Better way to edit PDF forms with any desired font using freeware?

On Sun, 5 May 2019 02:48:20 +0200, Carlos E. R. wrote:

The fact remains, Adobe Reader is the canonical tool to open a PDF. Any
other tool may or may not cope, but if it doesn't another tool might -
thus better save time and try directly with adobe Reader.

Hi Carlos,

As an adult, I always agree with reasonable logic, where I agree with you
that the Adobe Acrobat Reader DC suggestion by Keith Nuttle worked for the
stated purpose, as long as we know to use the "comments" toolbar and as
long as we know to preserve the comments (if desired) after they are
inserted.

Even though I've since deleted the product from my system, as a general
purpose solution for annotating "flat" PDF forms, it works as well as Keith
said it would, even down to the choice of almost any font & emphasis
imaginable.

Where the Adobe Acrobat Reader will fail is in edits to the existing text,
for which I have the Adobe Acrobat program (often termed the "writer"),
albeit an older version.

Another place the Adobe Acrobat Reader will fail will be in removing the
encryption or copy/print protections, which, again, requires a non Adobe
product to perform.

I admit it has been a long time since I've bothered to remove encryption
where this old tutorial popped up in a quick search of comp.text.pdf:
o DIY for freeware to remove the 128-bit encrypted user password in a PDF file
https://groups.google.com/d/msg/comp.text.pdf/6RgLQwgKP7o/mQ2nMKpsdv4J

I'd have to check if the Adobe Acrobat Reader DC does these things, where
this is an _old_ checklist that I found in my software archives of the
dozen main things I have needed to do with a PDF file over time, where I've
slightly modified the checklist, but I didn't check it for total accuracy.

Tasks commonly performed on PDFs:
[x]Fast PDF reader: (Sumatra PDF)
[x]Archive entire web sites (wkhtmltopdf, Acrobat Writer)
[x]Add pages (pdftk)
[x]Remove pages (pdfsam, pdftk)
[x]Rotate pages (Acrobat Reader)
[x]Renumber pages (Acrobat Reader)
[x]Remove restrictions (Ghostscript/Ghostview)
[x]Merge PDFs (pdfsam, pdftk)
[x]Extract images (PDF Exchange Viewer)
[x]Edit PDF existing text (Acrobat Writer)
[x]Fill out flat forms (Acrobat Reader)
[x]Save PDF sans username in the properties (Libre Office Writer)
[_]Print book format PDF (FinePrint payware)
[x]Tile PDFs of larger size than your printer can print (Posterazor)

What other tasks do people commonly perform when editing PDFs?

Dan, agree with you. I'll occasionally unplonk just to see if anything
changes. You just reah a point sometimes.

On Sun, 5 May 2019 02:50:17 +0200, Carlos E. R. wrote:

Arguably, a PDF is impossible to edit. It is "decompiled", then edited,
then another PDF is created.

Hi Carlos,

Point to a PDF on the net that you think I can't edit.
o Name just one

nospam wrote:
In article , Arlen G. Holder
wrote:

I don't remove the password for fun, so I haven't done it in a long time,
but _every_ password-protected PDF I ever ran into, in the past, was
trivial to remove the password.

then whatever you supposedly ran into was not properly encrypted.

The crypto options have changed over the years.

The initial versions of PDF spec were pretty weak,
and easily cracked.

* 40-bit encryption (method 40bit) in Acrobat 3 (PDF 1.1) and above
* 128-bit encryption (method 128bit) in Acrobat 5 (PDF 1.4) and above
* 128-bit AES encryption (method AES) in Acrobat 7 (PDF 1.6) and above
* 256-bit AES encryption (method AES256) in Acrobat 9 (PDF 1.7) –
this is deprecated – do not use for new documents (mistakes in key handling)
* 256-bit AES encryption (method AES256ISO) in PDF 2.0 - "key stretching"
" In the latest PDF versions, the encryption key is generated by a
single iteration of SHA-256 followed by a variable set of
key transformations using the algorithms SHA-256, SHA-384 and SHA-512."

