]> git.ipfire.org Git - thirdparty/HylaFAX.git/log
thirdparty/HylaFAX.git
17 years agoNLS update
Aidan Van Dyk [Mon, 15 Sep 2008 16:47:46 +0000 (16:47 +0000)] 
NLS update

17 years agoDebugging: cleanup and additions
Aidan Van Dyk [Mon, 8 Sep 2008 18:48:09 +0000 (18:48 +0000)] 
Debugging: cleanup and additions

This removes some old debug logging, and adds a -D <sleep time> to hfaxd.

Option sleeps after initial setup, allowing a debugger to attach before
actually processing client.

hfaxd: Remove extra debug syslogging

17 years agoconfigure: Modern C++ compilers no-longer support the STL <file.h> syntax
Aidan Van Dyk [Fri, 5 Sep 2008 20:57:09 +0000 (20:57 +0000)] 
configure:  Modern C++ compilers no-longer support the STL <file.h> syntax

It's been depricated for years now, and finally, support is being
removed from modern compilers.  I haven't found a compiler old enough
(even GCC 2.7 suppors the "new" syntax) to not support modern syntax.

17 years agodoc: Update RELEASENOTES for CVS
Aidan Van Dyk [Tue, 22 Jul 2008 19:14:19 +0000 (19:14 +0000)] 
doc: Update RELEASENOTES for CVS

17 years agodoc: Update RELEASENOTES for back branches
Aidan Van Dyk [Tue, 22 Jul 2008 18:54:33 +0000 (18:54 +0000)] 
doc: Update RELEASENOTES for back branches

17 years agofaxgetty/faxsend: Correct Include config file parsing
Aidan Van Dyk [Tue, 22 Jul 2008 17:55:40 +0000 (17:55 +0000)] 
faxgetty/faxsend: Correct Include config file parsing

If we're parsing a config file "Include" statement, we recursively call
readConfig, with the new file.  We need to make sure our "before/after
parsing" logic isn't defeated by this recursion.

17 years agoAllow for colons to appear in the value part of a qfile field
Patrice Fournier [Fri, 18 Jul 2008 19:23:48 +0000 (19:23 +0000)] 
Allow for colons to appear in the value part of a qfile field

17 years agonotify: have faxq *always* run notify
Patrice Fournier [Thu, 19 Jun 2008 20:42:17 +0000 (20:42 +0000)] 
notify: have faxq *always* run notify

From Aidan:

This allows the admin to use FaxNotify to directly manipulate when
notifications are sent.  The knowledge of when to send the notify emails
is now in notify.sh.in

17 years agoTMPDIR was not exported for external programs usage
Patrice Fournier [Fri, 13 Jun 2008 19:29:03 +0000 (19:29 +0000)] 
TMPDIR was not exported for external programs usage

17 years agoNow use hfExit and SetupPrivateTmp in util scripts
Patrice Fournier [Fri, 13 Jun 2008 19:06:28 +0000 (19:06 +0000)] 
Now use hfExit and SetupPrivateTmp in util scripts

Util scripts now source common-functions and use SetupPrivateTmp when
requiring a tmp dir and always exit with hfExit.

17 years agoRe-visited tmpdir creation
Patrice Fournier [Fri, 13 Jun 2008 19:04:59 +0000 (19:04 +0000)] 
Re-visited tmpdir creation

- Now create private tmpdir in $TMPDIR if it is set by the admin.
- Try to use mktemp first and fallback to directory named
  hylafaxtmp-$RANDOM-$RANDOM-$RANDOM-$$
- Use a private TMPDIR variable to make sure CleanupPrivateTmp only
  clean OUR tmpdir.

17 years agoUse traps to make sure cleanup is done on exit
Patrice Fournier [Fri, 13 Jun 2008 18:28:43 +0000 (18:28 +0000)] 
Use traps to make sure cleanup is done on exit

Admins adding exit calls in FaxDispatch or FaxNotify to stop processing
of the respective scripts won't be left with temporary files undeleted.

17 years agoUpdated POT template header
Patrice Fournier [Fri, 30 May 2008 22:32:43 +0000 (22:32 +0000)] 
Updated POT template header

- Added POT date and where to report bugs
- Added templates to be filled by the translator of a new PO file

17 years agocommon-functions: enhancements to parseQfile
Patrice Fournier [Fri, 30 May 2008 22:06:35 +0000 (22:06 +0000)] 
common-functions: enhancements to parseQfile

This enhances parseQfile to:
1) Allow a "prefix" that all variable names start with
2) Allow a specific file to be parsed instead of $QFILE
3) Exports *all* qfile variables

This built on what Lee's work started:
| commit e34950281e6245e19d03bf23ecc5db782f9ab4f2
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Mon Oct 8 16:47:44 2007 +0000
|
|     Small modification allows other parseQFile-using scripts to prevent variable
|     namespace collisions and "easy" exporting using:
|
|     source bin/common-functions
|     parseQfile "SOMETHING_"
|     export "${!SOMETHING_@}"

| commit 3e29292811f0ed0a9f1db6d3fd182c98ac0a725c
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Fri Oct 12 03:19:04 2007 +0000
|
|     Fix oops.

