| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 90.1 | A guess | STAR::BUTTLAR | Dick Buttlar | Fri Feb 24 1989 08:42 | 17 |
| ONLINE_CLEANUP.COM is a command procedure that creates a VAX BLISS image.
If you do not have VAX BLISS installed on your system the BLISS DCL command
may be giving you the 'unrecognized command verb' message -- and, of course,
the BLISS source isn't compiled, there's no .OBJ, and -- hence --
no ONLINE-CLEANUP.EXE.
Is it worthwhile installing BLISS to run ONLINE_CLEANUP? Probably not.
ONLINE_CLEANUP is *great* for inserting symbol-names in heads that do not
already have them. Most of the other stuff you can do pretty easily by hand.
There is a slower completely DCL variant of ONLINE_CLEANUP.COM that is described
in Note 17. Your syntax line is correct -- for this version too.
When you use it, though, be sure to keep a copy of your original files
somewhere.
-Dick
|
| 90.2 | incomplete directions were given | CASEY::BURACK | This train is bound for glory | Fri Feb 24 1989 11:32 | 6 |
| Thanks,
I had one of the DSM engineers compile the bliss code and he produced
the .exe file for me. I will give it a try.
Ruth-Ellen
|
| 90.3 | Can't get the program to work???? | CASEY::BURACK | This train is bound for glory | Mon Feb 27 1989 14:03 | 51 |
| Hi,
When I tried to run ONLINE_CLEANUP.COM I got various results depending on the
command line I used. None of the ways I tried worked. I am not running BLISS
on my system, but as I said in the previous reply, I have the .EXE and the
original program and the .COM file to run this.
Some of the problems follow with the log file information.
Ruth-Ellen
++++++++++++++++++++++++++++++++++++++++++++++++++++
Try Number 5:
$@ONLINE_CLEANUP VAXDSM_051_RNMAIN [BURACK.CLEANUP]
I ran the profile and it said it completed....
Cleanup completed ...
But I ended up with empty files.
..........
$@ONLINE_CLEANUP VAXDSM_051_RNMAIN [BURACK.CLEANUP.scratch]
I ran this using another subdirectory and got the following message, but no
files anywhere.
Processing file "ZORBA$DUB5:[BURACK.CLEANUP]VAXDSM_051_RNFRONT.SDML; "
Completed processing element file "ZORBA$DUB5:[BURACK.CLEANUP]VAXDSM_051_RNFRONT.SDML; "
Processing file "ZORBA$DUB5:[BURACK.CLEANUP]VAXDSM_051_RNPREF.SDML;; "
Completed processing element file "ZORBA$DUB5:[BURACK.CLEANUP]VAXDSM_051_RNPREF.SDML;; "
Processing file "ZORBA$DUB5:[BURACK.CLEANUP]VAXDSM_051_RNCH1.SDML;;; "
Completed processing element file "ZORBA$DUB5:[BURACK.CLEANUP]VAXDSM_051_RNCH1.SDML;;; "
Processing file "ZORBA$DUB5:[BURACK.CLEANUP]VAXDSM_051_RNCH2.SDML;;; "
Completed processing element file "ZORBA$DUB5:[BURACK.CLEANUP]VAXDSM_051_RNCH2.SDML;;; "
Cleanup completed ...
........
I didn't run this with a profile, but with just one chapter and got the
following errors:
ERROR - PROFILE not defined correctly - rerun procedure
%RMS-FNF, file not found
%TRACE-TRACEBACK, symbolic stack dump follows
module name routine name line rel PC abs PC
ONLINE_CLEANUP PROFILE_FILE_READ 300 0000004E 00001E67
ONLINE_CLEANUP MAIN 245 00000007 00001E07
|
| 90.4 | See if this works | STAR::BUTTLAR | Dick Buttlar | Tue Feb 28 1989 09:05 | 12 |
| Try copying your profile and book SDML files to the system that is
running BLISS, and then issuing the @ONLINE_CLEANUP ...
