build: Pikaur-safe hardening flags, opt-in sanitizers, -Wsign-compare

Re-introduce the build hardening that 5459e78e8 attempted, in a form that
coexists with Arch/Pikaur builds (the conflict that forced its revert in
0722dc9e3).

configure.ac:
- Add -D_FORTIFY_SOURCE=2 only when neither $CFLAGS nor $CPPFLAGS already
  define _FORTIFY_SOURCE (makepkg injects its own) and not under --coverage.
- Route glib/gio CFLAGS through -isystem so their macros don't trip our -W set.
- Add --enable-sanitizers (ASan + UBSan; unsigned-integer-overflow probed,
  skipped on gcc) for a dedicated CI job, off by default.
- Add --enable-hardening to gate -Wnull-dereference (false positives under
  -O2 on some toolchains), off by default.
- Promote -Wsign-compare from -Wno- to an active probed warning.

Also fix the sign-conversion sites that -Wsign-compare/-Wconversion cleanup
surfaced as safe (pointer-diff to gsize via new g_diff_to_gsize helper,
non-negative int casts), and validate the /vcard remove index through
strtoi_range so a negative argument can no longer become a huge unsigned
index.

Refs #112

src/command/cmd_funcs.c

@@ -10063,8 +10063,14 @@ cmd_vcard_remove(ProfWin* window, const char* const command, gchar** args)
     }
 
     if (args[1]) {
-        vcard_user_remove_element(atoi(args[1]));
-        cons_show("Removed element at index %d", atoi(args[1]));
+        int index;
+        auto_gchar gchar* err_msg = NULL;
+        if (!strtoi_range(args[1], &index, 0, INT_MAX, &err_msg)) {
+            cons_show("%s", err_msg);
+            return TRUE;
+        }
+        vcard_user_remove_element((unsigned int)index);
+        cons_show("Removed element at index %d", index);
         vcardwin_update();
     } else {
         cons_bad_cmd_usage(command);

src/common.c

@@ -294,7 +294,7 @@ str_replace(const char* string, const char* substr,
     wr = newstr;
 
     while ((tok = strstr(head, substr))) {
-        size_t l = tok - head;
+        size_t l = g_diff_to_gsize(tok, head);
         memcpy(wr, head, l);
         wr += l;
         memcpy(wr, replacement, len_replacement);
@@ -359,9 +359,9 @@ string_matches_one_of(const char* what, const char* is, gboolean is_can_be_null,
         char errmsg[256] = { 0 };
         size_t sz = 0;
         int s = snprintf(errmsg, sizeof(errmsg) - sz, "%s must be one of:", what);
-        if (s < 0 || s + sz >= sizeof(errmsg))
+        if (s < 0 || (size_t)s + sz >= sizeof(errmsg))
             return ret;
-        sz += s;
+        sz += (size_t)s;
 
         cur = first;
         va_start(ap, first);
@@ -375,12 +375,12 @@ string_matches_one_of(const char* what, const char* is, gboolean is_can_be_null,
                 errmsg[sz] = '\0';
                 s = snprintf(errmsg + sz, sizeof(errmsg) - sz, " or '%s'.", cur);
             }
-            if (s < 0 || s + sz >= sizeof(errmsg)) {
+            if (s < 0 || (size_t)s + sz >= sizeof(errmsg)) {
                 log_debug("Error message too long or some other error occurred (%d).", s);
                 s = -1;
                 break;
             }
-            sz += s;
+            sz += (size_t)s;
             cur = next;
         }
         va_end(ap);
@@ -431,7 +431,7 @@ str_xml_sanitize(const char* const str)
         return NULL;
     }
 
-    GString* sanitized = g_string_new_len(NULL, strlen(str));
+    GString* sanitized = g_string_new_len(NULL, (gssize)strlen(str));
     const char* curr = str;
 
     while (*curr != '\0') {
@@ -692,12 +692,13 @@ get_file_paths_recursive(const char* path, GSList** contents)
 gchar*
 get_random_string(int length)
 {
+    g_assert(length >= 0);
     GRand* prng;
     gchar* rand;
     gchar alphabet[] = "0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ";
     int endrange = sizeof(alphabet) - 1;
 
