The query_command option expects the command to return lines in the following format: <email address> <tab> <long name> <tab> <other info> <newline>
(From https://gitlab.com/muttmua/mutt/-/wikis/MuttGuide/Aliases) Before this patch the parser used strtok() to iterate through the Tab separators. The problem is that strtok() skips through multiple consecutive Tab characters at once. This imposed a hidden requirement on the format above, which is that <long name> must not be empty, otherwise the parser will have a weird bug, where it will interpret the <other info> token as the <long name> token, because it consumed two consecutive Tab separators in one call. To fix this, we replace strtok() with strsep() which is almost identical, except that it does not skip over consecutive tokens, therefore the above usecase scenario will work as expected, i.e. <long name> will be an empty string and <other info> will be the correct token. This allows us to provide <other info> while still leaving out the <long name> token which fixes #397. Example: Program outputs: f...@bar.com\t\tOther Info\n Before this patch: <long name> = "Other Info", <other info> = "" After this patch: <long name> = "", <other info> = "Other Info" This would mean that before this patch, in the query_command menu, mutt would insert "Other Info <f...@bar.com>" by mistake, whereas now it will correctly insert "<f...@bar.com>". Signed-off-by: Magnus Groß <magnus.gr...@rwth-aachen.de> --- query.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/query.c b/query.c index 5a5c07b0..5533ddc1 100644 --- a/query.c +++ b/query.c @@ -128,7 +128,7 @@ static QUERY *run_query (char *s, int quiet) msg = mutt_read_line (msg, &msglen, fp, &dummy, 0); while ((buf = mutt_read_line (buf, &buflen, fp, &dummy, 0)) != NULL) { - if ((p = strtok(buf, "\t\n"))) + if ((p = strsep(&buf, "\t\n"))) { if (first == NULL) { @@ -142,11 +142,11 @@ static QUERY *run_query (char *s, int quiet) } cur->addr = rfc822_parse_adrlist (cur->addr, p); - p = strtok(NULL, "\t\n"); + p = strsep(&buf, "\t\n"); if (p) { cur->name = safe_strdup (p); - p = strtok(NULL, "\t\n"); + p = strsep(&buf, "\t\n"); if (p) cur->other = safe_strdup (p); } -- 2.35.1