17 years agoSome grep and awk calls were not using the configured values of $GREP and $AWK
Patrice Fournier [Fri, 30 May 2008 18:58:35 +0000 (18:58 +0000)] 
Some grep and awk calls were not using the configured values of $GREP and $AWK

17 years agodebian: DIR_LOCALE needs to be set, not LOCALEDIR
Aidan Van Dyk [Thu, 8 May 2008 18:59:14 +0000 (18:59 +0000)] 
debian: DIR_LOCALE needs to be set, not LOCALEDIR

17 years agohfaxd: sorting fixups
Aidan Van Dyk [Thu, 8 May 2008 17:33:32 +0000 (17:33 +0000)] 
hfaxd: sorting fixups

I forgot the FILESORTFORMAT sorting last time.

And I don't know how I slipped mssing the Jprintf in Jobs.c++

17 years agoREADME: Remove unused hyphenation
Aidan Van Dyk [Wed, 7 May 2008 20:00:27 +0000 (20:00 +0000)] 
README: Remove unused hyphenation

17 years agoman page: fix up strange alignment in hfaxd.1m
Aidan Van Dyk [Fri, 2 May 2008 18:06:55 +0000 (18:06 +0000)] 
man page: fix up strange alignment in hfaxd.1m

17 years agohfaxd sorting based on the *SortFormat variables
Aidan Van Dyk [Fri, 2 May 2008 17:57:59 +0000 (17:57 +0000)] 
hfaxd sorting based on the *SortFormat variables

This introduces the following new protocol commands
        MDMSORTFMT <format>
        JOBSORTFMT <format>
        RCVSORTFMT <format>
        FILESORTFMT <format>

These commands take the exact same format specification as the related:
        MDMFMT <format>
        JOBFMT <format>
        RCVFMT <format>
        FILEFMT <format>

If one of them is set, when the server generates the listing for the
status, sendq/doneq, or recvq, it will sort the list using the specified
format before sending it to the client.  They all default to nothing,
which allows hfaxd to send output directly to the client with no sorting.
The administrator can change the default value for these in hfaxd.conf.

It also introduces a new hfaxd boolean config option:
        AllowSortFormat
This defaults to true to allow the client to specify the sorting
format, and the administrator can set it to false to not let the client
request a different sorting format.  This allows the adminstator to make
sure the server doesn't sort at all, or to only use the sorting set in
hfaxd.conf.

17 years agoAdd fxStackBuffer interface to Job/Modem/Recv *printf functions
Aidan Van Dyk [Fri, 2 May 2008 17:56:10 +0000 (17:56 +0000)] 
Add fxStackBuffer interface to Job/Modem/Recv *printf functions

This allows us to use the *printf functions to generate data for more than
just direct output to file descriptors.

17 years agopo: update po files
Aidan Van Dyk [Fri, 2 May 2008 17:21:08 +0000 (17:21 +0000)] 
po: update po files

Now po files have the path in them as well, not just the filename of where
the strings came from.

17 years agoMakefiles: SUBDIR support in defs.in
Aidan Van Dyk [Fri, 2 May 2008 17:13:30 +0000 (17:13 +0000)] 
Makefiles: SUBDIR support in defs.in

This reduces the ammount of duplication in the makefiles.  Currently every
(excpt for pkg) subdir makefile does:

        DEPTH=..
        SRCDIR=${DEPTH}/${TOPSRCDIR}/<name>
        VPATH=@DEPTH@/@SRCDIR@/<name>

This makes it just:
        DEPTH=..
        SUBDIR=<name>

SRCDIR and VPATH are handled by the included defs

17 years agoconfigure: Detect darwin for broken FD_ISSET
Aidan Van Dyk [Mon, 28 Apr 2008 20:37:52 +0000 (20:37 +0000)] 
configure: Detect darwin for broken FD_ISSET

Darwin seem to not like const fd_set args for FD_ISSET.

17 years agohfaxd: Fix RECV_START, RECV_PAGE, RECV_DOC triggers
Aidan Van Dyk [Mon, 28 Apr 2008 20:37:43 +0000 (20:37 +0000)] 
hfaxd: Fix RECV_START, RECV_PAGE, RECV_DOC triggers

The previous patch that fixed RECV_BEGIN, RECV_END and MODEM_CID triggers
actually brok RECV_START, RECV_PAGE, and RECV_DOC triggers.

This fixes the remaining 3, and combines the MODEM_ and RECV_ paths to
make it more clear what's going on.

This is the real fix for Lee's:
| commit 134af2ec4bbbc0be5e1c75f917b38c5236740a1c
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Sun Mar 16 08:08:21 2008 +0000
|
|   Another one of these... make sure that we stay within bounds.
|
|   It would seem that the FIFO messages were supposed to have some information
|   in them that is not there... for some reason.

