On 07-06-2012 22:35:14 +0200, Yagiz Kargin wrote: > registrar: mimiced seq_no because original data is not clean. > > > diffs (65 lines): > > diff --git a/sql/backends/monet5/miniseed/registrar.c > b/sql/backends/monet5/miniseed/registrar.c > --- a/sql/backends/monet5/miniseed/registrar.c > +++ b/sql/backends/monet5/miniseed/registrar.c > @@ -554,6 +554,7 @@ str mseed_register(str file_path, temp_c > BAT *aBAT = NULL; > int files_done = FALSE; > timestamp start_timestamp; > + int32_t seq_no_fake = 1;
Yagiz, These types are known to be portability problems. Any reason "int" isn't good enough? Since you use it with BUNappend, you better use types that GDK understands for consistency anyway, IMO. > lng st; > str ch = (str) GDKmalloc(2*sizeof(char)); > ch[1] = '\0'; > @@ -614,7 +615,8 @@ str mseed_register(str file_path, temp_c > > if ((aBAT = > BATdescriptor(ret_tc->tables_columns[1].column_bats[1])) == NULL) > throw(MAL, "mseed_register", RUNTIME_OBJECT_MISSING); > - BUNappend(aBAT, (ptr) &(msr->sequence_number), FALSE); > +// BUNappend(aBAT, (ptr) &(msr->sequence_number), FALSE); Please use /* ... */ comments, if I'm not mistaken // requires C99, and we don't specify this to the compiler for MonetDB's codebase (on purpose). On a side-note, you're using a version control system, so there is no good reason to comment out code without documenting why, as compared to simply removing it. While at it, please don't mix tabs with spaces. Only use tabs to indent in MonetDB code. Fabian
signature.asc
Description: Digital signature
_______________________________________________ Checkin-list mailing list Checkin-list@monetdb.org http://mail.monetdb.org/mailman/listinfo/checkin-list