-    rand = g_malloc0(length + 1);
+    rand = g_malloc0((gsize)length + 1);
 
     prng = g_rand_new();
 

src/common.h

@@ -158,6 +158,14 @@ gboolean create_dir(const char* name);
 gboolean copy_file(const char* const src, const char* const target, const gboolean overwrite_existing);
 char* str_replace(const char* string, const char* substr, const char* replacement);
 gboolean strtoi_range(const char* str, int* saveptr, int min, int max, char** err_msg);
+
+// Pointer-diff to gsize for callers that know end >= start by construction.
+static inline gsize
+g_diff_to_gsize(const void* end, const void* start)
+{
+    g_assert(end >= start);
+    return (gsize)((const char*)end - (const char*)start);
+}
 int utf8_display_len(const char* const str);
 gchar* str_xml_sanitize(const char* const str);
 

src/database_export.c

@@ -282,7 +282,7 @@ _export_one_contact(const char* contact)
         gchar* key = _make_dedup_key(pl->stanza_id, pl->timestamp_str, pl->from_jid, pl->message);
         g_hash_table_add(seen_keys, key);
     }
-    int existing_count = g_slist_length(existing);
+    int existing_count = (int)g_slist_length(existing);
 
     // 2. Query SQLite for this contact — get ALL messages via direct SQL
     GSList* sqlite_lines = db_sqlite_get_all_chat(contact);
@@ -581,7 +581,7 @@ log_database_import_from_flatfile(const gchar* const contact_jid)
         //    Wrap in a transaction for atomicity and performance
         int contact_imported = 0;
         int contact_skipped = 0;
-        int total_lines = g_slist_length(ff_lines);
+        int total_lines = (int)g_slist_length(ff_lines);
 
         db_sqlite_begin_transaction();
         gboolean import_ok = TRUE;

src/database_flatfile.c

@@ -88,7 +88,7 @@ _ff_cache_line_ids(ff_contact_state_t* state, const char* line)
         return;
 
     // Split metadata on unescaped '|'
-    auto_gchar gchar* meta_str = g_strndup(bracket + 1, close - bracket - 1);
+    auto_gchar gchar* meta_str = g_strndup(bracket + 1, g_diff_to_gsize(close, bracket + 1));
     char** parts = ff_split_meta(meta_str);
 
     char* stanza_id = NULL;
@@ -111,9 +111,9 @@ _ff_cache_line_ids(ff_contact_state_t* state, const char* line)
         const char* sender_start = close + 2;
         const char* colonspace = ff_find_unescaped_colonspace(sender_start);
         if (colonspace) {
-            const char* slash = memchr(sender_start, '/', colonspace - sender_start);
+            const char* slash = memchr(sender_start, '/', g_diff_to_gsize(colonspace, sender_start));
             const char* jid_end = slash ? slash : colonspace;
-            from_jid = g_strndup(sender_start, jid_end - sender_start);
+            from_jid = g_strndup(sender_start, g_diff_to_gsize(jid_end, sender_start));
         }
     }
 
@@ -140,7 +140,7 @@ _ff_maybe_index_line(ff_contact_state_t* state, const char* buf, off_t pos)
     if (!space)
         return;
 
-    auto_gchar gchar* ts_str = g_strndup(buf, space - buf);
+    auto_gchar gchar* ts_str = g_strndup(buf, g_diff_to_gsize(space, buf));
     GDateTime* dt = g_date_time_new_from_iso8601(ts_str, NULL);
     if (!dt) {
         log_warning("flatfile: unparsable timestamp in %s at offset %ld: %s",
@@ -771,7 +771,7 @@ _flatfile_get_previous_chat(const gchar* const contact_barejid, const gchar* sta
 
     // For "last N messages": back up from read_to using index
     if (!from_start) {
-        int margin_entries = (MESSAGES_TO_RETRIEVE * 3) / FF_INDEX_STEP + 2;
+        size_t margin_entries = (MESSAGES_TO_RETRIEVE * 3) / FF_INDEX_STEP + 2;
 
         // Find index entry closest to (but before) read_to
         size_t entry_idx = 0;
@@ -781,7 +781,7 @@ _flatfile_get_previous_chat(const gchar* const contact_barejid, const gchar* sta
             entry_idx = i;
         }
 
-        size_t start_entry = entry_idx > (size_t)margin_entries
+        size_t start_entry = entry_idx > margin_entries
                                  ? entry_idx - margin_entries
                                  : 0;
         off_t backed = state->entries[start_entry].byte_offset;

src/database_flatfile_parser.c

@@ -553,7 +553,7 @@ ff_split_meta(const char* meta)
             continue;
         }
         if (*p == '|' || *p == '\0') {
-            g_ptr_array_add(arr, g_strndup(start, p - start));
+            g_ptr_array_add(arr, g_strndup(start, g_diff_to_gsize(p, start)));
             if (*p == '\0')
                 break;
             start = p + 1;
@@ -710,7 +710,7 @@ ff_parse_line(const char* line)
 
     if (bracket_start && first_space && first_space < bracket_start) {
         // Standard format with metadata: {timestamp} [{meta}] {sender}: {msg}
-        result->timestamp_str = g_strndup(work, first_space - work);
+        result->timestamp_str = g_strndup(work, g_diff_to_gsize(first_space, work));
 
