home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!dtix!darwin.sura.net!wupost!usc!rpi!crdgw1!rdsunx.crd.ge.com!rodson!bloomer
- From: bloomer@rodson.crd.ge.com (John Bloomer)
- Newsgroups: comp.unix.internals
- Subject: recv() vs. read() (and send() vs. write())
- Keywords: socket buffering
- Message-ID: <1992Jul21.213133.23400@crd.ge.com>
- Date: 21 Jul 92 21:31:33 GMT
- Sender: usenet@crd.ge.com (Required for NNTP)
- Organization: GE Corporate R&D Center
- Lines: 14
- Nntp-Posting-Host: rodson.crd.ge.com
-
- If I interpret published unix internals books and manual pages
- correctly, it is possible to peek at the data waiting at a socket -
- using the MSG_PEEK flag.
-
- While a read() will return with whatever it has found available at a
- socket, does a recv() do the same, or does it peek to make sure the
- whole len message is available?
-
- I believe the same general question applies to send() vs. write().
- Some guidance would be appreciated.
- --
- John J. Bloomer 518-387-6964 (dialcomm 833) FAX -7332
- KWC317, General Electric Corporate Research and Development
- PO Box 8, Schenectady NY 12301
-