wherever
|
||||
# they sit in the cluster, so the i of -q%i is not the main-set -i either.
|
||||
run "$tmp/ord-qpi" "-q%i"
|
||||
accepted "$tmp/ord-qpi" "#615: -q%i after the URL wiped the URL list"
|
||||
run "$tmp/ord-qai" "-q@i"
|
||||
accepted "$tmp/ord-qai" "#615: -q@i after the URL wiped the URL list"
|
||||
|
||||
# -%q only forces quiet mode, which nothing here can see: a run whose stdout is
|
||||
# not a tty is quiet from the start (htscoremain.c:174). Just check it crawls.
|
||||
run "$tmp/ord-iqs" "-%q"
|
||||
accepted "$tmp/ord-iqs" "#615: -%q after the URL broke the crawl"
|
||||
|
||||
# The real -i still forces continue mode and drops the URL list, in every form:
|
||||
# a fix that merely stopped looking for 'i' would pass the checks above.
|
||||
run "$tmp/cont-s" -i
|
||||
continued "$tmp/cont-s" "#615: -i after the URL no longer forces continue mode"
|
||||
run "$tmp/cont-c" -iC2
|
||||
continued "$tmp/cont-c" "#615: -iC2 after the URL no longer forces continue mode"
|
||||
run "$tmp/cont-l" --continue
|
||||
continued "$tmp/cont-l" "#615: --continue after the URL no longer forces continue mode"
|
||||
run "$tmp/cont-u" --update
|
||||
continued "$tmp/cont-u" "#615: --update after the URL no longer forces continue mode"
|
||||
|
||||
# ...including where another set's letter precedes it in the cluster.
|
||||
run "$tmp/cont-pq" "-%qi"
|
||||
continued "$tmp/cont-pq" "#615: -%qi after the URL no longer forces continue mode"
|
||||
run "$tmp/cont-pm" "-%Mi"
|
||||
continued "$tmp/cont-pm" "#615: -%Mi after the URL no longer forces continue mode"
|
||||
|
||||
# Placed before the URL these always worked; they must keep working.
|
||||
run_only "$tmp/pre-bti" "-%i" "file://$tmp/index.html"
|
||||
accepted "$tmp/pre-bti" "#615: -%i before the URL broke the crawl"
|
||||
run_only "$tmp/pre-proto" "-@i2" "file://$tmp/index.html"
|
||||
accepted "$tmp/pre-proto" "#615: -@i2 before the URL broke the crawl"
|
||||
|
||||
exit 0
|
||||
|
||||
@@ -1,90 +0,0 @@
|
||||
#!/bin/bash
|
||||
#
|
||||
# lang/*.txt are line pairs: an English msgid, then its translation. A stray
|
||||
# blank or a drifted msgid silently unhooks every translation that follows,
|
||||
# rather than failing, so assert the pairing and the join against English.txt.
|
||||
# Pairs are physical lines here; a msgid ending in \ continues onto the next one
|
||||
# for the engine (linput_cpp) but not for us. None do today.
|
||||
|
||||
set -euo pipefail
|
||||
|
||||
langdir="${top_srcdir:-..}/lang"
|
||||
def="${top_srcdir:-..}/lang.def"
|
||||
eng="$langdir/English.txt"
|
||||
for f in "$eng" "$def"; do
|
||||
[ -f "$f" ] || {
|
||||
echo "cannot find $f"
|
||||
exit 1
|
||||
}
|
||||
done
|
||||
|
||||
tmp=$(mktemp -d)
|
||||
trap 'rm -rf "$tmp"' EXIT
|
||||
keys="$tmp/english.keys"
|
||||
|
||||
# Byte-wise: each file is in its own declared legacy charset, lang.def is CRLF.
|
||||
LC_ALL=C awk 'NR%2==1 { sub(/\r$/, ""); print }' "$eng" >"$keys"
|
||||
|
||||
# An empty parse would make every check below pass vacuously.
|
||||
nkeys=$(wc -l <"$keys")
|
||||
if [ "$nkeys" -lt 400 ]; then
|
||||
echo "only $nkeys msgids parsed from $eng; the parse is broken"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
nfiles=0
|
||||
fail=0
|
||||
for f in "$langdir"/*.txt; do
|
||||
nfiles=$((nfiles + 1))
|
||||
LC_ALL=C awk -v keys="$keys" -v name="${f##*/}" '
|
||||
BEGIN { while ((getline k < keys) > 0) known[k] = 1 }
|
||||
{ sub(/\r$/, "") }
|
||||
NR % 2 == 1 {
|
||||
if ($0 == "")
|
||||
printf "%s:%d: blank line where an English msgid belongs\n", name, NR
|
||||
else if (!($0 in known))
|
||||
printf "%s:%d: msgid absent from English.txt: %s\n", name, NR, $0
|
||||
else
|
||||
next
|
||||
bad++
|
||||
}
|
||||
END {
|
||||
if (NR % 2) {
|
||||
printf "%s: odd line count (%d): msgid/translation pairing is broken\n", name, NR
|
||||
bad++
|
||||
}
|
||||
exit(bad > 0)
|
||||
}
|
||||
' "$f" || fail=1
|
||||
done
|
||||
|
||||
if [ "$nfiles" -lt 25 ]; then
|
||||
echo "only $nfiles language files found in $langdir"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
# lang.def maps each LANG_* macro to the English string the GUI compiles in,
|
||||
# which is the lookup key into every lang/*.txt: no msgid, no translations.
|
||||
LC_ALL=C awk -v keys="$keys" '
|
||||
BEGIN { while ((getline k < keys) > 0) known[k] = 1 }
|
||||
{ sub(/\r$/, "") }
|
||||
NR % 2 == 1 { macro = $0; next }
|
||||
macro ~ /^(LANG_|LISTDEF_)/ {
|
||||
n++
|
||||
if (!($0 in known)) {
|
||||
printf "lang.def:%d: %s maps to \"%s\", which is not a msgid in English.txt\n", NR, macro, $0
|
||||
bad++
|
||||
}
|
||||
}
|
||||
END {
|
||||
if (n < 400) {
|
||||
printf "lang.def: only %d LANG_ entries parsed; the parse is broken\n", n
|
||||
bad++
|
||||
}
|
||||
exit(bad > 0)
|
||||
}
|
||||
' "$def" || fail=1
|
||||
|
||||
[ "$fail" -eq 0 ] || exit 1
|
||||
|
||||
echo "$nfiles language files consistent with $nkeys English msgids"
|
||||
@@ -136,7 +136,6 @@ TESTS = \
|
||||
55_local-chunked.test \
|
||||
56_local-proxy-noleak.test \
|
||||
57_local-proxy-connect.test \
|
||||
58_watchdog.test \
|
||||
62_lang-integrity.test
|
||||
58_watchdog.test
|
||||
|
||||
CLEANFILES = check-network_sh.cache
|
||||
|
||||
Reference in New Issue
Block a user