Hello! I think that sometimes it could help to keep downloaded Metalink's files which have a bad hash.
The default wget behaviour is to delete such files. This patch provides a way to keep files which have a bad hash through the option --keep-badhash. It appends the suffix .badhash to the file name, except without overwriting existing files. In the latter case, an unique suffix is appended after .badhash. I made this patch working on the following branch: master (latest 20cac2c5ab3d63aacfba35fb10878a2d490e2377) git://git.savannah.gnu.org/wget.git What do you think? -- Matthew White <[email protected]>
>From 94ceddf1019f32181a317cba6f8918d7d73100cf Mon Sep 17 00:00:00 2001 From: Matthew White <[email protected]> Date: Thu, 28 Jul 2016 20:21:48 +0200 Subject: [PATCH] Implement --keep-badhash to keep Metalink's files with a bad hash With --keep-badhash, append .badhash to Metalink's files with checksum mismatch, except without overwriting existing files. Without --keep-badhash, remove downloaded files with checksum mismatch (this conforms to the old behaviour). * src/init.c: keepbadhash * src/main.c: keep-badhash * src/options.h: keep_badhash * doc/wget.texi: --keep-badhash * src/metalink.h: badhash_suffix(), badhash_or_remove() * src/metalink.c: On error, append .badhash if opt.keep_badhash --- doc/wget.texi | 6 ++++++ src/init.c | 1 + src/main.c | 3 +++ src/metalink.c | 53 ++++++++++++++++++++++++++++++++++++++++++++++------- src/metalink.h | 3 +++ src/options.h | 1 + 6 files changed, 60 insertions(+), 7 deletions(-) diff --git a/doc/wget.texi b/doc/wget.texi index f6f0fbc..909696c 100644 --- a/doc/wget.texi +++ b/doc/wget.texi @@ -512,6 +512,12 @@ href if none was specified. Downloads files covered in local Metalink @var{file}. Metalink version 3 and 4 are supported. +@cindex keep-badhash +@item --keep-badhash +Keeps downloaded Metalink's files with a bad hash. It appends .badhash +to the name of Metalink's files which have a checksum mismatch, except +without overwriting existing files. + @cindex metalink-over-http @item --metalink-over-http Issues HTTP HEAD request instead of GET and extracts Metalink metadata diff --git a/src/init.c b/src/init.c index 06d2e44..dc2d9a6 100644 --- a/src/init.c +++ b/src/init.c @@ -237,6 +237,7 @@ static const struct { { "input-metalink", &opt.input_metalink, cmd_file }, #endif { "iri", &opt.enable_iri, cmd_boolean }, + { "keepbadhash", &opt.keep_badhash, cmd_boolean }, { "keepsessioncookies", &opt.keep_session_cookies, cmd_boolean }, { "limitrate", &opt.limit_rate, cmd_bytes }, { "loadcookies", &opt.cookies_input, cmd_file }, diff --git a/src/main.c b/src/main.c index 4d69e03..57c7703 100644 --- a/src/main.c +++ b/src/main.c @@ -343,6 +343,7 @@ static struct cmdline_option option_data[] = { "input-metalink", 0, OPT_VALUE, "input-metalink", -1 }, #endif { "iri", 0, OPT_BOOLEAN, "iri", -1 }, + { "keep-badhash", 0, OPT_BOOLEAN, "keepbadhash", -1 }, { "keep-session-cookies", 0, OPT_BOOLEAN, "keepsessioncookies", -1 }, { "level", 'l', OPT_VALUE, "reclevel", -1 }, { "limit-rate", 0, OPT_VALUE, "limitrate", -1 }, @@ -703,6 +704,8 @@ Download:\n"), --unlink remove file before clobber\n"), #ifdef HAVE_METALINK N_("\ + --keep-badhash keep files with checksum mismatch (append .badhash)\n"), + N_("\ --metalink-over-http use Metalink metadata from HTTP response headers\n"), N_("\ --preferred-location preferred location for Metalink resources\n"), diff --git a/src/metalink.c b/src/metalink.c