[B.A.T.M.A.N.] [PATCH] alfred: output valid JSON

[B.A.T.M.A.N.] [PATCH] alfred: output valid JSON

[Nils Schneider]

With this patch alfred -r will output JSON like this:

{
  "00:11:22:33:44:55" : "...",
  "22:33:44:55:66:FF" : "..."
}
---
 client.c | 18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

diff --git a/client.c b/client.c
index a535f5a..ec882d9 100644
--- a/client.c
+++ b/client.c
@@ -39,7 +39,7 @@ int alfred_client_request_data(struct globals *globals)
 	struct alfred_status_v0 *status;
 	struct alfred_tlv *tlv;
 	struct alfred_data *data;
-	int ret, len, data_len, i;
+	int ret, len, data_len, i, count;
 
 	if (unix_sock_open_client(globals, ALFRED_SOCK_PATH))
 		return -1;
@@ -59,6 +59,9 @@ int alfred_client_request_data(struct globals *globals)
 		fprintf(stderr, "%s: only wrote %d of %d bytes: %s\n",
 			__func__, ret, len, strerror(errno));
 
+	printf("{");
+
+	count = 0;
 	push = (struct alfred_push_data_v0 *)buf;
 	tlv = (struct alfred_tlv *)buf;
 	while ((ret = read(globals->unix_sock, buf, sizeof(*tlv))) > 0) {
@@ -98,9 +101,12 @@ int alfred_client_request_data(struct globals *globals)
 		if (ret < data_len)
 			break;
 
+		if (count > 0)
+			printf(",\n");
+
 		pos = data->data;
 
-		printf("{ \"%02x:%02x:%02x:%02x:%02x:%02x\", \"",
+		printf("\n  \"%02x:%02x:%02x:%02x:%02x:%02x\": \"",
 		       data->source[0], data->source[1],
 		       data->source[2], data->source[3],
 		       data->source[4], data->source[5]);
@@ -110,14 +116,18 @@ int alfred_client_request_data(struct globals *globals)
 			else if (pos[i] == '\\')
 				printf("\\\\");
 			else if (!isprint(pos[i]))
-				printf("\\x%02x", pos[i]);
+				printf("\\u00%02x", pos[i]);
 			else
 				printf("%c", pos[i]);
 		}
 
-		printf("\" },\n");
+		printf("\"");
+
+		count++;
 	}
 
+	printf("\n}\n");
+
 	unix_sock_close(globals);
 
 	return 0;
-- 
1.8.4

Re: [B.A.T.M.A.N.] [PATCH] alfred: output valid JSON

[Antonio Quartulli]

[-- Attachment #1: Type: text/plain, Size: 399 bytes --]

On Sat, Sep 07, 2013 at 10:19:46PM +0200, Nils Schneider wrote:
> With this patch alfred -r will output JSON like this:
> 
> {
>   "00:11:22:33:44:55" : "...",
>   "22:33:44:55:66:FF" : "..."
> }

is this patch fixing a broken behaviour? The commit message is not very clear
about this.

Regards,

-- 
Antonio Quartulli

..each of us alone is worth nothing..
Ernesto "Che" Guevara

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [B.A.T.M.A.N.] [PATCH] alfred: output valid JSON

[Nils Schneider]

The current output format isn't really broken, just tricky to parse.
I decided to make it valid JSOΝ as it was already pretty close. This
allows parsing with virtually any programming language.

At Sat, 7 Sep 2013 22:32:07 +0200,
Antonio Quartulli wrote:
>=20
> [1  <text/plain; utf-8 (quoted-printable)>]
> On Sat, Sep 07, 2013 at 10:19:46PM +0200, Nils Schneider wrote:
> > With this patch alfred -r will output JSON like this:
> >=20
> > {
> >   "00:11:22:33:44:55" : "...",
> >   "22:33:44:55:66:FF" : "..."
> > }
>=20
> is this patch fixing a broken behaviour? The commit message is not very c=
lear
> about this.
>=20
> Regards,
>=20
> --=20
> Antonio Quartulli
>=20
> ..each of us alone is worth nothing..
> Ernesto "Che" Guevara
> [2 Digital signature <application/pgp-signature (7bit)>]
>=20

Re: [B.A.T.M.A.N.] [PATCH] alfred: output valid JSON

[Marek Lindner]

On Sunday, September 08, 2013 04:56:39 Nils Schneider wrote:
> The current output format isn't really broken, just tricky to parse.
> I decided to make it valid JSOΝ as it was already pretty close. This
> allows parsing with virtually any programming language.

Please be more specific about what was broken and how your fix addresses the 
problem. An example would be nice too.

Cheers,
Marek

Re: [B.A.T.M.A.N.] [PATCH] alfred: output valid JSON

[Nils Schneider]

The current output format of alfred -r looks like this:

{ "fe:f1:00:00:01:01", "OpenWRT-node-1\x0a" },
{ "fe:f1:00:00:02:01", "OpenWRT-node-2\x0a" },
{ "fe:f1:00:00:03:01", "OpenWRT-node-3\x0a" },

It's not clear how one should parse this as there is no documentation
except the source code.

With my patch, it'll look like this:

{
  "fe:f1:00:00:01:01": "OpenWRT-node-1\u000a",
  "fe:f1:00:00:02:01": "OpenWRT-node-2\u000a",
  "fe:f1:00:00:03:01": "OpenWRT-node-3\u000a"
}

This is JSOΝ which is well documented and thus can be parsed
easily. It's a dictionary with the node's ID as the key and the data
as value.