17 years agohfaxd: Cleanup FIFO removal
Aidan Van Dyk [Mon, 28 Apr 2008 20:37:27 +0000 (20:37 +0000)] 
hfaxd: Cleanup FIFO removal

Based on:
| commit 18141065e3f30c95d4bb4b20d1a0d7a6414b2af4
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Mon Nov 26 21:59:09 2007 +0000
|
|     If the client was disconnecting when not being in the root directory then
|     the client's FIFO file was not getting deleted because it was looking
|     in the wrong place for it.

| commit de026506294a4b8fcb309d98d4485d8671f9cc19
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Tue Mar 11 17:44:11 2008 +0000
|
|     Reinhard Pfau submits some improvements to client FIFO cleanup.
|
|     Sourceforge tracker ID 1911860

17 years agofaxq: Be more careful in choosing jobs to batch
Aidan Van Dyk [Mon, 28 Apr 2008 20:37:20 +0000 (20:37 +0000)] 
faxq: Be more careful in choosing jobs to batch

Before we batch a job, we need to make sure it's "OK" to send, OK being:
- hasn't gover over it's maxdials/maxtries
- It fits in any maxpages restrictions
- JobControl hasn't "rejected" it
- JobControl hasn't "delayed" it with TOD restrictions

Noticed from work Lee did on the old scheduler:
| commit d79fed7a0c46142f9bf9fa1c87f521d40eeb6b27
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Thu Mar 13 18:28:43 2008 +0000
|
|     TimeOfDay needs to be considered before batching.

17 years agoCallID: Initialize answerlength on depricated CIDName/CIDNumber config options
Aidan Van Dyk [Mon, 28 Apr 2008 20:37:12 +0000 (20:37 +0000)] 
CallID: Initialize answerlength on depricated CIDName/CIDNumber config options

From Lee:
| commit 78e09ec251ea78608a2acab5d2476c0f95768fb3
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Thu Feb 7 21:09:48 2008 +0000
|
|     The CallID types are not being initialized when CIDName or CIDNumber are
|     being used in lieu of CallIDPattern.

17 years agoAdd "Restart" enhancement to init script
Aidan Van Dyk [Mon, 28 Apr 2008 20:37:05 +0000 (20:37 +0000)] 
Add "Restart" enhancement to init script

From Michel Petrazzo through Lee:
| commit bd45320fb137a813ae50b790f596fd3903e209fd
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Thu Feb 7 20:15:30 2008 +0000
|
|     Michele Petrazzo offers a "restart" enhancement to the init script.

| commit 269aa1dd4a40a48b4fb2b0918c41792ad1c939e6
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Mon Feb 11 18:53:33 2008 +0000
|
|     oops

17 years agoClass1: Be more lenient on handling NO CARRIER response
Aidan Van Dyk [Mon, 28 Apr 2008 20:36:59 +0000 (20:36 +0000)] 
Class1: Be more lenient on handling NO CARRIER response

From Lee:
| commit 1779a3ff4cf0e68f5eb67cbe54b8174e761d29fc
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Wed Jan 16 19:03:42 2008 +0000
|
|     The meaning of a direct NO CARRIER response/result to the AT+FRH=3
|     command can be interpreted two ways due to wording in ITU T.31.
|     The more traditional interpretation is that it is out-of-spec: that
|     a NO CARRIER response should only come after a CONNECT message.
|     However, T.31 gives sample sessions that illustrate precisely this
|     kind of out-of-spec sequence:
|
|      <-- AT+FRH=3
|      --> NO CARRIER
|
|     And the interpretation is clear to indicate that in this case the
|     modem detected a carrier, but it apparently was not V.21 HDLC, and
|     the NO CARRIER response indicates that the indeterminate carrier
|     has dropped.  This functionality is more cumbersome than it is
|     helpful (its value is not much different than that of an AT+FRS=1
|     command), and what it really means is that we just need to re-issue
|     the AT+FRH=3 command until we get something useful as this scenario
|     will most likely occur on incidence of Phase C premature carrier loss.

17 years agoconfigure: HP-UX fixes
Aidan Van Dyk [Mon, 28 Apr 2008 20:36:48 +0000 (20:36 +0000)] 
configure: HP-UX fixes

From Lee:
| commit 1503c6f60ebf22eaec61b751592046c38d82f288
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Fri Feb 1 01:57:21 2008 +0000
|
|     Bernd Krumböck helps get HP-UX socklen_t building tidied up.

17 years agofaxcron: Be more explicit in last 7 days
Aidan Van Dyk [Mon, 28 Apr 2008 20:36:35 +0000 (20:36 +0000)] 
faxcron: Be more explicit in last 7 days

From Lee:;
| commit 24d2150b1f467e2fbdf12fe9d487cb5d647323c6
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Wed Dec 26 18:24:27 2007 +0000
|
|     Max Mothwurf complained that "last week" means a Sun-Sat or Mon-Sun duration,
|     so to be clear that faxcron means "the last seven days" say it that way.

17 years agoPollLockPokeModem: check on modem when lock file is polled
Aidan Van Dyk [Mon, 28 Apr 2008 20:36:28 +0000 (20:36 +0000)] 
PollLockPokeModem: check on modem when lock file is polled

From Lee:
| commit 78114b9300badd714f44f8b25a919b09605f3ac6
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Thu Dec 6 18:19:28 2007 +0000
|
|     Adds PollLockPokeModem
|
|     When polling for the presence/removal of a UUCP lockfile, also test to make
|     sure the modem is still reachable.  This is useful for scenarios where the
|     modem is apt to become disconnected while the modem is idle, such as an
|     external modem that may lose power, where the administrator wants notification
|     of this (as a ‘‘wedged’’ condition).