ONLINE_CLEANUP.COM creates ONLINE-CLEANUP.EXE and may actually pass
the DCL parameters to it. (I'd have to take a look.) In other words,
having ONLINE-CLEANUP.EXE about while you @ONLINE_CLEANUP ... may
not be enough.
I'll see if I can dig up some more information.
-Dick
|
| 90.5 | Not the answer . . . | PROSE::CLOUTIER | Techno-weenies to the back of the bus. | Tue Feb 28 1989 10:02 | 22 |
|
> Try copying your profile and book SDML files to the system that is
> running BLISS, and then issuing the @ONLINE_CLEANUP ...
Dick, I've tried that myself and it doesn't seem to have any effect.
I copied all the files to TLE and run everything from there. The
same things that happened to Ruth Ellen happened to me. ONLINE_CLEANUP
deleted the contents of all the chapter files and left only the
first 4 lines of the FRONT_MATTER file.
Maybe we should rename the file ONLINE_CLEANOUT. :-)
If I run ONLINE_CLEANUP on a chapter file only, it won't process
the file (or if it does, it refuses to add symbolic names to header
tags).
Short of going through the file to add my own symbolic names, I'm
at a loss as to what to do.
--Leo
|
| 90.6 | A more recent version | STAR::BUTTLAR | Dick Buttlar | Tue Feb 28 1989 17:02 | 20 |
| Ruth Ellen and Leo-
I just copied a new version of ONLINE_CLEANUP.COM to BULOVA. This may or may
not fix the problems.
One thing that ONLINE_CLEANUP handles terribly is a head tag whose arguments
span more than one line. For instance:
<head1>(Creating, Displaying, and Correcting Documents\
create_display_sec)
causes ONLINE_CLEANUP to delete the rest of the file!
But the makers of ONLINE_CLEANUP never claimed that they supplied anything
but a potentially helpful tool for converting certain sorts of SDML
source files and reporting coding problems. In other words, if certain
problems are too time-consuming to fix, they won't get fixed.
After all, there are books out there to write.
-Dick
|
| 90.7 | | CLOSET::UTT | | Wed Mar 01 1989 08:45 | 9 |
| Interesting footnote to 90.6: The example that Dick cites:
<headn>(Long Title of Section\
symbol-name)
also breaks online processing. (The symptoms are hideous and
inscrutable TeX error messages.)
Mary
|
| 90.8 | someone should fix it | TLE::THORSTENSEN | | Wed Mar 01 1989 09:34 | 12 |
| We encountered problems with arguments that span more than one
line before and had the BLISS program looked at by the BLISS
development team. They commented that fixing the program would be
a rather tedious chore and they didn't have the time - understandably.
On the other hand, there are a large number of manuals that need
to use the cleanup program. If we added up the number of dollars spent
in wasted writer time, we would probably find that it would make
good economic sense to have someone spend the additional time to
make this convenience tool more reliable.
|
| 90.9 | My time is just as valuable... | CASEY::BURACK | This train is bound for glory | Wed Mar 01 1989 16:21 | 12 |
|
Hi,
I will copy the new file and try running it again.
It seems to me that if you provide a tool - then maybe it should
work! Is that asking too much? I have spent a lot of time trying
to make this work for myself and for other writers who have many books.
I should have just continued entering symbolic names by hand in
the 12 or so books I have.
Ruth-Ellen
|
| 90.10 | More about online-cleanup limitations | STAR::RUSSO | | Wed Mar 01 1989 17:23 | 5 |
| Online-cleanup cannot handle files that are included using logical
names and it cannot handle head tags that include the <reference>
tag in the head tag argument, just as it cannot handle head tags
that break across lines. We agree that the tool should be redesigned
to handle these situations but we do not have resources for doing it.
|
| 90.11 | Maybe a tool implies that online cleanup is more difficult than it is | STAR::BUTTLAR | Dick Buttlar | Thu Mar 02 1989 09:50 | 50 |
| Ruth-Ellen and Sue --
These are strictly my opinions.
