HTML2PDF for VMS?

[Clark Hayden]

Hi.

We produce HTML files which contain some graphics and the business would now like to convert these to PDF which will then be e-mailed to the clients as attachments.

Is there a way to convert the HTML files to PDF within VMS that is freeware? I have managed to compile a C module to convert text to pdf bu cannot find anything on converting HTML to PDF for the VMS platform.

Thanks.
Hayden.

[Brian Reiter]

There's HTMLDOC which is on the Freeware 7 set of disks and is also included as part of the T4 installation.

cheers

Brian

[P Muralidhar Kini]

Hi Clark,

You can make use of the HTMLDOC freeware utility.

You can find it in the below link -
http://h71000.www7.hp.com/openvms/freeware/v7_00freeware_abstract.txt
>> HTMLDOC Translate HTML pages into Postscript or PDF

To download HTMLDOC or other utilities from Freeware V7, Refer
http://h71000.www7.hp.com/openvms/freeware/

That should meet your requirement.

Regards,
Murali

[Clark Hayden]

The HTMLDOC runs but I keep getting the follwoing error message:

ERR005: Unable to open font width file HTMLDOC_DATA/fonts/Times-Roman.afm!

Any ideas to remedy the problem? Or is this something I need to take up with our system administrator?

Thanks.

[Volker Halle]

Clark,

you have the sources available, have a look for this error message...

The font files seem to exist on the Freeware V7 CD, may it's just a missing HTMLDOC_DATA logical ?

Volker.

[Bhadresh]

>>> ERR005: Unable to open font width file HTMLDOC_DATA/fonts/Times-Roman.afm!
>>>Any ideas to remedy the problem?

Error 5 indicates the specified file could not be found. HTMLDOC_DATA variable specifies the location of HTMLDOC's data and fonts directories. Have you set HTMLDOC_DATA?

Regards,
Bhadresh

[Clark Hayden]

Yes, HTMLDOC_DATA has been set up to map to the correct directory. The problem seems to be that the app tags on "/fonts/..." to the location.

I've been scouring the web trying to find a solution but so far I have had no luck.

Any assitance will be greatly appreciated.

Thanks.
Hayden.

[Volker Halle]

Hayden,

does $ DIR HTMLDOC_DATA:[FONTS] show the font files ? If not, can you copy them there ?

Volker.

[Andy Bustamante]

Would you post the results of

$ show logical HTMLDOC_DATA

[Clark Hayden]

Ok. I have extracted the files/directories from the Freeware V7 backups and have copied these to $DKE0:[HTMLDOC] on VMS, with $DKE0 being the disk.

I had to copy the HTMLDOC_IA64.EXE executable from another source since I could not find an executable image in the freeware files that would run.

I then set the HTMLDOC_DATA logical up as follows:
"HTMLDOC_DATA" = "$DKE0:[HTMLDOC]" (LNM$PROCESS_TABLE)

But I still get the ERR005 message that it cannot find the files.

Any tips?

PS: Volker, thanks for all your help so far.

Thanks.
Hayden.

[Joseph Huber_1 · Accepted solution]

I almost never used it, but htmldoc.com setup reads:
$! HTMLDOC.COM
$!
$! Customize this procedure as required for your installation.
$!
$! Note: use unix-style directory format for the HTMLDOC_DATA directory!
$!
$ define HTMLDOC_DATA "disk$www/htmldoc/
$!
$ HTMLDOC :== $DISK$WWW:[HTMLDOC]HTMLDOC.EXE
$!
$ exit

i.e defining htmldoc_data in VMS syntax is wrong.
Define it in Unix syntax:
define htmldoc_data "/device/htmldoc/"
to make it working.
A quick test on my system showed it is working.

[Joseph Huber_1]

In my above define command, there may be a variation of the Unix-style path definition,
depending on "device" being a logical name, and the definition of the Unix-compatibility level and the definition of
DECC$DISABLE_TO_VMS_LOGNAME_TRANSLATION

So either "/device/directory/"
or "device/directory".

[Clark Hayden]

Hi Joseph.

