I don't think I suggested that. Word formation is where quoted strings
are detected.
Henry Rich
On 10/19/2020 10:24 AM, 'Pascal Jasmin' via Programming wrote:
{{ }} will be used in a step between word formation and parsing
You're suggesting here that there will NEVER be a use/occasion to see a string
that contains these new characters with a need to interpret them how J would.
If such a raw string might be useful, then modifying some of the tokens it
contains could as well.
On Monday, October 19, 2020, 09:26:30 a.m. EDT, Henry Rich
<[email protected]> wrote:
{{. etc will NOT be defined as single words.
{{ }} will be used in a step between word formation and parsing, and
will never be seen in parsing. I don't think it matters much whether
{{. is defined as a word or not. The state table for word recognition
is the same size either way.
Henry Rich
On 10/18/2020 11:01 PM, 'Pascal Jasmin' via Programming wrote:
Both are possible, but existing state machine requires fewer new entries if }}.
is not a defined token.
On Sunday, October 18, 2020, 10:20:59 p.m. EDT, Raul Miller
<[email protected]> wrote:
That would be the simplest implementation (and it's kind of critical
if code which uses ;: to manipulate j expressions should be expected
to continue to work).
Me, I'm wondering if this means that }}: and }}. will be treated as
single words. It seems like that should be the case (because otherwise
it gets harder to explain to rules for : and . in reserved words), but
I do not know enough about where things stand.
Thanks,
--
This email has been checked for viruses by AVG.
https://www.avg.com
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