Online cleanup is not inherently difficult, but seems so. (Several of us are
working on a semi-official book that in part will demystify online cleanup.)
Most of it can be handled quite neatly if a writer reads the CUP Engineering
manual "Coding Documentation for the DECwindows Bookreader" and approaches
a book's source files with some traditional tools: the DCL SEARCH command,
TPU learn sequences, global replace, etc. A book should draft successfully
without a lot of effort if the guidelines in the abovementioned manual are
followed. As always, though, you should expect the usual trial-and-error.
Tailoring a book so that it views properly with the Bookreader is inevitably
time-consuming. Books that look fine in hardcopy may not look fine online.
There are a few things to check, but having a printout of a source file handy
while viewing should show where bad coding habits are interfering with
a viewing (such as the misplaced <p> tag). No tool (even ONLINE_CLEANUP)
can correct these problems -- you still have to hand-edit viewability
problems. Still, we're still learning about viewability. One thing is
certain -- the more time you spend viewing a book, the better you can
make it appear. In a pinch, one can check only to make sure that all
information is accessible; later checks, time permitting, can check its
appearance.
CUP Engineering decided to support online cleanup with "Coding Documentation
for the DECwindows Bookreader." Throughout the engineering effort
there has been but a single person assigned to the tag translator
and another assigned to the device converter. If there were another
person available to work on online software, I would wager that that
person would have been assigned to help one of these two people.
A VMS writer wrote ONLINE_CLEANUP.COM on his own time, testing it on VMS
documentation source files. VMS documentation source files all follow similar
coding conventions, so tailoring a tool to these source files regrettably
ignores other -- legitimate and illegitimate coding practices. Also,
writers are not always sophisticated programmers.
When it was clear that ONLINE_CLEANUP.COM took more time to run than "The Last
Emperor," a VMS engineer volunteered to translate it to BLISS in his spare
time. The people involved were told -- rightly, I think -- not to commit any
more time to the tool. I think there are various disclaimers on Topic 16 to
that effect.
The lack of a completely correct tool is not to say that there are not a lot of
people willing to help with the cleanup process. But, if you use
ONLINE_CLEANUP.COM -- or any tool -- please stash the original copies
of your files elsewhere.
-Dick
|
| 90.12 | The unexpected gotcha... | CASEY::BURACK | This train is bound for glory | Fri Mar 03 1989 16:41 | 14 |
| HI,
I appreciate the time and effort taken to help me. I finally got
it to work on my book. I had checked for the things mentioned in
the "Coding Documentation Source Files for the DECwindows Bookreader".
The thing that made my files bomb was an <INCLUDE> tag in a <comment>
section in the frontmatter of my book. It bombed because it couldn't
find a file to include! I have a better idea of what to look for
now and hopefully will have more succes.
Thanks again....
Ruth-Ellen
|
| 90.13 | Undocumented Restriction | AITG::ISEN | Joyce, 291-8230 DLB5-2 | Thu Jun 01 1989 12:01 | 17 |
| It sure would have saved me time if the restrictions on using
ONLINE_CLEANUP.COM listed in Note 90.10 had also appeared in the
booklet "Guide to Producing Online Documentation." Page 2-3 lists six
restrictions, but the problem with processing files that are included
using logical names was not listed there.
A profile that looks anything like the following one will cause
ONLINE_CLEANUP to die:
<PROFILE>
<ELEMENT>(S1163FM.SDML)
<INCLUDES_FILE>(vms_booknames\S1163VMS_BOOKNAMES.SDML)
<ELEMENT>(S1163C1.SDML)
.
.
