[ClusterLabs] Antw: Re: Antw: Re: Antw: Changes coming in Pacemaker 2.0.0

[Klaus Wenninger]

On 01/15/2018 10:17 AM, Ulrich Windl wrote:
>
>>>> Klaus Wenninger <kwenning at redhat.com> schrieb am 15.01.2018 um 08:41 in
> Nachricht <a5a3bfc9-ddfb-5cb6-797d-afb69e585ddb at redhat.com>:
>> On 01/15/2018 08:33 AM, Ulrich Windl wrote:
>>>>>> Ken Gaillot <kgaillot at redhat.com> schrieb am 11.01.2018 um 17:37 in Nachricht
>>> <1515688644.12807.1.camel at redhat.com>:
>>>> On Thu, 2018-01-11 at 08:54 +0100, Ulrich Windl wrote:
>>>>> On "--crm_xml -> --xml-text": Why not simply "--xml" (XML IS text)?
>>>> Most Pacemaker tools that accept XML can get it from standard input (
>>>> --xml-pipe), a file (--xml-file), or a literal string (--xml-text).
>>>>
>>>> Although, looking at it now, it might be nice to reduce it to one
>>>> option:
>>>>
>>>>  --xml -       standard input
>>>>  --xml '<cib>' anything starting with '<' is literal
>>>>  --xml file    anything else is a filename
>>> Sounds good. Maybe some "inverted microsoft logic" could be applied: In the 
>> past (MS-DOS times) for some parameters an argument staring with '@' meant to 
>> read the actual parameters from the file that follows '@' (mostly because the 
>> MS-DOS line length was limited to 127 bytes): For XML we should consider 
>> everything as a file name, with two exceptions:
>>> 1) "-" stands for stdin
>>> 2) "@" says the rest is to be taken as file contents (not file name)
>>>
>>> So you could have XML that does not start with '<' immediately.
>> What did you explicitly have in mind that wouldn't start with '<',
>> and that justifies introduction of a special-character like '@' that
>> might cause all sorts of other issues?
> I don't know, but when changing the syntax anyway, who will be unable to add a '@', especially after getting "no such file or directory" errors for the XML? You may argue whether '@' is needed or desired at all, but not about complicating the syntax by requiring to add '@' in front of literal XML.

The first ... Don't think it is needed to mark literal xml at all.
And especially not using a character that might have a special
meaning depending on how the commands get to their final destination
(thinking of the nightmare of (multiple) escapes and such).

Regards,
Klaus

>
> Regards,
> Ulrich
>
>> Regards,
>> Klaus
>>
>>> Regards,
>>> Ulrich
>>>
>>>> -- 
>>>> Ken Gaillot <kgaillot at redhat.com>
>>>>
>>>> _______________________________________________
>>>> Users mailing list: Users at clusterlabs.org 
>>>> http://lists.clusterlabs.org/mailman/listinfo/users 
>>>>
>>>> Project Home: http://www.clusterlabs.org 
>>>> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf 
>>>> Bugs: http://bugs.clusterlabs.org