17 years agoman pages: Small fixups
Aidan Van Dyk [Mon, 28 Apr 2008 20:36:14 +0000 (20:36 +0000)] 
man pages: Small fixups

17 years agoTime formats: Make them concistantly *always* HH:MM:SS
Aidan Van Dyk [Mon, 28 Apr 2008 20:36:05 +0000 (20:36 +0000)] 
Time formats: Make them concistantly *always* HH:MM:SS

From Lee:
| commit 640974099533468fbbc4844d9f0f0c88f6310505
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Thu Nov 15 21:31:28 2007 +0000
|
|     This causes the fmtTime() time format to always include hours.
|
|     Logs like this:
|
|     JOB 1086 (sleeping dest +1234 pri 126 tts 4:58 killtime 4:58:00): SLEEP FOR 4:58
|
|     Just aren't that intuitive with respect to the "4:58".  To some readers that
|     may normally mean 4 minutes and 58 seconds... but to others it may look like
|     four hours and 58 minutes.
|
|     One such "reader" is MySQL where "4:58" will be interpreted as nearly five hours
|     instead of nearly five minutes.  In order to communicate the proper time one must
|     use "0:04:58".
|
|     So to interface things one constantly needs to parse the fmtTime()
|     output to add the hours (if not present).  Let's save everyone a headache and
|     simply always include hours to the fmtTime() time format.
|
|     JOB 1086 (sleeping dest +1234 pri 126 tts 0:04:58 killtime 4:58:00): SLEEP FOR 0:04:58
|
|     That's a lot more clear to understand.

| commit 201b91d031bd987c453f6452a7bb6cc45d17638b
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Tue Dec 11 20:16:50 2007 +0000
|
|     Times in excess of 99:59:59 were getting formatted as ??:??:??.  This
|     will give us the full number of hours, up to a year's time.

17 years agofixup: hfaxd old protocol - man pages
Aidan Van Dyk [Mon, 28 Apr 2008 20:35:59 +0000 (20:35 +0000)] 
fixup: hfaxd old protocol - man pages

17 years agofixup: Remove old protocol - don't remove formats
Aidan Van Dyk [Mon, 28 Apr 2008 20:35:49 +0000 (20:35 +0000)] 
fixup: Remove old protocol - don't remove formats

17 years agohfaxd: Remove old protocol support
Aidan Van Dyk [Mon, 28 Apr 2008 20:35:39 +0000 (20:35 +0000)] 
hfaxd: Remove old protocol support

Based on orignal work from Lee:
| commit 32525756162d43e98d0df8761c94f9ba262ad37a
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Thu Jan 31 02:26:35 2008 +0000
|
|     Remove (unused) old protocol support.

| commit d8f9b19302fd662aba1b554778644e3c6cd2d746
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Thu Jan 31 02:27:22 2008 +0000
|
|     Remove old protocol support.

17 years agofaxsend/faxgetty: SaveRawImage
Aidan Van Dyk [Mon, 28 Apr 2008 20:35:14 +0000 (20:35 +0000)] 
faxsend/faxgetty: SaveRawImage

Internal debugging config option from Lee:
| commit fd5fde78af93d32d3b0e8ebbd832da2b32214c4a
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Tue Oct 9 22:35:49 2007 +0000
|
|     This adds a new configuration option, SaveRawImage.  Which basically instructs
|     faxsend and faxgetty to save received or sent image data to file in /tmp/out.fax
|     or /tmp/in.fax, respectively.
|
|     As this is not something that will have a useful purpose as-is to the general
|     user I have left it undocumented... although that's probably very bad of me.

17 years agofixup: hfaxd PCL
Aidan Van Dyk [Mon, 28 Apr 2008 20:35:03 +0000 (20:35 +0000)] 
fixup: hfaxd PCL

17 years agoComplete PCL support in the server.
Aidan Van Dyk [Mon, 28 Apr 2008 20:34:54 +0000 (20:34 +0000)] 
Complete PCL support in the server.

Based on work from Lee:
| commit 36c27b71defb2aec5d6597cca37c3ed03920e0b2
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Fri Jan 4 23:24:31 2008 +0000
|
|     This should complete the server-side PCL support implementation.
|
|     It also intentionally makes hfaxd reject PJL documents... which seem
|     to give some WHFC users fits when they use the wrong printer driver.

| commit ace9e11f2f8054e6de05aded856342216cbc7940
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Fri Jan 25 20:44:41 2008 +0000
|
|     This adopts the pcl2fax scripting used by Philip Gwyn.

| commit 9fe7c0ee4f77cbadce2926e8fcb2a07c6012a1d6
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Sat Jan 26 19:18:26 2008 +0000
|
|     In order to support PCL best on the server-side we need to begin to trust
|     the document type specification that the client supplied during the protocol
|     exchange (FORM).  This is because there is no sure-fire way to type a PCL
|     file.  So we need to trust whatever the client specified.
|
|     However, we can't trust it in the case of Postscript because that format is,
|     unfortunately, the default server setting, and thus clients have traditionally
|     been able to get away without specifying the document format when using
|     Postscript... and thus even PDF or TIFF documents could have been submitted
|     in this fashion with little to no fallout.

| commit 56a957ed2247864d46b89e81451b551ae498225f
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Sat Jan 26 19:21:17 2008 +0000
|
|     Support ESC-E detection for PCL.

| commit 3f3e69ed499e905247331fc26c072711d4ddc0b5
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Sat Jan 26 19:31:55 2008 +0000
|
|     ESC-* is also PCL

17 years agofixup: man/xferfaxlog merge
Aidan Van Dyk [Mon, 28 Apr 2008 20:34:40 +0000 (20:34 +0000)] 
fixup: man/xferfaxlog merge

17 years agoxferfaxlog: Add recvq fax filename to RECV records
Aidan Van Dyk [Mon, 28 Apr 2008 20:34:34 +0000 (20:34 +0000)] 
xferfaxlog: Add recvq fax filename to RECV records

Based on work from Lee:
| commit b1d0cf778a0435c5b2f6af7d16ae10109b8d215c
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Mon Dec 24 18:29:25 2007 +0000
|
|     Adds qfile to RECV xferfaxlog records.

17 years agoman: Fix NotifyCmd reference
Aidan Van Dyk [Mon, 28 Apr 2008 20:34:27 +0000 (20:34 +0000)] 
man: Fix NotifyCmd reference

From Lee:
| commit 7442d5178e6c1f38594ad31dca0ac97cb92f0356
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Fri Dec 14 19:21:51 2007 +0000
|
|     https://sourceforge.net/forum/message.php?msg_id=4675923
|
|     points out that NotifyCmd is misdocumented as being available in modem config files

17 years agoDon't write info file unless it has a name
Aidan Van Dyk [Mon, 28 Apr 2008 20:34:15 +0000 (20:34 +0000)] 
Don't write info file unless it has a name

From Lee:
| commit e8b509f041eae9af7e8ee022fc1cc9e2adcb88fe
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Tue Nov 6 04:32:35 2007 +0000
|
|     Aha, the destructor attempts to save the values to disk even if the file
|     is obviously invalidly named.

17 years agofaxgetty/faxsend: Remove extra forks
Aidan Van Dyk [Mon, 28 Apr 2008 20:34:09 +0000 (20:34 +0000)] 
faxgetty/faxsend: Remove extra forks

Removes all the fork() calls around the notify*()

This reverts commit part of a8330595137f04a70f6dc9310ffa7cb5acf20be9.
This reverts commit part of d3dc8a48d5463faa980742b028a5bbe5b4b6730c.

17 years agofaxq: Clieanup timeout accounting
Aidan Van Dyk [Mon, 28 Apr 2008 20:34:02 +0000 (20:34 +0000)] 
faxq: Clieanup timeout accounting

Spurred on by:
| commit 0225ec9ccf5fd73cb559d4519d62b6d9842388d0
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Thu Oct 18 21:10:12 2007 +0000
|
|     This extends the UNSENT accounting... and it absorbs the KILLED accounting.

17 years agolibhylafax: Don't tryncate commands at 100 characters at 100 characters
Aidan Van Dyk [Mon, 28 Apr 2008 20:33:55 +0000 (20:33 +0000)] 
libhylafax: Don't tryncate commands at 100 characters at 100 characters

Spurred on by:
| commit dc265ed79edcf6424d44d45e1d3f6ce78a50a77e
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Wed Oct 10 22:50:35 2007 +0000
|
|     Let's not truncate job params at 100 characters.  So let's make the buffer
|     bigger, but let's not send the truncated string to the server.

17 years agoClass1: prevent infinte looping in recvFrame
Aidan Van Dyk [Mon, 28 Apr 2008 20:33:49 +0000 (20:33 +0000)] 
Class1: prevent infinte looping in recvFrame

Based on Lee's commits:
| commit 3f335853e3c7610b0d9b8228988940b672a07ba4
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Fri Oct 5 01:21:51 2007 +0000
|
|     Looks like there is potential for infinite looping in recvFrame if the
|     modem behavior is just right.  We simply set up a timer to prevent it.

| commit 64c904b3e08e582177813746e018aae5cf1080d6
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Thu Oct 18 19:09:05 2007 +0000
|
|     t1Timer is in ms ... let's compare seconds with seconds not seconds with ms

17 years agofaxcover: Add from-mail-address feature
Aidan Van Dyk [Mon, 28 Apr 2008 20:33:38 +0000 (20:33 +0000)] 
faxcover: Add from-mail-address feature

Based on Lee's commit:
| commit 967221f223fa762a060f384f2c96145c57a2ddd9
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Tue Oct 23 17:45:13 2007 +0000
|
|     Riccardo Magliocchetti contributes "from-mail-address" feature for faxcover.

17 years agofixup: faxq: STDERR/STDOUT - runConverter()
Aidan Van Dyk [Mon, 28 Apr 2008 20:33:26 +0000 (20:33 +0000)] 
fixup: faxq: STDERR/STDOUT - runConverter()

17 years agofaxq: Don't take STDERR as STDOUT
Aidan Van Dyk [Mon, 28 Apr 2008 20:33:12 +0000 (20:33 +0000)] 
faxq: Don't take STDERR as STDOUT

Based on work from Lee:
| commit 71efb2e351dbfb5a052966f53c0a55e7f682280d
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Wed Nov 28 06:05:25 2007 +0000
|
|     Meaningful JobControl output shouldn't be coming in from stderr.  Thus, our
|     use of stderr only is apt to cause problems.  Ignore stderr.