The last one there would be a challenge. In the graph here,
brute-forcing RC4 40-bit can occur at 2 million attempts per
minute. While the AES256ISO can be attempted at 5000 attempts
per minute. (CPU = Intel T4200 2GHz, likely a laptop)

https://www.codeproject.com/KB/cs/10..._speedtest.png

( https://www.codeproject.com/Articles...-the-brute-and )

AES256ISO 5,000
AES256 2,500,000 "this is deprecated"
AES128 100,000
RC4 128 100,000
RC4 40 2,000,000

So part of the prevention of cracking the password,
is in slowing down the key processing. Even without discussing
how "crack-worthy" each scheme is, a couple of them allow
a lot of attempts to be made per minute.

Paul

Carlos E. R. wrote:
On 05/05/2019 01.49, nospam wrote:
In article , Arlen G. Holder
wrote:

I most often convert PDFs to MS Office documents, where even MS Word can
"edit" a PDF
converting a pdf to a different format is *not* 'editing a pdf'.


Agree.

Arguably, a PDF is impossible to edit. It is "decompiled", then edited,
then another PDF is created.


A PDF can be edited. (.pdf in, .pdf out).

There just aren't a lot of tools that would do
a good job.

I had a Postscript editor, on the Mac. Cost around
$200, only one version was ever made (company went out
of business). It's main claim to fame, was you could
"group" objects together, in a sense making layers,
and change their Z-axis priority to make them visible
or hide them behind other groups. The ability to group,
was essential for larger projects. You could drop in new
primitives if you wanted. But it lacked the full feature
set of Adobe Illustrator. There wasn't "spline curve text strings".

Illustrator is probably closest to high technical finesse,
but then there's the question of what the input and output
types are. I don't know if Illustrator has been consistent
over the years or not.

http://www.thegraphicmac.com/illustr...ats-explained/

Save as .ai, .eps, .pdf, .svg

.ai = "The data contained in the file is based on PDF,
but it isn’t a format that Acrobat can read correctly."

A list with fewer details is here.

https://helpx.adobe.com/ca/illustrat...lustrator.html

An editor is not "a tool that changes a few letters in an
existing sentence". That's an insult. You can do just about
anything with Illustrator.

There's CorelDraw, but after buying a copy and having it
crash in the first ten minutes, I would never trust them
again with one thin dime of mine.

http://apps.corel.com/partners_devel...ileformats.htm

Whatever LibreOffice does to PDFs, is not... anything.
Defies description in polite company. It makes CorelDraw
look positively heroic.

Portable Document Format and PostScript are computer
languages. You can write programs in them. An editor then,
is an "impossibility". If the editor recognizes a pattern,
it may treat a chunk of code as an "object". It's my guess,
if you put your mind to it, you could leave the editor
quite confused, or you could create objects for which no
simple editor operations are practical. (Like, say you
created a document consisting of "a million dots" sitting
in space, and the boss asked you to "change those two lines
of text there". Well, you couldn't, and keep within the
confines of the original concept of the document. The text
added that way, simply wouldn't look the same as the rest
of the document, so the edit would be a failure.)

An editor in such a situation, is only practical if the
original object follows some conventions.

For example, "if you think you're an editor, edit this".
Illustrator would likely work. The first file, was written
as if you were writing a computer program (you can open the
first file in a text editor, and learn how to write in
PostScript). I don't know if the second document is merely
a distilled version, or the author sat down and wrote the
same code using PDF constructs. But there are sufficient
curves in here, to keep "bozo" editors out of the running.

http://ecee.colorado.edu/~kuester/smith/smith.ps

http://ecee.colorado.edu/~kuester/smith/smith.pdf

I'd open the second document in LibreOffice, but I've
got better things to do.

