| From: | "Ansley, Michael" <Michael(dot)Ansley(at)intec(dot)co(dot)za> |
|---|---|
| To: | "'Tom Lane'" <tgl(at)sss(dot)pgh(dot)pa(dot)us>, "Ansley, Michael" <Michael(dot)Ansley(at)intec(dot)co(dot)za> |
| Cc: | "'Leon'" <leon(at)udmnet(dot)ru>, hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | RE: [HACKERS] Postgres' lexer |
| Date: | 1999-09-04 18:47:14 |
| Message-ID: | 1BF7C7482189D211B03F00805F8527F748C038@S-NATH-EXCH2 |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
>>
>> > Point taken. So, if the spaces are used, then a>-2 is not
>> the same as a>-
>> > 2. The latter should then generate an error, right?
>>
>> It wasn't real clear where you intended to insert whitespace in this
>> example... but in any case, it might or might not generate an error
>> depending on what operators have been defined. Both "a >- 2" (three
>> tokens) and "a > - 2" (four tokens) might be legal expressions.
>> If they are not, it's not the lexer's job to figure that out.
Yes, and -2 is two tokens, whether it's - 2 or -2. Taking the unary minus
and the number, and creating a single expression meaning two less than zero
is not the lexer's job. Or is it? Or am I missing the plot?
MikeA
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 1999-09-04 19:05:03 | Re: [HACKERS] md.c is feeling much better now, thank you |
| Previous Message | Bruce Momjian | 1999-09-04 17:27:50 | Re: [HACKERS] temp table oddness? |