At Sun, 8 Sep 2013 10:39:23 +0800,
Marek Lindner wrote:
>=20
> On Sunday, September 08, 2013 04:56:39 Nils Schneider wrote:
> > The current output format isn't really broken, just tricky to parse.
> > I decided to make it valid JSOΝ as it was already pretty close. This
> > allows parsing with virtually any programming language.
>=20
> Please be more specific about what was broken and how your fix addresses =
the=20
> problem. An example would be nice too.
>=20
> Cheers,
> Marek

Re: [B.A.T.M.A.N.] [PATCH] alfred: output valid JSON

[Simon Wunderlich]

[-- Attachment #1: Type: text/plain, Size: 1926 bytes --]

Hey Nils,

thanks for your patch!

On Sun, Sep 08, 2013 at 10:47:00AM +0200, Nils Schneider wrote:
> The current output format of alfred -r looks like this:
> 
> { "fe:f1:00:00:01:01", "OpenWRT-node-1\x0a" },
> { "fe:f1:00:00:02:01", "OpenWRT-node-2\x0a" },
> { "fe:f1:00:00:03:01", "OpenWRT-node-3\x0a" },
> 
> It's not clear how one should parse this as there is no documentation
> except the source code.

The output was first designed for debugging and scripts who should parse
that (binary programs like vis may access the unix sockets directly).

> 
> With my patch, it'll look like this:
> 
> {
>   "fe:f1:00:00:01:01": "OpenWRT-node-1\u000a",
>   "fe:f1:00:00:02:01": "OpenWRT-node-2\u000a",
>   "fe:f1:00:00:03:01": "OpenWRT-node-3\u000a"
> }
> 
> This is JSOΝ which is well documented and thus can be parsed
> easily. It's a dictionary with the node's ID as the key and the data
> as value.

I think your suggestion/patch is a good idea, it will result in a more
general format. I'd like to ask you to send your patch again with a
revised commit message (e.g. add your examples from this mail and a
short explanation why the old format was bad, and why JSON is better).

I'd then would like to wait for approx. 1 week to gather comments from
people who already parse the old format already (Guido?), and if there are
no objections merge it.

Thanks!
	Simon
> 
> At Sun, 8 Sep 2013 10:39:23 +0800,
> Marek Lindner wrote:
> > 
> > On Sunday, September 08, 2013 04:56:39 Nils Schneider wrote:
> > > The current output format isn't really broken, just tricky to parse.
> > > I decided to make it valid JSOΝ as it was already pretty close. This
> > > allows parsing with virtually any programming language.
> > 
> > Please be more specific about what was broken and how your fix addresses the 
> > problem. An example would be nice too.
> > 
> > Cheers,
> > Marek

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [B.A.T.M.A.N.] [PATCH] alfred: output valid JSON

[Gui Iribarren]

I'm already turning your output into proper json, but inside my lua script. So getting json out of alfred would simplify things :)
I.e.: +1 and thanks for the patch!

Simon Wunderlich <simon.wunderlich@s2003.tu-chemnitz.de> wrote:
>Hey Nils,
>
>thanks for your patch!
>
>On Sun, Sep 08, 2013 at 10:47:00AM +0200, Nils Schneider wrote:
>> The current output format of alfred -r looks like this:
>> 
>> { "fe:f1:00:00:01:01", "OpenWRT-node-1\x0a" },
>> { "fe:f1:00:00:02:01", "OpenWRT-node-2\x0a" },
>> { "fe:f1:00:00:03:01", "OpenWRT-node-3\x0a" },
>> 
>> It's not clear how one should parse this as there is no documentation
>> except the source code.
>
>The output was first designed for debugging and scripts who should
>parse
>that (binary programs like vis may access the unix sockets directly).
>
>> 
>> With my patch, it'll look like this:
>> 
>> {
>>   "fe:f1:00:00:01:01": "OpenWRT-node-1\u000a",
>>   "fe:f1:00:00:02:01": "OpenWRT-node-2\u000a",
>>   "fe:f1:00:00:03:01": "OpenWRT-node-3\u000a"
>> }
>> 
>> This is JSOΝ which is well documented and thus can be parsed
>> easily. It's a dictionary with the node's ID as the key and the data
>> as value.
>
>I think your suggestion/patch is a good idea, it will result in a more
>general format. I'd like to ask you to send your patch again with a
>revised commit message (e.g. add your examples from this mail and a
>short explanation why the old format was bad, and why JSON is better).
>
>I'd then would like to wait for approx. 1 week to gather comments from
>people who already parse the old format already (Guido?), and if there
>are
>no objections merge it.
>
>Thanks!
>	Simon
>> 
>> At Sun, 8 Sep 2013 10:39:23 +0800,
>> Marek Lindner wrote:
>> > 
>> > On Sunday, September 08, 2013 04:56:39 Nils Schneider wrote:
>> > > The current output format isn't really broken, just tricky to
>parse.
>> > > I decided to make it valid JSOΝ as it was already pretty close.
>This
>> > > allows parsing with virtually any programming language.
>> > 
>> > Please be more specific about what was broken and how your fix
>addresses the 
>> > problem. An example would be nice too.
>> > 
>> > Cheers,
>> > Marek