I have tried defining the logical as:
"HTMLDOC_DATA" = "/ALEX04$DKE0/HTMLDOC"
"HTMLDOC_DATA" = "/ALEX04$DKE0/HTMLDOC/"
"HTMLDOC_DATA" = "ALEX04$DKE0/HTMLDOC"
"HTMLDOC_DATA" = "ALEX04$DKE0/HTMLDOC/"

None of which has returned anything other than ERR005: Unable to open font width file.........

I have no previous Unix experience so this is all a learning experience for me. :)

Thanks.
Hayden.

[Oswald Knoppers_1]

Maybe the html2pdf tool doesn't like the $ sign in the (ALEX04$DKE0) logical name.

Oswald

[Volker Halle]

Hayden,

and there is a DKE0:[HTMLDOC.FONT] directory with all the required font files ?

Volker.

[Joseph Huber_1]

On my system (VMS 7.3-1 Alpha) it looks like this:

MPIW10_HUB>sh log htmldoc_data
"HTMLDOC_DATA" = "disk$www/htmldoc/" (LNM$PROCESS_TABLE)
MPIW10_HUB>sh log disk$www
"DISK$WWW" = "AVMP01$DKA100:" (LNM$SYSTEM_TABLE)
MPIW10_HUB>sh log DECC$DISABLE_TO_VMS_LOGNAME_TRANSLATION
"DECC$DISABLE_TO_VMS_LOGNAME_TRANSLATION" = "DISABLE" (LNM$PROCESS_TABLE)
MPIW10_HUB>dir disk$www:[htmldoc]

Directory DISK$WWW:[HTMLDOC]

COPYING.TXT;1 DATA.DIR;1 FONTS.DIR;1 HTMLDOC.COM;2
HTMLDOC.EXE;1 HTMLDOC_VMS.TXT;1 README.TXT;1

Total of 7 files.
MPIW10_HUB>htmldoc -f tmp.pdf ALL.HTML
PAGES: 4
BYTES: 19848

[Clark Hayden]

Ok. I figured out this problem. I used the unix style format you described and it worked. All I had to do was track down the junior system administrator who figured he needed some space and so deleted some of the files that I created yesterday. The files were copied back and HTMLDOC runs.

It now produces a PDF but Adobe can't read it. Is there a limit to which version of Adobe can be used to access the file? I have Adobe Reader 9.

Thanks.
Hayden.

[Joseph Huber_1]

Well I did not extensive tests, but my all.html in last reply is a HTML file containing some

tags, and an

list of links in between.
It produces a table of contentents for the

tags, and correctly lists everything.
XPDF on VMS and the Adobe reader on Linux have no problems with the file produced.
So maybe some HTML elements (graphics ?) produce the errors ?
Can You look at the acroread error ? Maybe run Adobe reader on the command-line to see errors ?

[Clark Hayden]

I was about to post a follwo up. I see that the conversion to PDF produced the following:
ERR014: Text on page 7 too large - truncation or overlapping may occur!

Would this have anything to do with the problem being experienced by Adobe Reader?

Thanks.

[Joseph Huber_1]

Hayden,
since You produce the PDF on VMS, but use Adobe reader on another system (Linux, windows ?, acroread standalone, in a web-browser ?),
how do You transfer the VMS file ?
From a VMS web-server ?
Transferred to the other system by which method ?

[Clark Hayden]

As you point out the PDF is created on VMS and is then FTP'ed as an ascii file to Windows where I use Adobe Reader 9 to access the file.

[Joseph Huber_1]

Try to transfer in BINARY mode.
The file produced on VMS is a stream_lf file (DIR/FULL), and transferring in ASCII mode may introduce additional line-feeds or carriage return characters (or omit those).

[Clark Hayden]

It seems we go from one problem to the next. :)

The file transferred as a binary and it then opens in Adobe. But now the file contains blank pages and when I scroll to the nex page then the zoom jumps up to over 4000%.

[Clark Hayden]

I now have the process producing a PDF file which contains HTML. The problem is that it contains the unconverted HTML for the first couple of pages. I attached the output PDF file.

[Joseph Huber_1]

For me it doesn't look like HTML, but more like graphics embedded in PDF.
But anyway, that's beyond my knowledge of PDF.