Paul

On Sun, 05 May 2019 04:10:59 -0400, Paul wrote:

But there are sufficient
curves in here, to keep "bozo" editors out of the running.
http://ecee.colorado.edu/~kuester/smith/smith.ps
http://ecee.colorado.edu/~kuester/smith/smith.pdf

Hi Paul,
You're not kidding that a Smith Chart PDF document had "curves".

I read in that PDF, sheared it to 15 degrees in both the X & Y direction,
and wrote out the resultant PDF file and uploaded it here for review:
http://s000.tinyupload.com/?file_id=19854826114495782092

Here's a screenshot of the 15 degree skewed results for your edification.
https://i.postimg.cc/XY3mdZS4/pdfedit10.jpg

I do agree with you that both PS and PDF are just "programs" though.

Arlen G. Holder wrote:
On Sun, 05 May 2019 04:10:59 -0400, Paul wrote:

But there are sufficient
curves in here, to keep "bozo" editors out of the running.
http://ecee.colorado.edu/~kuester/smith/smith.ps
http://ecee.colorado.edu/~kuester/smith/smith.pdf

Hi Paul,
You're not kidding that a Smith Chart PDF document had "curves".

I read in that PDF, sheared it to 15 degrees in both the X & Y direction,
and wrote out the resultant PDF file and uploaded it here for review:
http://s000.tinyupload.com/?file_id=19854826114495782092

Here's a screenshot of the 15 degree skewed results for your edification.
https://i.postimg.cc/XY3mdZS4/pdfedit10.jpg

I do agree with you that both PS and PDF are just "programs" though.

The original smith.pdf is 93KB, and if you zoom into it,
the curves are smooth all the way to 6400% magnification.

Your file is a bitmap, because I can see jaggies when I zoom in.

In principle, you can do certain types of matrix operations
that would preserve the "SVG" nature of the original document.
You can translate, rotate, scale with a matrix. Shear, I
don't know right off hand how I'd do that, and preserve
the vector nature of the source. But at least, PostScript
and PDF should have some matrix operations you could
experiment with.

This isn't exactly what you've done. Yours is better
looking than this one.

https://en.wikipedia.org/wiki/Shear_mapping

Paul

On 05/05/2019 03.27, Arlen G. Holder wrote:
On Sun, 5 May 2019 02:48:20 +0200, Carlos E. R. wrote:

The fact remains, Adobe Reader is the canonical tool to open a PDF. Any
other tool may or may not cope, but if it doesn't another tool might -
thus better save time and try directly with adobe Reader.

Hi Carlos,

As an adult, I always agree with reasonable logic, where I agree with you
that the Adobe Acrobat Reader DC suggestion by Keith Nuttle worked for the
stated purpose, as long as we know to use the "comments" toolbar and as
long as we know to preserve the comments (if desired) after they are
inserted.

Even though I've since deleted the product from my system, as a general
purpose solution for annotating "flat" PDF forms, it works as well as Keith
said it would, even down to the choice of almost any font & emphasis
imaginable.

Where the Adobe Acrobat Reader will fail is in edits to the existing text,
for which I have the Adobe Acrobat program (often termed the "writer"),
albeit an older version.

This is not a fail: it is intentional. If you want to edit the text with
adobe products, you have to pay.


Another place the Adobe Acrobat Reader will fail will be in removing the
encryption or copy/print protections, which, again, requires a non Adobe
product to perform.

Again, this is not a failure, it is intentional.



I admit it has been a long time since I've bothered to remove encryption
where this old tutorial popped up in a quick search of comp.text.pdf:
o DIY for freeware to remove the 128-bit encrypted user password in a PDF file
https://groups.google.com/d/msg/comp.text.pdf/6RgLQwgKP7o/mQ2nMKpsdv4J

128 bits is old.