.
|
| 90.14 | No elements processed: good or bad? | CASEE::THOMSON | Richard Thomson | Mon Oct 02 1989 07:47 | 38 |
| Hi,
I just ran ONLINE_CLEANUP on one of my files. No complaints from the
procedure, and a couple of minutes later it returned my command line
prompt. I looked at the file, and nothing seems to have changed. I
looked at the documentation Guide to Producing Online Documentation,
page 2-3, and I don't see any restrictions which would cause my file to
slip through the cracks. So I looked at the log file:
%CLEANUP-W-NOELEM, No elements processed
Cleanup completed ...
Hmmm. What does this mean? The first line looks like an error, but the
second seems to indicate a successful run. So where am I going wrong?
Regards
Richard
ps I agree with all the remarks in this topic about the restrictions
being irritating, and the lack of documentation being worse. Whilst I
have every sympathy for the poor overworked people who created the tool
(and in their own time by the sound of it), the arguments laid out here
suggest either a) it is an unsupported hack, and by rights should not
be documented at all, or b) it is a useful tool that would vastly repay
the efforts put in to make it flexible/reliable, and should therefore
be properly supported.
Sometimes out here in the "field" it is *very* hard for us to
understand why CUP and its sister organizations don't (appear to) try
to equate the lost "hidden time" of all its struggling users (just look
how busy all the tools notes conferences are), with the cost of an
extra body or two on tool development/maintenance.
Just another 2� worth...
rt
|
| 90.15 | maybe it's OK? | AITG::WARNER | It's only work if they make you do it | Mon Oct 02 1989 10:19 | 9 |
| Maybe there's nothing wrong with your file -- the .COM procedure doesn't thnk
it needs any help?
Try deleting one of the symbol names (from a head, for instance) and see what
happens.
(Just the most obvious approach.)
Ross
|
| 90.16 | you have to run it on a profile! | REORG::ROGOFF | Zen Software Documentation | Mon Oct 02 1989 18:46 | 0 |
| 90.17 | Yet another ONLINE_CLEANUP gotcha? | CASEE::THOMSON | Richard Thomson | Tue Oct 03 1989 07:45 | 15 |
| Note 90.16 by REORG::ROGOFF
-< you have to run it on a profile! >-
You do? Well, it wasn't, so that would explain the problem if it's
true. I didn't catch that "gotcha" in the documentation.
In reply to Ross (.15), I knew the <HEAD*> tags had no symbols, I hoped
I was running the cleanup on it to save me adding them by hand!
Meanwhile, I have cleaned up by hand, but I will try this fix next time
around. Thanks, as always, for the prompt support.
Regards
Richard
|
| 90.18 | access vio error, any hints? | MUNLEG::SCHERTENLEIB | | Wed Feb 28 1990 06:42 | 83 |
| Hello there,
since nobody so far seems to have found my reply to note 16, I am
trying again here to find someone able to help me: I tried running
ONLINE_CLEANUP on a profile, but it would always stop somewhere in the
middle and give me an access violation error, even on a node with lots
of memory space available.
I am appending the LOG file. Has anybody encountered similar problems?
