Jim Meyering <jim@meyering.net>
To:
libmicrohttpd@gnu.org
Date:
Today 10:17:27
I sent this change before, along with the s/are/is/ change,
but I guess you didn't see it:
Index: doc/microhttpd.texi
===================================================================
--- doc/microhttpd.texi (revision 15482)
+++ doc/microhttpd.texi (working copy)
@@ -269,7 +269,7 @@
@deftp {Enumeration} MHD_FLAG
Options for the @mhd{} daemon.
-Note that if neither @code{MHD_USER_THREAD_PER_CONNECTION} nor
+Note that if neither @code{MHD_USE_THREAD_PER_CONNECTION} nor
@code{MHD_USE_SELECT_INTERNALLY} is used, the client wants control over
the process and will call the appropriate microhttpd callbacks.
Index: src/include/microhttpd.h
===================================================================
--- src/include/microhttpd.h (revision 15482)
+++ src/include/microhttpd.h (working copy)
@@ -306,7 +306,7 @@
/**
* Options for the MHD daemon. Note that if neither
- * MHD_USER_THREAD_PER_CONNECTION nor MHD_USE_SELECT_INTERNALLY is
+ * MHD_USE_THREAD_PER_CONNECTION nor MHD_USE_SELECT_INTERNALLY is
* used, the client wants control over the process and will call the
* appropriate microhttpd callbacks.<p>
From:
Thomas Stalder <thomas@netsolux.ch>
To:
libmicrohttpd@gnu.org
Date:
Today 12:55:56
Attachments:
cygwin_plibc.patch
Spam Status: Spamassassin 0% probability of being spam.
Full report:
Probability=No, score=-2.6 required=7.0 tests=BAYES_00 autolearn=ham version=3.2.5-tuminfo_1
Hello,
Cygwin provides POSIX emulation.
I think libmicrohttpd don't need PLIBC under cygwin.
I have removed the PLIBC dependency under cygwin with the attached patch.
I have made some tests and all seem to work fine.
From:
Eivind Sarto <ivan@espial.com>
To:
"libmicrohttpd@gnu.org" <libmicrohttpd@gnu.org>
Date:
Today 09:32:21 pm
Spam Status: Spamassassin 0% probability of being spam.
Full report:
Probability=No, score=-2.6 required=7.0 tests=BAYES_00 autolearn=ham version=3.2.5-tuminfo_1
There appears to be a bug in MHD_create_response_from_fd_at_offset().
Calling this function with anything other than a zero offset will cause wrong data
or no data (sendfile fails if length < 0).
If you use this call with any application that uses ranges, this bug will trigger.
In src/daemon/daemon.c: send_param_adapter()
.....
/* can use sendfile */
offset = (off_t) connection->response_write_position + connection->response->fd_off;
#ifdef BUGFIX
/* correct */
left = connection->response->total_size - connection->response_write_position;
#else
left = connection->response->total_size - offset;
#endif
if (left > SSIZE_MAX)
left = SSIZE_MAX; /* cap at return value limit */
ret = sendfile (connection->socket_fd,
fd,
&offset,
left);
-eivind
Christian Grothoff
Matthias Wachs