I'd have to check if the Adobe Acrobat Reader DC does these things, where
this is an _old_ checklist that I found in my software archives of the
dozen main things I have needed to do with a PDF file over time, where I've
slightly modified the checklist, but I didn't check it for total accuracy.

Tasks commonly performed on PDFs:
[x]Fast PDF reader: (Sumatra PDF)
[x]Archive entire web sites (wkhtmltopdf, Acrobat Writer)
[x]Add pages (pdftk)
[x]Remove pages (pdfsam, pdftk)
[x]Rotate pages (Acrobat Reader)
[x]Renumber pages (Acrobat Reader)
[x]Remove restrictions (Ghostscript/Ghostview)
[x]Merge PDFs (pdfsam, pdftk)
[x]Extract images (PDF Exchange Viewer)
[x]Edit PDF existing text (Acrobat Writer)
[x]Fill out flat forms (Acrobat Reader)
[x]Save PDF sans username in the properties (Libre Office Writer)
[_]Print book format PDF (FinePrint payware)
[x]Tile PDFs of larger size than your printer can print (Posterazor)

What other tasks do people commonly perform when editing PDFs?

Editing PDFs is not something you are supposed to do. And don't get
started that you do it all the time. Of course we do. That's not the point.

If you have to edit a PDF, there is a failure somewhere.

--
Cheers,
Carlos E.R.

On 05/05/2019 05.08, Arlen G. Holder wrote:
On Sun, 5 May 2019 02:50:17 +0200, Carlos E. R. wrote:

Arguably, a PDF is impossible to edit. It is "decompiled", then edited,
then another PDF is created.

Hi Carlos,

Point to a PDF on the net that you think I can't edit.
o Name just one

Not the point. Not what I said. The program "emulates" edition so that
you /think/ you are editing it. It is not an editor in the sense MS Word
is.

Similarly, you can not edit a jpg. Now think it over and try to find out
why I say that, when every body knows they can edit photos and do it
every day. It is not obvious, but you are clever enough to find out. ;-)


But yes, there are PDFs that /you/ can not edit. You'd see the text
"word". But the 'w' is put there separate from the 'o' and 'r' and the
'd' so that the "word" can not be even selected. The file is
intentionally obfuscated. I don't have a sample handy.


--
Cheers,
Carlos E.R.

On 05/05/2019 10.10, Paul wrote:
Carlos E. R. wrote:
On 05/05/2019 01.49, nospam wrote:
In article , Arlen G. Holder
wrote:

I most often convert PDFs to MS Office documents, where even MS Word
can
"edit" a PDF
converting a pdf to a different format is *not* 'editing a pdf'.


Agree.

Arguably, a PDF is impossible to edit. It is "decompiled", then edited,
then another PDF is created.


A PDF can be edited. (.pdf in, .pdf out).

sort of :-)


There just aren't a lot of tools that would do
a good job.

I had a Postscript editor, on the Mac. Cost around
$200, only one version was ever made (company went out
of business). It's main claim to fame, was you could
"group" objects together, in a sense making layers,
and change their Z-axis priority to make them visible
or hide them behind other groups. The ability to group,
was essential for larger projects. You could drop in new
primitives if you wanted. But it lacked the full feature
set of Adobe Illustrator. There wasn't "spline curve text strings".

Illustrator is probably closest to high technical finesse,
but then there's the question of what the input and output
types are. I don't know if Illustrator has been consistent
over the years or not.

http://www.thegraphicmac.com/illustr...ats-explained/

Â*Â* Save as .ai, .eps, .pdf, .svg

Â*Â* .ai = "The data contained in the file is based on PDF,
Â*Â*Â*Â*Â*Â*Â*Â*Â* but it isn’t a format that Acrobat can read correctly."

A list with fewer details is here.

https://helpx.adobe.com/ca/illustrat...lustrator.html


An editor is not "a tool that changes a few letters in an
existing sentence". That's an insult. You can do just about
anything with Illustrator.