index f21f3aa..bae8135 100644 --- a/src/metalink.c +++ b/src/metalink.c @@ -117,13 +117,13 @@ retrieve_from_metalink (const metalink_t* metalink) retr_err = METALINK_RETR_ERROR; /* If output_stream is not NULL, then we have failed on - previous resource and are retrying. Thus, remove the file. */ + previous resource and are retrying. Thus, rename/remove + the file. */ if (output_stream) { fclose (output_stream); output_stream = NULL; - if (unlink (filename)) - logprintf (LOG_NOTQUIET, "unlink: %s\n", strerror (errno)); + badhash_or_remove (filename); xfree (filename); } @@ -489,15 +489,13 @@ gpg_skip_verification: #endif last_retr_err = retr_err == RETROK ? last_retr_err : retr_err; - /* Remove the file if error encountered or if option specified. + /* Rename the file if error encountered; remove if option specified. Note: the file has been downloaded using *_loop. Therefore, it is not necessary to keep the file for continuated download. */ if ((retr_err != RETROK || opt.delete_after) && filename != NULL && file_exists_p (filename)) { - logprintf (LOG_VERBOSE, _("Removing %s.\n"), quote (filename)); - if (unlink (filename)) - logprintf (LOG_NOTQUIET, "unlink: %s\n", strerror (errno)); + badhash_or_remove (filename); } if (output_stream) { @@ -515,6 +513,47 @@ gpg_skip_verification: return last_retr_err; } +/* Append the suffix ".badhash" to the file NAME, except without + overwriting an existing file with that name and suffix. */ +void +badhash_suffix (char *name) +{ + char *bhash, *uname; + bhash = malloc (strlen (name) + strlen (".badhash") + 1); + strcat (strcpy (bhash, name), ".badhash"); + uname = unique_name (bhash, false); + + logprintf (LOG_VERBOSE, _("Renaming â%sâ to â%sâ.\n"), name, uname); + + if (link (name, uname)) + logprintf (LOG_NOTQUIET, "link: %s\n", strerror (errno)); + else if (unlink (name)) + logprintf (LOG_NOTQUIET, "unlink: %s\n", strerror (errno)); + + xfree (bhash); + xfree (uname); +} + +/* Append the suffix ".badhash" to the file NAME, except without + overwriting an existing file with that name and suffix. + + Remove the file NAME if the option --delete-after is specified, or + if the option --keep-badhash isn't set. */ +void +badhash_or_remove (char *name) +{ + if (opt.delete_after || !opt.keep_badhash) + { + logprintf (LOG_VERBOSE, _("Removing %s.\n"), quote (name)); + if (unlink (name)) + logprintf (LOG_NOTQUIET, "unlink: %s\n", strerror (errno)); + } + else + { + badhash_suffix(name); + } +} + int metalink_res_cmp (const void* v1, const void* v2) { const metalink_resource_t *res1 = *(metalink_resource_t **) v1, diff --git a/src/metalink.h b/src/metalink.h index e98c210..020fdf5 100644 --- a/src/metalink.h +++ b/src/metalink.h @@ -47,6 +47,9 @@ uerr_t retrieve_from_metalink (const metalink_t *metalink); int metalink_res_cmp (const void *res1, const void *res2); +void badhash_suffix (char *name); +void badhash_or_remove (char *name); + bool find_key_value (const char *start, const char *end, const char *key, diff --git a/src/options.h b/src/options.h index b2e31a8..63b9bba 100644 --- a/src/options.h +++ b/src/options.h @@ -260,6 +260,7 @@ struct options bool cookies; /* whether cookies are used. */ char *cookies_input; /* file we're loading the cookies from. */ char *cookies_output; /* file we're saving the cookies to. */ + bool keep_badhash; /* Keep files with checksum mismatch. */ bool keep_session_cookies; /* whether session cookies should be saved and loaded. */ -- 2.7.3