         // Parse metadata section [...]
         const char* bracket_end = ff_find_unescaped_char(bracket_start + 1, ']');
@@ -721,7 +721,7 @@ ff_parse_line(const char* line)
             return NULL;
         }
 
-        auto_gchar gchar* meta_content = g_strndup(bracket_start + 1, bracket_end - bracket_start - 1);
+        auto_gchar gchar* meta_content = g_strndup(bracket_start + 1, g_diff_to_gsize(bracket_end, bracket_start + 1));
         char** parts = ff_split_meta(meta_content);
         if (parts) {
             _ff_parse_meta_parts(parts, result);
@@ -736,12 +736,12 @@ ff_parse_line(const char* line)
         // Find first *unescaped* ': ' which separates sender from message.
         const char* colon = ff_find_unescaped_colonspace(after_meta);
         if (colon) {
-            auto_gchar gchar* raw_sender = g_strndup(after_meta, colon - after_meta);
+            auto_gchar gchar* raw_sender = g_strndup(after_meta, g_diff_to_gsize(colon, after_meta));
 
             // Split sender into jid/resource, then unescape resource
             char* slash = strchr(raw_sender, '/');
             if (slash) {
-                result->from_jid = g_strndup(raw_sender, slash - raw_sender);
+                result->from_jid = g_strndup(raw_sender, g_diff_to_gsize(slash, raw_sender));
                 result->from_resource = ff_unescape_sender_resource(slash + 1);
             } else {
                 result->from_jid = g_strdup(raw_sender);
@@ -755,7 +755,7 @@ ff_parse_line(const char* line)
         }
     } else if (first_space) {
         // Legacy/simple format without metadata: {timestamp} - {sender}: {msg}
-        result->timestamp_str = g_strndup(work, first_space - work);
+        result->timestamp_str = g_strndup(work, g_diff_to_gsize(first_space, work));
         result->type = g_strdup("chat");
         result->enc = g_strdup("none");
 
@@ -767,10 +767,10 @@ ff_parse_line(const char* line)
 
         char* colon = strstr(rest, ": ");
         if (colon) {
-            char* sender = g_strndup(rest, colon - rest);
+            char* sender = g_strndup(rest, g_diff_to_gsize(colon, rest));
             char* slash = strchr(sender, '/');
             if (slash) {
-                result->from_jid = g_strndup(sender, slash - sender);
+                result->from_jid = g_strndup(sender, g_diff_to_gsize(slash, sender));
                 result->from_resource = g_strdup(slash + 1);
             } else {
                 result->from_jid = g_strdup(sender);

src/profanity.c

@@ -105,7 +105,7 @@ prof_run(gchar* log_level, gchar* account_name, gchar* config_file, gchar* log_f
                      * so we can be sure there's runtime left after executing
                      * the last command.
                      */
-                    min_runtime += waittime;
+                    min_runtime += (unsigned int)waittime;
                 } else {
                     log_error("%s", err_msg);
                     g_free(err_msg);

src/xmpp/jid.c

@@ -117,7 +117,7 @@ jid_is_valid(const gchar* const str)
 
     // Localpart validation
     if (at) {
-        size_t local_len = at - str;
+        size_t local_len = g_diff_to_gsize(at, str);
         if (local_len == 0 || local_len > JID_MAX_PART_LEN) {
             return FALSE;
         }
@@ -135,7 +135,7 @@ jid_is_valid(const gchar* const str)
 
     // Resourcepart validation if present
     if (slash) {
-        domain_len = slash - domain_start;
+        domain_len = g_diff_to_gsize(slash, domain_start);
         size_t resource_len = strlen(slash + 1);
         if (resource_len > JID_MAX_PART_LEN) {
             return FALSE;

src/xmpp/muc.c

@@ -707,7 +707,7 @@ muc_autocomplete(ProfWin* window, const char* const input, gboolean previous)
     } else {
         search_str = last_space + 1;
         if (!chat_room->autocomplete_prefix) {
-            chat_room->autocomplete_prefix = g_strndup(input, search_str - input);
+            chat_room->autocomplete_prefix = g_strndup(input, g_diff_to_gsize(search_str, input));
         }
     }