Re: Small changes to facilitate Win32 port
От | Thomas Lockhart |
---|---|
Тема | Re: Small changes to facilitate Win32 port |
Дата | |
Msg-id | 3CF77DED.202048DA@fourpalms.org обсуждение исходный текст |
Ответ на | Small changes to facilitate Win32 port (Katherine Ward <kward6@yahoo.com>) |
Ответы |
Re: Small changes to facilitate Win32 port
|
Список | pgsql-hackers |
> >> 2. Add _P to the following lex/yacc tokens to avoid collisions > >> CONST, CHAR, DELETE, FLOAT, GROUP, IN, OUT > I'm tempted to suggest that we should stick _P on *all* the lexer token > symbols, rather than having an inconsistent set of names where some of > them have _P and some do not. Or perhaps _T (for token) would be a more > sensible convention; I'm not sure why _P was used in the first place. "P" for "Parser". The symbols are used past the lexer, but are isolated to other places in the parser, and are (or should be) stripped out beyond there. > >> 3. Rename two local macros > >> a. MEM_FREE => MEM_FREE_IT in backend/utils/hash/dynahash.c > >> b. IGNORE => IGNORE_TOK in include/utils/datetime.h & > >> backend/utils/adt/datetime.c > It's fairly amazing that IGNORE is the only one of the datetime.h field > names that's bitten anyone (so far). Macros named TZ, YEAR, MONTH, DAY, > HOUR, MINUTE, SECOND, UNITS all look like trouble waiting to happen > (and UNKNOWN_FIELD looks like someone already had to beat a retreat from > calling it UNKNOWN ;-)). I'm inclined to suggest that these names > should be uniformly changed to DTF_FOO (DTF for "datetime field"). > The macro names appearing before the field name list look like trouble > as well --- anyone have an interest in changing them? Thomas, this is > pretty much your turf; what do you think? If the lexer/parser should have postfix qualifiers, let's use postfix for other naming conventions too (or switch everything to prefix, but be consistant in the conventions). No problem with qualifying the names, though you have likely overstated the case for it; "fairly amazing" after 6 years of use on over a dozen platforms probably qualifies as a good test of reality and we aren't quite to the point of having to invoke miracles and magic to explain why it works ;) In any case, we would certainly be open to accepting patches for the limited number of cases Katherine has identified, and would welcome patches which are more comprehensive if they were available. - Thomas
В списке pgsql-hackers по дате отправления: