On Wednesday, April 23, 2014 11:24:44 AM UTC-5, squeegee wrote: > > > Just FYI, some background on keywords that start with a number... > > > > The reader docs (http://clojure.org/reader) states that symbols (and > keywords follow symbols in these rules) must "begin with a non-numeric > character". > > > Thanks, Alex. The REPL-y and sjacket issues cited the Clojure issues in > Jira and I looked through them. > > There seems to be an assumption in the discussion I’ve seen that the > current description of keywords at http://clojure.org/reader means that > the characters *after* the leading colon in a keyword have to follow the > rules for a symbol. >
This was not an assumption - Rich told me to interpret the page in this way. > I think another reasonable way to interpret the description of keywords is > that the spelling of a keyword *including the leading colon* has to follow > the rules for a symbol — with the obvious exception that a symbol’s spelling *cannot* begin with a > colon. Since a colon is always non-numeric, keywords *always* pass the > “begins with a non-numeric character” test. > > When I looked in the past at what it takes to parse a token in Clojure, I > concluded that the syntax of numbers vs. non-numbers (including the syntax > for numbers with a radix component) was chosen such that a token represents > a number if and only if it has a leading numeric character. I’ve always > interpreted the “no leading numeric character” rule for the spelling of > symbols (and keywords, including the leading colon) as in support of that > simple distinction. More generally, the reader can classify a token by its > leading character, or leading 2 characters in the case of a leading #. > I think you are reasoning from implementation, which may change. The http://clojure.org/readers page should be the normative reference. (as much as anything in Clojure is normative :) > I don’t see a similarly compelling reason to disallow numeric characters > immediately after the leading colon in a keyword. > There is not any particularly compelling reason which is why we deemed it of lesser importance than breaking people's existing programs and reverted the change. > > —Steve > > -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en --- You received this message because you are subscribed to the Google Groups "Clojure" group. To unsubscribe from this group and stop receiving emails from it, send an email to clojure+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
