cvs commit: src/lib/libfetch fetch.3

2008-02-08 Thread Dag-Erling Smorgrav
des 2008-02-08 09:44:34 UTC FreeBSD src repository Modified files: lib/libfetch fetch.3 Log: s/MAXPATHLEN/PATH_MAX/ to reflect five-year old change to the code :) Submitted by: Joerg Sonnenberger <[EMAIL PROTECTED]> MFC after: 2 weeks Revision Chan

cvs commit: src/lib/libfetch fetch.3

2007-12-18 Thread Dag-Erling Smorgrav
des 2007-12-18 11:03:26 UTC FreeBSD src repository Modified files: lib/libfetch fetch.3 Log: Document NO_PROXY / no_proxy. MFC after: 3 weeks Revision ChangesPath 1.64 +17 -1 src/lib/libfetch/fetch.3 _

cvs commit: src/lib/libfetch fetch.3 ftp.c

2007-05-16 Thread Nate Lawson
njl 2007-05-16 07:04:02 UTC FreeBSD src repository Modified files:(Branch: RELENG_6) lib/libfetch fetch.3 ftp.c Log: MFC: the new FTP_COMBINE_CWDS define allows libfetch to attempt to change directories all at once, saving latency. Tested by: des

Re: cvs commit: src/lib/libfetch fetch.3 ftp.c

2007-05-09 Thread Nate Lawson
Dag-Erling Smørgrav wrote: > Nate Lawson <[EMAIL PROTECTED]> writes: >> Could it be possible to check the values from the URL used to derive >> TYPE for RETR and then if it's the same as the value for MDTM, don't >> reissue it? Same for MODE. > > I suggest you look at the code. We *do* derive th

Re: cvs commit: src/lib/libfetch fetch.3 ftp.c

2007-05-09 Thread Dag-Erling Smørgrav
Nate Lawson <[EMAIL PROTECTED]> writes: > Could it be possible to check the values from the URL used to derive > TYPE for RETR and then if it's the same as the value for MDTM, don't > reissue it? Same for MODE. I suggest you look at the code. We *do* derive the TYPE value from the URL. We do *n

Re: cvs commit: src/lib/libfetch fetch.3 ftp.c

2007-05-09 Thread Nate Lawson
Dag-Erling Smørgrav wrote: > Nate Lawson <[EMAIL PROTECTED]> writes: >> BTW, is there any reason to redo the "MODE S" and "TYPE I" commands >> after MDTM? > > MODE and TYPE commands are always issued before an MDTM (fetchStat*()) > or RETR (fetchGet*()), because the code assumes that they can vary

Re: cvs commit: src/lib/libfetch fetch.3 ftp.c

2007-05-08 Thread Dag-Erling Smørgrav
Nate Lawson <[EMAIL PROTECTED]> writes: > BTW, is there any reason to redo the "MODE S" and "TYPE I" commands > after MDTM? MODE and TYPE commands are always issued before an MDTM (fetchStat*()) or RETR (fetchGet*()), because the code assumes that they can vary from one call to another. This is t

Re: cvs commit: src/lib/libfetch fetch.3 ftp.c

2007-05-08 Thread Nate Lawson
Dag-Erling Smørgrav wrote: > [EMAIL PROTECTED] (Dag-Erling Smørgrav) writes: >> Still on-going. I forgot to add BATCH=YES to the make command line >> and had to start over. I expect it will take several more days to run >> to completion. > > Here are the results. I haven't had time to look thro

Re: cvs commit: src/lib/libfetch fetch.3 ftp.c

2007-05-08 Thread Dag-Erling Smørgrav
[EMAIL PROTECTED] (Dag-Erling Smørgrav) writes: > Still on-going. I forgot to add BATCH=YES to the make command line > and had to start over. I expect it will take several more days to run > to completion. Here are the results. I haven't had time to look through them yet, but judging from the p

Re: cvs commit: src/lib/libfetch fetch.3 ftp.c

2007-05-05 Thread Dag-Erling Smørgrav
Nate Lawson <[EMAIL PROTECTED]> writes: > Dag-Erling Smørgrav <[EMAIL PROTECTED]> writes: > > I'm going to set up a ports fetch run with FTP_COMBINE_CWDS enabled to > > see how it works out in the ugly, real world. If it works out well, I > > might replace the #ifdef with a command-line option or

Re: cvs commit: src/lib/libfetch fetch.3 ftp.c

2007-04-23 Thread Nate Lawson
Dag-Erling Smørgrav wrote: > Nate Lawson <[EMAIL PROTECTED]> writes: >> Log: >> Add back the original behavior of changing the entire directory path at >> once (CWD a/b/c vs. 3 CWDs). If an error occurs, we fall back to the >> default >> method of a single CWD per directory element. Sinc

Re: cvs commit: src/lib/libfetch fetch.3 ftp.c

2007-04-22 Thread Dag-Erling Smørgrav
Nate Lawson <[EMAIL PROTECTED]> writes: > Log: > Add back the original behavior of changing the entire directory path at > once (CWD a/b/c vs. 3 CWDs). If an error occurs, we fall back to the > default > method of a single CWD per directory element. Since this is technically > a violat

Re: cvs commit: src/lib/libfetch fetch.3 ftp.c

2007-04-22 Thread Nate Lawson
Nate Lawson wrote: > njl 2007-04-22 22:33:29 UTC > > FreeBSD src repository > > Modified files: > lib/libfetch fetch.3 ftp.c > Log: > Add back the original behavior of changing the entire directory path at > once (CWD a/b/c vs. 3 CWDs). If an error occurs, we fall

cvs commit: src/lib/libfetch fetch.3 ftp.c

2007-04-22 Thread Nate Lawson
njl 2007-04-22 22:33:29 UTC FreeBSD src repository Modified files: lib/libfetch fetch.3 ftp.c Log: Add back the original behavior of changing the entire directory path at once (CWD a/b/c vs. 3 CWDs). If an error occurs, we fall back to the default method of a sin

cvs commit: src/lib/libfetch fetch.3

2005-11-29 Thread Tom McLaughlin
tmclaugh2005-11-30 04:21:59 UTC FreeBSD src repository (ports committer) Modified files:(Branch: RELENG_5) lib/libfetch fetch.3 Log: MFC:Revision 1.60 Misspelling in Poul-Henning Kamp's email address under AUTHORS Approved by:ade Revision C

cvs commit: src/lib/libfetch fetch.3

2005-11-29 Thread Tom McLaughlin
tmclaugh2005-11-30 04:19:40 UTC FreeBSD src repository (ports committer) Modified files:(Branch: RELENG_6) lib/libfetch fetch.3 Log: MFC:Revision 1.60 Misspelling in Poul-Henning Kamp's email address under AUTHORS Approved by:ade Revision C

cvs commit: src/lib/libfetch fetch.3

2005-11-29 Thread Tom McLaughlin
tmclaugh2005-11-30 04:08:46 UTC FreeBSD src repository (ports committer) Modified files: lib/libfetch fetch.3 Log: Fix misspelling in Poul-Henning Kamp's email address under AUTHORS, from pkh@ to [EMAIL PROTECTED] Approved by:ade Revision ChangesPath