17 years agofaxaddmodem: Prefer Class 1.0
Aidan Van Dyk [Mon, 28 Apr 2008 20:33:04 +0000 (20:33 +0000)] 
faxaddmodem: Prefer Class 1.0

| commit d7cc4704e5c9188daccdc9eed3d9dfcd558a87bd
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Thu Oct 11 16:13:15 2007 +0000
|
|     Prefer Class 1.0 over 2 and 2.0

17 years agoman pages: Fix up incorrect docs
Aidan Van Dyk [Mon, 28 Apr 2008 20:32:56 +0000 (20:32 +0000)] 
man pages: Fix up incorrect docs

Based on work from Lee:
| commit 83ee24062883e6fed2bafdf502fc45903d79fb77
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Thu Feb 28 18:37:15 2008 +0000
|
|     %%c and %%C were backwards

17 years agofixup: tiffcheck black-on-white
Aidan Van Dyk [Mon, 28 Apr 2008 20:32:43 +0000 (20:32 +0000)] 
fixup: tiffcheck black-on-white

17 years agotiffcheck: Check for black-on-white
Aidan Van Dyk [Mon, 28 Apr 2008 20:32:35 +0000 (20:32 +0000)] 
tiffcheck: Check for black-on-white

| commit b4bb07f4aef5e038594692025143356b5a9056d0
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Wed Jan 2 20:38:34 2008 +0000
|
|     It would be an extremely rare possibility up until now (since apparently no
|     commonly-used TIFF creators make a file that passes tiffcheck but also
|     produces the file with min-is-black photometrics), but this makes tiffcheck
|     also verify that photometric interpretation is min-is-white... otherwise
|     a negative image gets faxed.

17 years agofixup: faxmail error discard
Aidan Van Dyk [Mon, 28 Apr 2008 20:32:29 +0000 (20:32 +0000)] 
fixup: faxmail error discard

17 years agofaxmail: print an error message when a part is discarded
Aidan Van Dyk [Mon, 28 Apr 2008 20:32:23 +0000 (20:32 +0000)] 
faxmail: print an error message when a part is discarded

| commit 8e9d78dcafdde09455b42c55f9f5bfa4f762d355
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Tue Oct 2 00:32:35 2007 +0000
|
|     output an error message if a part is discarded

17 years agoClass1: Be more lax in our timeouts wrating for Phase C
Aidan Van Dyk [Mon, 28 Apr 2008 20:32:15 +0000 (20:32 +0000)] 
Class1: Be more lax in our timeouts wrating for Phase C

| commit 2b6aa6612adb835a84fe012d7bb31a40f9efe2b0
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Wed Feb 6 01:28:57 2008 +0000
|
|     Although spec gives the sender T2 (7 seconds) to start sending Phase C image
|     data... recordings show us that some may wait even as much as 9.5 seconds.
|     In cases where Class1RMPersistence is non-zero (where +FCERROR works or AT+FAR=1)
|     it is acceptable to extend this timeout inversely as it is shortened for
|     cases where Class1RMPsersistence is zero.

17 years agoProtocol Clients: Initialize FORM to UNKNOWN
Aidan Van Dyk [Mon, 28 Apr 2008 20:32:00 +0000 (20:32 +0000)] 
Protocol Clients: Initialize FORM to UNKNOWN

| commit f8a05cb8865b4657f52d011e21836c9dda7e0ecf
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Fri Jan 25 21:38:43 2008 +0000
|
|     Deliberately intialize in an unknown format so that the first document
|     submitted will cause the FORM hfaxd parameter to be used... rather than
|     expecting the server's FORM default to be the same as the client's.

17 years agoAdd support for VCON in waitfor escape
Aidan Van Dyk [Mon, 28 Apr 2008 20:31:49 +0000 (20:31 +0000)] 
Add support for VCON in waitfor escape

| commit 6adf5fb4ade96622363068a5df8d5443bcb1d874
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Fri Feb 1 02:52:39 2008 +0000
|
|     Adds support for <waitfor:VCON> escape.

17 years agoAdd Class 1.0 for Multitech 5656ZDX
Aidan Van Dyk [Mon, 28 Apr 2008 20:31:38 +0000 (20:31 +0000)] 
Add Class 1.0 for Multitech 5656ZDX

| commit 2d8b541b88fec449a3cf9f0f0f2fa47febf6c066
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Mon Jan 14 06:00:39 2008 +0000
|
|     Adds Class 1.0 for the MT5656ZDX

17 years agoClass1: Initialize blockNumber
Aidan Van Dyk [Mon, 28 Apr 2008 20:31:13 +0000 (20:31 +0000)] 
Class1: Initialize blockNumber

| commit d1d12bd76d5f617e014b3f036985e44a56bf2037
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Sat Jan 5 21:43:28 2008 +0000
|
|     blockNumber was not properly initialized... thus when the compiler wasn't
|     initializing it for us, we'd have problems.

17 years agoClass1: Add another needed pause
Aidan Van Dyk [Mon, 28 Apr 2008 20:30:26 +0000 (20:30 +0000)] 
Class1: Add another needed pause

| commit 9942928cfd6672296234cc4e490d7fe691d03a62
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Mon Oct 1 18:39:42 2007 +0000
|
|     requisite pause

17 years agoClass1: Don't train if we are seeing adequate progress
Aidan Van Dyk [Mon, 28 Apr 2008 20:30:05 +0000 (20:30 +0000)] 
Class1: Don't train if we are seeing adequate progress

From Lee:
| commit fc0e1fa89d2566d10c95e6a816922b537ef193fb
| Author: Lee Howard <faxguy@howardsilvan.com>
| Date:   Sun Sep 9 15:32:30 2007 +0000
|
|     Training down can often complicate the matter.  Don't do it if we're seeing
|     adequate progress.