There's CorelDraw, but after buying a copy and having it
crash in the first ten minutes, I would never trust them
again with one thin dime of mine.

http://apps.corel.com/partners_devel...ileformats.htm


Whatever LibreOffice does to PDFs, is not... anything.
Defies description in polite company. It makes CorelDraw
look positively heroic.

Quite.


Portable Document Format and PostScript are computer
languages. You can write programs in them. An editor then,
is an "impossibility". If the editor recognizes a pattern,
it may treat a chunk of code as an "object". It's my guess,
if you put your mind to it, you could leave the editor
quite confused, or you could create objects for which no
simple editor operations are practical. (Like, say you
created a document consisting of "a million dots" sitting
in space, and the boss asked you to "change those two lines
of text there". Well, you couldn't, and keep within the
confines of the original concept of the document. The text
added that way, simply wouldn't look the same as the rest
of the document, so the edit would be a failure.)

An editor in such a situation, is only practical if the
original object follows some conventions.

Yes.


For example, "if you think you're an editor, edit this".
Illustrator would likely work. The first file, was written
as if you were writing a computer program (you can open the
first file in a text editor, and learn how to write in
PostScript). I don't know if the second document is merely
a distilled version, or the author sat down and wrote the
same code using PDF constructs. But there are sufficient
curves in here, to keep "bozo" editors out of the running.

http://ecee.colorado.edu/~kuester/smith/smith.ps

http://ecee.colorado.edu/~kuester/smith/smith.pdf

Wow! I wanted that thing when I was in college. The copies we got were ugly.

I'd open the second document in LibreOffice, but I've
got better things to do.

:-D


--
Cheers,
Carlos E.R.

On 05/05/2019 14.15, Paul wrote:
Arlen G. Holder wrote:
On Sun, 05 May 2019 04:10:59 -0400, Paul wrote:

But there are sufficient
curves in here, to keep "bozo" editors out of the running.
http://ecee.colorado.edu/~kuester/smith/smith.ps
http://ecee.colorado.edu/~kuester/smith/smith.pdf

Hi Paul,
You're not kidding that a Smith Chart PDF document had "curves".

I read in that PDF, sheared it to 15 degrees in both the X & Y direction,
and wrote out the resultant PDF file and uploaded it here for review:
http://s000.tinyupload.com/?file_id=19854826114495782092

Here's a screenshot of the 15 degree skewed results for your edification.
https://i.postimg.cc/XY3mdZS4/pdfedit10.jpg

I do agree with you that both PS and PDF are just "programs" though.

The original smith.pdf is 93KB, and if you zoom into it,
the curves are smooth all the way to 6400% magnification.

Your file is a bitmap, because I can see jaggies when I zoom in.

Indeed! So, not strictly an edit.

Now, with an svg graphic file that manipulation could be done "easily".

--
Cheers,
Carlos E.R.

In article , Paul
wrote:

I had a Postscript editor, on the Mac. Cost around
$200, only one version was ever made (company went out
of business).

postscript is text. any text editor will work. even free ones.

"Carlos E. R." wrote:
Editing PDFs is not something you are supposed to do. And don't get
started that you do it all the time. Of course we do. That's not the point.

If you have to edit a PDF, there is a failure somewhere.

Yes, like those manuals for consumer electronics etc. which are
supplied as a single multi-megabyte PDF in about 20 different
languages where every section contains paragraphs in each language.
An absolute pain, since you are continually scrolling around the
document. Being able to extract one language version is essential for
easy reading.

On Sun, 05 May 2019 08:15:21 -0400, Paul wrote:

In principle, you can do certain types of matrix operations
that would preserve the "SVG" nature of the original document.
You can translate, rotate, scale with a matrix.

Hi Paul,

Thanks for looking at the results & for analyzing them.

I understand what you're saying, where the bitmap conversion may have been
a result of how I read the PDF in, or, more likely, how I saved it.

