All of lore.kernel.org
 help / color / mirror / Atom feed
From: Markus Armbruster <armbru@redhat.com>
To: Eric Blake <eblake@redhat.com>
Cc: kwolf@redhat.com, qemu-devel@nongnu.org, qemu-block@nongnu.org,
	mreitz@redhat.com
Subject: Re: [PATCH for-5.1 2/8] qemu-options: Factor out get_opt_name_value() helper
Date: Tue, 14 Apr 2020 11:42:22 +0200	[thread overview]
Message-ID: <87blnuwght.fsf@dusky.pond.sub.org> (raw)
In-Reply-To: <5f39e636-f799-55f3-873f-08a3a0f53dca@redhat.com> (Eric Blake's message of "Thu, 9 Apr 2020 13:01:59 -0500")

Eric Blake <eblake@redhat.com> writes:

> On 4/9/20 10:30 AM, Markus Armbruster wrote:
>> The next commits will put it to use.
>>
>> Signed-off-by: Markus Armbruster <armbru@redhat.com>
>> ---
>>   util/qemu-option.c | 102 +++++++++++++++++++++++++--------------------
>>   1 file changed, 56 insertions(+), 46 deletions(-)
>>
>
>> +static const char *get_opt_name_value(const char *params,
>> +                                      const char *firstname,
>> +                                      char **name, char **value)
>> +{
>> +    const char *p, *pe, *pc;
>> +
>> +    pe = strchr(params, '=');
>> +    pc = strchr(params, ',');
>> +
>> +    if (!pe || (pc && pc < pe)) {
>> +        /* found "foo,more" */
>> +        if (firstname) {
>> +            /* implicitly named first option */
>> +            *name = g_strdup(firstname);
>> +            p = get_opt_value(params, value);
>
> Is this correct even when params is "foo,,more"?  But...
>
>>   static void opts_do_parse(QemuOpts *opts, const char *params,
>>                             const char *firstname, bool prepend,
>>                             bool *invalidp, Error **errp)
>>   {
>> -    char *option = NULL;
>> -    char *value = NULL;
>> -    const char *p,*pe,*pc;
>>       Error *local_err = NULL;
>> +    char *option, *value;
>> +    const char *p;
>>   -    for (p = params; *p != '\0'; p++) {
>> -        pe = strchr(p, '=');
>> -        pc = strchr(p, ',');
>> -        if (!pe || (pc && pc < pe)) {
>> -            /* found "foo,more" */
>> -            if (p == params && firstname) {
>> -                /* implicitly named first option */
>> -                option = g_strdup(firstname);
>> -                p = get_opt_value(p, &value);
>
> ...in this patch, it is just code motion, so if it is a bug, it's
> pre-existing and worth a separate fix.

If @p points to "foo,,more", possibly followed by additional characters,
then we have pc && pc < pe, and enter this conditional.  This means that

    foo,,more=42

gets parsed as two name=value, either as

    name        value
    -----------------------
    @firstname  "foo,more"
    ""          "42"

if @firstname is non-null, or else as

    name        value
    -----------------
    "foo,,more" "on"
    ""          "42"

Has always been that way; see commit e27c88fe9e "QemuOpts: framework for
storing and parsing options.", v0.12.0.

You might expect

    name        value
    -----------------
    "foo,,more" "42"

or

    name        value
    -----------------
    "foo,more"  "42"

instead.  Close, but no cigar.

A language without syntax errors is bound to surprise.

> Reviewed-by: Eric Blake <eblake@redhat.com>

Thanks!



  reply	other threads:[~2020-04-14  9:43 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-09 15:30 [PATCH for-5.1 0/8] qemu-option: Fix corner cases and clean up Markus Armbruster
2020-04-09 15:30 ` [PATCH for-5.1 1/8] tests-qemu-opts: Cover has_help_option(), qemu_opt_has_help_opt() Markus Armbruster
2020-04-09 17:50   ` Eric Blake
2020-04-14  9:10     ` Markus Armbruster
2020-04-14 13:13       ` Kevin Wolf
2020-04-14 13:36         ` Markus Armbruster
2020-04-14 14:29           ` Kevin Wolf
2020-04-14 20:13             ` Markus Armbruster
2020-04-15 10:00               ` Kevin Wolf
2020-04-15 14:45                 ` Markus Armbruster
2020-04-09 15:30 ` [PATCH for-5.1 2/8] qemu-options: Factor out get_opt_name_value() helper Markus Armbruster
2020-04-09 18:01   ` Eric Blake
2020-04-14  9:42     ` Markus Armbruster [this message]
2020-04-14 14:05       ` Kevin Wolf
2020-04-15  7:03         ` Markus Armbruster
2020-04-14 14:05   ` Kevin Wolf
2020-04-09 15:30 ` [PATCH for-5.1 3/8] qemu-option: Fix sloppy recognition of "id=..." after ", , " Markus Armbruster
2020-04-09 18:05   ` Eric Blake
2020-04-14 14:44   ` [PATCH for-5.1 3/8] qemu-option: Fix sloppy recognition of "id=..." after ",," Kevin Wolf
2020-04-09 15:30 ` [PATCH for-5.1 4/8] qemu-option: Avoid has_help_option() in qemu_opts_parse_noisily() Markus Armbruster
2020-04-09 18:07   ` Eric Blake
2020-04-14 10:04     ` Markus Armbruster
2020-04-09 15:30 ` [PATCH for-5.1 5/8] qemu-option: Fix has_help_option()'s sloppy parsing Markus Armbruster
2020-04-09 18:10   ` Eric Blake
2020-04-14 10:16     ` Markus Armbruster
2020-04-14 14:57       ` Kevin Wolf
2020-04-15  7:48         ` Markus Armbruster
2020-04-09 15:30 ` [PATCH for-5.1 6/8] test-qemu-opts: Simplify test_has_help_option() after bug fix Markus Armbruster
2020-04-09 18:13   ` Eric Blake
2020-04-14 14:58   ` Kevin Wolf
2020-04-09 15:30 ` [PATCH for-5.1 7/8] qemu-img: Factor out accumulate_options() helper Markus Armbruster
2020-04-09 18:15   ` Eric Blake
2020-04-14 15:00   ` Kevin Wolf
2020-04-09 15:30 ` [PATCH for-5.1 8/8] qemu-option: Move is_valid_option_list() to qemu-img.c and rewrite Markus Armbruster
2020-04-09 19:45   ` Eric Blake
2020-04-14  8:47     ` Markus Armbruster
2020-04-14 14:34       ` Markus Armbruster
2020-04-14 15:10         ` Kevin Wolf
2020-04-14 20:14           ` Markus Armbruster
2020-04-09 17:09 ` [PATCH for-5.1 0/8] qemu-option: Fix corner cases and clean up no-reply
2020-04-09 17:44   ` Eric Blake
2020-04-14  8:52     ` Markus Armbruster

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87blnuwght.fsf@dusky.pond.sub.org \
    --to=armbru@redhat.com \
    --cc=eblake@redhat.com \
    --cc=kwolf@redhat.com \
    --cc=mreitz@redhat.com \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.