17 years agoNLS: Some fixups for when NLS is disabled
Aidan Van Dyk [Mon, 28 Apr 2008 20:29:56 +0000 (20:29 +0000)] 
NLS:  Some fixups for when NLS is disabled

18 years agoAdd Vietnamese PO translation
Aidan Van Dyk [Thu, 20 Mar 2008 18:52:36 +0000 (18:52 +0000)] 
Add Vietnamese PO translation

18 years agoconfigure: Be more careful about replacing generated files
Aidan Van Dyk [Mon, 10 Mar 2008 14:43:19 +0000 (14:43 +0000)] 
configure: Be more careful about replacing generated files

configure is carefull to not change "generated" files if the only change
is the date.  This makes it a bit more generous about date lines.

18 years agoi18n: Updated cvsignore for nls generated files
Patrice Fournier [Fri, 29 Feb 2008 22:05:50 +0000 (22:05 +0000)] 
i18n: Updated cvsignore for nls generated files

18 years agoi18n: Regenerated pot templates to include the previously missing strings
Patrice Fournier [Fri, 29 Feb 2008 22:02:38 +0000 (22:02 +0000)] 
i18n: Regenerated pot templates to include the previously missing strings

18 years agoi18n: Missing translatable strings from .pot files
Patrice Fournier [Fri, 29 Feb 2008 22:01:49 +0000 (22:01 +0000)] 
i18n: Missing translatable strings from .pot files

N_() translated strings were not picked up by make update-po and some
source files were not checked at all for translatable strings.

18 years agoi18n: fixed typos in template header
Patrice Fournier [Fri, 29 Feb 2008 21:55:16 +0000 (21:55 +0000)] 
i18n: fixed typos in template header

18 years agoi18n: Missed some changes of _() to NLS::TEXT()
Patrice Fournier [Fri, 29 Feb 2008 21:51:51 +0000 (21:51 +0000)] 
i18n: Missed some changes of _() to NLS::TEXT()

Only the first occurence of _() was changed on a line... 4 lines had two
which have now been changed also.

18 years agoDebian: Add rules to include i18n support
Aidan Van Dyk [Fri, 29 Feb 2008 19:03:40 +0000 (19:03 +0000)] 
Debian: Add rules to include i18n support

18 years agoi18n: Install the language specifec catalog files
Aidan Van Dyk [Thu, 28 Feb 2008 22:09:58 +0000 (22:09 +0000)] 
i18n: Install the language specifec catalog files

18 years agoconfigure: LIBINTL to find libraries needed for gettext
Aidan Van Dyk [Thu, 28 Feb 2008 20:57:07 +0000 (20:57 +0000)] 
configure: LIBINTL to find libraries needed for gettext

18 years agoi18n: Add NL and ES translations
Aidan Van Dyk [Thu, 28 Feb 2008 20:56:57 +0000 (20:56 +0000)] 
i18n: Add NL and ES translations

These come from the work HylaFAX work at translationproject.org.

18 years agoi18n: Add built .pot infrastructure
Aidan Van Dyk [Thu, 28 Feb 2008 20:56:02 +0000 (20:56 +0000)] 
i18n: Add built .pot infrastructure

The 3 catalogs, and an updated master hylafax.pot

18 years agoi18n: Build infrastructure
Aidan Van Dyk [Thu, 28 Feb 2008 20:55:09 +0000 (20:55 +0000)] 
i18n: Build infrastructure

This allows us to "build" .po and .pot files for HylaFAX.

It builds 3 sets:
        libhylafax              Strings used in libhylafax
        hylafax-client          Strings used in hylafax client components
        hylafax-server          Strings used in hylafax server components

All are merged to one master "hylafax.pot" for translating.  The 3
separate catalogs translations are merged out of the master
hylafax.pot/$LANG.po to build the .mo for installation.

Makefile hackery makes me which we only supported "real" makes...

18 years agoi18n: Change NLS interface a bit - libhylafax *is* a C++ library
Aidan Van Dyk [Thu, 28 Feb 2008 20:49:25 +0000 (20:49 +0000)] 
i18n: Change NLS interface a bit - libhylafax *is* a C++ library

We also want libhylafax to be able to use it's own domain, so
it needs to have a way to get it's gettext strings with dgettext.

18 years agoconfigure: Fix an always true typo
Aidan Van Dyk [Thu, 28 Feb 2008 02:36:26 +0000 (02:36 +0000)] 
configure: Fix an always true typo

18 years agoSCO Fixups
Aidan Van Dyk [Wed, 27 Feb 2008 22:30:47 +0000 (22:30 +0000)] 
SCO Fixups

18 years agohfaxd: Compile fixup
Aidan Van Dyk [Fri, 22 Feb 2008 20:26:43 +0000 (20:26 +0000)] 
hfaxd: Compile fixup

Ooops - missing ;

18 years agoProtocol: add SITE LOCKWAIT command
Aidan Van Dyk [Fri, 22 Feb 2008 20:24:23 +0000 (20:24 +0000)] 
Protocol: add SITE LOCKWAIT command

This enhancement allows the client to specify how long
to block when flock() tries to lock a file which is already
locked.  It defaults to 0 for backwards compatibility.