I didn't write down the steps since it was only a quick proof of concept
attempt where there weren't any rules, so it was a freeforall. Normally I
write down all the steps of a process, in a scientific manner, but this was
just proof of concept since I knew I could "change" the PDF at will.

As I recall, I first opened the smith.pdf in Microsft Word 2007, but not
everything was there, where I didn't bother to debug. I also tried the
Adobe Acrobat Reader, but it wouldn't change the text (saying it didn't
have the font), which was the change that I had originally wanted to do (I
was gonna change the title to "Paul Chart" or something like that).

In the Adobe "writer", version 6, I could select the text in the word
"Smith Chart", which were recognized as Palatino-Bold text, but the error I
got when I tried it just now again to reproduce for you was:
"All or part of the selection has no available system font. You cannot
add or delete text using the currently selected font"
https://i.postimg.cc/QCvPjX78/pdfedit11.jpg

Bearing in mind I mostly change text in PDFs, I don't know this for a fact,
but, I suspect if I either edited the PDF with a text editor, or if I added
the required font, that I "might" have been able to change the "Smith" to
"Paul" directly in the PDF.

Since it was just a proof of concept, I didn't bother to try to work around
that problem, and simply decided to use an image editor, where I right
clicked on the 94KB PDF to arbitrarily "open in Krita" freeeware, where the
Krita "Import" dialog asked how many pages (where I said "all pages"). The
resolution defaults to 100dpi

Then I simply hit Krita:Image Shear Image" which asked for the
"Shear angle" for X & Y, each of which I arbitrarily set to 15 degrees.

Since that left white space at the edges, I filled it in and also increased
the canvas so that it was even at the top and bottom and then cropped.

The problem was that I couldn't _save_ as a PDF. Bummer. As I recall, I
saved ot Photoshop Image (smith.psd), and then read it into PhotoShop and
then simply saved it to "Photoshop PDF" without changing any of the default
options.

Since PS is simply a "language", I realize we could use a text editor on
the PostScript, but we'd have to know how to interpret the language, which
is best left to software, IMHO. I did look at the original smith.pdf in VIM
just now, where it's a combination of text & binary, apparently. Obviously
a hex editor would do the trick, but I think that is overstepping the
bounds of the quick test.

Vim doesn't show us the original text (which was what I was hoping to see),
but it does seem that the original PDF was created by Aladdin Ghostscript
6.01 based on what it says at line 550 almost at the end of the original
94KB smith.pdf file.

At line 410 of the original PDF is a definition of a font:
"URW Palladio L Roman"

At line 280 & 532 of the original PDF, are font descriptions of the sort:
/Type/FontDescriptor/FontName/Symbol
/Subtype/Type1/BaseFont/EKHBOO+Palatino-Bold/Type/Font/Name/R14/FontDescriptor 13 0 R/FirstChar 32/LastChar 251/
Where I "presume" that, if I knew the syntax, I could change that to a
system font that I have on my system.

At line 391 & 508 is another apparent font "Palatino-Roman", where at lines
293 & 407 there are copyright notices, apparently related to the use of
each of those fonts.

The rest is scrambled like the Franciscan sediments, where obviously things
would have been more descriptive had I looked at the postscript file in the
vim text editor instead of the PDF file.

I'm not going to spend the time to load or change the fonts, but I suspect
that there's a chance that a focused font-based approach perhaps would have
worked.

In short, you can edit (as in "change") any PDF, with varying degrees of
success, depending on a plethora of factors, not the least of which is what
you want to change.

On Sun, 5 May 2019 15:15:56 +0200, Carlos E. R. wrote:

Wow! I wanted that thing when I was in college. The copies we got were ugly.

We barely scraped the microwave realm in those days, many moons ago.

Just curious what color your Smith Charts were in college?

Mine were all in a specific "orange" color, for some reason unknown to me,
where, today, they seem to be in green.