Thanks in advance and regards
from stormy Munich
Ingrid
-------------------------------------------------------------
Processing file "PLATFORM:<USER.REF>4728FRONT.SDML;NE.ONL_CLEANUP.TEST]4728PRO.SDML; "
Inappropriate tag follows <P> tag on line number 191
Completed processing element file "PLATFORM:<USER.REF>4728FRONT.SDML;NE.ONL_CLEANUP.TEST]4728PRO.SDML; "
Processing file "PLATFORM:<USER.REF>4728INTRO.SDML;NE.ONL_CLEANUP.TEST]4728PRO.SDML; "
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 38
Inappropriate tag follows <P> tag on line number 109
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 116
Completed processing element file "PLATFORM:<USER.REF>4728INTRO.SDML;NE.ONL_CLEANUP.TEST]4728PRO.SDML; "
Processing file "PLATFORM:<USER.REF>4728OPENING.SDML;.ONL_CLEANUP.TEST]4728PRO.SDML; "
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 51
Inappropriate tag follows <P> tag on line number 203
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 210
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 350
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 373
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 411
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 428
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 470
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 494
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 577
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 605
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 667
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 726
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 763
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 798
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 896
Completed processing element file "PLATFORM:<USER.REF>4728OPENING.SDML;.ONL_CLEANUP.TEST]4728PRO.SDML; "
Processing file "PLATFORM:<USER.REF>4728TECHNIQUES.SDML;L_CLEANUP.TEST]4728PRO.SDML; "
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 135
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 169
Inappropriate tag follows <P> tag on line number 223
Inappropriate tag follows <P> tag on line number 238
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 245
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 341
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 411
Inappropriate tag follows <P> tag on line number 500
PARAGRAPH tag follows <P> tag on line number 502
Inappropriate tag follows <P> tag on line number 535
PARAGRAPH tag follows <P> tag on line number 537
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 629
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 671
Inappropriate tag follows <P> tag on line number 702
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 708
Inappropriate tag follows <P> tag on line number 717
Inappropriate tag follows <P> tag on line number 730
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 751
Completed processing element file "PLATFORM:<USER.REF>4728TECHNIQUES.SDML;L_CLEANUP.TEST]4728PRO.SDML; "
Processing file "PLATFORM:<USER.REF>4728FILES_DIR.SDML;;L_CLEANUP.TEST]4728PRO.SDML; "
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 79
Completed processing element file "PLATFORM:<USER.REF>4728FILES_DIR.SDML;;L_CLEANUP.TEST]4728PRO.SDML; "
Processing file "PLATFORM:<USER.REF>4728FILEVIEW.SDML;;;L_CLEANUP.TEST]4728PRO.SDML; "
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 48
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 296
Inappropriate tag follows <P> tag on line number 377
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 383
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 407
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 512
Inappropriate tag follows <P> tag on line number 541
Inappropriate tag follows <P> tag on line number 564
FIGURE_SPACE tag does not follow <FIGURE> tag on line number 571
%SYSTEM-F-ACCVIO, access violation, reason mask=01, virtual address=0001C600, PC=00002540, PSL=03C00008
%TRACE-F-TRACEBACK, symbolic stack dump follows
module name routine name line rel PC abs PC
ONLINE_CLEANUP PROCESS_RECORD 598 0000016A 00002540
ONLINE_CLEANUP PROFILE_FILE_READ 495 0000039F 000023B8
ONLINE_CLEANUP MAIN 245 00000007 00002007
|
| 90.19 | not optimistic... | RAGMOP::UTT | | Wed Feb 28 1990 06:55 | 12 |
| Ingrid,
I have forwarded your note to a couple of people here in Spit Brook
who I know have had something to do with ONLINE_CLEANUP in the past.
The person who wrote it originally has left the group. As noted
elsewhere in this notes conference, ONLINE_CLEANUP is an unsupported
midnight hack, so I'm not too optimistic about whether help will be
forthcoming, but we'll see...
Regards,
Mary
|
| 90.20 | I tried it and gave up... | ZORBA::BURACK | Not Fade Away | Thu Mar 22 1990 12:28 | 22 |
| Hi,
I had a lot of trouble using ONLINE_CLEANUP. I wouldn't recommend using
it.
I seem to recall getting some sort of fatal error ... not sure
if it was an access violation... but the problem was that in my
frontmatter section I had some <include> tags that were in a section of
text that was commented out. ONLINE_CLEANUP reads every line of your
file - it doesn't ignore comments. The error was caused because of
the <include> tags.
I also tried ONLINE_CLEANUP on a book that used <reference> tags
in <head> tags. When it tried to create the symbolic name on a head
like that, it created quite a mess.
If you want it to add the <figure_file>(bookreader...) tag for you,
it won't if the <figure_file>(postscript..) tag already exists.
Just a few of the bugs I ran into.
Ruth-Ellen
|