18 years agohfaxd: Don't update job's last modified time unless it's actually successfull
Aidan Van Dyk [Fri, 22 Feb 2008 20:23:32 +0000 (20:23 +0000)] 
hfaxd: Don't update job's last modified time unless it's actually successfull

If the qfile can't be locked the updateJobFromDisk() will fail. In that
case, *don't* change the job's last modified time, so that next time,
it will be tried again.

18 years agohfaxd lockJob() - don't blindly O_CREAT job we're trying to lock
Aidan Van Dyk [Fri, 22 Feb 2008 20:23:08 +0000 (20:23 +0000)] 
hfaxd lockJob() - don't blindly O_CREAT job we're trying to lock

It happens that in some cases, we are only tryhing to lock a job to
updateJobFromDisk(), and this can happen *before* the qfile was renamed (to the
doneq), but after it was stated (successfully).  Opening wiht O_CREAT here
means that we end up "creating" these 0 length files when we shouldn't have
been able to lock it.

So we carefully create the file if we might need it, the lockJob can only
open it O_RDWR

18 years agoRemove DestControls cruft. OK Patrice.
Tim Rice [Sun, 3 Feb 2008 01:12:23 +0000 (01:12 +0000)] 
Remove DestControls cruft.  OK Patrice.

18 years agofallout from libhylafax reorg
Tim Rice [Thu, 31 Jan 2008 23:44:14 +0000 (23:44 +0000)] 
fallout from libhylafax reorg

18 years ago man/callid.4f has been gone almost 2 years
Tim Rice [Thu, 31 Jan 2008 22:54:54 +0000 (22:54 +0000)] 
man/callid.4f has been gone almost 2 years

18 years agoAdd missing modem config entries
Tim Rice [Wed, 30 Jan 2008 06:15:09 +0000 (06:15 +0000)] 
Add missing modem config entries

18 years agoAFM fonts were removed a long time ago
Tim Rice [Tue, 29 Jan 2008 04:31:43 +0000 (04:31 +0000)] 
AFM fonts were removed a long time ago

18 years agoQuiet compiler warnings like
Tim Rice [Thu, 24 Jan 2008 05:18:27 +0000 (05:18 +0000)] 
Quiet compiler warnings like
"../src/faxd/ModemConfig.h", line 178: warning: extra ";" ignored
"../src/faxd/ModemConfig.h", line 179: warning: extra ";" ignored
"../src/faxd/Class1.h", line 159: Warning: Identifier expected instead of "}".
"../src/faxd/FaxSendStatus.h", line 40: Warning: Identifier expected instead of "}".

18 years agoBug 883: faxmail error with newline in headers
Aidan Van Dyk [Thu, 24 Jan 2008 02:05:50 +0000 (02:05 +0000)] 
Bug 883: faxmail error with newline in headers

If it's a continuation header, we shouldn't be putting an extra \n in it.

18 years agoSolaris: Needs some support for new code
Aidan Van Dyk [Wed, 23 Jan 2008 19:55:26 +0000 (19:55 +0000)] 
Solaris:  Needs some support for new code

inet_aton used in PASV transfer
ngettext needed for NLS

18 years agoSolaris DSO: fixups after the DSO work
Aidan Van Dyk [Wed, 23 Jan 2008 19:55:10 +0000 (19:55 +0000)] 
Solaris DSO: fixups after the DSO work

If only I had access to IRIX...

18 years agoSolaris/SunWSPro: Don't use empty array definitions
Aidan Van Dyk [Wed, 23 Jan 2008 19:55:00 +0000 (19:55 +0000)] 
Solaris/SunWSPro:  Don't use empty array definitions

SunWSPro doesn't like empty array definitions:

> "../src/faxd/faxQueueApp.c++", line 3396: Error: The type "faxQueueApp::booltag[]" is incomplete.

18 years agoFaxClient: Quote any " in the string before passing it to server
Aidan Van Dyk [Tue, 22 Jan 2008 21:23:30 +0000 (21:23 +0000)] 
FaxClient: Quote any " in the string before passing it to server

When issuing the command:
        JPARM <param> "<string value>"
We should be quoting any of the " in the string value, or it get's
interpreted strangely by the server.

18 years agofaxmail: Use X-Fax-To for coverpage name
Aidan Van Dyk [Tue, 22 Jan 2008 19:59:51 +0000 (19:59 +0000)] 
faxmail: Use X-Fax-To for coverpage name

18 years agofaxmail: Remove double-quotes, etc
Aidan Van Dyk [Tue, 22 Jan 2008 19:59:29 +0000 (19:59 +0000)] 
faxmail: Remove double-quotes, etc

When the "to" field is used as the coverpage page name, try and remove any
of the "seperator" stuff that can surround them, like quotes, brackets,
etc.

18 years agoFix faxq modem Triggers
Aidan Van Dyk [Wed, 16 Jan 2008 15:44:03 +0000 (15:44 +0000)] 
Fix faxq modem Triggers

If "extra information" was included in the Trigger::post call, the actual
modem information wasn't been included in the trigger.  It should always
be included in the trigger.